Br HeartJ 1993;70:357-362

Pathogenesis of oedema in chronic severeanaemia: studies of body water and sodium, renalfunction, haemodynamic variables, and plasmahormones

Inder S Anand, Y Chandrashekhar, Roberto Ferrari, Philip A Poole-Wilson,Peter C Harris

Department ofCardiology,Postgraduate InstituteofMedical Educationand Research,Chandigarh, IndiaI S AnandY ChandrashekharCattedra diCardiologia,Universita di Brescia,Spedali Civili,Brescia, ItalyR FerrariDepartment ofCardiac Medicine,National Heart andLung Institute,Dovehouse Street,LondonP A Poole-WilsonP C HarrisCorrespondence to:Professor I S Anand,Department ofCardiovascular Medicine,University of MinnesotaMedical School, VAMedical Center i1i CMinneapolis MN 55417,USA.Accepted for publication5 May 1993

AbstractBackground-Patients with chronicsevere anaemia often retain salt andwater. Fluid retention in these patients isnot caused by heart failure and the exactmechanisms remain unclear. This studywas designed to examine some of thepossible mechanisms.Methods and results-Haemodynamicvariables, body fluid compartments,renal function, and plasma hormoneswere measured in four patients withoedema caused by chronic severeanaemia (mean (SE) haematocrit 13(1-7)) who had never received any treat-ment. Cardiac output was increased (6-1(0.6) lminIm2) and right atrial (7.8 (1)mm Hg), mean pulmonary arterial (20-5(2.0) mm Hg), and mean pulmonaryarterial wedge (13 (2.7) mm Hg) pres-sures were slightly increased. The meansystemic arterial pressure (81 (1.3) imm Hg)and systemic vascular resistance(12.3 (1.1) mm Hg x min x m21 were low.There were significant increases in totalbody water (+ 14%), extracellular volume(+ 32%), plasma volume (+70%), andtotal body exchangeable sodium (+ 30%).Renal blood flow was moderatelydecreased (- 46%) and the glomerularfiltration rate was slightly reduced( - 24%). There were significant increasesin plasma noradrenaline (2.1-fold), reninactivity (15-fold), aldosterone (3.2-fold),growth hormone (6.3-fold), and atrialnatriuretic peptide (12-fold).Conclusion-In patients with oedemacaused by chronic severe anaemia thereis retention of salt and water, reductionof renal blood flow and glomerular filtra-tion rate, and neurohormonal activationsimilar to that seen in patients withoedema caused by myocardial disease.However, unlike patients with myocar-dial disease, patients with anaemia havea high cardiac output and a low systemicvascular resistance and blood pressure.It is suggested that the low concentrationof haemoglobin in patients with anaemiacauses a reduced inhibition of basalendothelium-derived relaxing factoractivity and leads to generalised vaso-dilatation. The consequent low bloodpressure may be the stimulus for neuro-

hormonal activation andretention.

salt and water

(Br HeartJf 1993;70:357-362)

Chronic severe anaemia is often associatedwith various degrees of salt and water reten-tion.'-" When fluid retention is severe thecondition is often referred to as "congestiveheart failure".2' Though the pathogenesis ofsalt and water retention in congestive heartfailure caused by low output states has beenwell studied,45 the mechanisms of fluid reten-tion in patients with chronic severe anaemiaremain unclear. Patients with anaemia and"congestive heart failure" have a high cardiacoutput which increases with exercise evenmore than in healthy subjects.'2 The intra-cardiac filling pressures of these patients areusually normal or only slightly raised2 3 6 7 andmyocardial contractility is normal orincreased.8 All the haemodynamic changesare rapidly reversed after the anaemia iscorrected.'267 Because there is little evidencethat myocardial dysfunction is a cause of saltand water retention in chronic severeanaemia, it has been suggested that a betterterm for the condition is a "non-cardiac con-gestive state".9 Few detailed physiologicalstudies have been made in patients withchronic severe anaemia and fluid retention.238The mechanism of salt and water retentionhas remained elusive because it is rare to findsuch patients who have not received treat-ment for "congestive heart failure", whichitself may affect the mechanisms beingstudied.0"11We present data on haemodynamic vari-

ables, body fluid compartments, renal func-tion, and plasma hormones in four untreatedpatients with oedema caused by chronicsevere anaemia and compare the results withthose found in patients with untreated con-gestive heart failure caused by myocardialdisease.

Patients and methodsPATIENTSThe studies were carried out at thePostgraduate Institute of Medical Educationand Research, Chandigarh, India, on fourpatients (one man and three women) admit-ted with chronic severe anaemia (haematocrit9-16%) who had had symptoms for 2 to 4

357 on M

arch 30, 2021 by guest. Protected by copyright.

http://heart.bmj.com

/B

r Heart J: first published as 10.1136/hrt.70.4.357 on 1 O

ctober 1993. Dow

nloaded from

http://heart.bmj.com/

Anand, Chandrashekhar, Ferrari, Poole-Wilson, Hanis

months (mean (SE) 2'9 (0 8) months). Theycomplained of fatigue and shortness of breath(mean New York Heart Association class, 2-7(0-2)). All looked pale and had mild to mod-erate ankle oedema and one (A2) had mildascites. The jugular venous pressure wasraised (6 (1 5) cm) and the liver was enlargedto an average (SD) of 5-7 (0 5) cm below theright costal margin. No patients had lymphnode or splenic enlargement. A chest radio-graph showed a slightly increased cardio-thoracic ratio (0-52 (0-02)) but no signs ofpulmonary venous congestion. The electro-cardiogram showed non-specific ST segmentand T wave changes. The cause of anaemiawas severe iron deficiency caused by chronichookworm infection in two (Al and A2),dietary iron deficiency in one (A3), andchronic bleeding haemorrhoids in one (A4).Routine clinical chemistry, serum creatinine,liver function tests and in particular serumalbumin were normal. None of the patientshad been treated for anaemia or "congestiveheart failure". They were on a normal dietand salt was not restricted. Sodium balancewas not studied.

PROTOCOLThe studies were performed after weobtained written, informed consent from thepatients and the study was approved by thelocal ethics committee. Patients were studiedon two consecutive days. Hormones andhaemodynamic variables were measured onthe first day and body fluid compartmentsand renal blood flow measurements on thesecond day. Two patients (Al and A2) wererestudied after anaemia had improved withtreatment.

HAEMODYNAMIC VARIABLESHaemodynamic measurements were made inthe postabsorptive state with a Swan-Ganzthermodilution balloon catheter placed in thepulmonary artery. Arterial pressure was mea-sured by cannulation of the left brachialartery with a 3F Teflon catheter (Seldicath).Pressures were measured with Hewlett-Packard 1290C transducers and a Hewlett-Packard 78354A monitor. The cardiac outputwas determined by thermodilution (ModelSP 1445 Gould, Cleveland, Ohio).4

WATER AND SODIUM SPACES AND RENALFUNCTIONBody water, sodium spaces, and renal func-tion were measured by standard isotope dilu-tion techniques.412 Briefly, plasma volumewas calculated from the volume of distribu-tion of 5 mCi 125I-labelled human serum albu-min 10 minutes after intravenous injection.The extracellular volume and glomerular fil-tration rate were measured simultaneouslywith 100 mCi 5"Cr-labelled ethylenediamine-tetraacetic acid (EDTA) and calculated bynumerical analysis. Effective renal plasmaflow (ERPF) was estimated with 100 mCi125I-labelled sodium iodohippurate (Hippuran)and also calculated by numerical analysis.This method of measuring ERPF is indepen-

dent of assumed volumes of distribution and,therefore, partcularly useful when oedema ispresent. Renal blood flow was calculatedfrom ERPF and the haematocrit was cor-rected for trapped plasma. Total body waterwas determined with 100 mCi tritiated waterand total body exchangeable sodium with 20mCi 22Na administered orally. A speciallydesigned protocol allowed all measurementsto be made in one 24 hour period.412

PLASMA HORMONESPlasma hormones were assayed in a 20 mlblood sample drawn from a forearm vein afterthe patients had been recumbent for 30 min-utes. The techniques are described in detailelsewhere.413 Plasma noradrenaline andadrenaline were measured by high-perfor-mance liquid chromatography with electro-chemical detection. Plasma renin activity,aldosterone, vasopressin, cortisol, growthhormone, and atrial natriuretic peptide(ANP) were measured by radioimmunoassay.

STATISTICAL ANALYSISThe data are presented as mean (SEM). Thesignificance of the difference between variousvariables in patients with anaemia, controls,and patients with congestive heart failure wasestimated with the unpaired Student's t test.

ResultsTables 1-3 give the haemodynamic, bodycompartment, renal function, and plasmahormone data. Earlier data from healthy indi-viduals and from patients with oedemacaused by myocardial disease, obtained byidentical methods,4 are also provided forcomparison.

HAEMODYNAMIC VARIABLESThe resting heart rate was slightly increased(p < 0-01) (table 1). The cardiac index wasincreased in all the patients, the average (6- 13(0 56) being 150% above normal (p < 0-01).The mean pulmonary arterial and wedgepressures were raised in one patient (A2) butwere within the normal range in the others.Right atrial pressure was slightly raised inthree and was even higher in one (A2). Themean systemic blood pressure was signifi-cantly reduced (p < 0-01). The most impres-sive finding was considerable reduction in thesystemic vascular resistance (p < 0-01).Arterial blood gases were normal.

BODY FLUID COMPARTMENTS AND RENALFUNCTIONAll body fluid compartments were increased(table 2). Total body water was, on average,14% greater than normal and the increasewas accommodated almost entirely in theextracellular space. Total body waterincreased by 77 ml/kg and the extracellularspace by 72 ml/kg. The increase in extracellu-lar space was divided between the extravascu-lar and intravascular compartments but not inproportion to their normal volumes: theplasma volume increased by a much greater

358 on M

arch 30, 2021 by guest. Protected by copyright.

http://heart.bmj.com

/B

r Heart J: first published as 10.1136/hrt.70.4.357 on 1 O

ctober 1993. Dow

nloaded from

http://heart.bmj.com/

Studies in anaemia

Table 1 Haemodynamic variables and arterial blood gases in patients with oedema caused by chronic severe anaemia, in controls, and in patients withoedema caused by myocardial disease

PVRI SVRIBSA HR RAP PAP PAWP AOP Cl (mm Hg x P02 PCo2 0,sat

Patients Sex Age (Mi) (beatslmin) (mm Hg) (mm Hg) (mm Hg) (mm Hg) (i/minim2) min x M211) (kPa) (kPa) pH (%)

Patients with anaemia:Al F 17 1-36 96 8 16 9 82 7-7 0 90 9-6 9-57 3-86 7-51 96A2 F 27 1-43 88 10 28 21 81 5-4 1-29 13-1 10-77 4 39 7-45 96A3 F 18 1-42 92 7 20 12 78 6-2 1-29 11-5 11-70 4-66 7-42 99A4 M 38 2-16 80 6 18 10 84 5-2 1-54 15-0 11-97 4-52 7-44 99Mean 25-0 1-59 89-0 7-8 20-5 13-0 81-3 6-13 1-26 12-30 11-01 4-36 7-45 98SEM 4-2 0-17 4-2 1-0 2-6 2-7 1-3 0-56 1-13 1-14 0 53 0-29 0.03 2n 4 4 4 4 4 4 4 4 4 4 4 4 4 4

Controls:Mean 22-4 1-72 71-3 2-2 16-1 7-9 97-1 3-92 2-25 25-74 - - - -SEM 2-0 0-02 3-0 0-5 0-6 0-6 2-2 0-28 0-23 1-59 - - - -n 16 16 16 16 16 16 16 16 16 16 - - - -p (v anaemia) NS

Anand, Chandrashekhar, Ferrari, Poole-Wilson, Haris

Table 4 Haemodynamic variables, body fluid comphormones in two patients before and after treatment

Haemodynamic variable:Heart rate (beats/min)Right atrial pressure (mm Hg)Mean pulmonary artery pressure (mm Hg)Mean pulmonary wedge pressure (mm Hg)Mean arterial pressure (mm Hg)Cardiac index (1/min x M2)Pulmonary vascular resistance(mm Hg x min x m2/l)

Systemic vascular resistance(mm Hg x min x m2/l)

Body fluid compartment and renal function:Extracellular volume (mi/kg)Plasma volume (ml/mg)Packed cell volume (%)Blood volume (ml/kg)Total body water (ml/kg)Total body exchangeable sodium (mmol/kg)Glomerular filtration rate (ml/min x 1 73m2)Effective renal plasma flow (ml/min x 1 73m2)Effective renal blood flow (ml/min x 1-73m')

Hormones:Adrenaline (pg/ml)Noradrenaline (pg/ml)Plasma renin activity (ng/ml x h)Aldosterone (pg/ml)Vasopressin (pg/ml)Atrial natriuretic peptide (pg/ml)Growth hormone (ng/ml)Cortisol (ng/ml)

within the normalso increased ivalue was 15 tterone was raisecin one. Plasmapatient and thenormal. Growthand normal in iwere variable.

EFFECT OF TREA1All patients imprtwo (Al and A2patient (A4) waother (A3) did Xpatients were giron supplemen

artments, renalfunction, andplasma ties of sodium and water (nearly 4 0 1 ofexcess water and 748 mmol sodium). Despite

Before treatment After treatment a high cardiac output and a reduced systemic(mean) (mean) vascular resistance, the renal blood flow was

reduced to nearly 50% of normal. The92 74 glomerular filtration rate was relatively pre-9 4gomrarrltvy22 15 served, averaging 75% of normal, suggesting2 104 greater vasoconstriction of the efferent renal6-6 3-9 arterioles than the afferent ones. The reduc-1.1 1.9 tion in renal function that we saw in our

patients has been described elsewhere.3141511-4 25-7 Renal blood flow falls in proportion to the

severity of anaemia, and patients with severe325 243 anaemia and massive oedema have lower81-3 61-112-5 390 renal blood flows and higher renal vascular90 9251 resistance than patients with lesser degrees of638-0 588-058-1 51-6 oedema.68 154 The hormonal response consisted of an332 790

375 1184 increase in plasma noradrenaline; activationof the renin-angiotensin-aldosterone system;

36 108 and an increase in atrial natriuretic peptide,949 036 vasopressin, and growth hormone. There are

66 30 almost no reported data on the neuroen-18-6 90 docrine response of patients with chronic

120 104-4 2-3 severe anaemia. Wassermann et al found high

232 50 resting plasma noradrenaline concentrationsin anaemic dogs and an exaggerated responseto moderate exercise.'6 Iron deficiency causes

ial range. Renin activity was an increased excretion of urinary cate-in every patient; the mean cholamines in humans'7 and in rats which isimes normal. Serum aldos- corrected by treatment.'8 The increases ini in three patients but normal noradrenaline, renin, and aldosterone proba-ANP was raised in every bly reflect a reflex response to the reducedmean was nearly 12 times blood pressure and decreased renal bloodhormone was raised in three flow. The high concentration of ANP is sur-one. Cortisol concentrations prising in view of the rather modest increase

in filling pressures. Though plasma osmolalitywas not measured, the normal serum sodium

rMENT indicates that it is likely to have been normal.roved with treatment but only Therefore the most likely cause of the) could be reinvestigated: one increase in vasopressin is a non-osmoticls lost to follow up and the release. An increase in growth hormone hasnot wish to be restudied. All never been reported in anaemia and is similariven blood transfusions and to that reported by us in other syndromes ofLts. None was treated with salt and water retention.45 '9

diuretics or cardiovascular drugs. Table 4shows the results in the two patients. Bothpatients became symptom free and lostweight and oedema: patient A2 became freeof ascites. The haemodynamic variablesreturned to normal. In particular, the bloodpressure and systemic vascular resistanceincreased to normal values as the haematocritimproved. Clinical improvement and a reduc-tion in oedema was associated with a sub-stantial decrease in extracellular volume, totalbody water, and total body exchangeablesodium and an increase in effective renalplasma and blood flow and glomerular filtra-tion rate. Though the plasma volume fell con-siderably, the blood volume did not changebecause the haematocrit increased from 12-5to 39%. The plasma hormones returned tonormal values.

DiscussionWe investigated patients with chronic severeanaemia and oedema who had never beentreated. They had retained substantial quanti-

MECHANISM OF REDUCED SYSTEMIC VASCULARRESISTANCE IN CHRONIC SEVERE ANAEMIAMeasurement of the haemodynamic variablesconfirmed the presence of increased cardiacoutput; low arterial blood pressure; muchreduced systemic- vascular resistance; andmildly raised right atrial, pulmonary arterial,and wedge pressures shown previously insuch patients.267 The mechanisms that leadto a low systemic vascular resistance inpatients with chronic severe anaemia are notknown but vasodilatation and low blood vis-cosity have been implicated.62021 Viscosity,however, does not fully explain the low sys-temic vascular resistance.202' We have previ-ously postulated22 that enhanced activity ofendothelium-derived relaxing factor (EDRF)may be responsible for the vasodilatation seenin patients with chronic severe anaemia.Haemoglobin is an important inhibitor ofEDRF.2324 It binds avidly to nitric oxide andregulates the activity of EDRF.2'26 Infusionof stroma-free haemoglobin as replacement inhaemorrhagic shock is associated with an

360 on M

arch 30, 2021 by guest. Protected by copyright.

http://heart.bmj.com

/B

r Heart J: first published as 10.1136/hrt.70.4.357 on 1 O

ctober 1993. Dow

nloaded from

http://heart.bmj.com/

Studies in anaemia

immediate and pronounced increase in sys-temic vascular resistance that persists as longas haemoglobin remains in circulation.27 28 Itis, therefore, possible that low concentrationsof haemoglobin in patients with chronicsevere anaemia increase basal EDRF activity,leading to generalised vasodilatation and tothe hyperdynamic circulatory state character-istic of chronic severe anaemia. The resultingincrease in blood flow could further amplifythe flow mediated release of EDRF.29-" Anincrease in EDRF activity is responsible forthe hyperdynamic circulation in patients withcirrhosis of liver,'23' and for hypotension inpatients with septic shock.'4

COMPARISON OF FINDINGS IN CHRONIC SEVEREANAEMIA WITH THOSE IN PATIENTS WITHOEDEMA FROM MYOCARDIAL DISEASEThe congestive state seen occasionally inpatients with chronic severe anaemia is simi-lar to that seen in patients with oedemacaused by myocardial disease, but the mecha-nisms of salt and water retention in the twoconditions may be different. We thereforecompared the findings in patients withchronic severe anaemia with those, reportedby us previously in patients with oedemafrom myocardial disease4 (tables 1-3).The haemodynamic variables in these two

conditions were entirely different (table 1).While the right-sided pressures were muchraised in patients with myocardial disease,they were either normal or only slightlyincreased in patients with chronic severeanaemia. The most striking difference wasseen in the systemic circulation. The cardiacoutput was considerably reduced in conges-tive heart failure but was increased in chronicsevere anaemia. Also, while patients withcongestive heart failure were able to maintaina normal arterial blood pressure by a consid-erable increase in the systemic vascular resis-tance, patients with chronic severe anaemiacould not maintain a normal blood pressurebecause of systemic vasodilatation. Theincrease in cardiac output in anaemia- waslargely the result of an increase in the strokevolume with only a modest increase in heartrate, suggesting normal myocardial function.The hormone response was similar in bothconditions. Thus despite increased sympa-thetic stimulation and activation of the renin-angiotensin-aldosterone system, patients withoedema from chronic severe anaemia hadmarkedly reduced systemic vascular resis-tance.

Despite the striking differences in thehaemodynamic variables in these two condi-tions, the changes in the body fluid compart-ments and renal function were similar. Therenal plasma and blood flow and glomerularfiltration rate were reduced in both condi-tions, though the decrease was greater inpatients with congestive heart failure. Thesystemic vascular resistance fell to nearly halfnormal in patients with anaemia, while itincreased by 80% in patients with congestiveheart failure. Despite this the renal vascularresistance increased in both conditions: by

more than 40% in anaemia and more than200% in congestive heart failure. It is, there-fore, clear that patients with chronic severeanaemia constrict their renal vasculaturewhile they have an overall systemic vasodi-latation.

PATHOPHYSIOLOGY OF SALT AND WATERRETENTION IN CHRONIC SEVERE ANAEMIAThe neurohormonal response, changes inrenal circulation, and degree of salt and waterretention in patients with oedema fromchronic severe anaemia are qualitatively simi-lar to those in patients with oedema frommyocardial disease, conditions with funda-mentally different haemodynamic mecha-nisms. We have previously shown that asimilar pathophysiology is seen in patientswith oedema from chronic obstructive pul-monary disease who, like patients withoedema from chronic severe anaemia, havereduced systemic vascular resistance.9 Thecommon factor in all these conditions seemsto be a tendency towards a low arterial bloodpressure. Blood pressure is "threatened" incongestive heart failure by low cardiac outputand in chronic severe anaemia and chroniccor pulmonale by decreased systemic vascularresistance. Low or "threatened" blood pres-sure evokes a reflex increase in sympatheticactivity through the baroreceptors and acti-vates the renin-angiotensin-aldosterone sys-tem through the juxtaglomerular cells, in thesame way as do low cardiac output states.45The increased sympathetic-stimulationreduces the renal blood flow and to a lesserextent the glomerular filtration rate and maycontribute to the non-osmotic release ofvasopressin. The net result is salt and waterretention. The activated renin angiotensin-aldosterone axis and vasopressin further helpto retain salt and water. The extracellular vol-ume expands and leads to release of ANPwhose vasodilator properties may be expectedto reduce the systemic vascular resistancefurther. Replacement of the red blood cellsrestores haemoglobin concentrations, inhibitsEDRF activity, and reverses the process thatcontributed to salt and water retention. Thissequence of events is consistent with thehypothesis35 that the syndrome of congestiveheart failure is the result of sustainedattempts to maintain arterial blood pressureby neurohormonal activation.

LIMITATIONS OF THIS STUDYThis study was carried out on only fourpatients. The major objective was to investi-gate patients with chronic severe anaemiawho had salt and water retention and whohad never been treated. Treatment, especiallywith diuretics has such profound effects onthe neurohormones that it becomes difficultto distinguish between the effects of treat-ment and those of the disease itself.'01'However, it is difficult to find such patients inhospital practice and during a period ofnearly 18 months only four such patients whohad not received any diuretic treatment wereadmitted to a large teaching hospital.

361 on M

arch 30, 2021 by guest. Protected by copyright.

http://heart.bmj.com

/B

r Heart J: first published as 10.1136/hrt.70.4.357 on 1 O

ctober 1993. Dow

nloaded from

http://heart.bmj.com/

Anand, Chandrashekhar, Ferrari, Poole-Wilson, Harris

This study was supported by the Indian Council of MedicalResearch} British Heart Foundation, and EEC contract NST2J-0071-1 -i.

1 Varat MA, Adolph RJ, Fowler NO. The cardiovasculareffects of anaemia. Am HeartJ' 1972;83:415-24.

2 Graettinger JS, Parsons RL, Campbell JA. A correlation ofclinical and haemodynamic studies in patients with mildand severe anaemia with and without congestive heartfailure. Ann Intern Med 1963;58:617-26.

3 Singhvi LM, Sharma VS, Mansharamani G. Mechanismof oedema in chronic severe anaemia. Br Heart J 1963;25:89-96.

4 Anand IS, Ferrari R, Kalra GS, Wahi PL, Poole-WilsonPA, Harris P. Oedema of cardiac origin: Studies of bodywater & sodium, renal function, haemodynamics andplasma hormones in untreated congestive heart failure.Circulation 1989;80:299-305.

5 Anand IS, Ferrari R, Kalra GS, Wahi PL, Poole-WilsonPA, Harris P. Pathogenesis of oedema in constrictivepericarditis. Studies of body water & sodium, renalfunction, haemodynamics and plasma hormones beforeand after pericardiectomy. Circulation 1991;83: 1880-7.

6 Duke M, Abelman WH. The haemodynamic response tochronic anaemia. Circulation 1969;39:503-13.

7 Roy SB, Bhatia ML, Mathur VS, Virmani S. Haemo-dynamic effects of chronic severe anaemia. Circulation1963;28:346-56.

8 Florenzano F, Diaz G, Regonesi C, Escobar E. Left ven-tricular function in chronic anaemia: evidence of non-catecholamine, positive inotropic factor in serum. Am JCardiol 1984;54:638-45.

9 Echina LV. Circulatory congestion and heart failure.Circulation 1960;22:864-86.

10 Bayliss J, Norell M, Canepa-Anson R, Sutton G, Poole-Wilson P. Untreated heart failure: Clinical and neuro-endocrine effects of introducing diuretics. Br Heart J1987;57: 17-22.

11 Kubo SH, Clark M, Laragh JH, Borer JS, Cody RJ.Identification of normal neurohormonal activity in mildcongestive heart failure and the stimulating effect ofupright posture and diuretics. Am J Cardiol 1987;60:1322-9.

12 Anand IS, Veal NB, Kalra GS, Ferrari R, Sutton G,Lipkin D, Harris P, Poole-Wilson PA. Treatment ofheart failure with diuretics: Body compartments, renalfunctions, and plasma hormones. Eur Heart J7 1989;10:445-50.

13 Ferrari R, Ceconi C, Signorini C, Anand IS, Harris P,Albertini A. Sample treatment for long distance airtransport of human plasma for hormone assay. ClinChem 1989;35:331-2.

14 Bradley SE, Bradley GP. Renal function during chronicanaemia in man. Blood 1947-2:192-202.

15 Renbi RD, Vorberger C, Keller HM. The renal excretionof sodium paraaminohippurate in anaemic subjectsbefore and after red cell transfusion. Clin SCi 1962;23:213-9.

16 Wassermann DH, Lavina H, Likely A, Vranic M. Effect ofhaematocrit reduction on hormonal and metabolicresponse to exercise. JAppl Physiol 1985;58: 1257-6 1.

17 Wagner A, Fortier N, Giroux A, Lukes J, Snyder LM.Catecholamines in adult iron deficient patients.Experentia 1979;35:681-2.

18 Vorrhess LM, Stuart MJ, Stoclknan SA, Oski FA. Irondeficiency anaemnia and increase in urinary norepineph-rine. Pediatrics 1975;86:542-7.

19 Anand IS, Chandrashekhar Y, Ferrari R, Sarma P,Guleria R, Jindal SK, Wahi PL, Poole-Wilson PA,Harris P. Pathogenesis of congestive state in chronicobstructive pulmonary disease. Studies of body waterand sodium, renal function, haemodynamics andplasma hormones during oedema and after recovery.Circulation 1992;86: 12-21.

20 Fowler NO, Holmes JC. Blood viscosity and cardiac out-put in acute experimental anaemia. JI Appl Physiol1975;39:453-6.

21 Escobar E, Jones NL Rapaport E, Murray JF. Ventricularperformance in acute normovolemic anaemia andeffects of beta blockade. AmJ Physiol 1966;211:877-84.

22 Anand IS, Chandrashekhar Y. Reduced inhibition ofendothelial-derived relaxing factor causes the hyper-dynamic circulation in chronic severe anaemia. MedicalHypotheses 1993 (in press).

23 Martin W, Smith JA, White DG. The mechanisms bywhich haemoglobin inhibits the relaxation of rabbitaorta induced by nitrovasodilators, nitric oxide, orbovine retractor penis inhibitory factor. Br Jf Pharmacol1986;89:563-7 1.

24 Martin W, Villani GM, Jothianandan D, Furchgott RF.Selective blockade of endothelium-dependent and glyc-eryl trdnitrate-induced relaxation by haemoglobin and bymethylene blue in the rabbit aorta. J Pharmacol Exp Ther1985;232:708-16.

25 Gibson QH, Roughton FJW. The kinetics and equilibriaof the reactions of nitric oxide with sheep haemoglobin.J Physiol 1957;136:507-26.

26 Murad F, Mittal CK, Arnold WP, Katsuki S, Kimura H.Guanylate cyclase: activation by azide, nitro com-pounds, nitric oxide and hydroxyl radical and inhibitionby haemoglobin and myoglobin. Adv Cyclic NucleotideRes 1978;9: 145-58.

27 Nees JE, Hauser CJ, Shippy C, State D, Shoemaker WC.Comparison of the cardiorespiratory effects of crys-tallised haemoglobin, whole blood, albumin and ringerlactate in the resuscitation of shock in dogs. Surgery1978;83:639-47.

28 Hauser CJ, Shoemaker WC. Haemoglobin solutions in thetreatment of haemorrhagic shock. Crit Care Med 1982;10:283-7.

29 Pohl U, Busse R, Kuon E, Bassenge E. Pulsatile perfusionstimulates the release endothelial autacoids. J ApplCardiol 1986;1:215-35.

30 Lamontagne D, Pohl U, Busse R. Mechanical deforma-tion of vessel wall and shear stress determine basalrelease of endothelial derived relaxing factor in theintact rabbit coronary vascular bed. Circ Res 1992;70:123-30.

31 Miller VM, Vanhoutte PM. Enhanced release of endo-thelium derived factor(s) by chronic increases in bloodflow. AmI Physiol 1988;255:H446-51.

32 Vallance P, Moncada S. Hypothesis: induction of nitricoxide synthase in the vasculature underlines the hyper-dynamic circulation of cirrhosis. Lancet 1991;337:776-8.

33 Pizcueta P, Pique JM, Bosch J, Whittle BJR, Moncada S.Effects of endogenous nitric oxide inhibition on thehaemodynamic changes of portal hypertensive rats(abstr]. Gastroenterology 1991;100:785.

34 Kilbourn RG, Gross SS, Jurbran A, Adams J, GriffithOW, Levi R, Lodato RF. NG-methyl-L-arginine inhibitstumor necrosis factor-induced hypotension: implicationsfor the involvement of nitric oxide. Proc Nat! Acad SciUSA 1990;87:3639-2.

35 Harris P. Role of arterial pressure in the oedema of heartdisease. Lancet 1988;i:1036-8.

362 on M

arch 30, 2021 by guest. Protected by copyright.

http://heart.bmj.com

/B

r Heart J: first published as 10.1136/hrt.70.4.357 on 1 O

ctober 1993. Dow

nloaded from

http://heart.bmj.com/