medications and diabetes risk mechanisms and approach to risk reduction

www.ebook3000.com

http://www.ebook3000.org

Medications and Diabetes Risk: Mechanisms and Approach to Risk Reduction

O A P NOX F O R D A M E R I C A N P O C K E T N O T E S

Jack_Prelims.indd i 11/15/2010 9:20:08 PM

www.ebook3000.com


This material is not intended to be, and should not be considered, a substitute for medical or other professional advice. Treatment for the conditions described in this material is highly dependent on the indi-vidual circumstances. While this material is designed to off er accurate information with respect to the subject mat-ter covered and to be current as of the time it was written, research and knowledge about medical and health issues is constantly evolving, and dose schedules for medications are being revised continually, with new side eff ects recognized and accounted for regularly. Readers must therefore always check the product information and clinical procedures with the most up-to-date published product information and data sheets provided by the manufacturers and the most recent codes of conduct and safety regulation. Oxford University Press and the authors make no representations or warranties to readers, express or implied, as to the accuracy or com-pleteness of this material, including without limitation that they make no representations or warranties as to the accuracy or effi cacy of the drug dosages mentioned in the material. The authors and the publishers do not accept, and expressly disclaim, any responsibility for any liability, loss, or risk that may be claimed or incurred as a consequence of the use and/or application of any of the contents of this material.

The Publisher is responsible for author selection and the Publisher and the Author(s) make all editorial decisions, including decisions regarding content. The Publisher and the Author(s) are not responsible for any product information added to this publication by companies purchasing copies of it for distribution to clinicians.

Jack_Prelims.indd ii 11/15/2010 9:20:08 PM

www.ebook3000.com


Medications and Diabetes Risk: Mechanisms and Approach to Risk Reduction

Samuel Dagogo-Jack, MD, FRCPMullins Professor in Translational ResearchProfessor of Medicine & ChiefDivision of Endocrinology, Diabetes & MetabolismUniversity of Tennessee Health Science CenterMemphis, TN

O A P NOX F O R D A M E R I C A N P O C K E T NO T E S

1

Jack_Prelims.indd iii 11/15/2010 9:20:08 PM

www.ebook3000.com


1Oxford University Press, Inc., publishes works that further Oxford University’s objective of excellencein research, scholarship, and education.

Oxford New YorkAuckland Cape Town Dar es Salaam Hong Kong KarachiKuala Lumpur Madrid Melbourne Mexico City NairobiNew Delhi Shanghai Taipei Toronto

With offi ces inArgentina Austria Brazil Chile Czech Republic France GreeceGuatemala Hungary Italy Japan Poland Portugal SingaporeSouth Korea Switzerland Thailand Turkey Ukraine Vietnam

Copyright © 2011 by Oxford University Press, Inc.

Published by Oxford University Press, Inc.198 Madison Avenue, New York, New York 10016www.oup.com

Oxford is a registered trademark of Oxford University Press

All rights reserved. No part of this publication may be reproduced,stored in a retrieval system, or transmitted, in any form or by any means, electronic, mechanical, photocopying, recording, or otherwise, without the prior permission of Oxford University Press.

ISBN: 978-0-19-973431-3

9 8 7 6 5 4 3 2 1Printed in the Chinaon acid-free paper

Jack_Prelims.indd iv 11/15/2010 9:20:08 PM

www.ebook3000.com

www.oup.com


MEDICATIONS AND

DIABETES RISK

DISCLOSURES

Dr. Dagogo-Jack has served as a consultant for Eli Lilly and GlaxoSmithKline. He has also served on the speaker’s bureaus of Merck and Sanofi -Aventis.

Jack_Prelims.indd v 11/15/2010 9:20:09 PM

www.ebook3000.com


This page intentionally left blank

www.ebook3000.com


MEDICATIONS AND

DIABETES RISK

PREFACE

More than 23 million Americans currently have diabetes, and approximately 54 million have prediabetes. People with diabetes often also require medications for several comorbid conditions (including hypertension, dyslipidemia, depres-sion, heart disease, and pain syndromes). Yet, a vast litera-ture abounds on the potential adverse eff ects of numerous medications on glucose metabolism. There is, thus, genuine clinical concern that certain medications used for treatment of comorbid conditions and other indications (such as hor-mone replacement, contraception, infections) might worsen glycemic control in diabetic patients or trigger diabetes in others. These concerns infl uence therapeutic decisions in a manner that sometimes emphasizes avoidance of possible dysglycemia over eff ective control of the comorbid condi-tions. The same concerns may also weigh against the other-wise appropriate use of necessary medications.

The purpose of this concise book is to provide clinicians with actionable knowledge regarding the eff ects of vari-ous medications on glucose regulation and diabetes risk. Beginning with a brief overview of diabetes pathophysiol-ogy, the diff erent drugs have been organized by class, and the scientifi c evidence for the diabetes risk and possible mechanisms have been presented for each drug. The agents discussed include widely prescribed medication classes: antibiotics, antidepressants, antihypertensives, bronchodi-lators, estrogens and oral contraceptives, glucocorticoids, lipid-lowering agents, nonsteroidal anti-infl ammatory drugs (NSAIDs), and thyroid hormone. Although less widely pre-scribed than the foregoing list, atypical antipsychotics,

Jack_Prelims.indd vii 11/15/2010 9:20:09 PM

www.ebook3000.com


OAPNOAPNOAPN

human immunodefi ciency virus (HIV) anti retrovirals, immunomodulatory agents, and human growth hormone have also been included because of the interest generated by their link to diabetes risk. In addition to medications used in ambulatory practice, this work includes a discussion of total parenteral nutrition (TPN)-induced hyperglycemia, which is associated with increased morbidity and mortality among hospitalized patients. For completeness, an account of the growing link between use of recreational drugs (alco-hol, nicotine, cannabinoids, opioids, cocaine) and glucose abnormalities has been included, because of the possible intersection between these addictive agents and the growing diabetes epidemic.

With some medications, the data presented should help debunk myths, clarify misperceptions, and provide reassur-ance to the practicing clinicians. Wherever the evidence sup-ports increased diabetes risk, clear suggestions are given on how to reduce the risk. There are also diabetes management guidelines for special situations, including new-onset diabetes after transplant (NODAT) and TPN-induced hyperglycemia. The fi nal chapter provides suggestions for a general approach to the prevention or attenuation of diabetes risk, focusing on risk stratifi cation, lifestyle modifi cation, and the emerging concept of diabetes “pharmacoprevention” in high-risk persons receiving treatment with potentially diabetogenic agents.

This book should serve two essential functions—to enable clinicians to confi dently prescribe therapeutic regimens that embody the best risk–benefi t profi le with regard to glyce-mia, and to equip them with the know-how for preventing and managing drug-induced hyperglycemia.

Jack_Prelims.indd viii 11/15/2010 9:20:09 PM

www.ebook3000.com


MEDICATIONS AND

DIABETES RISK

TABLE OF CONTENTS

1 Pathophysiology of Type 1 and Type 2 Diabetes 1

2 General Mechanisms of Medications and Diabetes Risk 8

3 Steroids and Immunomodulatory Agents 14

4 Antihypertensive Agents 26

5 Catecholamines, �-Adrenergic Agonists, and Bronchodilators 33

6 Lipid-lowering Drugs 35

7 Antimicrobial Drugs 40

8 Atypical Antipsychotic and Antidepressant Agents 47

9 Recreational Drugs 56

10 Miscellaneous Agents 59

11 General Approach to Risk Reduction 66

References 70

Jack_Prelims.indd ix 11/15/2010 9:20:09 PM

This page intentionally left blank

1

MEDICATIONS AND

DIABETES RISK

Section 1: Pathophysiology of Type 1 and Type 2 Diabetes

DIABETES MELLITUS

Diabetes mellitus (DM) refers to a group of metabolic dis-orders that result in hyperglycemia. These disorders have diff erent underlying processes but their common mani-festation is hyperglycemia, regardless of underlying pro-cess. More than 20 million Americans have diabetes and approximately 5 million of them are undiagnosed. Diabetes is a major public health problem because of the long-term complications (such as blindness, amputations, kidney fail-ure, heart disease, and stroke) that could occur if the condi-tion is inadequately controlled. Fortunately, as demonstrated in the Diabetes Control and Complications Trial (DCCT) study and other landmark clinical trials, the complications of diabetes can be prevented by maintaining excellent glyce-mic control using diet, exercise, and medications.

Diagnosis of Diabetes The diagnosis of diabetes can be established using any of the following American Diabetes Association criteria:

Plasma glucose of 126 mg/dL or greater after an over- ■

night fast. A repeat test on a diff erent day is required to confi rm the diagnosis of diabetes (American Diabetes Association, 2009).Symptoms of diabetes and a random (nonfasting) plasma ■

glucose of 200 mg/dL or greater.A standard Oral Glucose Tolerance Test showing a ■

plasma glucose level of 200 mg/dL or greater at 2 hours after ingestion of a 75 g glucose load, provided all testing protocols are followed.A hemoglobin A1c level of 6.5% or greater (International ■

Expert Committee, 2009).

Jack_Chapter.indd 1 11/15/2010 10:02:07 PM

OAPN

2

Diagnosis of Prediabetes The term “prediabetes” is used to describe persons with im-paired glucose tolerance (IGT) or impaired fasting glucose (IFG). Impaired glucose tolerance is defi ned by a 2-hour oral glucose tolerance test (OGTT) plasma glucose level greater than 140 mg/dL but less than 200 mg/dL, and IFG is defi ned by a fasting plasma glucose level of 100 mg/dL or greater, but less than 126 mg/dL (ADA, 2009). Approximately 57 million Americans have prediabetes, and studies have shown that people with prediabetes tend to develop type 2 diabetes within 10 years. Lifestyle modifi cations (dietary restriction and exercise) and certain medications can prevent the de-velopment of diabetes in persons with prediabetes (Diabetes Prevention Program Research Group, 2002).

CLASSIFICATION OF DIABETES

Type 1 diabetes accounts for less than 10% of all cases of DM, occurs in younger persons, and is caused by absolute insulin defi ciency resulting from an immune-mediated destruction of the insulin-producing cells of the pan-creas, known as �-cells. Type 2 diabetes mellitus (T2DM) accounts for more than 90% of all cases of DM. Usually a disease of adults, type 2 DM is being diagnosed increasingly in younger age groups. Obesity, insulin resistance, and rela-tive insulin defi ciency are characteristic fi ndings. The risk factors for T2DM are listed in Table 1. Insulin resistance refers to a decreased ability of insulin to drive the uptake of glucose from the blood into the cells and to produce the other appropriate metabolic eff ects of insulin. Persons with insulin resistance alone do not develop diabetes, because their pancreatic �-cells compensate by increasing


3

MEDICATIONS AND

DIABETES RISK

insulin secretion to levels that can maintain normoglyce-mia. Therefore, a second defect—an inability of the � cells to maintain compensatory hyperinsulinemia—is required to precipitate type 2 diabetes.

Other specifi c types of diabetes include those due to surgi-cal resection or diseases of the pancreas (that compromise insulin secretion) and glandular disorders (e.g., Cushing syndrome, acromegaly), and rare syndromes. Gestational diabetes occurs during pregnancy and tends to resolve once the baby and placenta have been delivered (although the mother remains at high risk for future type 2 diabetes).

Table 1 Risk Factors for Type 2 Diabetes

Having a fi rst-degree relative with diabetes mellitus ■

Increasing age ■

Being overweight (body mass index >25) ■

Having dyslipidemia (elevated triglycerides or decreased high-density ■

lipoprotein cholesterol)

Having hypertension ■

Belonging to a high-risk ethnic group (African American, Hispanic, ■

Native American, Asian)

History of gestational diabetes or birth of child weighing ≥9 lbs ■

History of polycystic ovary syndrome ■

Habitual physical inactivity ■

History of vascular disease ■

History of impaired fasting glucose or impaired glucose tolerance ■

From American Diabetes Association. (2009). Standards of Medical Care in Diabetes—2009. Diabetes Care, 32(Suppl 1):S13–S61, with permission of the publisher.


OAPN

4

THE GENETIC BASIS OF TYPE 2 DIABETES

A family history of diabetes in fi rst-degree relatives (parents, siblings, and off spring) is one of the strongest risk factors for the development of diabetes. However, the genetic transmission of diabetes risk from parents to off -spring is rather complex. Among monozygotic (identical) twins, the concordance rate of type 2 diabetes (that is, the chances of diabetes occurring in the second twin if one has diabetes) is approximately 80%, and the lifetime risk of developing type 2 diabetes among off spring and siblings of aff ected patients has been estimated at approximately 40% (Granner & O’Brien, 1992). If both parents are aff ected, the risk approaches 80% in off spring (Martin et al., 1992). Current understanding indicates that multiple genes may be involved in this process, and rarely have single genes been discovered that explained the entire processes underlying the development of diabetes.

PATHOPHYSIOLOGY OF TYPE 2 DIABETES

In genetically susceptible persons, the development of type 2 diabetes is characterized ultimately by three underlying mechanisms: impaired insulin action (also known as insu-lin resistance), which is expressed in skeletal muscle and fat cells; impaired insulin secretion by the pancreatic �-cells; and increased hepatic (liver) glucose production (HGP) (Dagogo-Jack & Santiago, 1997; Moneva & Dagogo-Jack, 2002). The transition from the normal state to type 2 diabe-tes is punctuated by a variable interlude (usually lasting sev-eral years) in the intermediate state of prediabetes (IGT and IFG). There is general agreement that insulin resistance and impaired insulin secretion are present in most individuals


5

MEDICATIONS AND

DIABETES RISK

before the onset of diabetes (Dagogo-Jack & Santiago, 1997; Moneva & Dagogo-Jack, 2002). Longitudinal studies in which initially healthy Pima Indians underwent metabolic assessments repeatedly over several years showed that sub-jects who progressed from the normal state to prediabetes (IGT) had lost approximately 12% of their insulin sensitiv-ity, but 27% of their insulin secretion; the further progres-sion from prediabetes to type 2 diabetes was preceded by a 31% decline in insulin sensitivity and a 78% decline in insu-lin secretion (Weyer et al., 1999).

INSULIN RESISTANCE

The binding of insulin to its receptor triggers a series of phosphorylation reactions in the cytosol. The initial phos-phorylation occurs on tyrosine residues within the cytoplas-mic tail of the insulin receptor, followed by phosphorylation of multiple other intracellular proteins, including insulin receptor substrates (IRS)-1, 2, 3, and 4. Phosphorylation of the IRS proteins activates the enzyme phosphatidylinositol 3-kinase (PI3K), leading to the translocation of an intracel-lular pool of glucose transporter molecules (GLUT4) to the plasma membrane, where they form vesicles that mediate glucose transport into the cell. Thus, insulin lowers blood glucose by stimulating the transport of glucose across cell membranes through a series of complex chemical reac-tions. Failure of this mechanism at any level between the binding of insulin to its cell membrane receptor and the eventual translocation of GLUT4 and internalization of glu-cose results in insulin resistance. Insulin resistance can be inherited or acquired. Obesity, aging, physical inactivity, overeating, and accumulation of free fatty acids are known


OAPN

6

causes of insulin resistance. Phosphorylation of serine or threonine residues (instead of tyrosine) interferes with the insulin signaling, and is a common molecular mechanism that leads to insulin resistance.

INSULIN SECRETION BY THE �-CELLS

Under normal conditions, the pancreatic � cells secrete insulin in response to glucose stimulation through a series of transmembrane electrical reactions. Glucose metabolism in � cells generates bursts of action potentials that ultimately lead to calcium infl ux. The adenosine triphosphate (ATP)-sensitive potassium channel (KATP) normally maintains the �-cell resting membrane potential, thereby preventing cal-cium entry. The KATP channel is closed when the ratio of ATP to adenosine diphosphate rises within the � cells, as occurs during glucose metabolism. The resultant depolarization (change in electrical charge) of the �-cell membrane drives calcium into the cell, which then triggers insulin secretion. Agents that open the KATP channel (e.g., diazoxide) reverse the depolarization and inhibit insulin secretion by the � cells.The �-cell mass is reduced in type 2 diabetic patients as a result of apoptosis induced by islet amyloid deposition, oxida-tive stress, infl ammatory cytokines, and other mechanisms (Haataja et al., 2008).

HEPATIC GLUCOSE PRODUCTION

In the prospective study of Pima Indians, hepatic glucose output remained normal during the transition from nor-mal glucose tolerance to IGT, but increased by 15% with further progression to diabetes (Weyer et al., 1999). In patients with established type 2 diabetes, the rate of hepatic


7

MEDICATIONS AND

DIABETES RISK

gluconeogenesis is not suppressed postprandially (as occurs normally). Thus, the upregulated endogenous glucose pro-duction becomes a key determinant of fasting, as well as postprandial glucose excursions in type 2 diabetes. The increased endogenous glucose production is triggered by an increased fl ux of lipolytic products and other glucose precursors, and is due, at least in part, to hepatic insulin resistance.

GLUCAGON AND INCRETINS

Glucagon secretion by the pancreatic � cells also tends to be elevated in type 2 diabetes subjects, which further stimu-lates hepatic glucose production. Incretin hormones (GLP-1 and GIP), normally secreted by the enterocytes in response to food, serve to amplify postprandial insulin secretion and suppress glucagon secretion. Emerging data indicate that type 2 diabetes is associated with impaired incretin secretion and relative resistance to the action of incretin hormones. The major pathophysiological defects in T2DM are summa-rized in Table 2.

Table 2 Pathophysiological Defects in Type 2 Diabetes

Insulin resistance ■

Impaired insulin secretion ■

Increased glucagon secretion ■

Incretin defi ciency/resistance ■

Impaired glucagon suppression ■

Increased hepatic glucose production ■


OAPN

8

Section 2: General Mechanisms of Medications

and Diabetes Risk

Diabetes has been reported in association with exposure to a wide variety of medications. As summarized in Table 3, in many cases, the mechanism relating the particular drug to hyperglycemia is explainable, whereas the mech-anism of the association is less clear with regard to sev-eral other medications (Comi, 2000). Hyperglycemia can result from drugs that induce hypoinsulinemia through destruction of the pancreatic � cells (e.g., pentamidine, Vacor), or by inhibiting insulin secretion (e.g., diazox-ide). Hypokalemia impairs insulin secretion and also desensitizes the insulin receptor, and it is one mecha-nism underlying the dysglycemia associated with thiaz-ides, loop diuretics, and hyperaldosteronism (Zillich et al., 2006). Induction of insulin resistance is a classic mechanism for steroid-induced diabetes, but the under-lying processes are complex (with contributions from hyperglucagonemia, glycogenolysis, lipolysis, and glu-coneogenesis). It has been suggested that drugs that re-strict blood f low could impair the normal delivery of substrates to insulin-sensitive tissues (especially skeletal muscle) and, by that mechanism, reduce glucose disposal (McCullen & Ahmed, 2007). Theoretically, vasoconstric-tors and �-blockers (unopposed �-adrenergic activity) could induce dysglycemia through this mechanism. In many instances, multiple mechanisms appear to mediate the drug effect. It is important to recognize drug-induced diabetes promptly, because withdrawal of the offending drug should result in resolution of the diabetes, in the absence of permanent cellular damage to the insulin-secreting � cells.


9

MEDICATIONS AND

DIABETES RISK

Table 3 Mechanisms and Examples of Iatrogenic Hyperglycemia

Mechanism Examples

A. Interference with insulin secretion

Pancreatotoxic•

Diazoxide, �-blockers Diuretics Pentamidine, Vacor

B. Interference with insulin action

Diuretics (via hypokalemia) Glucocorticoids, antiretrovirals �-agonists, growth hormone

C. Impaired insulin secretion and action

Thiazide diuretics cyclosporin, tacrolimus

D. Increased nutrient fl ux and gluconeogenesis

Nicotinic acidtotal parenteral nutrition�-interferon

E. Unknown mechanism Nonsteroidal anti-infl ammatory drugsantipsychoticsantidepressants

RISK FACTOR VERSUS CAUSATION: THE BRADFORD HILL’S CRITERIA

Originally established by Sir Austin Bradford Hill (Hill, 1965) and later elaborated by others, Hill’s criteria form the basis for establishing scientifi cally valid causal connections between potential disease agents and the many diseases that affl ict humankind. The criteria are (a) temporal relationship, an essential requirement that exposure to the agent must necessarily always precede the occurrence of the disease; (b) strength of association; (c) dose–response, a requirement that exposure to increasing amounts of the agent should result in increasing severity of disease; (d) consistency; (e) biological


OAPN

10

plausibility, with regard to currently accepted understanding of disease mechanisms; (f) consideration of alternate explana-tions; (g) experimental insight, a requirement that the observed association be reproducible and modifi able through delib-erate experimentation; (h) specifi city of the association; and (i) coherence, which requires that the association be compat-ible with existing theory and knowledge.

Applying the Hill’s criteria, only a few exceptions (e.g., pancreatic toxins and steroid-induced diabetes) among the myriad medications that have been reported to alter blood glucose levels meet the scientifi c threshold for establish-ing direct causality. For the majority of medications, the occurrence of diabetes is observed only in a minority of patients exposed, and the reported link to diabetes suff ers from paucity of randomized control data, weak association, inconsistent pattern, indeterminate temporal relationship (due to lack of baseline glucose data in many cases), unclear mechanism(s), and uncertain dose–response relationship.

Apart from a direct causal eff ect, treatment-emergent dia-betes could arise from preexisting undiagnosed diabetes; increased susceptibility to diabetes from underlying genetic or environmental risk factors; an indirect eff ect mediated by known risk factors, such as weight gain; a coincidental fi nd-ing; or an idiosyncratic reaction in susceptible persons that is inherently unpredictable.

DRUGS ASSOCIATED WITH TYPE 1 DIABETES

Type 1 diabetes is a disease of absolute or severe insulin defi -ciency. The vast majority of patients with type 1 diabetes have autoimmune destruction of the pancreatic islet �-cells as the underlying mechanism for the insulin defi ciency. In a small


11

MEDICATIONS AND

DIABETES RISK

minority of patients with type 1 diabetes, evidence for auto-immunity is lacking and the etiology of islet destruction is unclear. Drug-induced type 1 diabetes is rare in clinical prac-tice. Experimental insulin-defi cient diabetes can be induced by treating rodents with the pancreatic toxins streptozotocin or alloxan. Ingestion of the rat poison Vacor (N-3 pyridylm-ethyl-N’ 4 nitrophenol urea), which is structurally related to alloxan and streptozotocin, has been associated with human cases of diabetes (Miller et al., 1978; Pont et al., 1979). The diabetes induced by Vacor poisoning can have a delayed onset (≥1 week), but is usually severe and often presents with ketoacidosis. Prophylactic treatment with the antidote nicotinamide should be started as soon as possible following ingestion of Vacor, even in euglycemic persons (Miller et al., 1978; LeWitt, 1980). The antiprotozoal drug pentamidine, widely used for the treatment of refractory Pneumocystis carnii pneumonia (PCP) in human immunodefi ciency virus (HIV)-infected patients, can induce acute insulinopenic diabetes in some patients through destruction of the pancreatic � cells (Liegl et al., 1994; Coyle et al., 1996). An initial phase of hypoglycemia (refl ecting �-cell degranulation and transient hyperinsulinemia) may precede the development of pent-amidine-induced diabetes (Bouchard et al., 1982). The risk factors and exact mechanisms for �-cell destruction by pen-tamidine are unknown, and the resultant diabetes tends to persist after withdrawal of pentamidine (Liegl et al., 1994). Although pentamidine-induced hypoglycemia can be reversed by treatment with oral diazoxide (Fitzgerald & Young, 1984), it is not known whether such therapy would alter the course of progressive �-cell destruction by pentamidine.

Alcoholism and chronic pancreatitis have been associ-ated with the development of insulin-defi cient diabetes


OAPN

12

(Ito et al., 2007). It is possible that other as yet unidentifi ed environmental toxins may be involved in the pathogenesis of the rare nonautoimmune cases of type 1 diabetes.

DRUGS ASSOCIATED WITH TYPE 2 DIABETES

In contrast to the rarity of drug-associated typical type 1 dia-betes, numerous medications in common use have been asso-ciated with the development of what can best be classifi ed as type 2 diabetes. The latter classifi cation is based loosely on the absence of absolute insulinopenia together with evidence for insulin resistance or related mechanisms for the disrup-tion of glucoregulation. However, depending on the sever-ity and acuteness of the perturbation, some patients with drug-related diabetes may present with diabetic ketoacidosis (DKA). The fact that DKA is more characteristic of type 1 than type 2 diabetes may lead physicians to classify such pre-sentations as type 1 diabetes. However, it should be noted that approximately 25% of type 2 diabetes patients in the general population present with DKA (Johnson et al., 1980). After initial stabilization with insulin therapy and fl uid repletion, the majority of such patients respond to oral antidiabetic agents, as is typical of type 2 diabetes. Therefore, it is more important to stabilize the patient with drug-related hyper-glycemic crisis than to be distracted by a quest for an exact classifi cation in the acute setting. In fact, in many instances, the exact mechanism of the medication-related hyperglyce-mia remains obscure, and the condition is best assigned to the “Other” category.

Approximately 50% of patients with diabetes have hyper-tension; other chronic comorbidities include dyslipidemia, degenerative joint disease, chronic obstructive pulmonary disease (COPD), sleep apnea, congestive heart failure,


13

MEDICATIONS AND

DIABETES RISK

aff ective disorders, and peptic ulcer disease. Persons with diabetes also are at risk for infections. These various condi-tions often require chronic or recurrent treatment with a wide array of medications, some of which could aff ect insulin sensitivity, �-cell function, or other aspects of glucoregula-tion. Whenever feasible, preference should be given to those agents that are either neutral or benefi cial in their eff ects on carbohydrate and lipid metabolism.

In the sections that follow, diff erent classes of medications will be discussed with regard to their impact on diabetes risk. These medication classes were selected for discussion based either on (a) their historical association with dysgly-cemia in clinical practice, (b) extensive utilization for the management of comorbid conditions (e.g., hypertension, dyslipidemia) in diabetic patients, or (c) existing or emerging reports of possible association with dysglycemia.


OAPN

14

Section 3: Steroids and Immunomodulatory Agents

GLUCOCORTICOID STEROIDS

In persons with diabetes, systemic glucocorticoid steroid therapy impairs glycemic control by multiple mechanisms. Glucocorticoids induce insulin resistance, inhibit peripheral glucose utilization, stimulate lipolysis, and increase hepatic glucose production (Chan & Cockram, 1991; Andrews et al., 2002). In addition, these steroids inhibit insulin secretion and stimulate glucagon release (Chan & Cockram, 1991; Fallo et al., 2006). For persons with prediabetes and those at high risk for type 2 diabetes, prolonged steroid therapy could worsen glucose tolerance and induce diabetes. The diabetes risk is dependent on the dose and duration of glucocorticoid therapy, but even single doses of potent agents, such as dexam-ethasone, can induce transient hyperglycemia (Dagogo-Jack et al., 1997). Genetic factors play a prominent role in deter-mining susceptibility: In one study, a family history of diabetes increased the risk of steroid-induced diabetes tenfold (Fajans & Conn, 1954). Other risk factors for steroid-induced diabe-tes include the potency of steroid preparation, age, weight, decreased �-cell capacity, and a history of gestational diabetes (Heazell et al., 2005; McCullen & Ahmed, 2007) (Table 4).

Compared with systemic therapy, inhaled corticosteroids are far less likely to produce adverse eff ects on glucoregu-lation. However, high doses of topical steroids can induce hyperglycemia (Heazell et al., 2005).

Approach to Risk ReductionWhen systemic glucocorticoid therapy is unavoidable, as in patients with acute severe asthma or transplant recipients, blood glucose levels should be monitored frequently and


15

MEDICATIONS AND

DIABETES RISK

the antidiabetic regimen should be optimized. Insulin sen-sitizer drugs and insulin augmentation, alone or in combina-tion, can help restore glycemic control in most patients with steroid-induced diabetes. Because the metabolic eff ects of glucocorticoids are dose-related, use of the minimum eff ec-tive dose for treatment of the primary condition is recom-mended. For persons with prediabetes and those at high risk for type 2 diabetes, lifestyle modifi cation (Table 5) has been shown to be eff ective in preventing diabetes, and should be advocated empirically, although there are no specifi c data for the steroid-treated population. In experimental animals,

Table 4 Risk Factors for Steroid-induced Diabetes

Family history of diabetes ■

Type, dose, and duration of steroid therapy ■

History of gestational diabetes ■

Overweight/obesity ■

Older age ■

Decreased insulin secretory capacity ■

Table 5 Approach to Prevention of Steroid-induced Diabetes

Identify risk factors for diabetes (family history, body mass index, etc.) ■

Monitor blood glucose frequently in high-risk subjects ■

Recommend lifestyle modifi cation for high-risk persons ■

Use minimum eff ective dose of glucocorticoid steroid ■

Consider alternative-day regimen, if feasible ■

Consider metformin for persons with prediabetes (impaired fasting glu- ■

cose, impaired glucose tolerance)


OAPN

16

treatment with etomoxir (an inhibitor of fatty acid oxida-tion) improves insulin sensitivity and reverses glucocorti-coid-induced insulin resistance (Guillaume-Gentil et al., 1993). Theoretically, prophylactic use of insulin sensitizers (e.g., thiazolidinediones [TZDs] and metformin) to prevent diabetes during prolonged steroid therapy is appealing (Willi et al., 2002). However, that would be an “off -label” use of metformin and TZDs; besides, the number needed to treat is unknown, and the merit of such an approach over that of careful monitoring and lifestyle modifi cation is unproven. Nonetheless, for high-risk patients who show evidence of pre-diabetes (IFG and IGT) before or during prolonged steroid therapy, metformin can be added to lifestyle modifi cation to prevent progression to diabetes (Nathan et al., 2007).

ORAL CONTRACEPTIVES AND ESTROGEN REPLACEMENT

Oral contraceptive (OC) drugs containing estrogen and progesterone combinations have been reported to cause glucose intolerance, with elevations in plasma glucose and insulin levels. Depending on the dose and formulation, mild to moderate fl uctuations in blood glucose can occur follow-ing initiation of OC therapy (Gaspard, 1987). In one large study, OGTTs were performed in 1,060 women taking one of nine types of oral contraceptives for at least 3 months and 418 women who took none. Two of the formulations were progestin-only agents, and seven were combinations of pro-gestin (150–1,000 μg) and ethinyl estradiol (30–40 μg). The combination OCs were associated with a worsening of glucose tolerance. Compared to the control group, women taking OCs had an approximately 50% increase in the OGTT plasma glucose and an approximately 30% increase


17

MEDICATIONS AND

DIABETES RISK

in insulin and C-peptide levels, which indicates that expo-sure to OCs had induced insulin resistance (Godsland et al., 1990). In another report (Nader et al., 1997), 16 hyperan-drogenic women tested before and 6 months after receiving desogestrel-containing OC showed deterioration of glucose tolerance (two of them developed diabetes). Progestin-containing formulations are believed to be more liable to induce hyperglycemia, whereas low-dose triphasic OCs appear to be less so (Xiang et al., 2006; Spellacy et al., 1991). However, there are confl icting reports regarding whether a dose–response relationship exists between progestins and glycemia (Godsland et al., 1990; Rosenthal et al., 2004). In general, the OC-induced insulin resistance usually is mild and reversible; it is most likely a steroid eff ect, and can be minimized by selecting preparations with low estrogen and progesterone content.

Estrogen replacement therapy in menopausal women has the same potential eff ects on carbohydrate metabolism. In practice, however, diabetes patients on oral contraceptives or estrogen replacement therapy rarely develop signifi cant perturbations in glycemic control.

Approach to Risk ReductionAntidiabetic drug doses should be optimized in women with diabetes who experience blood glucose fl uctuations during treatment with OCs (Shawe & Lawrenson, 2003). Other women receiving OCs should be advised to undergo blood glucose testing if they develop symptoms suggestive of diabetes. The prevalence of diabetes in women treated with OCs is unknown, and causality has not been estab-lished. Therefore, diabetes (or the risk for diabetes) would not be a rational contraindication to the appropriate use


OAPN

18

of OCs. Use of low-dose triphasic formulations may be associated with less metabolic perturbations than higher- dose OCs.

FLORINEF

The mineralocorticoid agent, 9-�-fl uorohydrocortisone (Florinef), is widely prescribed for a variety of indications, including primary adrenal insuffi ciency, orthostatic hypo-tension, type IV renal tubular acidosis, and other conditions associated with aldosterone defi ciency and hyperkalemia. To date, the clinical use of Florinef has not been associated with diabetes risk in published reports. However, glucose intolerance is a feature of primary hyperaldosteronism, in which the mechanism appears to be mediated by hypokale-mia (Fallo et al., 2006).

IMMUNOMODULATORY AGENTS

Calcineurin Inhibitors and Posttransplant DiabetesOrgan transplantation is an expanding area of modern med-ical practice, and diabetes is being increasingly diagnosed in organ recipients. Posttransplant diabetes (also known as new-onset diabetes after transplantation, NODAT) refers to the occurrence of diabetes in previously nondiabetic subjects following transplantation. The development of NODAT is associated with adverse clinical outcomes, including renal allograft loss, posttransplant infections, cardiovascular disease, and increased mortality (Cosio et al., 2002; Chapman et al., 2005; Hjelmesaeth et al., 2006). Variable incidence rates of NODAT have been reported over variable intervals among recipients of diff erent organ transplants. The estimated rates of NODAT at 12 months


19

MEDICATIONS AND

DIABETES RISK

or longer posttransplant are approximately 20% for kidney transplants, 9%–21% for liver transplants, and approxi-mately 20% for lung transplants (Kasiske et al., 2003; Chadban, 2008; Bonato et al., 2009). Most of the data in this fi eld are derived from analysis of renal transplants, as there is limited information on other types of transplants. The peak incidence of NODAT appears to occur around 3 months posttransplant, but the risk of diabetes persists for much longer.

MechanismsThe clinical presentation of NODAT is consistent with type 2 diabetes, and studies have identifi ed insulin-resistance and impaired �-cell function as the underlying mechanisms (Hjelmesaeth et al., 2001). The insulin resistance can be in-duced in previously normoglycemic subjects and aggravated in prediabetic subjects, following organ transplantation. Medications used for posttransplant immunosuppression have been implicated in the pathogenesis of NODAT. The calcineurin inhibitors (tacrolimus and cyclosporine) and steroids have been the most documented drugs associated with the induction of NODAT. However, these agents also are the most widely used immunosuppressive agents in the transplant population, and many other risk factors appear to infl uence the development of NODAT (Table 6) (Vesco et al., 1996; Kasiske et al., 2003). The mechanisms for ste-roid-induced insulin resistance are well-known and have been discussed in the preceding section; the mechanism whereby calcineurin inhibitors induce insulin resistance is unclear, but these agents are known to inhibit insulin gene transcription and decrease insulin secretion by the � cells (Oetjen et al., 2003). The reported approximately 20% incidence of NODAT for kidney transplant recipients at 1


OAPN

20

year means that about 80% of graft recipients can expect to be free from diabetes (Kasiske et al., 2003; Chadban, 2008; Bonato et al., 2009). Thus, immunosuppressive agents induce NODAT in a substantial minority of post-transplant patients, whereas a sizeable majority of such patients escape the diabetes risk, which underscores the contributory roles of other risk factors (Table 6). Notably, the risk of NODAT does not appear to be a class eff ect for the calcineurin inhibitors: In one study, treatment with tacrolimus was associated with a two fold higher incidence of NODAT (16.8% vs. 8.9%) than cyclosporine (Vincenti et al., 2007). Prediabetes events also were more frequent with tacrolimus.

Approach to Management and Risk ReductionThe occurrence of NODAT provides opportunity for devel-oping evidence-based strategies for diabetes prevention, yet randomized controlled studies are lacking. Pretransplant lifestyle counseling would be most appropriate for high-risk patients, including those who are overweight or obese, and

Table 6 Risk Factors for Posttransplant Diabetes Mellitus

Nonmodifi able Modifi able

AgeRaceEthnicityFamily history of T2DM

Immunosuppressive regimenTacrolimus ■

Glucocorticoid steroids; dose and duration ■

Concurrent use of tacrolimus and steroid ■

Acute rejectionObesitySource of organ (cadaver vs. living)Hepatitis-C infection


21

MEDICATIONS AND

DIABETES RISK

persons with a family history of T2DM (Table 1). A screen-ing fasting plasma glucose level or OGTT would help identify subjects with prediabetes (IFG or IGT) during the pretrans-plant period, so that diabetes prevention counseling could be better targeted (Table 7). The goal of glycemic control in patients with NODAT is similar to the target (A1c <7%) for diabetic patients in the general populace (Wilkinson et al., 2005). For kidney transplant recipients, it is especially imperative to maintain excellent glycemic control, to pro-tect the allograft from diabetic nephropathy.

Blood glucose monitoring, diabetes education, dietary coun-seling, physical activity, and selective use of antidiabetic medications constitute the basis of comprehensive diabe-tes management. The mnemonic, MEDEM (Monitoring, Education, Diet, Exercise, Medication) can be used to recall the key elements of diabetes care. Depending on the organ transplanted and the state of renal, hepatic, and car-diac function, certain oral antidiabetic medications may be contraindicated in a given patient with NODAT, and insulin

Table 7 Approach to Prevention of Posttransplant Diabetes

Pretransplant Period

Document baseline FPG ■

Identify high-risk subjects ■

Initiate lifestyle intervention ■

Dietary counseling• Exercise counseling•

Posttransplant Period

Immunosuppressive regimen ■

Minimize steroid dose ■


OAPN

22

often is the drug of choice. In NODAT patients with intact renal, hepatic, and cardiac function, use of insulin sensitizers (metformin and TZDs) alone or in combination with secre-tagogues (sulfonylureas, glinides [repaglinide, nateglinide], gliptins [sitagliptin, saxagliptin]) may be tried before pro-ceeding to insulin. The �-glucosidase inhibitors (acarbose, miglitol), which act predominantly within the intestinal lumen to improve glycemic control, may be a consideration when systemic toxicity from other drugs must be avoided. Management of NODAT should be undertaken in close col-laboration with the transplant team; whenever feasible, the minimum daily or alternate-day dose of steroid should be used, to minimize steroid-induced dysglycemia (Wilkinson et al., 2005). Routine withdrawal or substitution of other-wise eff ective antirejection immunosuppressive therapy, solely because of NODAT, is ill-advised and could jeopardize the survival of the allograft. Nonetheless, small studies have suggested that substitution of tacrolimus with cyclosporine may be associated with improved glycemic control, and even resolution of NODAT in a few patients (Dumortier et al., 2006; Ghisdal et al., 2007). Since tacrolimus inhibits insulin secretion more potently (and is associated with higher risk for NODAT) than cyclosporine (Redmon et al., 1996), prefer-ential use of cyclosporine is a tempting approach to diabetes risk reduction, provided, of course, that it is as eff ective as tacrolimus in organ protection. A meta-analysis concluded that “treating 100 recipients with tacrolimus instead of ciclosporin for the fi rst year after transplantation avoids 12 patients having acute rejection and two losing their graft but causes an extra fi ve patients to develop insulin-dependent dia-betes” (Webster et al., 2005). This is a complex risk–benefi t calculation that has been well discussed by Chadban (2008).


23

MEDICATIONS AND

DIABETES RISK

TUMOR NECROSIS FACTOR-� INHIBITORS

Tumor necrosis factor (TNF)-� is a proinfl ammatory cytokine that has been implicated in the etiology of myriad immunological disorders, including rheumatoid arthritis, Crohn disease, psoriasis, and refractory asthma. Since the late 1990s, several monoclonal antibodies and fusion proteins that inhibit TNF-� by blocking its receptor have been intro-duced to clinical practice. These agents, collectively known as TNF-� inhibitors, include adalimumab, certolizumab pegol, etanercept, golimumab, and infl iximab, and are being increasingly prescribed for autoimmune disorders. There have been case reports indicating that exposure to TNF-� inhibi-tors might improve glycemic control or even induce hypo-glycemia in persons with diabetes (Kiortsis et al., 2005; van Eijk et al., 2007; Wambier et al., 2009; Cheung & Bryer-Ash, 2009). TNF-� induces pancreatic �-cell apoptosis and has been linked to insulin resistance and diabetes (Zinman et al., 1999; Obayashi et al., 2000). Thus, inhibition of TNF-� conceivably could improve glucose tolerance in diabetic sub-jects. The likely mechanisms for such improvement include amelioration of insulin resistance (Kiortsis et al., 2005; Martínez-Abundis et al., 2007) and preservation of �-cell function (Mastrandrea et al., 2009). Given the increased prevalence of diabetes in patients with autoimmune disorders (Brauchli et al., 2008; Cohen et al., 2008), close blood glucose monitoring and adjustment of antidiabetic regimen (as neces-sary) are advisable in patients receiving TNF-� inhibitors.

INTERLEUKIN-1 RECEPTOR ANTAGONIST

Interleukin (IL)-1 is overexpressed in pancreatic � cells of patients with T2DM, and is known to impair insulin secretion


OAPN

24

and induce �-cell apoptosis (Bendtzen et al., 1986; Maedler et al., 2002). Because progressive �-cell dysfunction under-lies the pathophysiology of T2DM, interventions that ame-liorate �-cell apoptosis can be expected to decrease diabetes risk. In recent randomized placebo-controlled trials, short-term treatment with anakinra (100 mg QD SC for 13 weeks), a recombinant human IL-1 receptor antagonist, was shown to reduce infl ammatory markers, improve �-cell function, and improve glycemic control in patients with T2DM (Larsen et al., 2007; Larsen et al., 2009). The improvement in �-cell function was demonstrable 39 weeks after the last dose of anakinra (Larsen et al., 2009), which suggests a cellular rather than humoral mechanism, and raises hope for IL-1 inhibition as a future strategy for diabetes prevention.

INTERFERONS

The interferons (IFNs) belong to a cytokine family of reg-ulatory peptides that exhibit antiviral, antiproliferative, and immunomodulatory properties (Dagogo-Jack, 1991). Advances in basic and translational research have led to clini-cal applications of some members of the interferon family in the treatment of viral infections and certain malignancies. For example, IFN-� is well established as an eff ective treat-ment for hepatitis-C infection. Following its wide clinical use, reports of diabetes or dysglycemia in patients treated with IFN-� have appeared (Fabris et al., 1998; Chedin et al., 1996). The cellular eff ects associated with IFNs include destruction of pancreatic islet cells (via an autoimmune mechanism) and possible induction of insulin resistance at the postreceptor level (Foulis et al., 1987; Pankewycz et al., 1995; Konrad et al., 2000).


25

MEDICATIONS AND

DIABETES RISK

The incidence of treatment-emergent diabetes in patients receiving chronic IFN treatment for hepatitis C infection appears modest. In one report, 7 of 202 (3.4%) patients developed diabetes and 33 (16.3%) developed prediabetes (IFG) during 5–16 years (median 8 years) of chronic IFN-� therapy (Giordanino et al., 2008). The diabetes associated with IFN therapy often requires treatment with insulin; the insulin requirement may be transient or permanent. The fact that patients with hepatitis C infection have an increased risk for diabetes (independent of treatment) complicates the interpretation of reports of IFN-associated diabetes in such patients (Ozylikau & Arslan, 1997; Huang et al., 2008).


OAPN

26

Section 4: Antihypertensive Agents

Antihypertensive therapy is generally well tolerated, such that no particular class of agents is specifi cally contrain-dicated for use in patients with diabetes. Diuretics and �-blockers have been well documented to reduce cardiovas-cular morbidity and mortality in the general hypertensive population (Alderman, 1992). Moreover, secondary pre-vention studies have confi rmed the effi cacy of �-blockers in preventing reinfarction and sudden death in patients with a previous history of myocardial infarction (Yusuf et al., 1990), as well as the survival advantage of angiotensin-converting enzyme (ACE) inhibitors in patients with heart failure (SOLVD Investigators, 1991).

DIURETICS

Several studies have shown that thiazide diuretics induce insulin resistance in hypertensive patients (Pollare et al., 1989; Lithell, 1991), and may be associated with treat-ment-emergent diabetes (Sowers, 1994). In both the Systolic Hypertension in the Elderly Patients (SHEP) and the Antihypertensive and Lipid-Lowering Treatment to Prevent Heart Attack Trial (ALLHAT) trials, treatment with chlorthalidone was associated with increased rates of incident diabetes compared to placebo (SHEP) or other active agents (ALLHAT) during a 3- to 4-year follow-up period (Sierra & Ruilope, 2003). The magnitude of diabe-tes risk was approximately 18%–49% for chlorthalidone versus comparators. In the ASCOT, the rate of incident diabetes was approximately 30% higher in the thiazide-containing treatment arm compared to the calcium channel blocker-ACE inhibitor arm (Dahlof et al., 2005). Diuretics


27

MEDICATIONS AND

DIABETES RISK

also have been associated with unfavorable changes in serum lipoprotein fractions, principally elevated trig-lycerides and decreased high-density lipoprotein (HDL) cholesterol levels, that appear to be transient. Insulin resistance and impaired insulin secretion, probably medi-ated by hypokalemia, appear to underlie the mechanism of the dysglycemia associated with thiazide and thiazide-like diuretics (Gulliford et al., 2005; Zillich et al., 2006). Other proposed mechanisms include increased free fatty acids and increased hepatic glucose output (Flamenbaum, 1983; Weir & Moser, 2000). As the metabolic eff ects of thiazide diuretics may be dose-related, use of smaller doses (e.g., 6.25–25 mg) is advisable.

�-ADRENERGIC BLOCKERS

Both selective and nonselective �-blockers can induce insulin resistance (Pollare et al., 1989; Lithell, 1991) and increase diabetes risk. In the Atherosclerosis Risk In Communities (ARIC) study, the rate of incident dia-betes was 28% higher in hypertensive patients treated with �-blockers compared to an untreated control group (Gress et al., 2000). Reports from other recent trials also document the increased risk of diabetes associated with use of �-blockers (reviewed by McCullen & Ahmed, 2007). During hyperinsulinemic euglycemic clamp stud-ies, nondiabetic patients treated with metoprolol (200 mg/day) or atenolol (50–100 mg/day) for 4–6 months had an approximate 25% reduction in insulin sensitivity (Lithell, 1991). There is also evidence that �-blockers can alter insulin clearance (Chan et al., 1996), reduce insulin secretion (Van Bortel & Ament, 1995), promote weight


OAPN

28

gain (Pischon & Sharma, 2001), and decrease peripheral blood fl ow (McCullen & Ahmed, 2007), all of which are mechanisms that could increase diabetes risk. However, these reports have not been widely replicated, and the glycemic eff ects of the reported alterations in insulin sen-sitivity and secretion are either modest or unapparent in the majority of patients treated with �-blockers. Thus, the exact mechanisms for induction of diabetes by �-blockers remain to be elucidated. Parenthetically, in diabetic patients, �-blockers can mask hypoglycemic symptoms (Hirsch et al., 1991) and prolong recovery from insulin-induced hypoglycemia. Furthermore, �-blockers have been associated with unfa-vorable changes in serum lipoprotein fractions, includ-ing elevated triglycerides and decreased HDL cholesterol levels. Drugs with combined �- and �-adrenergic block-ing activity (e.g., labetalol, carvedilol) might have a less adverse metabolic profi le compared to pure �-blockers (Bakris et al., 2004).

�-ADRENERGIC BLOCKERS

The �1-blockers, such as prazosin and terazosin, and com-bined �- and �-blockers (labetalol and carvedilol) have been reported consistently to decrease insulin resistance and improve glucose tolerance (Alderman, 1992; Yusuf et al., 1990).

ANGIOTENSIN-CONVERTING ENZYME INHIBITORS AND ANGIOTENSIN RECEPTOR BLOCKERS

Overactivity of the renin-angiotensin-aldosterone system has been associated with cellular and molecular changes


29

MEDICATIONS AND

DIABETES RISK

that lead to insulin resistance and impaired �-cell func-tion (Ferrannini et al., 2003; Jandeleit-Dahm et al., 2005). Moreover, there is evidence that ACE inhibitors can improve insulin sensitivity (Lithell, 1991; Giordano et al., 1995). Consonant with these reports, post hoc anal-yses from several multicenter randomized controlled stud-ies have observed signifi cant reductions in the incidence of type 2 diabetes in hypertensive patients treated with either ACE inhibitors or angiotensin receptor blockers (ARBs) compared to other agents, such as thiazide diuret-ics or �-blockers (Yusuf et al., 2000; Jandeleit-Dahm et al., 2005; Dahlof et al., 2005; Barzilay et al., 2006). However, a study specifi cally designed to test the protec-tive eff ects of ramipril (vs. placebo) on diabetes showed an increased rate of reversion from prediabetes to normo-glycemia but not a reduction in new-onset diabetes among initially prediabetic subjects in the ramipril arm (Bosch et al., 2006). Elliott and Meyer (2007) performed a network meta-analysis of 22 clinical trials of several antihypertensive drugs that enrolled a total of 143,153 participants who did not have diabetes at randomiza-tion. They found that the association of antihypertensive drugs with incident diabetes was lowest for ARB and ACE inhibitors followed by calcium channel blockers, placebo, �-blockers, and diuretics in that order (Elliott & Meyer, 2007) (Fig. 1). Thus, it is certain that ACE inhibitors and ARBs do not increase diabetes risk, and it is quite probable that they decrease risk.

The ACE inhibitors and ARBs are generally well toler-ated by diabetic patients, and are specifi cally indicated for nephroprotection in patients with microalbuminuria.


OAPN

30

These agents also confer signifi cant cardiovascular bene-fi ts. However, it must be noted that angiotensin inhibition may precipitate acute renal failure in patients with bilat-eral atherosclerotic renovascular disease, or exacerbate hyperkalemia in patients with diabetes-associated type IV renal tubular acidosis. Furthermore, these agents should be used with caution in patients with diabetic autonomic

Figure 1 Metaanalysis of incident diabetes in antihypertensive drug trials.Modifi ed from Elliott WJ, Meyer PM. (2007). Incident diabetes in clinical trials of antihypertensive drugs: a network meta-analysis. Lancet, 369:201–207, with per-mission of the publisher.

0.50 0.80 1.00 1.25 2.00

ARBs

ACEi

Placebo

CCBs

Betablocker

Diuretic

Treatment Odds Ratio (95% CI)

Odds Ratio for New-Onset Diabetes

1.347 (1.133-1.632)

0.822 (0.679-0.999)

0.889 (0.765-1.036)

Reference

1.051 (0.893-1.263)

1.250 (1.055-1.503)


31

MEDICATIONS AND

DIABETES RISK

neuropathy, who may respond with a worsening of ortho-static symptoms.

CALCIUM-CHANNEL BLOCKERS

There have been anecdotal reports of hyperglycemia (Roth et al., 1989) or improved glucose tolerance (Andersson & Rojdmark, 1981) in some patients treated with the vera-pamil. However, in general, clinical use of calcium-channel blockers does not pose a signifi cant diabetes risk and is asso-ciated with neutral eff ects on glucose tolerance and insulin sensitivity (Lithell, 1991; Giordano et al., 1995; Elliott & Meyer, 2007; Lam & Owen, 2007).

DIAZOXIDE AND OTHER PERIPHERAL VASODILATORS

The potent arteriolar vasodilator diazoxide causes hyperg-lycemia principally by inhibiting insulin secretion (Zunkler et al., 1988). Other suggested mechanisms for induction of hyperglycemia include activation of sympathetic discharge and increased gluconeogenesis. Diazoxide is indicated as an adjunctive treatment for insulinoma, and has also been used to reverse pentamidine-induced hypoglycemia. Diazoxide inhibits insulin secretion by opening the KATP channels on the �-cell membrane, which maintains hyperpolarization and prevents calcium entry into the cell. Although the acute eff ect of diazoxide inhibition of insulin secretion is hyper-glycemia, over time diazoxide inhibits pancreatic �-cell apoptosis and actually decreases diabetes risk (Guldstrand et al., 2002; Hansen et al., 2004; Huang et al., 2007). The acute eff ect of diazoxide has been exploited therapeuti-


OAPN

32

cally for the reversal of pentamidine-induced hypoglycemia (Fitzgerald & Young, 1984).

Minoxidil has been associated with exacerbation of glucose intolerance in diabetic patients and persons with impaired glucose tolerance (Lederballe Pedersen, 1977). Like diaz-oxide, the vasodilatory eff ect of minoxidil provokes an adrenergic response that potentially could lead to cate-cholamine-induced hyperglycemia. No such dysglycemic eff ect has been reported for hydralazine, which in fact has neutral or favorable eff ects on plasma lipoprotein profi les (Ames, 1986; Alderman, 1992; Wandell et al., 1996).

OTHER AGENTS

The centrally acting antihypertensive agents guanabenz, guanfacine, and clonidine are not associated with adverse eff ects on carbohydrate or lipid metabolism.


33

MEDICATIONS AND

DIABETES RISK

Section 5: Catecholamines, �-Adrenergic Agonists,

and Bronchodilators

Some of the drugs used for treatment of asthma and chronic obstructive pulmonary disease (COPD), such as �-adrenergic agonists and methylxanthines, can have appreciable eff ects on carbohydrate metabolism.

�-ADRENERGIC AGONISTS

Catecholamines and �2-adrenergic agonists stimulate gly-cogenolysis and lipolysis, and transiently increase insulin and glucagon secretion (Mayer et al., 1961; Young & Langsberg, 1983; Halter et al., 1984). These agents also induce insu-lin resistance and decrease peripheral glucose utilization (Rizza et al., 1980). The net eff ect of these actions is elevated plasma concentrations of glucose, lactate, and free fatty acids (FFAs). The catecholamine-induced insulin resistance is mediated, in part, by increased FFAs from lipolysis and by direct interference with the insulin receptor and postrecep-tor signaling mechanisms (Kirsch et al., 1983). Epinephrine, isoproterenol, and terbutaline, as well as over-the-counter decongestants (containing sympathomimetic drugs) are capable of elevating blood glucose. Inhaled agents are far less likely to aff ect glucose metabolism than are oral or paren-teral medications.

THEOPHYLLINE

Therapeutic concentrations of theophylline can increase plasma catecholamine levels and thereby lead to increased lipolysis and release of FFAs (Vestal et al., 1983). Thus, theo-phylline can induce hyperglycemia through the well-known diabetogenic eff ects of catecholamines and increased FFAs.


OAPN

34

Theophylline-induced hypokalemia also could be a contribu-tory factor, especially in cases of overdose (Chan & Cockram, 1991). These actions of theophylline and other xanthines may be mediated by inhibition of phosphodiesterase, leading to potentiation of catecholamine action (Hall et al., 1984). Very high doses of theophylline stimulate insulin secretion, probably through an indirect �-adrenergic mechanism, but this is not seen with standard doses (Vestal et al., 1983).


35

MEDICATIONS AND

DIABETES RISK

Section 6: Lipid-lowering Drugs

With regard to blood glucose regulation, the collective evidence indicates that fi brates and bile acid sequestrants may have benefi cial eff ects and statins have a neutral eff ect, whereas nicotinic acid increases the risk for dysglycemia. Furthermore, emerging data suggest that, in addition to their well-known cardioprotective eff ects, statins and fi brates might decrease the risk for diabetic microvascular complications (retinopathy, nephropathy, neuropathy).

NICOTINIC ACID

Some patients with dyslipidemia receiving chronic treat-ment with nicotinic acid experience a worsening of glucose tolerance (Tornvall & Walldius, 1991). Persons with pre-diabetes or diabetes appear to be at greater risk of glycemic exacerbation following exposure to nicotinic acid (Garg & Grundy, 1990). Nicotinic acid has complex interactions with glucoregulatory physiology. Acutely, it inhibits lipoly-sis, decreases FFA levels, and improves insulin sensitivity by stimulating peripheral glucose uptake and oxidation (Balasse & Neef, 1973). However, long-term treatment with nicotinic acid results in rebound lipolysis, increased FFA levels, increased gluconeogenesis, with resultant dys-glycemia (Thiebaud et al., 1982; Garg & Grundy, 1990; O’Byrne & Feely, 1990). These adverse metabolic eff ects of nicotinic acid are dose-dependent (Henkin et al., 1991), and the clinical impact on glycemic control in patients receiv-ing antidiabetic medications appears to be modest (~0.2 % increase in HbA1c) (Mohan & Mohan, 1997). Moreover, the sustained-release formulation may be associated with less perturbation of glucose tolerance (Fulcher et al., 1988; Grundy et al., 2002).


OAPN

36

Thus, nicotinic acid is strictly not contraindicated in patients with diabetes. Nicotinic acid has a uniquely benefi cial profi le in its ability to raise plasma high-density lipoprotein (HDL) cholesterol levels and decrease the levels of total cholesterol, low-density lipoprotein (LDL) cholesterol, and triglycer-ides. These eff ects help decrease cardiovascular disease risk, which is most desirable in patients with diabetic dyslipi-demia. With careful monitoring and appropriate adjustment of antidiabetic medications, nicotinic acid can be used to treat dyslipidemia in diabetic patients. Owing to the dose-related dysglycemic eff ect (Henkin et al., 1991), use of lower doses (~1,000–1,500 mg/day) of nicotinic acid is recom-mended in persons with diabetes (Grundy et al., 2002). It is also prudent to obtain plasma glucose levels prior to initia-tion of nicotinic acid therapy, to identify persons with predi-abetes. Because persons with prediabetes have an increased risk for hyperglycemia following treatment with nicotinic acid (Garg & Grundy, 1990), they should receive diabetes prevention counseling and should preferably be treated with submaximal doses of nicotinic acid (Table 8).

STATINS AND INCIDENT DIABETES

In addition to the well-known cholesterol-lowering eff ects, statin therapy has been associated with benefi cial

Table 8 Lipid-Lowering Drugs and Risk of Hyperglycemia

Increased risk Neutral Decreased Risk

Nicotinic acid Statins Colesevelam

Ezetimibe Bile acid sequestrants

Fibrates


37

MEDICATIONS AND

DIABETES RISK

“pleiotropic” eff ects, including reduction of circulating infl ammatory and oxidative stress markers, and improve-ment of endothelial function. The expectations that these eff ects would favorably impact diabetes risk have not been borne out by clinical trials. In a meta-analysis that included 42,860 patients without preexisting diabetes, the rates of incident type 2 diabetes were similar among patients ran-domized to statins compared to placebo during a 4-year follow-up period (Rajpathak et al., 2008). A similar con-clusion was reached by Susan et al. (2004) in nested case-control analysis of the UK-based General Practice Research Database (GPRD). In a recent trial that demonstrated supe-rior effi cacy of rosuvastatin over placebo in decreasing car-diovascular events among subjects with elevated C-reactive protein, the rosuvastatin group had a higher incidence of physician-reported diabetes than placebo (3.0% vs. 2.4% over a median follow-up period of 1.9 yr) (Ridker et al., 2008). A similar trend had been reported for other statins (Sasaki, et al., 2006). It must be noted that diabetes was not prespecifi ed as a primary outcome, and the diagnosis was not rigorously adjudicated, in the statin trials that reported increased diabetes rates. Thus, statins probably have a neu-tral eff ect on glucose metabolism, but the possibility of an adverse eff ect deserves further scrutiny.

FIBRATES

Hypertriglyceridemia is a hallmark of diabetic dyslipidemia, and fi brates (gemfi brozil, fenofi brate, bezafi brate) are fre-quently prescribed to control it. A retrospective analysis of a large database showed that exposure to bezafi brate was specifi cally associated with a reduced risk of incident diabetes (Flory et al., 2009). The database comprised 12,161


OAPN

38

patients treated with bezafi brate and 4,191 patients treated with other fi brates; baseline characteristics were similar be-tween the two groups. The hazard ratio for incident diabetes was 0.66 (95% confi dence interval [CI] 0.53–0.81) among bezafi brate users compared to users of other fi brates. The protective eff ect of bezafi brate became stronger with in-creasing duration of therapy. The potential mechanisms for prevention of diabetes might be amelioration of insulin resistance (Tenenbaum et al., 2006) through dual activa-tion of peroxisome proliferator-activated receptor (PPAR) � and � receptors (Tenenbaum et al., 2005). Clearly, ran-domized controlled trials are needed before bezafi brate or other fi brates can be recommended specifi cally for diabetes prevention.

COLESEVELAM AND BILE ACID SEQUESTRANTS

The bile acid sequestrant colesevelam is the only cholesterol-lowering agent that has been approved by the U.S. Food and Drug Administration for the adjunctive treatment of type 2 diabetes. In a randomized, placebo-controlled study in patients with T2DM who were poorly controlled with met-formin, the addition of colesevelam (3.75 g/day) reduced HbA1c by 0.54% compared to placebo during a 26-week follow-up period (Bays et al., 2008). The mechanism(s) whereby colesevelam improves glycemic control are unclear. Bile acid sequestrants are not absorbed systemi-cally, but exert their lipid-lowering eff ects by binding bile acids in the gut. The resultant exclusion of bile acids from the enterohepatic circulation leads to increased hepatic bile acid synthesis. The latter process is catalyzed by choles-terol 7-�-hydroxylase, which utilizes cholesterol as sub-strate, resulting in a depletion of the LDL-cholesterol pool.


39

MEDICATIONS AND

DIABETES RISK

Recent advances in our understanding of bile acids indicate that these moieties exert diverse metabolic eff ects through tightly regulated signaling pathways (Staels, 2009). It is plausible that interference with bile acid signal transduc-tion plays a role in the eff ects of colesevelam and other bile acid sequestrants on glucose metabolism (Staels, 2009).

DIABETIC COMPLICATIONS

The protective eff ects of statins and fi brates on cardiovascu-lar disease have been well documented. Emerging data also suggest that use of statins and fi brates may be associated with decreased risk for diabetic microvascular complications. In an observational study, Davis et al. (2007) reported that the risk of developing peripheral neuropathy among dia-betic patients treated with statins was decreased by 35%, and by 48% in those taking fi brates, compared to patients not treated with statins or fi brates. The benefi ts exerted by statins and fi brates on diabetic neuropathy appeared specifi c to each drug class and were independent of their eff ects on plasma lipid profi les. In the Fenofi brate and Event Lowering in Diabetes (FIELD) study, diabetic patients treated with fenofi brate experienced reductions in the risks of develop-ing microalbuminuria, proliferative retinopathy, and minor amputations (Keech et al., 2007; Rajamani et al., 2009). The exact mechanisms whereby statins and fi brates decrease microvascular complications are unknown. Putative media-tors include eff ects on infl ammatory cytokines, nitric oxide synthase, endothelial function, and angiogenesis. Diabetic neuropathy is a major risk factor for amputation among dia-betic patients; thus, the reported protective eff ects of statins and fi brates, if confi rmed by randomized controlled trials, could be of tremendous public health signifi cance.


OAPN

40

Section 7: Antimicrobial Drugs

ANTIBIOTICS

Several reports linked the fl uoroquinolone gatifl oxacin (withdrawn from the U.S. market in 2006) to increased risks for dysglycemia (i.e., hyperglycemia or hypoglyce-mia) (Catero, 2007). Acutely, gatifl oxacin stimulates insu-lin secretion (leading to hypoglycemia), whereas chronic therapy leads to inhibition of insulin synthesis and secretion (Yamada et al., 2006; Tomita et al., 2007). The fl uoroquino-lones share structural similarities with antimalarials (such as quinine, chloroquine, and mefl oquine) that are known to stimulate insulin secretion by closing the KATP channel on pancreatic � cells (Gribble et al., 2000). However, individ-ual fl uoroquinolone antibiotics have diff ering affi nities for the KATP channel: studies have shown that gatifl oxacin and temafl oxacin are more avid in this regard than ciprofl oxacin and levofl oxacin (Saraga et al., 2004). The risk of dysglyce-mia does not appear to be a class eff ect, as currently available quinolones (ciprofl oxacin, levofl oxacin, and moxifl oxacin) have been associated with less glucose abnormalities than gatifl oxacin (Aspinall et al., 2009). Among outpatients, the frequency of dysglycemic events (number of events divided by total number of treated patients) was 1.1% for gatifl oxacin, 0.3% for ciprofl oxacin, 0.3 % for levofl oxa-cin, and 0.2 % for moxifl oxacin (Park-Wyllie et al., 2006; Catero, 2007). By contrast, the dysglycemia rate was 0.2% for second-generation cephalosporins and 0.1 % for mac-rolides (Park-Wyllie et al., 2006). An earlier report among inpatients showed diff erent rates of dysglycemia: gatifl oxa-cin (1.01%), levofl oxacin (0.93%), and ciprofl oxacin (0%) (Mohr et al., 2005). The complicated illnesses and variable


41

MEDICATIONS AND

DIABETES RISK

caloric intake among the inpatients make direct comparison with the outpatient rates problematic.

The majority of patients who experience glucose disrup-tion with fl uoroquinolones have preexisting diabetes. Hypoglycemia typically occurs within 3 days of drug exposure, but has been reported following a single dose. Fluoroquinolone-induced acute insulin secretion can potentiate the eff ects of concomitant antidiabetic agents and thereby increase the risk of hypoglycemia among dia-betes patients. Overall, hyperglycemia is more prevalent than hypoglycemia in patients treated with quinolones. The mechanism of the hyperglycemia likely involves deple-tion of insulin granules and inhibition of insulin biogenesis, at least in the case of gatifl oxacin (Yamada et al., 2006). Fluoroquinolone-associated hyperglycemia usually presents after several days to weeks of drug exposure. Ambrose and colleagues have identifi ed renal dysfunction as a major risk factor for hyperglycemia in patients with toxic overdosage with fl uoroquinolone (Ambrose et al., 2003).

In a case-control analysis of a large outpatient database (cov-ering 1.4 million patients), the risk of treatment-emergent hyperglycemia was specifi c to gatifl oxacin, and was no higher for levofl oxacin, moxifl oxacin, or ciprofl oxacin than for other antibiotic classes such as macrolides or cephalosporins (Park-Wyllie et al., 2006). However, in a smaller chart review study covering approximately 17,000 inpatients, 92 patients (almost all with preexisting diabetes) had random blood glucose levels of 200 mg/dL or higher within 3 days of receiving levofl oxacin, gatifl oxacin, or ceftriaxone (Mohr et al., 2005). Preexisting diabetes and renal insuffi ciency were characteristic fi ndings among those who developed fl uoroquinolone-associated hyperglycemia (Mohr et al.,


OAPN

42

2005). In yet another retrospective analysis involving more than 64,000 patients in the Veterans Administration data-base, use of fl uoroquinolones was not associated with hyper-glycemia (Coblio et al., 2004).

These confl icting fi ndings refl ect the inherent weakness of the retrospective study design, as compared with random-ized controlled trials. Nonetheless, with regard to dysglyce-mia, the attributes of temporality, biological plausibility, and possibly dose–response have been demonstrated, at least for one member of the fl uoroquinolone family. The incidence of fl uoroquinolone-associated glucose abnormalities is low, and the risk appears much lower for agents in current use than for gatifl oxacin. Patients with diabetes, the elderly, and those with renal dysfunction should be monitored closely for hyperglycemia (and hypoglycemia) during therapy with fl uo-roquinolones. However, concern over the possible risk of dysglycemia should not override the use of fl uoroquinolones for the appropriate indications in patients with or without diabetes.

ANTIRETROVIRAL AGENTS

The use of combination highly active antiretroviral therapy (HAART) has dramatically improved clinical outcomes for people with human immunodefi ciency virus (HIV) infec-tion (Palella et al., 1998; Dagogo-Jack, 2008a). However, adverse metabolic eff ects, including insulin resistance, diabetes, dyslipidemia, and lipodystrophy (Dagogo-Jack, 2006) have been associated with HAART. The clinical presentation of antiretroviral-associated diabetes (ARAD) is consistent with type 2 diabetes, and the underlying mechanisms include insulin resistance and impaired �-cell function. HIV-1 protease inhibitors (PIs) acutely induce


43

MEDICATIONS AND

DIABETES RISK

insulin resistance and concurrently impair insulin secretion (Yarasheski et al., 1999; Dube et al., 2001; Schütt et al., 2004). The PIs share structural similarities with glucose transporter molecules (GLUT4; the major intracellular glucose transporter), and probably induce insulin resistance by interfering with glucose transport (Hertel et al., 2004). In cultured islet cells from an insulinoma cell line, exposure to indinavir and other PIs such as amprenavir, nelfi navir, and ritonavir signifi cantly inhibits glucose-stimulated insu-lin secretion (Koster et al., 2003). The risk factors for insu-lin resistance and diabetes in patients with HIV infection treated with PIs include positive family history of diabetes, weight gain, lipodystrophy, older age, and coinfection with hepatitis C (Table 9) (Mehta et al., 2003).

Nucleoside analogues (reverse transcriptase inhibitors) stavudine, zidovudine, and didanosine also are associ-ated with the risk of incident diabetes during long-term follow-up (De Wit et al., 2008). The increased risk persists

Table 9 Risk Factors for Diabetes in Human Immunodefi ciency Virus (HIV)-infected Patients

Family history of diabetes ■

Older age ■

Medications ■

Protease inhibitors•

Nucleoside analogues ■

Weight gain ■

Lipodystrophy ■

Coinfection with hepatitis C ■


OAPN

44

after adjustment for diabetes risk factors, which suggests a possible direct eff ect of the nucleoside analogues on gluco-regulation. Indeed, nucleoside analogues have been shown to induce insulin resistance, lipodystrophy, and mitochondrial dysfunction (Brown et al., 2005; Fleischman et al., 2007), which could be mechanisms for the development of diabe-tes. In their analysis of 33,389 HIV-infected patients who were followed for incident diabetes, De Wit et al. (2008) observed that exposure to stavudine conferred the great-est risk for incident diabetes, compared with other agents, including PIs. The current evidence indicates that PIs confer acute metabolic risks, whereas the nucleoside analogs con-fer cumulative risk for diabetes over time, and concurrent exposure to PIs and nucleoside analogs poses additional risk for diabetes among HIV-infected patients.

Physicians who treat patients with HIV-acquired immune defi ciency syndrome (AIDS) need to be alert to the adverse metabolic eff ects of the expanding antiretroviral armamen-tarium. Furthermore, as a result of the effi cacy of HAART and improved nutritional status, many HIV-infected patients in remission experience signifi cant weight gain, which is an additional risk factor for insulin resistance, diabetes, and dyslipidemia.

Approach to Management and Risk Reduction

Lifestyle Modifi cationThe standards of care for diabetes in HIV-infected patients are the same as those for the general diabetes population. The general glycemic goals are: average preprandial blood glucose values of 80–120 mg/dL, bedtime blood glucose of 100–140 mg/dL, and A1c level of 7% or lower (ADA, 2009).


45

MEDICATIONS AND

DIABETES RISK

A comprehensive approach incorporating lifestyle modifi ca-tion (diet and exercise) and antidiabetic medications is advo-cated (Roubenoff et al., 1999). Patients with HIV disease and diabetes require an individualized dietary intervention, guided by current weight, state of glycemic control, lipid profi le, blood pressure, and presence or absence of catabolic drive from opportunistic infections.

MedicationsThe approved antidiabetic drugs are eff ective in controlling hyperglycemia in HIV patients. However, there is potential for adverse drug interactions among some of the concomi-tant medications. Any preexisting hepatic injury (likely from hepatitis C or drug toxicity) precludes the use of thiazoli-dinediones (TZDs), and patients with HIV nephropathy are not candidates for metformin therapy. Moreover, the gas-trointestinal side eff ects of metformin and �-glucosidase inhibitors (acarbose, miglitol, voglibose) may be limiting in patients with HIV-associated enteropathy. Nucleoside ana-logues and other agents used in HAART regimens have spe-cifi c warnings regarding potential risks of lactic acidosis and severe hepatic dysfunction (Table 10). Therefore, close mon-itoring of liver enzymes is required in patients on HAART who require treatment with TZDs, metformin, or other oral antidiabetic medications. Whenever such complex toxico-logical considerations exist, exogenous insulin becomes the therapy of choice, but care must be taken regarding needle disposal.

Switching MedicationsThere are reports that switching from PIs to regimens substituted with other agents such as nevirapine may


OAPN

46

improve hyperglycemia (Martinez et al., 1999). However, it is generally not prudent to abruptly discontinue otherwise effi cacious antiretroviral agents, and any such decision must be taken in close consultation with the physicians managing the HIV disease.

Table 10 Antiretroviral Drugs Associated with Lactic Acidosis and Hepatic Dysfunction

Lactic Acidosis/Hepatomegaly

Elevated Transaminases

Nucleoside analoguesAbacavirEmtricitabineLamivudineStavudine Zidovudine

Nucleoside RT inhibitorsTenofovirZalcitabine

Non-nucleoside RT inhibitorsDelavirdineEfavirenz

Protease inhibitorsFosamprenavir LopinavirRitonavirSaquinavir

RT, reverse transcriptase.From Physicians Desk Reference 2004, pp. 481, 496, 1114, 1392, 1467, 1489, 1493, 1608, 2630, 2905, 2913.


47

MEDICATIONS AND

DIABETES RISK

Section 8: Atypical Antipsychotic and Antidepressant Agents

ATYPICAL ANTIPSYCHOTIC AGENTS

The second-generation or atypical antipsychotic agents (SGAs) improve clinical outcomes and are less likely to induce extrapyramidal side eff ects, compared to older agents (Freedman, 2003). However, their use has been asso-ciated with adverse metabolic eff ects, including diabetes, weight gain, and dyslipidemia. Diabetes has been reported in patients receiving all SGAs in current use (Dagogo-Jack, 2009). Interestingly, older reports from the last century also associated use of fi rst-generation (typical) antipsychotic agents with the development of diabetes (Meduna et al., 1942; Thonnard-Neumann, 1968; Keskiner et al., 1973). The clinical pattern of antipsychotic associated diabetes (APAD) is consistent with type 2 diabetes; the mean age at presentation is approximately 40 years. About two-thirds of patients present with new-onset diabetes, one-third present with exacerbation of preexisting diabetes, and 25% pres-ent with diabetic ketoacidosis (DKA) or metabolic acidosis (Dagogo-Jack, 2006). The frequency of DKA at presenta-tion in patients with APAD is comparable to the 23% rate reported for the general type 2 diabetes population (Johnson et al., 1980).

Putative explanations for treatment-emergent diabetes include preexisting undiagnosed diabetes, increased suscep-tibility to diabetes due to underlying psychiatric disorders (Dixon et al., 2000; Fernandez-Egea et al., 2008), a direct drug eff ect, an indirect eff ect mediated by weight gain, or an inherently unpredictable idiosyncratic reaction in suscep-tible persons. It must be stressed that a causal link between


OAPN

48

specifi c antipsychotic agents and diabetes remains to be proven. The fact that diabetes is reported only in a minority of patients exposed to antipsychotic drugs weakens the case for direct causality. A mechanism driven purely by insulin resistance from antipsychotic-induced weight gain also is unlikely, because compensatory insulin secretion by the � cells normally restores glucose homeostasis (Polonsky et al., 1996; Sowell et al., 2002). Indeed, most of Hill’s criteria for causality have not been met with regard to the association between antipsychotic drug exposure and incident diabetes, because of paucity of data from randomized controlled stud-ies, inconsistencies in the literature, incomplete documen-tation of baseline glycemic status, and lack of mechanistic insight, among other reasons (Dagogo-Jack, 2006; Dagogo-Jack, 2009).

The Clinical Antipsychotic Trials of Intervention Eff ectiveness (CATIE) was a National Institutes of Health (NIH)-funded prospective study that documented metabolic data in 1,460 patients with schizophrenia randomized to treatment with four atypical antipsychotic agents (olanzap-ine, risperidone, quetiapine, ziprasidone) and perphenazine (a typical antipsychotic agent). The primary aim of CATIE was to compare the overall eff ectiveness of the fi ve diff er-ent treatments on clinical outcome of schizophrenia during a mean follow-up of 18 months (Lieberman, et al., 2005). Table 11 shows the changes in glycosylated hemoglobin during a median 18 months of follow-up of patients in the CATIE study.

Approximately 11% of the patients in CATIE had preex-isting diabetes, and the proportions were fairly evenly distributed across the drug treatment groups. The CATIE


49

MEDICATIONS AND

DIABETES RISK

study did not report the frequency of new treatment-emergent cases of diabetes, but the investigators reported the numbers (and percentages) of persons who required new added medications for diabetes, as follows: olan-zapine 12 (4%), quetiapine 7 (2%), risperidone 8 (2%), perphenazine 5 (2%), and ziprasidone 4 (2%). These rates were not significantly different from each other. In summary, this landmark prospective, randomized study showed that chronic treatment with different antipsy-chotic agents was associated with variable but modest f luctuations in weight and glucose levels. However, the rate of treatment-emergent diabetes was modest, perhaps 2%–4%, and was similar between typical and atypical antipsychotic agents.

Table 11 Changes in Weight and Glycosylated Hemoglobin (GHb) During Treatment with Antipsychotic Agents

Antipsychotics Mean Change in GHb (%)

Median Change in Weight (lb/mo)

Olanzapine 0.40 + 0.07 0.8 (–1.4 to 9.5)

Quetiapine 0.04 + 0.08 0.1 (–4.4 to 6.3)

Risperidone 0.07 + 0.08 0 (–4.6 to 5.7)

Ziprasidone 0.11 + 0.09 –0.3 (–5.3 to 5.9)

Perphenazine 0.09 + 0.09 –0.1 (–4.9 to 4.0)

Data are mean change from baseline + SE for GHb and median (range) for weight.Adapted from Lieberman JA, Stroup TS, McEvoy JP, et al. (2005). Clinical Antipsychotic Trials of Intervention Eff ectiveness (CATIE) Investigators. Eff ectiveness of antipsychotic drugs in patients with chronic schizophrenia. N Engl J Med, 353:1209–1223, with permission of the publisher.


OAPN

50


MonitoringThe risks of suicide, homicide, and other serious harm to self and society in persons with psychotic illness mandate that eff ective control of psychosis remains the top priority. Therefore, physicians should utilize agents that best accom-plish that goal, based on proven effi cacy and eff ectiveness data (Lieberman et al., 2005). As recommended by the U. S. Food and Drug Administration and the American Diabetes Association Consensus Panel, screening for diabetes risk factors and documentation of baseline glu-cose levels, weight, waist circumference, body mass index, blood pressure, and lipid profi le should precede initiation of antipsychotic therapy (ADA, 2004). The fasting plasma glucose and lipid levels should be repeated after 12 weeks of initiation of antipsychotic therapy, and annually thereafter (Table 12).

Table 12 American Diabetes Association (ADA) Screening Guidelines for Patients on Second-generation Antipsychotics

Baseline 4 wk 8 wk 12 wk 12 mo 5 yr

Personal/family history

X

Weight (BMI) X X X

Waist circumference X

Blood pressure X

Fasting glucose X X X

Fasting lipids X X X

From American Diabetes Association. (2004). Consensus Development Conference on antipsychotic drugs and obesity and diabetes. Diabetes Care,

27:596–601, with permission of the publisher.


51

MEDICATIONS AND

DIABETES RISK

Education and Lifestyle Modifi cationPatients and their families should be educated regarding the metabolic risks associated with antipsychotic agents (Keck et al., 2003). Body weight should be recorded at baseline and every 4 weeks during the fi rst 3 months of treatment with SGAs (ADA, 2004). Patients with rapid weight gain (≥3 lb during the fi rst month of drug exposure) should receive prompt lifestyle intervention to reverse or limit weight gain. Caloric reduction, food portion control, sub-stitution of beverages with water or sugar-free alternatives, increased physical activity, and close and frequent monitor-ing of weight and food intake are some of the proven strat-egies for prevention of antipsychotic-induced weight gain (Menza et al., 2004; Hoff mann et al., 2005). Patients with preexisting diabetes should be monitored for possible gly-cemic exacerbation, and appropriate adjustments to antidi-abetic regimens should be implemented.

MedicationsThe actual goals and interventions for control of diabetes in patients receiving antipsychotic drugs are similar to those of the general populace. The approved antidiabetic agents retain their effi cacy in psychiatric patients, and no adverse interactions with SGAs have been reported. Agents that pro-mote weight loss or are weight neutral may be preferable as initial therapy in obese patients, but most patients eventually require multiple medications for eff ective glycemic control. Insulin is indicated for severe hyperglycemia, DKA, and hyperosmolar crisis, and for patients whose diabetes is inad-equately controlled on oral agents.

Switching MedicationsRoutine withdrawal of an otherwise eff ective SGA is impru-dent, because of the risk of psychotic relapse. However, drug


OAPN

52

substitution may be compelling under certain circumstances (e.g., severe hyperglycemia, DKA, diabetic coma), but this must be undertaken in close consultation with a psychiatrist. Any drug substitution would be on an empirical basis, as there are currently no scientifi c data on diff erential diabetes risks among individual SGAs to guide such practice.

ANTIDEPRESSANTS

Depression, as well as use of antidepressant medications, has been associated with increased risk for diabetes (Eaton et al., 1996; Knol et al., 2006; Andersohn et al., 2009). In a randomized placebo-controlled double-blind trial among patients with diabetes and depression, treatment with fl uoxetine for 8 weeks improved depression scores and showed a trend toward improvement of glycemic control (glycated hemoglobin: –0.7% vs. –0.4%, P=0.13) (Lustman et al., 2000). However, antidepressant use has been associ-ated with increased risk of incident diabetes among initially nondiabetic subjects.

The exact nature of the relationship between antidepressant medications and diabetes risk is unclear. Tricyclic antide-pressants have been associated with weight gain (Fava, 2000; Aronne & Segal, 2003) and worsening glycemia in diabetic patients (Lustman et al., 1997). However, selective sero-tonin reuptake inhibitors (SSRIs) and other antidepressants are largely weight-neutral; some have even been associated with weight loss and improved insulin sensitivity (Maheux et al., 1997). Notably, depression occurs more frequently among people with diabetes than in the general population (Egede & Zheng, 2003), and is associated with worse out-comes (more diabetic complications, increased mortality)


53

MEDICATIONS AND

DIABETES RISK

(de Groot et al., 2001; Katon et al., 2005). Activation of the hypothalamic-pituitary-adrenal axis with resultant hyper-cortisolemia is a well known mechanism in depression and major psychiatric disorders (Yehuda et al., 1993). Thus, endogenous steroid-induced dysglycemia could be a factor linking depression to diabetes risk.

The Diabetes Prevention Program (DPP) tracked 3,000 prediabetic subjects (randomized to treatment with lifestyle intervention, metformin, or placebo) for incident type 2 dia-betes during 2.8 years of follow-up. At baseline, 0.3% of DPP participants had symptoms of depression (Beck Depression Index, BDI scores >11) and 5.7% were taking antidepres-sants (Diabetes Prevention Program [DPP] Research Group, 2008). Subjects with elevated BDI scores at baseline and those who ever had elevated BDI scores during the DPP were identifi ed. Cox proportional hazard models were then used to evaluate whether elevated depression symptoms or use of antidepressants predicted progression from prediabetes to type 2 diabetes. After controlling for multiple predictors of diabetes risk (age, gender, fasting glucose, weight change, and ethnicity), elevated BDI scores at baseline or during the DPP study did not predict diabetes risk in any treatment arm. By contrast, use of antidepressant at baseline more than doubled the risk of incident diabetes among participants in the placebo (hazard ratio 2.25 [95% confi dence interval (CI) 1.38–3.66]) and lifestyle (3.48 [1.93–6.28]) arms (DPP Research Group, 2008). Continuous antidepressant use dur-ing the study (vs. never users) also signifi cantly predicted a similar magnitude of diabetes risk in the same subgroups of DPP subjects. Interestingly, subjects in the metformin arm did not show a relationship between antidepressant use and diabetes risk.


OAPN

54

The association between antidepressant use and diabetes risk in the DPP was seen across all drug classes, including SSRIs and selective norepinephrine reuptake inhibitors (SNRIs). The DPP fi ndings are consistent with recent data from a nested case-control study in a cohort of 165,958 patients with depression in the United Kingdom General Practice Research Database (Andersohn et al., 2009). The U.K. data showed increased diabetes rates for persons receiving long-term (>24 months) treatment with moderate to high daily doses of tricy-clic antidepressants (TCAs) (incidence rate ratio = 1.77, 95% CI = 1.21–2.59) and SSRIs (incidence rate ratio = 2.06, 95% CI = 1.20–3.52). No increase in diabetes risk was observed for shorter courses or lower daily doses of antidepressants. Another study reported higher risk of diabetes for persons receiving combined treatment with SSRIs and TCAs than those treated with TCAs alone (Brown et al., 2008).

The exact mechanism(s) for the reported increased risk of diabetes from antidepressants are unclear; the weight gain sometimes associated with TCA use is not suffi cient explana-tion for the development of diabetes. (In the DPP, use of anti-depressants predicted increased diabetes risk after controlling for weight gain.) Moreover, in the DPP, the predominant anti-depressants used were SSRIs and related agents, medications generally considered to be weight-neutral or even associated with weight loss. One plausible explanation is that antide-pressant use probably refl ects severe or chronic of depression, which has a stronger association with diabetes risk than mild depression (Palinkas et al., 1991; Wing et al., 1990).

Approach to Risk ReductionUntil the exact mechanisms are elucidated, it is prudent to screen for diabetes at baseline and monitor blood glucose


55

MEDICATIONS AND

DIABETES RISK

levels at reasonable intervals during treatment with all anti-depressants. The emerging data also suggest that treatment with lower doses of antidepressants and for shorter durations (<24 months) may be associated with lower risk of incident diabetes. Therefore, whenever clinically feasible, clinicians should aim to minimize the dose and duration of antidepres-sant therapy, without jeopardizing the overall mental health of the patient. Persons with risk factors for diabetes (Table 1) who require long-term antidepressant therapy should proba-bly receive lifestyle counseling for diabetes prevention (DPP Research Group, 2002; Dagogo-Jack, 2008b)on empirical grounds. Although the DPP showed that metformin treat-ment was associated with apparent protection from antide-pressant-associated diabetes (DPP Research Group, 2008), further studies are needed before metformin prophylaxis can be recommended routinely.


OAPN

56

Section 9: Recreational Drugs

ALCOHOL

The relationship between alcohol consumption and diabetes risk is complex. Habitual excessive alcohol intake is a major risk factor for chronic pancreatitis, which is associated with increased diabetes risk. In one longitudinal study compris-ing 656 patients with chronic pancreatitis, the cumulative rate of diabetes over an 8-year period was 50.4% (Ito et al., 2007). Alcoholic pancreatitis accounted for two-thirds of all new diabetes cases, and continuous alcoholic intake aggra-vated chronic pancreatitis and increased the risk of diabe-tes in the study cohort (Ito et al., 2007). In contrast, the metabolic benefi ts of moderate alcohol intake have been known for many years. In the Diabetes Prevention Program, subjects who reported moderate alcohol intake had lower incidence rates of diabetes compared to those without a history of alcohol consumption (Crandall et al., 2009). However, the benefi cial eff ects of moderate alcohol intake were confi ned to subjects randomized to the lifestyle or metformin, but not placebo, arm of the study (Crandall et al., 2009). The mechanisms by which moderate alcohol con-sumption appears to reduce diabetes risk have not been fully explored.

NICOTINE

Diabetic patients with a current history of cigarette smoking tend to have higher HbA1c and lipoprotein levels compared with nonsmokers (Stamler et al., 1993). Cigarette smoking also is associated with metabolic syndrome (Miyatake et al., 2006) and new-onset diabetes (Mozaff arian et al., 2009). The mechanisms for the association between smoking and


57

MEDICATIONS AND

DIABETES RISK

dysglycemia include induction of insulin resistance, in-creased hepatic lipase activity, chronic elevation of stress hormones, endothelial dysfunction, and the vasoconstric-tive eff ect of nicotine (Facchini et al., 1992; Heizer et al., 1996; Kong et al., 2001).

The return of blood pressure, heart rate, and blood fl ow and skin temperature of hands and feet to normal within 20 minutes after smoking cessation indicates rapid reversibility of the acute vasoconstrictive eff ects of nicotine. However, rigorous intervention studies testing the eff ect of smoking cessation on dysglycemia or incident diabetes are yet to be reported. Nonetheless, there are compelling reasons for pro-moting smoking cessation among diabetic and diabetes-prone subjects. These include the expected pulmonary, cardiovas-cular, and cerebrovascular benefi ts that accompany smoking cessation (Solberg et al., 2001). Current approaches include cognitive behavioral therapy, use of tapered transdermal or buccal nicotine, and prescription medications (bupropion, varenicline) to decrease craving during the transitional pe-riod (Prochaska et al., 2004).

CANNABINOIDS AND OPIOIDS

Limited data indicate that opioids could have a deleterious eff ect on insulin secretion (Giugliano, 1984; Passariello et al., 1986). Furthermore, a phenotype resembling insu-lin resistance has been described among heroin addicts (Passariello et al., 1983). With regard to cannabinoids, their eff ect on appetite stimulation and food intake would predict that chronic use could result in weight gain and increase the risk for dysglycemia. This notion is strengthened by recent data showing weight reduction and improvement in


OAPN

58

glycemic control following treatment with the endocannab-inoid receptor blocker rimonabant (Scheen et al., 2006). Nonetheless, primary data are lacking on the use of can-nabinoids and incident diabetes in humans.

COCAINE

There are limited data on the eff ects of cocaine on diabe-tes risk. Sporadic reports have indicated that use of cocaine might be associated with hyperglycemic crises (including diabetic ketoacidosis and nonketotic hyperosmolar syn-drome) in persons with preexisting diabetes (Warner et al., 1998; Abraham & Khardori, 1999; Nyenwe et al., 2007). Mechanisms for hyperglycemic crisis include cocaine-induced discharge of counterregulatory hormones, such as catecholamines (Chiueh & Kopin, 1978) and glucocorticoids (Teoh et al., 1995; Heesch et al., 1995). Thus, a history of cocaine abuse in a diabetic patient should be considered a risk factor for hyperglycemic crisis; however, further stud-ies are needed to determine whether habitual use of cocaine increases the risk of incident diabetes, especially among high-risk populations.

Overall, there is inadequate information regarding the use of recreational drugs and diabetes risk. Given the widespread drug habit in society and the escalating diabetes epidemic, well-designed studies on exposure to hard drugs and diabe-tes risk are sorely needed.


59

MEDICATIONS AND

DIABETES RISK

Section 10: Miscellaneous Agents

NONSTEROIDAL ANTI-INFLAMMATORY DRUGS AND ANTIARTHRITIC AGENTS

Hyperglycemia is rarely reported in patients receiving non-steroidal anti-infl ammatory drugs (NSAIDs) (Tkach, 1982). The mechanism for this is obscure, and a causal relationship is doubtful. Medically prescribed, as well as over-the-counter products containing glucosamine sulfate, are being increas-ingly used for the treatment or palliation of osteoarthritis. In animal studies, administration of glucosamine results in im-paired secretion and action of insulin (Giaccari et al., 1995). Increased substrate fl ux through the hexosamine pathway could be a mechanism for insulin resistance following ex-posure to glucosamine (Monauni et al., 2000). However, studies in humans have failed to show signifi cant alterations in glucose tolerance, insulin sensitivity, or insulin action fol-lowing treatment with oral glucosamine (500 mg t.i.d.) for 4 weeks (Yu et al., 2003). In patients receiving medications for diabetes, exposure to glucosamine did not alter glyce-mic control (Scroggie et al., 2003). However, persons with impaired glucose tolerance appear to be at increased risk for worsening dysglycemia after ingestion of glucosamine (Biggee et al., 2007). Based on the human data, it is unlikely that use of glucosamine in the recommended dosage would pose a signifi cant risk for diabetes in persons with normal glucose tolerance. It would be advisable for persons with risk factors for diabetes (Table 1) to undergo blood glucose screening before initiating glucosamine supplementation. If the screen-ing test indicates prediabetes, lifestyle modifi cation for diabetes risk reduction should be considered.


OAPN

60

Ingestion of high doses of aspirin (4–6 g/day) or toxic overdose can cause hypoglycemia (Cotton & Fahlberg, 1964; Gilgore, 1960). Salicylates are known to improve glucose tolerance via complex mechanisms, including enhanced insulin sensitivity, deceased gluconeogenesis, and decreased infl ammatory cytokines (Hundal et al., 2002). These properties suggest that salicylates should decrease diabetes risk. The risk of hypoglycemia is mini-mal with the lower doses of aspirin used for prophylaxis or routine pain control.

THYROID HORMONE

Thyroid hormone is one of the most widely prescribed medications in the general population and in patients with diabetes. There is little evidence that physiological doses of thyroid hormone replacement signifi cantly alter glu-cose regulation. In hyperthyroid patients, total serum T3 levels have been reported to correlate with glycosylated hemoglobin levels (Saito et al., 1982). The mechanism for hyperthyroidism-associated dysglycemia probably involves increased glucose-6-phosphatase activity, increased hepatic glucose production, and questionable alterations in insulin action and secretion (Taylor et al., 1985; Cavallo-Perin et al., 1988; Karlander et al., 1989). The clinical import of the association between hyperthyroidism and glucose intoler-ance (even in persons with established diabetes) is moot, as the hyperthyroidism would not be left untreated. Thus, no special precautions or concerns regarding dysglycemia are called for in the routine treatment of hypothyroidism with thyroid hormone replacement in diabetic and nondiabetic patients.


61

MEDICATIONS AND

DIABETES RISK

PHENYTOIN

Phenytoin is indicated for the control of generalized tonic-clonic and complex partial seizures. It is also widely pre-scribed for prophylaxis and treatment of seizures following head trauma or neurosurgery. Phenytoin has a narrow safety margin, and surveillance of drug levels is required dur-ing its use. Toxic doses of phenytoin have been associated with hyperglycemia, probably mediated by impaired insulin secretion (Peters & Samaan, 1969; Fariss & Lutcher, 1971). Plasma levels of phenytoin within the therapeutic range (10–20 μg/mL) do not alter glucoregulation (Callaghan et al., 1977). However, because alterations in the pharma-cokinetics of phenytoin might occur in patients with diabetes (Adithan et al., 1991), careful monitoring of drug levels and glycemic control is warranted in diabetic patients.

GROWTH HORMONE

Patients with acromegaly have an increased risk of develop-ing type 2 diabetes due to the metabolic eff ects of growth hormone (GH). During the past few decades, clinical indi-cations for treatment with recombinant human GH have expanded to include short stature from GH defi ciency in childhood, human immunodefi ciency virus (HIV) cachexia, and GH defi ciency in older adults, among others. The expanded uses of GH have been associated with reports of treatment-emergent diabetes and dysglycemia in both pedi-atric and elderly populations (Czernichow, 1993; Bramnert et al., 2003). Growth hormone dose-dependently stimulates lipolysis, increases free fatty acids, induces insulin resis-tance, and decreases glucose oxidation (Bratusch-Marrain et al., 1982; Salgin et al., 2009). The glucoregulatory


OAPN

62

system is exquisitely sensitive to the antagonistic eff ects of GH on insulin action (Holly et al., 1988). For example, the physiological early morning increases in plasma GH levels do decrease insulin sensitivity in healthy subjects (Rizza et al., 1982; Bratusch-Marrain et al., 1982), and also explain the dawn phenomenon of morning hyperglycemia in diabetic patients (Møller & Jørgensen, 2009).

Because of its appeal in popular culture (antiaging move-ment), possible surreptitious use, and imprecisions in the current methods for dosing GH, the scope of GH-induced hyperglycemia is likely to increase in future. The metabolic eff ects of GH predict that persons with prediabetes or other risk factors for diabetes (Table 1) would be particularly at risk for hyperglycemia following exposure to pharmacologi-cal doses of GH. It is prudent, therefore, to obtain informa-tion on diabetes risk factors and to establish baseline fasting plasma glucose levels before initiating GH therapy. Persons with diabetes or prediabetes should be monitored closely and managed appropriately for any glycemic escalation.

TOTAL PARENTERAL NUTRITION AND INPATIENT HYPERGLYCEMIA

Although not strictly a drug, total parenteral nutrition (TPN) is given almost routinely to patients with criti-cal illness and to other inpatients undergoing prolonged restriction of oral intake. Hyperglycemia is a frequent fi nding in patients receiving TPN, and has been associated with adverse outcomes in retrospective studies (Cheung et al., 2005; Lin et al., 2007). In one study, 457 patients receiving TPN were stratifi ed by mean plasma glucose levels into quartiles: quartile 1 (<114 mg/dL), quartile 2 (114 to 137


63

MEDICATIONS AND

DIABETES RISK

mg/dL), quartile 3 (137 to 180 mg/dL), and quartile 4 (>180 mg/dL) (Lin et al., 2007). Logistic regression analysis showed that the odds ratio of death increased progressively by two- to fi vefold from quartile 2 to quartile 4, compared to quartile 1. Each 10 mg/dL increase in mean blood glucose level was asso-ciated with increased risks for sepsis, cardiac complications, and renal failure (Lin et al., 2007). A similar pattern of com-plications, and an even greater odds ratio of death (tenfold for quartile 4 compared to quartile 1), was reported by Cheung et al. (2007) in their retrospective study of 111 critically ill patients receiving TPN who developed hyperglycemia.

MechanismsThe mechanisms underlying TPN-induced hyperglycemia are probably similar to the general mechanisms proposed for hyperglycemia in critically ill patients (stress of illness, glucocorticoids and other medications, prolonged immo-bilization, exacerbation of preexisting dysglycemia, etc.) (Dagogo-Jack & Alberti, 2002). Furthermore, intravenous (IV) delivery of nutients via TPN bypasses the intestinal incretin-secreting cells. Incretins, released by gut cells in response to oral food intake, augment insulin secretion and suppress glucagon. Without the help of incretins, IV nutrients fail to elicit adequate insulin secretion or glucagon suppres-sion, leading to exaggerated postprandial hyperglycemia. In addition, the TPN solution often contains varying concentra-tions of dextrose that can contribute to hyperglycemia. The further mechanisms linking TPN-induced hyperglycemia to increased mortality and morbidity are not known precisely, although infl ammatory cytokines and oxidative stress may play a role (Esposito et al., 2002). Clearly, observational studies cannot separate the eff ects of underlying illness from


OAPN

64

those of TPN-induced hyperglycemia. Therefore, a prospec-tive, randomized, controlled study (with an intervention arm that achieves normoglycemia in TPN-treated patients) will be required to confi rm causal associations between hy-perglycemia and increased mortality and morbidity.


Glycemic TargetsNo randomized controlled studies have been conducted specifi cally to determine optimal glycemic targets for patients receiving TPN. In the absence of such data, an empirical approach, based on extrapolations from studies on inpatient hyperglycemia is a viable option. A reasonable and evidence-based target for surgical patients in critical care units is a blood glucose level of 80–100 mg/dL in (van den Berghe et al., 2001, 2003, 2006). Care must be taken to avoid hypoglycemia, the risk of which is increased threefold among inpatients undergoing aggressive glycemic control. Currently, there is controversy regarding the optimal gly-cemic target for patients in the general medical or surgi-cal wards. The chances of achieving impeccable glycemic control similar to that achievable in the outpatient setting are obviated by numerous inpatient factors, including stress hyperglycemia, variable nutritional intake, and concomi-tant medications, among others. Therefore, a reasonable goal for patients on the general medical or surgical wards is a fasting glucose level of less than 126 mg/dL and random blood glucose values of less than 180–200 mg/dL (Wiener et al., 2008; ADA, 2009; Moghissi et al., 2009).

Choice of Antidiabetic AgentIn most cases, the underlying indications for suspension of oral intake and institution of TPN would also preclude the use of


65

MEDICATIONS AND

DIABETES RISK

oral hypoglycemic agents for managing TPN-induced hyperg-lycemia. Similarly, the prolonged action of subcutaneous insu-lin makes that a risky option in a patient whose sole caloric intake is dependent on TPN. Any technical problems that lead to interruption of TPN (e.g., loss of venous access) would expose such a patient to the risk of severe hypoglycemia from the continued absorption of insulin from subcutaneous sites. For these reasons, IV insulin infusion is the regimen of choice for controlling TPN-induced hyperglycemia, whenever ade-quate skilled nursing staff is available to monitor the patient. Regular, short-acting insulin should be used for IV infusion, since there is no advantage to using the more expensive rapid-acting analogs, which are designed for faster absorption from subcutaneous (SC) sites. The insulin can be administered via a separate infusion or mixed with the TPN solution. The separate infusion allows greater fl exibility but requires closer monitoring by skilled nurses (Knapke, 1989).

Where adequate nursing coverage is unavailable, multiple injections of SC insulin may be used to maintain desirable blood glucose targets. Because of their shorter duration of action, SC administration of rapid-acting insulin analogs may pose a lower risk of prolonged hypoglycemia in the event of TPN interruption, compared to regular insulin. Bedside blood glucose should be monitored frequently (Q 1–2 hr during IV insulin infusion and Q 3–4 hr for SC regimen). A written protocol (algorithm) for adjustment of insulin infu-sion rates in relation to ambient glucose levels should be cre-ated, to help direct nursing staff in the safe management of hyperglycemia. Numerous algorithms have been published, and many hospitals have developed their own local ones, but the best algorithm is one that is fl exible, individualized, and tailored to the changing needs and condition of the patients.


OAPN

66

Section 11: General Approach to Risk Reduction

The fi rst step toward reduction of the risk of iatrogenic hy-perglycemia is a thoughtful evaluation of the indications for any prescription, which should be based on compelling evidence. Whenever feasible, those medications of proven effi cacy on the primary condition that exhibit a low risk for dysglycemia should be given preference, particularly in at-risk or diabetic patients. One example is treatment of hypertension, where a drug selection policy that favors angiotensin-converting enzyme inhibitors, angiotensin re-ceptor blockers, or calcium channel blockers over thiaz-ides and �-blockers in high-risk prediabetic subjects can be defended. When a drug with an adverse glucose pro-fi le is unavoidable, use of the lowest eff ective dose for the minimum duration is a reasonable strategy for diabetes risk reduction. This concept is most applicable to the fi eld of transplant immunosuppression and other steroid-requiring conditions. However, minimization of dose and/or dura-tion of drug treatment is a reasonable strategy for diabetes risk reduction with regard to antidepressants, nicotinic acid, �-agonists, thiazides, �-blockers, and other medi-cations. In some instances, the potential adverse glycemic eff ects of a given drug class can be mitigated by choosing a compound with composite activity. For example, use of a combined �- and �-blocker often obviates the adverse metabolic eff ects of a pure �-blocker. Furthermore, in the management of dyslipidemia in persons with diabetes or prediabetes, inclusion of agents that enhance glucose toler-ance (e.g., fi brates and bile acid sequestrants) aff ords oppor-tunity for risk reduction on empirical grounds. Whenever the evidence and the clinical condition permit, switching from a suspect drug to another equally effi cacious agent


67

MEDICATIONS AND

DIABETES RISK

that is known to have a better metabolic profi le should be considered (Table 13). If diabetes develops during treat-ment with a medication that cannot be withdrawn or substituted, a multimodality approach to glycemic control is advocated. That approach incorporates self-monitoring of blood glucose, diabetes education, dietary counseling, physical activity, and selective use of antidiabetic medica-tions. The mnemonic, MEDEM (Monitoring, Education, Diet, Exercise, Medication) can be used to recall the key elements of diabetes care.

Table 13 Drug-induced Diabetes: Approach to Risk Reduction

Before drug exposureScreening and identifi cation of at-risk subjects ■

FH of diabetes• Overweight/obesity• History of prediabetes (impaired fasting glucose/impaired glucose • tolerance)Other (see Table 1)•

Documentation of baseline fasting blood glucose ■

During drug exposure Periodic monitoring of fasting blood glucose and symptoms ■

Dietary and exercise counseling ■

Minimization of dose and duration of therapy ■

Substitution or switching of diabetogenic medications ■

Future directions: Pharmacoprophylaxis?Insulin sensitizers (metformin, thiazolidinediones) ■

Acarbose ■

Orlistat ■

Colesevelam ■

Fibrates ■

Dipeptidyl peptidase (DPP)-IV inhibitors ■


OAPN

68

LIFESTYLE INTERVENTION FOR DIABETES PREVENTION

In persons who require long-term treatment with a poten-tially diabetogenic drug, consideration should be given to lifestyle counseling for prevention of diabetes. The Diabetes Prevention Project (DPP) and other studies have shown that lifestyle modifi cation in persons with prediabetes can reduce the risk of progression to diabetes by approximately 60% (DPP Research Group, 2002; Tuomilehto et al., 2001). Other studies have shown that lifestyle modifi cation pre-vents weight gain in patients with schizophrenia treated with atypical antipsychotic drugs (Menza et al., 2004; Hoff mann et al., 2005). The DPP lifestyle intervention utilized four key approaches that provide a general construct for counsel-ing patients at risk for drug-induced hyperglycemia. The key approaches are enumerated for easier reference:

1. Selection of persons at risk. The DPP utilized the risk-factor approach (Table 1) to identify persons who may benefi t from intervention. Specifi cally targeted were persons with a family history of type 2 diabetes who have a high BMI (>24 kg/m2) and fasting plasma glucose in the range of 96–125 mg/dL. The BMI cut-off was lowered to greater than 22 kg/m2 for Asians.

2. Exercise intervention. The DPP exercise goal was 30 minutes of moderate-intensity aerobic activity (equivalent to brisk walking) for 5 days each week (total 150 min/week).

3. Dietary intervention. The participants were coached by dietitians to decrease fat calories to less than 30% and total calories by 500–700 kcals/day by selectively reducing the intake of saturated fats and excessive carbohydrates.


69

MEDICATIONS AND

DIABETES RISK

4. Self-monitoring. The participants recorded their food in-take and minutes spent performing physical activity each day. Self-monitoring behavior has been shown to predict long-term maintenance of weight loss, and could well have been an important adjunct to the excellent results obtained in the DPP.

MEDICATIONS FOR PREVENTION OF DRUG-INDUCED DIABETES

In the DPP, prediabetic subjects assigned to metformin treatment experienced a 31% reduction in the rate of pro-gression to diabetes (DPP Research Group, 2002). Acarbose and orlistat also have been reported to signifi cantly reduce progression to diabetes, but, like metformin, the effi cacy of these medications was weaker than that of lifestyle inter-vention. In contrast, thiazolidinediones (TZDs) have been reported to reduce progression by more than 60%, which approximates to the eff ect of lifestyle modifi cation alone (DREAM Investigators, 2006). Also, as already discussed, fi brates have been reported to decrease diabetes risk, and colesevelam improves glycemic control in diabetic patients. However, none of these medications has been approved for prevention of diabetes.

The question often arises as to whether high-risk patients receiving chronic treatment with potentially diabetogenic agents (e.g., steroids) should receive prophylactic therapy with medications that have been shown to prevent diabetes. Unfortunately, there is limited specifi c information from randomized controlled trials to inform such a practice. Clearly, randomized controlled studies are needed to demon-strate whether diabetes “pharmacoprophylaxis” is a rational


OAPN

70

strategy in high-risk persons receiving treatment with po-tentially diabetogenic drugs. Obvious candidates for such a trial would be metformin, TZDs, acarbose, fi brates, or colesevelam (all of which have pertinent preliminary data), but the newer dipeptidylpeptidase (DPP)-IV inhibitors (sita-gliptin and saxagliptin) would also be attractive candidates because of their low risk for iatrogenic hypoglycemia.

Although no drug has been approved for diabetes preven-tion, in subjects with documented prediabetes (impaired fasting glucose or impaired glucose tolerance), an Expert Panel of the American Diabetes Association (ADA) has rec-ommended that treatment with metformin be considered as an adjunct to diet and exercise for the prevention of type 2 diabetes (Nathan et al., 2007). The ADA recommendation was directed at the general population, but there is no rea-son why it should not apply to prediabetic patients at risk for drug-induced diabetes, although its effi cacy in preventing drug-induced diabetes has not been tested specifi cally.

REFERENCES

Abraham MR, Khardori R. (1999). Hyperglycemic hyperosmolar nonketotic syndrome as initial presentation of type 2 diabetes in a young cocaine abuser. Diabetes Care, 22:1380–1381.

Adithan C, Srinivas B, Indhiresan J, et al. (1991). Infl uence of type I and type II diabetes mellitus on phenytoin steady-state levels. Int J Clin Pharmacol Ther Toxicol, 29:310–313.

Alderman MH. (1992). Which antihypertensive agent fi rst—and why! JAMA, 267:2786–2787.

Ambrose PG, Bhavnani SM, Cirincione BB, et al. (2003). Gatifl oxacin and the elderly: pharmacokinetic-pharmacodynamic rationale for a potential age-related dose reduction. J Antimicrob Chemother, S2:434–444.


71

MEDICATIONS AND

DIABETES RISK

Ames RP. (1986). The effects of antihypertensive drugs on serum lipids and lipoproteins II. Non-diuretic drugs. Drugs, 32:335–357.

American Diabetes Association. (2004). Consensus Development Conference on antipsychotic drugs and obesity and diabetes. Diabetes Care, 27:596–601.

American Diabetes Association. (2009). Standards of Medical Care in Diabetes—2009. Diabetes Care, 32(Suppl 1):S13–S61.

Andersohn F, Schade R, Suissa S, Garbe E. (2009). Long-term use of antidepressants for depressive disorders and the risk of dia-betes mellitus. Am J Psychiatry, 166:591–598.

Andersson DE, Rojdmark S. (1981). Improvement of glucose tol-erance by verapamil in patients with non-insulin-dependent diabetes mellitus. Acta Med Scand, 210:27–33.

Andrews RC, Herlihy O, Livingstone DE, et al. (2002). Abnormal cortisol metabolism and tissue sensitivity to cortisol in patients with glucose intolerance. J Clin Endocriol Metab, 87:5587–5593.

Aronne LJ, Segal KR. (2003). Weight gain in the treatment of mood disorders. J Clin Psychiatry, 64:22–29.

Aspinall SL, Good CB, Jiang R, et al. (2009). Severe dysglyce-mia with the fl uoroquinolones: a class eff ect? Clin Infect Dis,

49:402–408. Bakris GL, Fonseca V, Katholi RE, et al. (2004). Metabolic eff ects

of carvedilol vs. metoprolol in patients with type 2 diabetes mellitus and hypertension: a randomized controlled trial. JAMA, 292:2227–2236.

Balasse EO, Neef MA. (1973). Infl uence of nicotinic acid on the rates of turnover and oxidation of plasma glucose in man. Metab

Clin Exp, 22:1193–1204.Barzilay JI, Davis BR, Cutler JA, et al. (2006). Fasting glucose

levels and incident diabetes mellitus in older nondiabetic adults randomized to receive 3 diff erent classes of


OAPN

72

antihypertensive treatment: a report from the Antihyper-tensive and Lipid- Lowering Treatment to Prevent Heart Attack Trial (ALLHAT). Arch Intern Med, 166:2191–2201.

Bays HE, Goldberg RB, Truitt KE, Jones MR. (2008). Colesevelam hydrochloride therapy in patients with type 2 diabetes mellitus treated with metformin: glucose and lipid eff ects. Arch Intern

Med, 168:1975–1983. Bendtzen K, Mandrup-Poulsen T, Nerup J, et al. (1986). Cytotoxicity

of human pI 7 interleukin-1 for pancreatic islets of Langerhans. Science, 232:1545–1547.

Bhatnagar SK, Amin MM, Al-Yusuf AR. (1984). Diabetogenic eff ects of nifedipine. Br Med J (Clin Res Ed), 289:19.

Biggee BA, Blinn CM, Nuite M, et al. (2007). Eff ects of oral glucosamine sulphate on serum glucose and insulin during an oral glucose tolerance test of subjects with osteoarthritis. Ann Rheum Dis, 66:260–262

Bonato V, Cataldo D, Dotta F, Carmellini M. (2009). Diagnosis and approach to posttransplant diabetes. Curr Diab Rep, 9:317–323.

Bosch J, Yusuf S, Gerstein HC, et al. (2006). Eff ect of ramipril on the incidence of diabetes. N Engl J Med, 355:1551–1562.

Bouchard PH, Saï P, Reach G, et al. (1982). Diabetes mellitus following pentamidine-induced hypoglycemia in humans. Diabetes, 31:40–45.

Bramnert M, Segerlantz M, Laurila E, et al. (2003). Growth horm-one replacement therapy induces insulin resistance by activating the glucose-fatty acid cycle. J Clin Endocrinol Metab, 88:1455–1463.

Bratusch-Marrain PR, Smith D, DeFronzo RA. (1982). The eff ect of growth hormone on glucose metabolism and insulin secre-tion in man. J Clin Endocrinol Metab, 55:973–982.

Brauchli YB, Jick SS, Meier CR. (2008). Psoriasis and the risk of incident diabetes mellitus: a population-based study. Br J

Dermatol, 159:1331–1337.


73

MEDICATIONS AND

DIABETES RISK

Brown LC, Majumdar SR, Johnson JA. (2008). Type of antidepre-ssant therapy and risk of type 2 diabetes in people with depres-sion. Diabetes Res Clin Pract, 79:61–67.

Brown TT, Cole SR, Li X, et al. (2005). Antiretroviral therapy and the prevalence and incidence of diabetes mellitus in the multi-center AIDS cohort study. Arch Intern Med 165:1179–1184.

Callaghan N, Feely M, O’Callaghan M, et al. (1977). The eff ects of toxic and non-toxic serum phenytoin levels on carbo-hydrate tolerance and insulin levels. Acta Neurol Scand, 56:563–571.

Catero M. (2007). Dysglycemia and fl uoroquinolones: are you put-ting patients at risk? J Fam Pract, 56:101–107.

Cavallo-Perin P, Bruno A, Bozzo C, et al. (1988). Insulin binding to monocytes and in vivo peripheral insulin sensitivity are nor-mal in Graves’ disease. J Endocrinol Invest, 11:795–800.

Chadban S. (2008). New-onset diabetes after transplantation—should it be a factor in choosing an immunosuppressant regi-men for kidney transplant recipients. Nephrol Dial Transplant,

23:1816–1818.Chan JC, Cockram CS. (1991). Drug-induced disturbances of

carbohydrate metabolism. Adverse Drug React Toxicol Rev, 10:1–29.

Chan JC, Cockram CS, Critchley JA. (1996). Drug-induced dis-orders of glucose metabolism: mechanisms and management. Drug Safety, 15:135–157.

Chapman JR, O’Connell PJ, Nankivell BJ. (2005). Chronic renal allograft dysfunction. J Am Soc Nephrol, 16: 3015–3026

Chedin P, Cahen-Varsaux J, Boyer N. (1996). Non-insulin depen-dent diabetes mellitus development during interferon � ther-apy for chronic viral hepatitis. Ann Intern Med, 125:521.

Cheung D, Bryer-Ash M. (2009). Persistent hypoglycemia in a patient with diabetes taking etanercept for the treatment of psoriasis. J Am Acad Dermatol, 60:1032–1036.


OAPN

74

Cheung NW, Napier B, Zaccaria C, Fletcher JP. (2005). Hyper-glycemia is associated with adverse outcomes in patients receiving total parenteral nutrition. Diabetes Care 28:2367–2371.

Chiueh CC, Kopin IJ. (1978). Centrally mediated release by cocaine of endogenous epinephrine and norepinephrine from the sympathoadrenal medullary system of unanesthetized rats. J Pharmacol Exp Ther, 205:148–154.

Coblio NA, Mowrey K, McCright P, et al. (2004). Use of a data warehouse to examine the eff ect of fl uoroquinolones on glu-cose metabolism. Am J Health Syst Pharm, 61:2545–2548.

Cohen AD, Sherf M, Vidavsky L, et al. (2008). Association between psoriasis and the metabolic syndrome. A cross-sectional study. Dermatology, 216:152–155.

Comi RJ. Drug-induced diabetes mellitus (2000). In: LeRoith, Taylor, Olefsky, eds. Diabetes Mellitus, 2nd ed. (pp. 582–588) Philadelphia: Lippincott Williams & Wilkins

Cosio FG, Pesavento TE, Kim S, et al. (2002). Patient survival after renal transplantation: IV. Impact of post-transplant dia-betes. Kidney Int, 62:1440–1446.

Cotton EK, Fahlberg YI. (1964). Hypoglycemia with salicylate poisoning. Am J Dis Child, 108:171–173.

Coyle P, Carr AD, Depczynski BB, Chisholm DJ. (1996). Diabetes mellitus associated with pentamidine use in HIV-infected patients. Med J Aust, 165:587–588.

Crandall JP, Polsky S, Howard AA, et al. (2009). Alcohol con-sumption and diabetes risk in the Diabetes Prevention Program. Am J Clin Nutr, 90:595–601.

Czernichow P. (1993). Growth hormone administration and car-bohydrate metabolism. Horm Res, 39:102–103.

Dagogo-Jack S. (2008a). HIV therapy and diabetes risk. Diabetes

Care, 31:1267–1268.Dagogo-Jack S. (2008b). Lifestyle intervention to reduce meta-

bolic and cardiovascular risks. In: Fonseca V, ed., Therapeutic


75

MEDICATIONS AND

DIABETES RISK

Strategies in Metabolic Syndrome (pp. 1–19). Oxford, UK: Medical Publishing Ltd.

Dagogo-Jack S. (2006). New drugs and diabetes risk: antipsychotic and antiretroviral agents. In: Fonseca V, ed., Clinical Diabetes (pp. 569–581). Philadelphia: Saunders.

Dagogo-Jack S. (1991). Peptide regulatory factors: physiological and clinical implications. Niger Med J, 21:144–149.

Dagogo-Jack S. (2009). The role of antipsychotic agents in the development of diabetes. Nat Clin Pract Endocrinol Metab, 5:22–23.

Dagogo-Jack S, Alberti KGMM. (2002). Diabetes mellitus in sur-gical patients. Diabetes Spectrum, 15:44–48.

Dagogo-Jack S, Santiago JV. (1997). Pathophysiology of type 2 diabetes and modes of action of therapeutic interventions. Arch

Intern Med,157:1802–1817.Dagogo-Jack S, Selke G, Melson AK, Newcomer JW. (1997).

Robust leptin secretory responses to dexamethasone in obese subjects. J Clin Endocrinol Metab, 82:3230–3233.

Dahlof B, Sever PS, Poulter NR, et al. (2005). Prevention of cardio-vascular events with an antihypertensive regimen of amlodipine adding perindopril as required versus atenolol adding bendro-fl umethiazide as required, in the Anglo-Scandinavian Cardiac Outcomes Trial-Blood Pressure Lowering Arm (ASCOT-BPLA): a multicentre randomised controlled trial. Lancet, 366:895–906.

Davis TM, Yeap B, Bruce DG, Davis WA. (2007). Lipid-lowering therapy protects against peripheral sensory neuropathy in type 2 diabetes. American Diabetes Association 2007 Scientifi c Sessions. Abstract 0004–OR. Chicago, IL.

de Groot M, Anderson R, Freedland KE, et al. (2001). Association of depression and diabetes complications: a meta-analysis. Psychosom Med, 63:619–630.

De Wit S, Sabin CA, Weber R, et al. (2008). Incidence and risk factors for new onset diabetes mellitus in HIV infected patients: the D:A:D study. Diabetes Care, 31:1224–1229.


OAPN

76

Diabetes Prevention Program Research Group. (2005). Depression symptoms and antidepressant medicine use in Diabetes Prevention Program participants. Diabetes Care, 28:830–837.

Diabetes Prevention Program Research Group. (2008). Elevated depression symptoms, antidepressant medicine use and risk of developing diabetes during the Diabetes Prevention Program. Diabetes Care 31:420–426.

Diabetes Prevention Program Research Group. (2002). Reduction in the incidence of type 2 diabetes with lifestyle intervention or metformin. N Engl J Med 346:393–403.

Dixon L, Weiden PJ, Delahanty J, et al. (2000). Prevalence and correlates of diabetes in national schizophrenia samples. Schizophr Bull, 26:903–912.

DREAM (Diabetes Reduction Assessment with ramipril and rosiglitazone Medication) Trial Investigators. (2006). Eff ect of rosiglitazone on the frequency of diabetes in patients with impaired glucose tolerance or impaired fasting glucose: a ran-domised controlled trial. Lancet, 368:1096–1105.

Dube MP, Edmondson-Melancon H, Qian D, et al. (2001). Prospective evaluation of the eff ect of initiating indinavir-based therapy on insulin sensitivity and B-cell function in HIV-infected patients. J Acquir Immune Defi c Syndr, 27:130–134.

Dumortier J, Bernard S, Bouff ard Y, et al. (2006). Conversion from tacrolimus to cyclosporine in liver transplanted patients with diabetes mellitus. Liver Transpl, 12: 659–664

Eaton WW, Armenian H, Gallo J, et al. (1996). Depression and risk for onset of type II diabetes. A prospective population-based study. Diabetes Care, 19:1097–1102.

Egede LE, Zheng D. (2003). Independent factors associated with major depressive disorder in a national sample. Diabetes Care,

26:104–111.


77

MEDICATIONS AND

DIABETES RISK

Elliott WJ, Meyer PM. (2007). Incident diabetes in clinical tri-als of antihypertensive drugs: a network meta-analysis. Lancet,

369:201–207.Esposito K, Nappo F, Marfella R, et al. (2002). Infl ammatory

cytokine concentrations are acutely increased by hyper-glycemia in humans: role of oxidative stress. Circulation, 106:2067–2072.

Fabris P, Betterle C, Greggio NA, et al. (1998). Insulin-dependent diabetes mellitus during � interferon therapy for chronic viral hepatitis. J Hepatol, 28:514–517.

Facchini FS, Hollenbeck CB, Jeppesen J, et al. (1992). Insulin resistance and cigarette smoking. Lancet, 339:1128–1130.

Fajans SS, Conn JW (1954). An approach to the prediction of dia-betes mellitus by modifi cation of the glucose tolerance test with cortisone. Diabetes, 3:296–302.

Fallo F, Veglio F, Bertello C, et al. (2006). Prevalence and charac-teristics of the metabolic syndrome in primary aldosteronism. J Clin Endocrinol Metab, 91:454–459.

Fariss BL, Lutcher CL. (1971). Diphenylhydantoin-induced hyperglycemia and impaired insulin release. Eff ect of dosage. Diabetes, 20:177–181.

Fava M. (2000). Weight gain and antidepressants. J Clin Psychiatry,

61(Suppl 11):37–41.Fernandez-Egea E, Bernardo M, Parellada E, et al. (2008). Glucose

abnormalities in the siblings of people with schizophrenia. Schizophr Res, 103:110–113.

Ferrannini E, Gastaldelli A, Miyazaki Y, et al. (2003). Predomi-nant role of reduced �-cell sensitivity to glucose over insulin resistance in impaired glucose tolerance. Diabetologia, 46:1211–1219.

Fitzgerald DB, Young IS. (1984). Reversal of pentamidine-induced hypoglycaemia with oral diazoxide. J Trop Med Hyg, 87:15–19.


OAPN

78

Flamenbaum W. (1983). Metabolic consequences of antihyperten-sive therapy. Ann Intern Med, 98: 875–880.

Fleischman A, Johnsen S, Systrom DM, et al. (2007). Eff ects of a nucleoside reverse transcriptase inhibitor, stavudine, on glu-cose disposal and mitochondrial function in muscle of healthy adults. Am J Physiol Endocrinol Metab, 292:E1666–1673.

Flory JH, Ellenberg S, Szapary PO, et al. (2009). Antidiabetic action of bezafi brate in a large observational database. Diabetes

Care, 32;547–551.Foulis AK, Farquharson MA, Meager A. (1987). Immunoreactive

�-interferon in insulin secreting betacells in type 1 diabetes mellitus. Lancet, 2:1423–1427.

Freedman R. (2003). Schizophrenia. N Eng J Med, 349:1738–1749.Fulcher GR, Jones IR, Alberti KG. (1988). Improvement in glucose

tolerance by reduction of lipid concentrations in non-insulin dependent diabetes mellitus. Diabetes News, 9:4–8.

Garg A, Grundy SM. (1990). Nicotinic acid as therapy for dys-lipidemia in non-insulin dependent diabetes mellitus. JAMA,

264:723–726.Gaspard UJ. (1987). Metabolic eff ects of oral contraceptives. Am J

Obstet Gynecol, 157:1029–1041.Ghisdal L, Bouchta NB, Broeders N, et al. (2007). Conversion

from tacrolimus to cyclosporine a for new-onset diabetes after transplantation: a single centre experience in renal transplanted patients and review of the literature. Transpl Int,

21:146–151.Giaccari A, Morviducci L, Zoretta D, et al. (1995). In vivo eff ects

of glucosamine on insulin secretion and insulin sensitivity in rat: possible relevance to the maladaptive responses to chronic hyperglycemia. Diabetalogia, 38:518–524.

Gilgore SG. (1960). The infl uence of salicylate on hyperglycemia. Diabetes, 9:392–393.


79

MEDICATIONS AND

DIABETES RISK

Giordanino C, Bugianesi E, Smedile A, et al. (2008). Incidence of type 2 diabetes mellitus and glucose abnormalities in patients with chronic hepatitis C infection by response to treatment: results of a cohort study. Am J Gastroenterol,

103:2481–2487.Giordano M, Matsuda M, Sanders L, et al. (1995). Eff ects of angio-

tensin-converting enzyme inhibitors, Ca2+ channel antagonists, and �-adrenergic blockers on glucose and lipid metabolism in NIDDM patients with hypertension. Diabetes, 44:665–671.

Giugliano D. (1984). Morphine, opioid peptides, and pancreatic islet function. Diabetes Care, 7:92–98.

Godsland IF, Crook D, Simpson R, et al. (1990). The eff ects of diff erent formulations of oral contraceptive agents on lipid and carbohydrate metabolism. N Engl J Med, 323:1375–1381.

Granner DK, O’Brien RM. (1992). Molecular physiology and genetics of NIDDM. Diabetes Care, 15:369–388.

Gress TW, Nieto FJ, Shahar E, et al. (2000). Hypertension and antihypertensive therapy as risk factors for type 2 diabetes mel-litus. N Engl J Med, 342:905–912.

Gribble FM, Davis TME, Higham CE, et al. (2000). The antima-larial agent mefl oquine inhibits ATP-sensitive K-channels. Br J

Pharmacol, 131:756–760.Grundy SM, Vega GL, McGovern ME, et al. (2002). Effi cacy,

safety and tolerability of once-daily Niacin for treatment of dyslipidemia associated with type 2 diabetes. Arch Intern Med,

162:1568–1573.Guillaume-Gentil C, Assimacopoulos-Jeannet F, Jeanrenaud B.

(1993). Involvement of non-esterifi ed fatty acid oxidation in glucocorticoid-induced peripheral insulin resistance in vivo in rats. Diabetologia, 36:899–906.

Guldstrand M, Grill V, Bjorklund A, Lins PE, Adamson U. (2002). Improved B-cell function after short-term treatment with


OAPN

80

diazoxide in obese subjects with type 2 diabetes. Diabetes Metab,

28:448–456.Gulliford MC, Charlton J, Latinovic R. (2005). Trends in antihy-

pertensive and lipid lowering therapy in subjects with type II diabetes: clinical eff ectiveness or clinical discretion? J Human

Hypertens, 19:111–117.Haataja L, Gurlo T, Huang CJ, Butler PC. (2008). Islet amyloid in

type 2 diabetes, and the toxic oligomer hypothesis. Endocr Rev,

29:303–316 Hall KW, Dobson KE, Dalton JG, et al. (1984). Metabolic abnor-

malities associated with intentional theophylline overdose. Ann

Intern Med, 101:457–462.Hansen JB, Arkhammer PO, Bodvarsdottir TB, Wahl P. (2004).

Inhibition of insulin secretion as a new drug target in the treat-ment of metabolic disorders. Curr Med Chem 11:1595–1615.

Halter JB, Beard JC, Porte D Jr. (1984). Islet function and stress hyperglycemia: plasma glucose and epinephrine interaction. Am J Physiol Endocrinol Metab, 247:E47–E52.

Heazell AE, Shina A, Bhatti NR. (2005). A case of gestational diabetes arising following treatment with glucocorticoids for pemphigoid gestations. J Maternal-Fetal Neonatal Med, 18:353–355.

Heesch C, Negus B, Keff er J, Snyder R, Risser R, Eichhorn E. (1995). Eff ects of cocaine on cortisol secretion in humans. Am J

Med Sci, 310:61–64.

Heizer T, Yla-Herttuala S, Luoma J, et al. (1996). Cigarette smok-ing potentiates endothelial dysfunction of forearm resistance vessels in patients with hypercholesterolemia: role of oxidized LDL. Circulation, 9:1346–1353

Henkin Y, Oberman A, Hurst DC, Segrest JP. (1991). Niacin revisited: clinical observation on an important but underuti-lized drug. Am J Med, 91:239–246.


81

MEDICATIONS AND

DIABETES RISK

Hertel J, Struthers H, Horj CB, Hruz PW. (2004). A structural basis for the acute eff ects of HIV protease inhibitors on GLUT4 intrinsic activity. J Biol Chem, 279:55147–55152.

Hill AB. (1965). The environment and disease: association or cau-sation? Proc R Soc Med, 58:295–300.

Hirsch IB, Boyle PJ, Craft S, Cryer PE. (1991). Higher glycemic thresholds for symptoms during �-adrenergic blockade in IDDM. Diabetes, 40:1177–1186.

Hjelmesaeth J, Hartmann A, Leivestad T, et al. (2006). The impact of early diagnosed new-onset post-transplantation dia-betes mellitus on survival and major cardiac events. Kidney Int,

69:588–595.Hjelmesaeth J, Midtvedt K, Jenssen T, et al. (2001). Insulin

resistance after renal transplantation: impact of immuno-suppressive and antihypertensive therapy. Diabetes Care, 24:2121–2126.

Hoff mann VP, Ahl J, Meyers A, et al. (2005). Wellness interven-tion for patients with serious and persistent mental illness. J Clin Psychiatry, 66:1576–1579.

Holly JMP, Amiel SA, Sandhu RR, et al. (1988). The role of growth hormone in diabetes mellitus. J Endocrinol, 118: 353–364.

Huang JF, Yu ML, Dai CY, et al. (2008). Reappraisal of the charac-teristics of glucose abnormalities in patients with chronic hepa-titis C infection. Am J Gastroenterol, 103:1933–1940.

Huang Q, Bu S, Yu Y, et al. (2007). Diazoxide prevents diabe-tes through inhibiting pancreatic B-cells from apoptosis via Bcl-2/Bax rate and p38–B mitogen-activated protein kinase. Endocrinology, 148:81–91.

Hundal RS, Petersen KF, Mayerson AB, et al. (2002). Mechanism by which high-dose aspirin improves glucose metabolism in type 2 diabetes. J Clin Invest, 109: 1321–1326.


OAPN

82

International Expert Committee Report on the Role of the A1C Assay in the Diagnosis of Diabetes. (2009). Diabetes Care, 32:1327–1334.

Ito T, Otsuki M, Itoi T, et al. (2007). Research Committee of Intractable Diseases of the Pancreas. Pancreatic diabetes in a follow-up survey of chronic pancreatitis in Japan. J Gastroenterol,

42:291–297.Jandeleit-Dahm KA, Tikellis C, Reid CM, et al. (2005). Why

blockade of the renin-angiotensin system reduces the incidence of new-onset diabetes. J Hypertens, 23:463–473.

Johnson DD, Palumbo PJ, Chu CP. (1980). Diabetic ketoacidosis in a community-based population. Mayo Clin Proc, 55:83–88.

Karlander SG, Khan A, Wajngot A, et al. (1989). Glucose turnover in hyperthyroid patients with normal glucose tolerance. J Clin

Endocrinol Metab, 68:780–786.Kasiske B, Snyder JJ, Gilbertson D, et al. (2003). Diabetes mel-

litus after kidney transplantation in the United States. Am J

Transplant, 3:178–185.Katon WJ, Rutter C, Simon G, et al. (2005). The association of

comorbid depression with mortality in patients with type 2 diabetes. Diabetes Care, 28:2668–2672.

Keck PE, Jr., Buse JB, Dagogo-Jack S, et al. (2003). Managing metabolic concerns in patients with severe mental illness: a special report. Postgrad Med, Minneapolis, MN, McGraw-Hill, pp 1–92.

Keech AC, Mitchell P, Summanen PA, et al. (2007). FIELD study investigators. Eff ect of fenofi brate on the need for laser treat-ment for diabetic retinopathy (FIELD study): a randomised controlled trial. Lancet, 370:1687–1697.

Keskiner A, el Toumi A, Bousquet T. (1973). Psychotropic drugs, diabetes and chronic mental patients. Psychosomatics, 14:176–181.


83

MEDICATIONS AND

DIABETES RISK

Kiortsis DN, Mavridis AK, Vasakos S, et al. (2005). Eff ects of inf-liximab treatment on insulin resistance in patients with rheu-matoid arthritis and ankylosing spondylitis. Ann Rheum Dis,

64:765–766.Kirsch DM, Baumgarten M, Deufel T, et al. (1983). Catecholamine-

induced insulin resistance of glucose transport in isolated rat adipocytes. Biochem J, 216:737–745.

Knapke CM, Owens JP, Mirtallo JM. (1989). Management of glu-cose abnormalities in patients receiving total parenteral nutri-tion. Clin Pharm, 8:136–144.

Knol MJ, Twisk JWR, Beekman ATF, et al. (2006). Depression as a risk factor for the onset of type 2 diabetes: a meta-analysis. Diabetologia, 49:837–845.

Kong C, Nimmo L, Elatrozy T, et al. (2001). Smoking is associ-ated with increased hepatic lipase activity, insulin resistance, dyslipidaemia and early atherosclerosis in type 2 diabetes. Atherosclerosis, 156:373–378.

Konrad T, Zeuzem S, Vicini P, et al. (2000). Evaluation of factors controlling glucose tolerance in patients with HCV infection before and after 4 months therapy with interferon-�. Eur J Clin

Invest, 30:111–121.Koster JC, Remedi MS, Qui H, et al. (2003). HIV Protease inhibi-

tors acutely impair glucose-stimulated insulin release. Diabetes,

52:1695–1700.Lam SK, Owen A. (2007). Incident diabetes in clinical trials of

antihypertensive drugs. Lancet, 369:1513–1514. Larsen CM, Faulenbach M, Vaag A, et al. (2007). Interleukin-

1–receptor antagonist in type 2 diabetes mellitus. N Engl J Med,

356:1517–1526.Larsen CM, Faulenbach M, Vaag A, et al. (2009). Sustained eff ects

of interleukin-1 receptor antagonist treatment in type 2 diabe-tes. Diabetes Care, 32:1663–1668.


OAPN

84

Lederballe Pedersen O. (1977). Long-term experience with minoxidil in combination treatment of severe arterial hyper-tension. Acta Cardiol, 32:283–293.

LeWitt PA. (1980). The neurotoxicity of the rat poison vacor. A clinical study of 12 cases. N Engl J Med, 302:73–77.

Lieberman JA, Stroup TS, McEvoy JP, et al. (2005). Clinical Antipsychotic Trials of Intervention Eff ectiveness (CATIE) Investigators. Eff ectiveness of antipsychotic drugs in patients with chronic schizophrenia. N Engl J Med, 353:1209–1223.

Liegl U, Bogner JR, Goebel FD. (1994). Insulin-dependent dia-betes mellitus following pentamidine therapy in a patient with AIDS. Clin Invest, 72:1027–1029.

Lin LY, Lin HC, Lee PC, et al. (2007). Hyperglycemia correlates with outcomes in patients receiving total parenteral nutrition. Am J Med Sci, 333:261–265.

Lithell HOL. (1991). Eff ect of antihypertensive drugs on insulin, glucose, and lipid metabolism. Diabetes Care, 14:203–209.

Lustman PJ, Freedland KE, Griffi th LS, Clouse RE. (2000). Fluoxetine for depression in diabetes: a randomized, double-blind, placebo-controlled trial. Diabetes Care, 23:618–623.

Lustman PJ, Griffi th LS, Clouse RE, et al. (1997). Eff ects of nor-triptyline on depression and glycemic control in diabetes: results of a double-blind, placebo controlled trial. Psychosom

Med, 59:241–250.Maedler K, Sergeev P, Ris F, et al. (2002). Glucose-induced � cell

production of IL-1beta contributes to glucotoxicity in human pancreatic islets. J Clin Invest, 110:851–860.

Maheux P, Ducros F, Bourque J, et al. (1997). Fluoxetine improves insulin sensitivity in obese patients with non-insulin-depen-dent diabetes mellitus independent of weight loss. Int J Obes

Relat Metab Disord, 21:97–102.Martin BC, Warram JH, Krolewski AS, et al. (1992). Role of

glucose and insulin resistance in the development of type 2


85

MEDICATIONS AND

DIABETES RISK

diabetes mellitus: results of a 25-year follow-up study. Lancet,

340:925–929. Martinez E, Conget I, Lazano L, et al. (1999). Reversion of meta-

bolic abnormalities after switching from HIV-1 protease inhib-itors to nevirapine. AIDS, 13:805–810.

Martínez-Abundis E, Reynoso-von Drateln C, Hernández-Salazar E, González-Ortiz M. (2007). Eff ect of etanercept on insulin secretion and insulin sensitivity in a randomized trial with pso-riatic patients at risk for developing type 2 diabetes mellitus. Arch Dermatol Res, 299:461–465.

Mastrandrea L, Yu J, Behrens T, et al. (2009). Etanercept treat-ment in children with new-onset type 1 diabetes: pilot random-ized, placebo-controlled, double-blind study. Diabetes Care,

32:1244–1249.Mayer S, Moran NC, Fain J. (1961). The eff ects of adrenergic

blocking agents on some metabolic actions of catecholamines. J Pharmacol Exp Therapeutics, 134:18–27.

McCullen MK, Ahmed I. (2007). Drug-induced hyperglycemia and diabetes mellitus. In: Goldstein BJ, Müller-Wieland, D, eds., Type 2 diabetes, principles and practice, 2nd ed. (pp. 513–528). New York: Taylor & Francis.

Meduna LJ, Gerty FJ, Urse VG. (1942). Biochemical disturbances in mental disorders. Arch Gen Psychiatry, 47:38–52.

Mehta SH, Moore RD, Thomas DL, et al. (2003).The eff ect of HAART and HCV infection on the development of hypergly-cemia among HIV-infected persons. J Acquir Immune Defi c Syndr,

33:577–584.Menza M, Vreeland B, Minsky S, et al. (2004). Managing atypical

antipsychotic-associated weight gain: 12-month data on a mul-timodal weight control program. J Clin Psychiatry, 65:471–477.

Miller LV, Stokes JD, Silpipat C. (1978). Diabetes mellitus and autonomic dysfunction after vacor rodenticide ingestion. Diabetes Care, 1:73–76.


OAPN

86

Miyatake N, Wada J, Kawasaki Y, et al. (2006). Relationship between metabolic syndrome and cigarette smoking in the Japanese population. Intern Med, 45:1039–1043.

Moghissi ES, Korytkowski MT, DiNardo M, et al. (2009). American Association of Clinical Endocrinologists and American Diabetes Association consensus statement on inpa-tient glycemic control. Diabetes Care, 32:1119–1131.

Mohan RJ, Mohan R. (1997). Drug induced diabetes mellitus. JAPI

45:876–878.Mohr JF, McKinnon PS, Peymann PJ, et al. (2005). A retrospec-

tive, comparative evaluation of dysglycemias in hospitalized patients receiving gatifl oxacin, levofl oxacin, ciprofl oxacin, or ceftriaxone. Pharmacother, 25:1303–1309.

Møller N, Jørgensen JO. (2009). Eff ects of growth hormone on glucose, lipid, and protein metabolism in human subjects. Endocr Rev, 30:152–177.

Monauni T, Zenti MG, Cretti A, et al. (2000). Eff ects of glu-cosamine infusion on insulin secretion and insulin action in humans. Diabetes, 49:926–935.

Moneva M., Dagogo-Jack S. (2002). Multiple drug targets in the management of type 2 diabetes mellitus. Current Drug Targets,

3:203–221.Mozaff arian D, Kamineni A, Carnethon M, et al. (2009). Lifestyle

risk factors and new-onset diabetes mellitus in older adults: the cardiovascular health study. Arch Intern Med, 169:798–807.

Nader S, Riad-Gabriel MG, Saad MF. (1997). The eff ect of a des-ogestrel-containing oral contraceptive on glucose tolerance and leptin concentrations in hyperandrogenic women. J Clin

Endocrinol Metab, 82:3074–3077.Nathan DM, Davidson MB, DeFronzo RA, et al. (2007). Impaired

fasting glucose and impaired glucose tolerance: implications for care. Diabetes Care, 30:753–759.


87

MEDICATIONS AND

DIABETES RISK

Nyenwe EA, Loganathan RS, Blum S, et al. (2007). Active use of cocaine: an independent risk factor for recurrent diabetic ketoacidosis in a city hospital. Endocr Pract, 13:22–29.

Obayashi H, Hasegawa G, Fukui M, et al. (2000). Tumor necrosis factor microsatellite polymorphism infl uences the develop-ment of insulin dependency in adult-onset diabetes patients with the DRB1*1502–DQB1*0601 allele and anti-glutamic acid decarboxylase antibodies. J Clin Endocrinol Metab,

85:3348–3351.O’Byrne S, Feely J. (1990). Eff ects of drugs on glucose toler-

ance in non-insulin dependent diabetics (Part II). Drugs, 40:6–18.

Oetjen E, Baun D, Beimesche S, et al. (2003). Inhibition of human insulin gene transcription by the immunosuppressive drugs cyclosporine and tacrolimus in primary, mature islets of trans-genicmice. Mol Pharmacol, 63:1289–1295.

Ozylikau E, Arslan M. (1997). Increased prevalence of diabetes mellitus in patients with chronic hepatitis C virus infection. Am

J Gastroenterol, 91:1480–1481.Palella FJ Jr., Delaney KM, Moorman AC, et al. (1998). Declining

morbidity and mortality among patients with advanced immunodefi ciency virus infection. HIV Outpatient Study Investigators. N Engl J Med, 338:853–860.

Palinkas LA, Barrett-Connor E, Wingard DL. (1991). Type 2 dia-betes and depressive symptoms in older adults: a population-based study. Diabet Med, 8:532–539.

Pankewycz OG, Guan J-X, Benedict JF. (1995). Cytokines as mediators of autoimmune diabetes and diabetic complications. Endocrine Rev, 16:164–171.

Park-Wyllie LY, Juurlink DL, Kopp A, et al. (2006). Outpatient gatifl oxacin therapy and dysglycemia in older adults. N Engl J

Med, 354:1352–1361.


OAPN

88

Passariello N, Giugliano D, Ceriello A, et al. (1986). Impaired insulin response to glucose but not to arginine in heroin addicts. J Endocrinol Invest, 9:353–357.

Passariello N, Giugliano D, Quatraro A. (1983). Glucose toler-ance and hormonal responses in heroin addicts. A possible role for endogenous opiates in the pathogenesis of non-insulin-dependent diabetes. Metabolism, 32:1163–1165.

Peters BH, Samaan NA. (1969). Hyperglycemia with relative hypoinsulinemia in diphenylhydantoin toxicity. N Engl J Med,

281:91–92.Physicians Desk Reference 2004, pp. 481, 496, 1114, 1392, 1467,

1489, 1493, 1608, 2630, 2905, 2913Pischon T, Sharma AM. (2001). Use of �-blockers in obesity, hyper-

tension: potential role of weight gain. Obes Rev, 2:275–280.Pollare T, Lithell H, Berne C. (1989). A comparison of the

eff ects of hydrochlorothiazide and captopril on glucose and lipid metabolism in patients with hypertension. N Engl J Med,

321:868–873.Polonsky KS, Sturis J, Bell GI. (1996). Non-insulin-dependent

diabetes mellitus: a genetically programmed failure of the � cell to compensate for insulin resistance. N Engl J Med,

334:777–783. Pont A, Rubino JM, Bishop D, Peal R. (1979). Diabetes mellitus

and neuropathy following vacor ingestion in man. Arch Intern

Med, 139:185–187.Prochaska JO, Velicer WF, Prochaska JM, Johnson JL. (2004).

Size, consistency, and stability of stage eff ects for smoking cessation. Addict Behav, 29:207–213.

Rajamani K, Colman PG, Li LP, et al. (2009).Eff ect of fenofi brate on amputation events in people with type 2 diabetes mellitus (FIELD study): a prespecifi ed analysis of a randomised con-trolled trial. Lancet, 373:1780–1788.


89

MEDICATIONS AND

DIABETES RISK

Rajpathak S, Kumbhani DJ, Crandall JP. (2008). Statin use and risk of developing type 2 diabetes: a meta-Analysis. [Abstract.] Circulation,118:S1152.

Redmon JB, Olson LK, Armstrong MB, et al. (1996). Eff ects of tacrolimus (FK506) on human insulin gene expression, insu-lin mRNA levels, and insulin secretion in HIT-T15 cells. J Clin

Invest 98:2786–2793.Ridker PM, Danielson E, Fonseca FAH, et al. (2008). Rosuvastatin

to prevent vascular events in men and women with elevated C-reactive protein. N Eng J Med, 359:2195–2207.

Rizza RA, Cryer PE, Haymond MW, Gerich JE. (1980). Adrenergic mechanisms of catecholamine action on glucose homeostasis in man. Metabolism, 29(11 Suppl 1):1155–1163.

Rizza RA, Mandarino LJ, Gerich JE. (1982). Eff ects of growth hormone on insulin action in man. Mechanisms of insulin resis-tance, impaired suppression of glucose production, and impaired stimulation of glucose utilization. Diabetes, 31:663–669.

Rosenthal AD, Shu X, Jin F, et al. (2004). Oral contraceptive use and risk of diabetes among Chinese women. Contraception, 69: 251–257.

Roth AM, Belhassen B, Laniado S. (1989).Slow-release verapamil and hyperglycemic metabolic acidosis. Ann Intern Med, 110:171–172.

Roubenoff R, Weiss L, McDermott A, et al. (1999). A pilot study of exercise training to reduce trunk fat in adults with HIV-associated fat redistribution. AIDS, 13:1373–1375.

Saito T, Sato T, Yamamoto M, et al. (1982). Hemoglobin A1 levels in hyperthyroidism. Endocrinol Jpn, 29:137–140.

Salgin B, Marcovecchio ML, Williams RM, et al. (2009). Eff ects of growth hormone and free fatty acids on insulin sensitiv-ity in patients with type 1 diabetes. J Clin Endocrinol Metab,

94:3297–3305.


OAPN

90

Saraga A, Yokokura M, Gonoi T, et al. (2004). Eff ects of fl uo-roquinolones on insulin secretion and ß-cell ATP-sensitive K+ channels. Eur J Pharm, 497:111–117.

Sasaki J, Iwashita M, Kono S. (2006). Statins: benefi cial or adverse for glucose metabolism. J Atheroscler Thromb, 13:123-129.

Scheen AJ, Finer N, Hollander P, et al. (2006). RIO-Diabetes Study Group. Effi cacy and tolerability of rimonabant in over-weight or obese patients with type 2 diabetes: a randomised controlled study. Lancet, 368:1660–1672.

Schütt M, Zhou J, Meier M, Klein HH. (2004). Long-term eff ects of HIV-1 protease inhibitors on insulin secretion and insulin signaling in INS-1 � cells. J Endocrinol, 183:445–454.

Scroggie DA, Albright A, Harris MD. (2003). The eff ect of glucosamine-chondroitin supplementation on glycosylated hemoglobin levels in patients with type 2 diabetes mellitus. A placebo-controlled, double-blinded, randomized clinical trial. Arch Intern Med, 163:1587–1590.

Shawe J, Lawrenson R. (2003). Hormonal contraception in women with diabetes mellitus: special consideration. Treat Endocrinol,

2:321–330.Sierra C, Ruilope LM. (2003). New-onset diabetes and antihy-

pertensive therapy: comments on ALLHAT trial. J Renin-

Angiotensin-Aldosterone Syst, 4:169–170.Solberg LI, Boyle RG, Davidson G, et al. (2001). Patient satisfac-

tion and discussion of smoking cessation during clinical visits. Mayo Clin Proc, 76:138–143.

SOLVD Investigators. (1991). Eff ect of enalapril on survival in patients with reduced left ventricular ejection fractions and congestive heart failure. N Engl J Med, 325:293–302.

Sowell MO, Mukhopadhyay N, Cavazzoni P, et al. (2002). Hyperglycemic clamp assessment of insulin secretory responses in normal subjects treated with olanzapine, risperidone, or placebo. J Clin Endocrinol Metab, 87:2918–2923.


91

MEDICATIONS AND

DIABETES RISK

Sowers JR. (1994). Hypertension in type 2 diabetes: update on therapy. J Clin Hypertens, 1:41–4.

Spellacy WN, Tsibris JC, Ellindson AB. (1991). Carbohydrate metabolic studies in women using a levonorgestrel/ethinyl estradiol containing triphasic oral contraceptive for eighteen months. Int J Gynaecol Obstet, 35:69–71.

Staels B. (2009). A review of bile acid sequestrants: potential mechanism(s) for glucose-lowering eff ects in type 2 diabetes mellitus. Post Grad Med, 121(3 Suppl 1): 25–30.

Stamler J, Vaccaro O, Neaton JD, et al. (1993). Diabetes, other risk factors, and 12–year mortality for men screened in the Multiple Risk Factor Intervention Trial. Diabetes Care, 16: 434–444.

Susan S, Jick SS, Bradbury BD. (2004). Statins and newly diag-nosed diabetes. Br J Clin Pharmacol, 58:303–309.

Taylor R, McCulloch AJ, Zeuzem S, et al. (1985). Insulin secre-tion, adipocyte insulin binding and insulin sensitivity in thyro-toxicosis. Acta Endocrinol (Copenh), 109:96–103.

Tenenbaum A, Fisman EZ, Boyko V, et al. (2006). Attenuation of progression of insulin resistance in patients with coro-nary artery disease by bezafibrate. Arch Intern Med, 166:737– 741.

Tenenbaum A, Motro M, Fisman EZ. (2005). Dual and pan-peroxisome proliferator-activated receptors (PPAR) co-agonism: the bezafi brate lessons. Cardiovasc Diabetol, 4:14.

Teoh SK, Sarnyai Z, Mendelson JH, et al. (1995). Eff ects of cocaine on cortisol secretion in humans. Am J Med Sci, 310:61–64.

Thiebaud D, DeFronzo RA, Jacot E, et al. (1982). Eff ect of long chain triglyceride infusion on glucose metabolism in man. Metab Clin Exp, 31:1128–1136.

Thonnard-Neumann E. (1968). Phenothiazines and diabetes in hospitalized women. Am J Psychiatry, 124:138–142.


OAPN

92

Tkach JR. (1982). Indomethacin-induced hyperglycemia in psori-atic arthritis. J Am Acad Dermatol, 7:802–803.

Tomita T, Onishi M, Sato E, et al. (2007). Gatifl oxacin induces augmented insulin release and intracellular insulin depletion of pancreatic islet cells. Biol Pharm Bull, 30:644–647.

Tornvall P, Walldius D. (1991). A comparison between nicotinic acid and acipimox in hypertryglyceridemia—eff ects on serum lipids, lipoproteins, glucose tolerance and tolerability. J Intern

Med, 230:415–421.Tuomilehto J, Lindstrom J, Eriksson JG, et al. (2001). Finnish

Diabetes Prevention Study Group. Prevention of type 2 dia-betes mellitus by changes in lifestyle among subjects with impaired glucose tolerance. N Engl J Med, 344:1343–1350.

Van Bortel LM, Ament AJ. (1995). Selective versus nonselective � adrenoceptor antagonists in hypertension. Pharmacoecon,

8:513–523.van den Berghe G, Wilmer A, Hermans G, et al. (2006).

Intensive insulin therapy in the medical ICU. N Engl J Med,

354:449–461.van den Berghe G, Wouters P, Weekers F, et al. (2001). Intensive

insulin therapy in the critically ill patients. N Engl J Med,

345:1359–1367.van den Berghe G, Wouters PJ, Bouillon R, et al. (2003). Out-

come benefi t of intensive insulin therapy in the critically ill: insulin dose versus glycemic control. Crit Care Med,

31:359–366.van Eijk IC, Peters MJ, Nurmohamed MT, et al. (2007). Decrease

of fructosamine levels during treatment with adalimumab in patients with both diabetes and rheumatoid arthritis. Eur J

Endocrinol, 156:291–293.Vesco L, Busson M, Bedrossian J, et al. (1996). Diabetes mellitus

after renal transplantation: characteristics, outcomes, and risk factors. Transplantation, 61:1475–1478.


93

MEDICATIONS AND

DIABETES RISK

Vestal RE, Eriksson CE, Musser B, et al. (1983).Eff ect of intra-venous aminophylline on plasma levels of catecholamines and related cardiovascular and metabolic responses in man. Circulation, 67:162–171.

Vincenti F, Friman S, Scheuermann E, et al. (2007). Results of an international, randomized trial comparing glucose metabolism disorders and outcome with cyclosporine versus tacrolimus. Am J Transplant,7:1506–1514.

Wambier CG, Foss-Freitas MC, Paschoal RS, et al. (2009). Severe hypoglycemia after initiation of anti-tumor necrosis factor ther-apy with etanercept in a patient with generalized pustular psoria-sis and type 2 diabetes mellitus. J Am Acad Dermatol, 60:883–885.

Wandell PE, Brorsson B, Aberg H. (1996). Drug use in patients with diabetes. Diabetes Care,19:992-994.

Warner EA, Greene GS, Buchsbaum MS, et al. (1998). Diabetic ketoacidosis associated with cocaine use. Arch Intern Med,

158:1799–1802.Webster AC, Woodroff e RC, Taylor RS, et al. (2005). Tacrolimus

versus ciclosporin as primary immunosuppression for kidney transplant recipients: meta-analysis and meta-regression of randomised trial data. Br Med J, 331:810.

Weir MR, Moser M. (2000). Diuretics and �-blockers: is there a risk for dyslipidemia? Am Heart J, 139:174–183.

Weyer C, Bogardus C, Mott DM, Pratley R. (1999). The natural history of insulin secretory dysfunction and insulin resistance in the pathogenesis of type 2 diabetes mellitus. J Clin Invest,

104:787–794.Wiener RS, Wiener DC, Larson RJ. (2008). Benefi ts and risks of

tight glucose control in critically ill adults: a meta-analysis. JAMA, 300:933–944.

Wilkinson A, Davidson J, Dotta F, et al. (2005). Guidelines for the treatment and management of new-onset diabetes after trans-plantation. Clin Transplant, 19:291–298.


OAPN

94

Willi SM, Kennedy A, Brant BP, et al. (2002). Eff ective use of thi-azolidinediones for the treatment of glucocorticoid-induced diabetes. Diabetes Res Clin Pract, 58:87–96.

Wing RR, Marcus MD, Blair EH, et al. (1990). Depressive symp-tomatology in obese adults with type II diabetes. Diabetes Care,

13:170 –172.Xiang AH, Kawakubo M, Kjos SL, Buchanan TA. (2006). Long-

acting injectable progestin contraception and risk of type 2 diabetes in Latino women with prior gestational diabetes. Diabetes Care, 29:613–617.

Yamada C, Nagashima K, Takahashi A, et al. (2006). Gatifl oxacin acutely stimulates insulin secretion and chronically suppresses insulin biosynthesis. Eur J Pharmacol, 553:67–72.

Yarasheski KE, Tebas P, Sigmund C, et al. (1999). Insulin resis-tance in HIV protease inhibitor-associated diabetes. J AIDS,

21:209–216.Yehuda R, Boisoneau D, Mason JW, Giller EL. (1993). Glucocorticoid

receptor number and cortisol excretion in mood, anxiety, and psychotic disorders. Biol Psychiatry, 34:18–25.

Young JB, Langsberg L. (1983). Adrenergic infl uence on periph-eral hormone secretion. In: Kunos G, ed. Adrenoceptors and

catecho l-amine action (Part B, pp. 157–217). New York: John Wiley & Sons.

Yu JG, Boies SM, Olefsky JM. (2003). The eff ect of oral glu-cosamine sulfate on insulin sensitivity in human subjects. Diabetes Care, 26:1941–1942.

Yusuf S, Sleight P, Pogue J, et al. (2000). Eff ects of an angiotensin-converting-enzyme inhibitor, ramipril, on cardiovascular events in high-risk patients. The Heart Outcomes Prevention Evaluation Study Investigators. N Engl J Med, 342:145–153.


95

MEDICATIONS AND

DIABETES RISK

Yusuf S, Wittes J, Probstfi eld J. (1990). Evaluating eff ects of treat-ment in subgroups of patients within a clinical trial: the case of non-Q-wave myocardial infarction and �-blockers. Am J Cardiol,

66:220–222.Zillich AJ, Garg J, Basu S, et al. (2006). Thiazide diuretics, potas-

sium, and the development of diabetes: a quantitative review. Hypertension, 48:219–224.

Zinman B, Hanley AJ, Harris SB, et al. (1999). Circulating tumor necrosis factor-� concentrations in a native Canadian popula-tion with high rates of type 2 diabetes mellitus. J Clin Endocrinol

Metab, 84:27–28.Zunkler BL, Lenzen S, Manner K, et al. (1988). Concentration-

dependent eff ects of tolbutamide, meglitinide, glipizide, glibencl-amide and diazoxide on ATP-regulated K+ currents in pancreatic B-cells. Naunyn Schmiedebergs Arch Pharmacol, 337:225–230.


medications and diabetes risk mechanisms and approach to risk reduction

Documents