pyrogenic reactions and hemorrhage associated with intrinsic exposure to endotoxin‐contaminated...

Pyrogenic Reactions and Hemorrhage Associated With Intrinsic Exposure to Endotoxin‐Contaminated Intravenous Solutions • Author(s): Luciane Zappelini Daufenbach , RN; Waneska A. Alves , MV, MSc;Jaime B. de Azevedo; Matthew J. Arduino , MS, DrPH; Terri S. Forster , BS;Eduardo H. Carmo , MD, DrPH; Douglas L. Hatch , MD, MPHSource: Infection Control and Hospital Epidemiology, Vol. 27, No. 7 (July 2006), pp. 735-741Published by: The University of Chicago Press on behalf of The Society for Healthcare Epidemiologyof AmericaStable URL: http://www.jstor.org/stable/10.1086/504360 .

Accessed: 14/05/2014 12:38

Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at .http://www.jstor.org/page/info/about/policies/terms.jsp

.JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range ofcontent in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new formsof scholarship. For more information about JSTOR, please contact [email protected].

.

The University of Chicago Press and The Society for Healthcare Epidemiology of America are collaboratingwith JSTOR to digitize, preserve and extend access to Infection Control and Hospital Epidemiology.

http://www.jstor.org

This content downloaded from 194.29.185.114 on Wed, 14 May 2014 12:38:23 PMAll use subject to JSTOR Terms and Conditions

http://www.jstor.org/action/showPublisher?publisherCode=ucpress

http://www.jstor.org/action/showPublisher?publisherCode=shea

http://www.jstor.org/action/showPublisher?publisherCode=shea

http://www.jstor.org/stable/10.1086/504360?origin=JSTOR-pdf

http://www.jstor.org/page/info/about/policies/terms.jsp


infection control and hospital epidemiology july 2006, vol. 27, no. 7

o r i g i n a l a r t i c l e

Pyrogenic Reactions and Hemorrhage Associated With IntrinsicExposure to Endotoxin-Contaminated Intravenous Solutions

Luciane Zappelini Daufenbach, RN; Waneska A. Alves, MV, MSc; Jaime B. de Azevedo;Matthew J. Arduino, MS, DrPH; Terri S. Forster, BS; Eduardo H. Carmo, MD, DrPH; Douglas L. Hatch, MD, MPH

objective. An epidemiological investigation was conducted to determine risk factors for adverse reactions among patients in hospitalsand the possibility of extrinsic or intrinsic contamination of intravenous solutions.

design. A retrospective cohort study was conducted to identify solutions associated with adverse reactions. Implicated lots were culturedfor bacteria, and endotoxin concentrations were measured.

setting. Five hospitals in the state of Pernambuco, Brazil, were investigated from February through March 2002.

patients. Surgical inpatients or outpatients receiving intravenous solutions during the study period.

results. Of 355 surgical patients or outpatients treated at hospitals, 28 (8%) developed illness within a mean of 2.5 hours after exposureto intravenous solutions implicated in adverse reactions; 5 (17.9%) of the case patients died. Laboratory testing of bottles from the lots ofRinger’s lactate solution implicated in deaths demonstrated a high mean endotoxin concentration of 88.3 endotoxin units (EU)/mL (range,9.7-298.0 EU/mL), compared with the permitted limit in Brazil of !0.5 EU/mL. Testing of metronidazole implicated in adverse reactionsat another hospital and produced by the same company that manufactured the lots Ringer’s lactate solution also showed high endotoxinconcentrations (mean level, 8.3 EU/mL [range, 5.0-58.3 EU/mL]). The outbreak was controlled after a national recall of the implicatedbrand of intravenous solutions.

conclusions. Case patient status was associated with use of Ringer’s lactate solution and metronidazole from large bottles, both ofwhich were produced by the same company. High endotoxin concentrations were demonstrated in unopened bottles of implicated products,which is consistent with intrinsic contamination. The high mortality rate may have been compounded by the fact that clinicians administeredadditional volumes of contaminated 0.9% isotonic sodium chloride solution in response to hypotension or bleeding to some surgicalpatients. No additional case patients were identified after a national recall of products, inspection, closure of the implicated company’smanufacturing facility, and establishment of random quality-control testing of intravenous solutions.

Infect Control Hosp Epidemiol 2006; 27:735-741

From the Field Epidemiology Training Program (L.Z.D., W.A.A., D.L.H.) and the Department of Epidemiological Surveillance (E.H.C.), GeneralCoordination of Communicable Diseases, Ministry of Health, Brasılia, and the Department of Sanitary and Epidemiological Surveillance (J.B.A.), Recife,Brazil; and the Environmental and Applied Microbiology Section, Division of Healthcare Quality Promotion, National Center for Infectious Diseases (M.J.A.,T.S.F.), and the Division of International Health, Office of Global Health (D.L.H.), Centers for Disease Control and Prevention, Atlanta, Georgia.

Received December 13, 2004; accepted June 27, 2005; electronically published June 22, 2006.� 2006 by The Society for Healthcare Epidemiology of America. All rights reserved. 0899-823X/2006/2707-0016$15.00.

Endotoxin, a normal component of the cell wall of gram-negative bacteria, was first shown to cause pyrogenic reactionsin 1923.1 The term “injection fever” was used to describe thesereactions, which were observed after intravenous administra-tion of solutions shown to be free of bacteria.2 Minute amountsof endotoxin administered intravenously can cause pyrogenicreactions, and more-substantial amounts can cause hypoten-sion, shock, and disseminated intravascular coagulation.3

During March 2002, clinicians in Caruaru City (estimated2000 population, 253,634 persons), Pernambuco, Brazil, re-ported that several patient deaths due to severe, unexplainedbleeding occurred shortly after the administration of intra-venous solutions during elective surgery. These events werepromptly reported to the local surveillance system for sus-

pected adverse reactions to medications, and a joint inves-tigation by state health officials, university experts, the FieldEpidemiology Training Program of the Brazil Ministry ofHealth, and the Centers for Disease Control and Prevention(CDC) was begun to determine the cause of illness.

methods

Inclusion and Exclusion Criteria

Medical records and pharmacy logbooks were reviewed in 5hospitals that had reported possible adverse reactions to in-travenous solutions or medications among patients to theofficial surveillance system in the state of Pernambuco. Wedefined case patients as those with adverse reactions of un-



736 infection control and hospital epidemiology july 2006, vol. 27, no. 7

explained fever, pyrogenic reactions, or hemorrhage (eg, sur-gical wound, gastrointestinal bleeding, or vaginal bleeding)between February 1 and March 15, 2002, after admission tothe emergency department or recent surgery and within 12hours of intravenous administration of solution(s) or med-ication(s). Exclusion criteria included chronic liver disease(eg, cirrhosis or chronic hepatitis), history of preexisting co-agulation disorder or abnormal bleeding, previous treatmentwith heparin or warfarin, and other known cause of fever,pyrogenic reaction, and hemorrhage (eg, bacterial infectionor sepsis) diagnosed during hospitalization.

Cohort Studies

Hospital A. A retrospective cohort study design was used toinvestigate data for all patients with possible adverse reactionswho were treated in the surgical ward at hospital A betweenFebruary 1 and March 15, 2002. Data collected included pa-tient demographic characteristics, diagnosis received, medicalhistory, type of surgery performed and anesthesia received,staff present during surgery, and type and volume of intra-venous solution(s) and medications administered during andafter surgery. Information about manufacturers’ lot numbersfor each intravenous solution or medication administered wascollected, if available, from pharmacy logbooks.

Hospitals B, C, and D. Similar information was collectedfor patients with possible adverse reactions who received in-travenous solutions or medications after surgery or admissionto the emergency department at hospitals B, C, and D duringthe study period. Each of these hospitals had documented atleast 1 possible adverse event in a patient who had receivedintravenous solution or medication during the study period.We also investigated the circumstances associated with thedeath of a patient after surgery at hospital E.

Laboratory Analysis

Unopened bottles of intravenous solutions or medicationsfrom lots implicated in the outbreak were randomly selectedfor microbiological testing, determination of physiochemicalcharacteristics, and determination of endotoxin concentrationusing a Limulus amebocyte lysate assay. The levels of endotoxinin large (ie, 500-mL) bottles of intravenous Ringer’s lactatesolution or 0.9% isotonic sodium chloride [ISC] solution andin small (ie, 100-mL) bottles of intravenous metronidazole wereevaluated with respect to the levels permitted in Brazil (lessthan 0.5 endotoxin units [EU]/mL for 500-mL bottles and lessthan 0.35 EU/mL for 100-mL bottles4).

Initial testing for detection of endotoxin was performed atthe Pernambuco State Public Health Laboratory by means ofa Limulus amebocyte lysate assay, and subsequent testing wasperformed at the National Institute for Quality Control inHealth, Oswaldo Cruz Institute–Ministry of Health.5 Intra-venous solution from previously unopened large bottles ofimplicated lots were also cultured via membrane filtration(the total volume of each bottle was analyzed) at the National

Centers for Infectious Diseases, Centers for Disease Controland Prevention. Filters were subsequently placed on R2A agarplates for 7 days to identify possible bacterial growth. Solu-tions were also analyzed using a kinetic Limulus amebocytelysate assay (Kinetix and Pyros Software; Associates of CapeCod) to determine the endotoxin concentration. The con-centration of the reagents was 7 EU/ng.

Statistical Analysis

Data abstracted from medical records were entered into a com-puterized database and analyzed using Epi Info, version 6.04d.6

The x2, Fisher exact, and Student t tests (for parametric data)and the rank sum test (for nonparametric data) were performedas appropriate. Relative risks (RRs), 95% confidence intervals(CIs), and P values were calculated, with a P value of less than.05 considered to be statistically significant.

results

Overall, 28 case patients were identified in a group of 355patients who underwent a surgical procedure or were treatedin the emergency department at study hospitals. Sixty (17%)of these 355 patients underwent surgical procedures at hos-pital A, of whom 7 (11.7%) were identified as case patients.

Hospital A

In hospital A, 7 (11.7%) of 60 patients who underwent sur-gery in March 2002 were identified as case patients, and 4(57.1%) of the 7 case patients died. The date the first casepatients were identified (March 1, 2002) coincided with thedate that a new lot of Ringer’s lactate solution produced bycompany A was administered to patients on the surgical wardof hospital A (Figures 1 and 2). The reason for use of thisnew brand was that intravenous solutions normally used andproduced by another company (company B) were in shortsupply in the marketplace. Pharmaceutical staff reported thatnearly all of the Ringer’s lactate solution from company Bhad been used on the surgical ward before March 1, 2002,but the exact date on which the last bottle of solution pro-duced by company B was used could not be confirmed duringreview of nursing records.

Case patients were older (mean age, 49.9 years [range, 23-81 years]) than non-case patients (mean age, 36.5 years[range, 15-80 years]) ( ), and there were no differ-P p .049ences between the groups according to sex ( ) (TableP p .61). Unexplained severe bleeding occurred in 5 (22.7%) of 22patients who received general anesthesia, compared with 2(5.3%) of 38 who received local anesthesia ( by theP p .056Fisher exact test). The mean interval between intraoperativeor postoperative administration of implicated intravenous so-lution(s) and onset of symptoms in case patients was 2.5hours (range, 1.3-4.0 hours).

The only intravenous solution in large bottles that wasadministered to all 7 case patients was Ringer’s lactate so-lution. Medical records for individual patients did not rou-



adverse reactions linked to contamination by endotoxin 737

figure 1. Date of illness onset or death for case patients after intravenous administration of Ringer’s lactate (RL) solution intrinsicallycontaminated with endotoxin, Hospital A, Pernambuco State, Brazil, 2003. A cross indicates that the case patient died.

tinely include details about the brand or lot number of in-travenous solution administered.

Table 2 summarizes the clinical characteristics and labo-ratory findings for the 7 case patients from hospital A. Allcase patients had hemorrhage at the surgical site or othersites and abnormal results of coagulation tests for prothrom-bin time and activated partial thromboplastin time.

A dose-response effect was observed: case patients receiveda significantly greater number of large bottles of intravenousRinger’s lactate solution intraoperatively (mean, 6.8 large bot-tles [range, 2-13 large bottles]), compared with non-case pa-tients (mean, 3.2 large bottles [range, 1-10 large bottles])( by the Student t test). An analysis of the quantityP p .001of all intravenous fluids administered during hospitalization(either during or after the operation) revealed that case pa-tients received significantly more Ringer’s lactate solution in-travenously (mean, 23.0 large bottles [range, 10-43 large bot-tles]), compared with non-case patients (mean, 3.7 largebottles [range, 1-13 large bottles]) ( by the StudentP ! .001t test). No similar dose-response effect was noted with othersolutions administered intravenously during or after the op-eration, including large bottles of ISC solution or 5% dextrosesolution ( ).P 1 .05

Several medications administered intravenously from bot-tles were associated with an increased risk of adverse reactionsat hospital A, but these medications were not given to all casepatients. For example, receipt of benzodiazepine and mor-phine were associated with an increased risk of adverse events,but 3 (42.9%) of 7 case patients who experienced hemorrhagedid not receive these medications. Medications in small bot-tles that were intravenously administered to surgical patientsat hospital A were not significantly associated with unex-plained hemorrhage.

Case patient status was not significantly associated with thepresence of any individual nursing, anesthesia, or surgicalstaff during surgery (data not shown). It was not possible toevaluate the risk for hemorrhage associated with each op-erating room, because information about operating rooms

was not systematically recorded. Anesthesia records duringsurgery were often incomplete with regard to vital sign mea-surements (eg, temperature, respiratory rate, heart rate, andblood pressure), which limited the possibility of accuratelyanalyzing these variables.

Other Hospitals

In hospital B, 74 patients received intravenous treatment withnondiluted metronidazole from small bottles, 5% dextrosesolution from large bottles, and/or Ringer’s lactate solutionfrom large bottles during the study period. Twenty-one pa-tients received nondiluted metronidazole intravenously, and9 (12.2%) had an adverse reaction (Figure 2). All 9 casepatients also received 5% dextrose solution and Ringer’s lac-tate solution. A dose-response effect was also noted with in-travenous metronidazole: case patients received significantlygreater volumes of metronidazole during the first 24 hoursof hospitalization ( by the Student t test) and duringP ! .01the entire hospitalization ( by the Student t test)P p .029(data not shown).

In hospital C, 4 (2.5%) of 161 patients who were treatedin the emergency department with intravenous fluids hadadverse reactions (Figure 2). The 4 case patients were froma group of 46 outpatients who received ISC solution intra-venously; none of the case patients died (Table 3). The onlyintravenous medication received by 100% of the case patientswas ISC solution from large bottles. Four (8.7%) of 46 out-patients treated intravenously with ISC solution experiencedadverse reactions. Case patients received a mean of 1.8 largebottles of intravenous ISC solution (range, 1-2 large bottles),whereas non-case patients received a mean of 1.1 large bottles(range, 1-2 large bottles) ( by the Student t test).P p .001

In hospital D, 7 (11.7%) of 60 patients undergoing surgeryexperienced adverse reactions (Figure 2), and 1 (14.3%) ofthe 7 case patients died (Table 3). Case patients received amean of 4.3 large bottles (range, 1-12 large bottles) of intra-venous Ringer’s lactate solution, whereas non-case patients




figure 2. Date of onset of adverse reactions or death for case patients after intravenous administration of solution intrinsically con-taminated with endotoxin, Hospitals A-D, Pernambuco State, Brazil, 2003. A cross indicates that the case patient died.

table 1. Demographic Characteristics and Type of Surgical Pro-cedure Among Case Patients and Non-Case Patients at Hospital Ain Pernambuco, Brazil, March 2002

CharacteristicCase patients

(n p 7)Non-case patients

(n p 53)

Age in years, mean (range) 50 (23-81) 37 (15-81)Female sex 60 60Type of surgical procedure

Gastrointestinal 14 35Gynecological and/or

obstetricalAny 43 31Cesarean section 100 47

Orthopedic 0 12Urologic 29 5Other 14 17

Overall 100 100

note. Data are percentage of patients, unless otherwise indicated. SeeMethods for definitions of “case patients” and “non-case patients.”

received a mean of 2.6 large bottles (range, 0-7 large bottles)( by the Student t test). A dose-response effect wasP p .05also observed with ISC solution: case patients received sig-nificantly more intravenous ISC solution (mean, 2.4 largebottles [range, 1-5 large bottles]), compared with non-casepatients (mean, 1.5 large bottles [range, 1-5 large bottles])( by the Student t test). All 7 of the case patientsP p .03received either Ringer’s lactate solution or ISC solution fromlarge bottles manufactured by company A. The attack rate ofadverse reactions was 13.5% among patients who receivedintravenous ISC solution from large bottles.

In hospital E, 1 case patient who underwent a cesareansection died on March 22, 2002. The patient had no historyof a bleeding disorder before surgery but experienced hem-orrhage during surgery. The only known exposure to intra-venous solution produced by company A involved a 10-mLbottle of distilled water. Company A had collected all intra-venous solutions and bottles at hospital E before this inves-tigation, so no products were available for laboratory testing.No additional adverse reactions were identified at hospital E.

Product Recall

On the basis of the epidemiological linkage between casepatient status and use of Ringer’s lactate solution producedby company A, on March 9, 2002, pharmaceutical staff col-lected all large bottles that had not been administered. Nosubsequent case patients were identified. On March 9, 2002,Pernambuco issued a state-level recall of all intravenous so-

lutions and metronidazole produced by company A. The man-ufacturing facility of company A, which was located in an-other Brazilian state, was inspected by federal investigators.A national recall was announced, and the factory was closed.

Endotoxin Testing

Limulus amebocyte lysate testing of previously unopenedlarge bottles from the implicated lot of Ringer’s lactate so-



table 2. Clinical Characteristics of and Laboratory Findings for Case Patients at Hospital A, Pernambuco, Brazil, 2002

Case patient Age, y DiagnosisType ofsurgery

Site(s) of hemorrhageafter surgery

Time fromRL use to

hemorrhage, h

Coagulation measurements

Clinicaloutcome

PC, #103

platelets/mL HT, % PT, s PTT, sFL,

mg/dL

Patient 1Episode 1 23 Pregnancy C-section Vagina 11.6 86 34.0 16 … 130 …Episode 2 … … … … 57.0 62 17.0 16 58 225 Cure

Patient 2 25 Pregnancy C-section Vagina 39.4 92 33.0 25 51 355 CurePatient 3 31 Pregnancy C-section Vagina 1.4 76 20.0 20 58 195 DeathPatient 4 43 Breast microcalcification Mastectomy Breast (female) 20.7 58 30.0 20 36 225 CurePatient 5

Episode 1 70 Adenocarcinoma of prostate Prostatectomy Rectum 17.0 77 25.0 21 160 225 …Episode 2 … … … … 24.3 92 18.0 26 58 100 Death

Patient 6 76 Benign prostatic hyperplasia Prostatectomy Rectum, urinary tract,and mouth

1.7 77 NA 24 62 130 Death

Patient 7 81 Diverticulitis of colon Colectomy Rectum 29.0 309 21.0 16 46 180 Death

note. See Methods for the definition of “case patients.” C-section, cesarean section; FL, fibrinogen level (normal range, 190-510 mg/dL); HT, hematocrit (normal range, 28%-42%;NA, not available; PC, platelet count (normal range, to platelets/mL; PT, prothrombin time (normal range, !12 s); PTT, activated partial thromboplastin time3 3150,000 # 10 440,000 # 10(normal range, 10-30 s).




table 3. Rates of Adverse Reactions (AEs) and Mortality and Data on Exposure to ParenteralSolutions or Medications Implicated in Case Patient Deaths at Hospitals A-D in Pernambuco, Brazil,2002

Hospital

Outcome, proportion (%) Solution or medication implicated in deatha

Patientswith an AE

Case patientswho died Typeb

Case patientsexposed, % P

A 7/60 (11.7) 4/7 (57.1) Ringer’s lactate 100 .08B 9/74 (12.2) 0 (0) Metronidazole 100 !.001C 4/161 (2.5) 0 (0) Isotonic sodium chloride (0.9%) 100 .006Dc 7/60 (11.7) 1/7 (14.3) Ringer’s lactate 100 .52Dc 7/60 (11.7) 1/7 (14.3) Isotonic sodium chloride (0.9%) 100 .39a The relative risk of exposure is equal to infinity for each solution or medication, because all case patients wereexposed to the implicated solutions (ie, the denominator is 0).b All implicated parenteral solutions were produced by company A.c The same patients received both solutions.

lution produced by company A showed very high concen-trations of endotoxin. The mean concentration was 88.3 EU/mL (range, 9.7-298.0 EU/mL), compared with the limit ofless than 0.5 EU/mL permitted in Brazil. Testing of largebottles of ISC solution showed no endotoxin contamination.Bacterial cultures of solution from unopened large bottles ofRinger’s lactate solution and ISC and cultures of samples frombottles of distilled water were negative at laboratories in bothBrazil and the United States. Implicated lots of metronidazoleproduced by company A and administered intravenously athospital B also had markedly elevated concentrations of en-dotoxin. The mean endotoxin concentration in unopenedbottles from the implicated lot was 28.3 EU/mL (range, 5.0-58.3 EU/mL), compared with the legal limit in Brazil of lessthan 0.35 EU/mL.

discussion

The pyrogenic reactions and life-threatening hemorrhagingin surgical patients and outpatients at several hospitals inPernambuco described in this report can most likely be at-tributed to intrinsic endotoxin contamination of bottles ofintravenous solution produced by company A. In surgicalpatients, severe hemorrhage was observed; and in multiplecase patients, fluid challenges and transfusion of blood prod-ucts did not have the expected effect of controlling bleedingand correcting severe hypotension. The lack of accurate in-traoperative recording of vital signs (including body tem-perature, heart rate, and blood pressure) limited the abilityof investigators to determine the frequency of documentedpyrogenic reactions that occurred before the unexplainedcases of hemorrhage in surgical case patients.

Endotoxin is a normal component of the cell wall of gram-negative bacteria and typically produces febrile type reactionswhen introduced intravenously at dosages exceeding 1 ng/kgof body weight per hour.7,8 Endotoxin may also initiate severebleeding by activating coagulation factor XII in the intrinsiccomponent of the complement system.9,10 Gram-negative bac-teria are another possible cause of coagulopathy, but no bac-teremia was documented in the case patients in this study,

and laboratory testing of implicated intravenous solutionsshowed no bacterial contamination. One might speculate thatthe high mortality rate observed in hospital A may have beenthe result of initial normal consumption of platelets and co-agulation factors during surgery, which was compounded bythe intravenous administration of intravenous solutions withintrinsic endotoxin contamination, with an outcome of un-controllable hemorrhage. Treatment of hypotension or shockwould normally be corrected by intravenous challenge withadequate isotonic solution,11 but administration of endo-toxin-contaminated fluids intraoperatively in hospital A mayhave actually compounded the severity of symptoms, includ-ing bleeding.

Laboratory testing confirmed high endotoxin concentra-tions in lots of Ringer’s lactate solution and metronidazolemanufactured by company A that were epidemiologicallylinked to case patients. The state-level surveillance system forpharmaceutical products was very useful in promptly iden-tifying the outbreak and facilitated notification of hospitalsstatewide by health authorities. A national recall was an-nounced by the National Agency for Sanitary Surveillance(the Brazilian equivalent of the US Food and Drug Admin-istration) for company A’s products distributed to Pernam-buco and other states in Brazil. An inspection of the plantwas conducted, and the manufacturing facility was closed,but a specific reason for contamination was not identified.None of the products manufactured by company A had beenexported, but the recall likely prevented additional morbidityand mortality, and no additional case patients were identifieddespite continued surveillance activities.

Pharmaceutical manufacturers of intravenous solutions inBrazil are required to analyze random samples of lots beingproduced to demonstrate compliance with federal regulationsregarding endotoxin levels. However, this outbreak was iden-tified by clinicians, rather than during quality control testingat the manufacturing level. Shortly after the outbreak, a rou-tine program was initiated in Pernambuco to monitor thequality of intravenous solutions for contamination, by testing



adverse reactions linked to contamination by endotoxin 741

a random sample of lots available in the commercial mar-ketplace (including those produced by other states).

Address reprint requests to Luciane Zappelini Daufenbach, Secretaria deVigilancia em Saude, Esplanada dos Ministerios, Bloco “G,” Edifıcio Sede, 1�

Andar-Ala Norte-Sala 137, Brasılia, DF 70058-900, Brazil ([email protected]).

acknowledgments

We thank the following persons from Pernambuco for their essential contri-butions to the study: Zuleide D. Wanderley, Maria Aparecida S. Elisbao, RosaA. Vasconcelos, Odin B. da Silva, Maria Helena D. G. dos Santos, TatianaBruscky, Lucia de Fatima, F. da Silva (State Secretary of Health, Pernambuco,Brazil), and the clinical staff at Casa de Saude Santa Efigenia, Hospital Jaboataodos Prazeres, Hospital Regional do Agreste, Hospital Santa Clara, and HospitalSao Lucas. In addition, we acknowledge the strong support of the followinghealth professionals at the national level: Reginaelena F. da Silva (Oswaldo CruzFoundation, Rio de Janeiro, Brazil); Mirtha Susana, Yamada Tanaka, and FlaviaQueiros Leite (National Agency of Sanitary Surveillance, Brasilia, Brazil); andDaniel Jernigan (National Centers for Infectious Diseases).

Use of the manufacturer name “Kinetix and Pyros Software” does notimply endorsement by the Ministry of Health (Brazil) or the Departmentof Health and Human Services (United States).

references

1. Seibert FB. Fever-producing substance found in some distilled waters.Am J Physiol 1923; 64:90-104.

2. Seibert FB. Pyrogens from an historical viewpoint. Transfusion 1963; 80:245-249.

3. Suffredini AF, Fromm RE, Parker MM, et al. The cardiovascular responseof normal humans to the administration of endotoxin. N Engl J Med1989; 321:280-287.

4. Santos F, Santos AG, Biernat JC, et al. Deteccao de Endotoxina pelo Li-mulus Amoebocyte Lysate (LAL) em Unidades de Hemodialise. RevistaVirtual de Medicina. Edicao 8. 2000. Available at: http://www.medonline.com.br/med_ed/med8/endotox.htm. Accessed June 1, 2006.

5. Ministerio da Saude. Manual de Ensaio para endotoxina bacteriana. Riode Janeiro, Brazil: Fundacao Oswaldo Cruz; 2001.

6. Dean AG, Dean JA, Coulombier D, et al. Epi Info, Version 6: A Word-Processing, Database, and Statistics Program for IBM-Compatible Micro-computers. Atlanta: Centers for Disease Control and Prevention; 1995.

7. Tsuji K, Steindler KA, Harrison SJ. Limulus amoebocyte lysate assay fordetection and quantitation of endotoxin in a small-volume parenteralproduct. Appl Environ Microbiol 1980; 40:533-538.

8. Hurley JC. Reappraisal of the role of endotoxin in the sepsis syndrome.Lancet 1993; 341:1133-1135.

9. Young LS. Sepsis syndrome. In: Mandell GL, Bennett JE, Dolin R, eds.Principles and Practice of Infectious Diseases. 5th ed. New York: ChurchillLivingstone; 2000:806-819.

10. Bone RC. The pathogenesis of sepsis. Ann Intern Med 1991; 115:457-469.

11. Food and Drug Administration. Guideline on validation of the limulusamebocyte lysate test as an end-product endotoxin test for human andanimal parenteral drugs, biological products, and medical devices. De-cember 1987. Available at: http://www.fda.gov/cber/gdlns/lal.pdf. Ac-cessed June 1, 2006.



pyrogenic reactions and hemorrhage associated with intrinsic exposure to endotoxin‐contaminated...

Documents