- Email: [email protected]
Bacteriologic and endotoxin analysis of salvaged blood used in autologous transfusions during cardiac operations
Bacteriologic and endotoxin analysis of salvaged blood used in autologous transfusions during cardiac operations Autologous blood transfusion is a common method of reducing the need for heterologous blood transfusion during cardiac operations. Recently we investigated an outbreak of severe, nonsurgical postoperative bleeding among patients undergoing heart operations and receiving intraoperative transfusion of blood from a cell conservation device (Cell Saver System, Haemonetics Corp., Braintree, Mass.), As a result of this investigation, we conducted a prospective study to determine if bacterial or endotoxin contamination of the blood collected in the Cell Saver System and used for reinfusion during heart operations contributes to postoperative bleeding complications. Patients' blood samples were collected immediately before operation, at the end of cardiopulmonary bypass, 1 hour postoperatively, and from the Cell Saver System. All blood samples were cultured for bacteria, and all plasma samples were assayed for endotoxin. Preoperatively all patients having heart operations were without signs of infection, 33 of 37 blood cultures taken were negative, and none of the plasma samples had detectable endotoxin. After discontinuance of cardiopulmonary bypass but before delivery of blood from the Cell Saver System, bacteria and endotoxin were detected in 11 of 36 (30.6 %) and five of 35 (14.3%) of the patients' blood samples, respectively. The blood aspirated from the open chest and collected by the Cell Saver System was culture positive in 30 of 31 (96.8%) samples, and seven of 29 (24.1 %) contained endotoxin. One of 28 blood samples collected 1 hour postoperatively was culture positive, and five of 25 samples contained endotoxin. Of 61 total microorganisms isolated, 50 (82 %) were coagulase-negative staphylococci, four (6.6 %) aerobic diphtheroids, five (8.2 %) anaerobic "diphtheroids" (Propionibacterium acnes), and two (3.2 %) gram-negative bacilli. Plasma endotoxin concentrations ranged from 10 to 765 pgjml. No signs of endotoxemia or unusual bleeding were observed intraoperatively or postoperatively in any of the 38 patients. Although blood collected in the Cell Saver System and used for reinfusion during heart operations often was contaminated with gram-positive bacterial commensals of the skin and low concentrations of endotoxin, no adverse effects were noted in the patients. (J THoRAe CARDIOVASC SURG 1992;103:582-8)
Lee A. Bland, MA, MPH,a Margarita E. Villarino, MD, MPH,b Matthew J. Arduino, MS, DrPH,a Sigrid K. McAllister, BS,a Steven M. Gordon, MD,b Charles T. Uyeda, PhD,c Carol Valdon, MS, RN,d Diana Potts, MS, RN,d William R. Jarvis, MD,b and Martin S. Favero, Phl)," Atlanta, Ga., and Palo Alto, Calif.
From the Nosocomial Infections Laboratory Branch" and the Epidemiology Branch," Hospital Infections Program, Centers for Disease Control, Public Health Service, U.S. Department of Health and Human Services, Atlanta, Ga., and the Microbiology Laboratory" and Infection Control/' Palo Alto Veterans Administration Medical Center, Palo Alto, Calif. Received for publication July 5, 1990. Accepted for publication Dec. 12, 1990. Address for reprints: Lee A. Bland, MA, MPH, Nosocomial Infections Laboratory Branch, Hospital Infections Program, COl, Centers for Disease Control, Atlanta, GA 30333.
12/1/28363
582
Since the 1970s the number of patients undergoing cardiac operations with cardiopulmonary bypass (CPB) has rapidly increased. Currently more than 350,000 cardiac operations are performed annually in the United States. I Volume replacement during heart operations often requires the use of banked blood components. Fear of transfusion-related diseases (e.g., human immunodeficiency virus, hepatitis C virus, cytomegalovirus, and Yersinia enterocolitica) and reactions (isosensitization or hemolytic and febrile reactions) from heterologous blood transfusions have led to research on artificial blood and increased use of autologous transfusion.
Volume 103 Number 3 March 1992
Salvaged blood used in autologous transfusions
583
Some investigators have had success with autologous blood transfusion both clinically and in reducing blood banking expenses, whereas others have found autologous blood transfusion safe but with little to no reduction in the use of blood bank components.v'" Still others have questioned the suitability of using salvaged blood as an alternative to donated blood. 1I, 12 Adan and coworkers 13 found that salvaged autologous blood for reinfusion contained a considerable amount of small-sized debris; they concluded that the dynamics of the coagulation process in the mediastinum do not favor autologous blood transfusion and that it should not be performed on a routine basis. Another study reported the detection of circulating endotoxin in the extracorporeal circuit during CPB, but no intraoperative or postoperative complications were noted in the 10 patients studied.!" The endotoxin contamination was determined to be a result of the CPB procedure and the intraoperatively administered fluids. Schwieger and colleagues'> found no risk of infection in cardiac surgical patients from blood collected in a cell conservation apparatus. We recently investigated an outbreak of massive nonsurgical postoperative bleeding after cardiac operations in which salvaged autologous blood was routinely transfused. As a result of this outbreak, we conducted a prospective study of cardiac operations to determine the prevalence of microbial and endotoxin contamination of the patients' blood collected during preoperative, intraoperative, and postoperative periods and of blood aspirated from the open chest cavity during the operation and collected by the Cell Saver System (Haemonetics Cell Saver System, Haemonetics Corp., Braintree, Mass.* Use of autologous blood transfusion was assessed as to its association with bacteremia, postoperative bleeding, and endotoxic shock.
Bacterial culture. Bacterial cultures were performed by inoculating 3 to 5 ml of sample into Bactec (Johnson Laboratories, Towson, Md.) aerobic 16B and anaerobic 17D blood culture bottlescontaining resin.These bottles were incubated at 35° C for 7 days and read on the Bactec 460 instrument as follows: The aerobic bottles were tested twice during the first 24 hours of incubation and onceon days 2, 3, 4,5, and 7; anaerobic bottles were tested once on days 2, 3, 4,5, and 7. Bottles registering thresholdgrowth indicesgreater than or equal to 20 were subjected to Gram stain and subcultured to a tube of enriched thioglycolate. Gram stain-"positive" bottles were subsequently subcultured to chocolate,phenylethanol,and MacConkey agar plates. Coagulase tests on staphylococci were performed with reconstituted rabbit plasma (BBL Microbiology Systems, Cockeysville, Md.) by the slide and tube methods. Certain staphylococci were speciated by using the gram-positive identification cards on the AutoMicrobic System (Vitek Systems, Hazelwood, Mo.) or by the Staph-Ident (Analytab Products, Plainview, N.Y.) method. Gram-negativebacilliwere identified by the gram-negative identification cards (Vitek Systems) on the AutoMicrobic System and anaerobes by the An-Ident (Analytab Products, Inc.) methods. Endotoxin analysis. Detection and quantification of endotoxin were done by the Limulus amebocyte lysate (LAL) turbidimetric assay method (LAL-5000; Associatesof Cape Cod, Woods Hole, Mass.) (Associates of Cape Cod, LAL Update 5:2, 1987).16,17 An Escherichia coli 0113 control standard endotoxin (lot 36; Associates of Cape Cod) was used for all standard curves, positivecontrol procedures, and product inhibition determinations.lv 19 Endotoxin-free polypropylene microcentrifuge tubes, containing I % heparin sodiuminjection(USP) as an anticoagulant, were used for collectionof blood samples. Blood samples were centrifuged for 3 minutes, and the plasma was removed. The plasma wasdiluted I: I0 with steriledistilledwater for irrigation (Abbott Laboratories, Chicago, Ill.) and heated at 70° C for 10 minutes to remove plasma endotoxin test inhibitors.Pj" After being heated, the samples were again diluted 1:10 with sterile distilledwater (Abbott) to make a final dilution of I: I00. CPB priming fluidwas collectedin endotoxin-freepolystyrenetubes. The priming fluid was diluted I:10 with sterile distilled water (Abbott) before analysis.
Materials and methods
Results
BetweenSeptember 1988and February 1989,bloodsamples from all patients undergoing cardiac operations were collected and assayed for bacterial and endotoxin contamination. The following samples were collectedfrom each patient: (I) preoperative blood drawn through an arterial line in the operating room before the operation and before prophylactic antibiotics were given; (2) blood drawn intraoperatively through an arterial lineimmediatelyafter the discontinuationof CPB; (3) blood collected percutaneously in the surgical intensive care unit I hour postoperatively; and (4) blood from the reservoirbag, collected by the Cell Saver System during the operation, immediately before reinfusion. In addition, samplesof Ringer's lactate solution used to prime the extracorporeal circuit of the CPB equipment were collectedfor bacterial and endotoxin analysis.
Thirty-eight patients were enrolled in the study. Patients were supported by CPB for a mean time of 125 minutes (range 20 to 243 minutes), and the mean length of operation was 299 minutes (range 155 to 470 minutes). Bacteria were isolated from 34.8% (46/132) of all collected blood samples, and 13.7% (17/124) had detectable endotoxin (i.e., > 10 pg/rnl) (Table I). Bacterial cultures were positive in 10.8% (4/37) of the preoperative samples, 30.6% (11/36) of the intraoperative samples, 3.6% (1/28) of the postoperative samples, and 96.8% (30/31) of the blood samples taken from the Cell Saver System. The predominant microorganisms isolated were coagulase-negative staphylococci, which included Sta-
"Use of trade names and commercial sources is for identification only and does not imply endorsement by the Public Health Service or the U.S. Department of Health and Human Services.
phylococcus epidermidis, S. capitis, S. warnerii, S. haemolyticus, S. hominis, S. simulans, and S. saccharolyticus (Table II). Several of the cultures were polymicrobic,
584
The Journal of Thoracic and Cardiovascular
Bland et al.
Surgery
Table I. Bacterial and endotoxin assay ofpatient and Cell Saver System blood and priming fluid during cardiac operations
Blood Preoperative Intraoperative Cell Saver System Postoperative All blood Priming fluid Total
Bacterial cultures
Endotoxin assay
No. positive/No. of tests (%)
No. positive/No. of tests (%)
4/37 (10.8) 11/36 (30.6) 30/31 (96.8) 1/28 (3.6) 46/132 (34.8) 0/38 (0) 46/169 (27.2)
and many consisted of more than one type of coagulasenegative staphylococci, coagulase-negative staphylococci and diphtheroids, or coagulase-negative staphylococci and gram-negative bacilli. Of the two gram-negative bacilli polymicrobic cultures isolated from the Cell Saver System, one was positive for Enterobacter agglomerans and the other for Pseudomonas aeruginosa. All bacterial cultures of the CPR priming fluid were negative. Endotoxin was detected in 10.7% of all samples collected (Table I). None of the patients' blood samples had detectable plasma endotoxin before the operation. Endotoxin was detected in five of 35 (14.8%) of the intraoperative samples (15 to 26 pg/rnl), five of 25 (20%) of the postoperative samples (12 to 44 pg/rnl), and seven of 29 (24.1 %) of the blood samples from the Cell Saver System (10 to 765 pg/rnl) (Table 11). Endotoxin was not detected in the CPB priming fluid. No adverse clinical signs or symptoms, including postoperative bleeding or endotoxemia, were detected postoperatively in any of the 38 study patients despite the frequent isolation of bacteria and detection of endotoxin, especially in blood from the Cell Saver System. Discussion
Recovery and reinfusion of blood aspirated from the open chest cavity during cardiac operations has many benefits, including reduction in the risk of infectious disease transmission from donor blood to the host patient, avoidance of isosensitive, hemolytic, and febrile reactions, and possible reduction in blood bank services.21• 22 One recovery method involves the use of a cell conservation apparatus, which washes and removes excess water, heparin, serum hemoglobin, bacteria, potassium, and cellular debris from the recovered blood cells while maintaining blood sterility.' However, several studies have isolated bacteria from blood retrieved from a cell conservation
0/35 (0) 5/35 (14.3) 7/29 (24.1) 5/25 (20) 17/124 (13.7) 0/35 (0) 17/159 (10.7)
apparatus or other autologous blood collection de-
vices.U: 15.22
Blood collected from the reservoir bag of the Cell Saver System during operations rarely was found to be sterile in this study. Ninety-seven percent of the samples taken of the blood aspirated from the open chest and recovered by the Cell Saver System were culture positive. Kluge, Calia, and Mclaughlin-' found that up to 60% of cultures taken from the open chest cavity from patients undergoing cardiac operations were contaminated with microorganisms. The most common organisms isolated were diphtheroids and coagulase-negative staphylococci. Other studies report 20% or less of salvaged blood yields positive cultures. I I. 22 Kluge, Calia, and Mclaughlin-' suggested that the frequent contamination of the perioperative site during heart operations was probably due to airborne contaminants. Blakemore and coworkers-" found that during cardiac operations the greatest density of airborne microbes occurred directly above the operative site and that coagulase-negative staphylococci were the predominant organisms isolated. Other organisms isolated in their study included Enterobacter sp., Mima sp., P. aeruginosa, Serratia sp., Aspergillus sp., and Candida sp. The chief source of these organisms was the operating room personnel. The number of isolates increased with the number of persons in the operating room, the degree of traffic, and the number of times the door to the operating room was opened. Other investigators suggest that organisms isolated from drained mediastinal blood result from the manipulation of the drained blood before reinfusion or specimen contamination by the laboratory.P II In our study the majority of organisms isolated from blood collected by the Cell Saver System were common skin commensals; 35 of 44 (79.5%) were coagulase-negative staphylococci, and nine of 44 (20.5%) were aerobic or anaerobic diphtheroids. Such microorganisms are part of the normal skin flora and commonly are found in the
Volume 103 Number 3 March 1992
Salvaged blood used in autologous transfusions
585
Table II. Bacterial isolates and endotoxin (pgjml) concentrations in patient and Cell Saver System blood and priming fluid during cardiac operations Patient No.
Preop. blood
lntraop. blood
Bact.
Bact.
1 2 3 4 5 6 7 8
Endo.
Cell Saver System blood Endo.
CNS
CNSt
*
P. acnes
P. acnes
ND:j:
9
CNS (3 strains) CNS ND
S. epidermidis S. epidermidis
10
26
CNS ND
ND
ND
S. capitis S. epidermidis
CNS
S. capitis CNS, P. acnes S. haemolyticus S. epidermidis S. saccharolyticus S. simulans S. epidermidis S. epidermidis (2 strains) S. epidermidis P. acnes
S. epidermidis
ND
S. warnerii, S. epidermidis S. epidermidis ND
25 26 27 28 29 30 31 32 33 34 35 36 37 38
ND
S. hominis S. hominis, S. epidermidis CNS, E. agglomerans Diphtheroids, S. haemolyticus, S. capitis S. epidermidis S. epidermidis, P. aeruginosa
Diphtheroids CNS
11
12 13 14 15 16 17 18 19 20 21 22 23 24
Bact.
ND
S. capitis, S. epidermidis, S. warnerii
ND CNS, P. acnes
CNS CNS
Diptheroids ND
S. epidermidis
S. epidermidis, S. warnerii
17
CNS ND ND
20 15
S. epidermidis, diphtheroids S. epidermidis
18
S. epidermidis S. epidermidis
ND
ND ND
Postop. blood Endo.
Bact.
14 18
Endo.
Prime fluid
Bact. Endo.
25
ND
ND S. epidermidis
19
ND
ND
ND ND ND ND ND ND
ND 21 ND ND ND ND ND ND
ND
ND
ND
ND
ND 14 765 ND ND 205 10
81
ND ND
ND ND ND ND
ND ND ND ND 44 12 ND ND
*Blank entry indicates a negative result. tCoagulase-negative staphylococci (unspeciated). iAssay not done.
air and on environmental surfaces, even in the operating room.23-25 Such ubiquitous organisms are also common laboratory isolates frequently considered contaminants.l? 26 However, Kohman, Coleman, and Parker.F' in a recent study on bacteremia after coronary artery bypass grafting, found that coagulase-negative staphylococci accounted for 55% of the positive postoperative blood cultures. Gram-negative bacilli were isolated in
only two of the samples taken from the Cell Saver System. We could not determine if the samples from the Cell Saver System were contaminated during the surgical procedures by endogenous or exogenous sources or if the samples were contaminated during collection and subsequent laboratory manipulation. However, because blood samples obtained through the arterial lines preoperative-
5 8 6 Bland et al.
Iy, intraoperatively, and postoperatively were assayed in the same manner but did not have the same high contamination rate, the samples from the Cell Saver System probably were not contaminated during collection or laboratory manipulation but during the actual recovery of the blood from the open chest cavity. The fact that all samples of CPB priming fluid were culture negative would also support this hypothesis. One patient had three successive blood cultures (preoperative, intraoperative, and Cell Saver System) positive for coagulase-negative staphylococci and suggestive of bacteremia, but, unfortunately, blood for culture was not drawn immediately postoperatively on this patient. The patient received antimicrobial prophylaxis and did not exhibit any clinical sign of bacteremia postoperatively, however. Blood cultures taken 6 days later were negative. No other patient had successive positive cultures between preoperative and intraoperative samples or intraoperative and postoperative samples. Intraoperative cultures obtained at the end of CPB were positive in 30.6% of the cases studied. No correlation was found between positive intraoperative cultures and the duration of CPB support or the duration of the surgical procedure. Although several patients had positive blood cultures, none had adverse clinical signs or symptoms. Similar findings have been reported in other studies. 10,22 This may be due to the relatively low frequency of reported sternal wound complications after cardiac operations. In a recent study by Loop and coworkers'" of 6504 patients who underwent isolated coronary artery bypass grafting, 1.1% had sternal wound complications. They reported even lower wound infection rates for valve operations (0.4%), and when coronary artery bypass grafting was combined with other cardiac procedures wound infection rates were only 1.0%. It is not surprising that such complications were not seen in our study involving only 38 patients. Plasma endotoxin was not detected in preoperative blood samples or samples of CPB priming fluid. A similar study by Andersen and colleagues 14 of 10 cardiac operations did not find detectable endotoxin in patients preoperatively but did find that 30% of the CPB priming fluid (Ringer's lactate solution) samples had measurable endotoxin concentrations (mean 11 pg/rnl). In another study of nine patients having heart operations, it was found that, the mean preoperative plasma endotoxin concentration was 128 ± 32 pgjml, with concentration increasing during the surgical procedure to 428 ± 60 pgjml and finally decreasing to baseline levels postoperatively.-? A similar trend was noticed in our study, in which endotoxin concentrations, although generally low,
The Journal of Thoracic and Cardiovascular Surgery
increased during the procedure and decreased by the end of the procedure and postoperative recovery. The greatest percentage of positive assays and the highest plasma endotoxin levels (765,205, and 81 pgjml) were in blood samples taken from the Cell Saver System. It has been suggested that, during cardiac operations, depression of the reticuloendothelial system allows endotoxin that leaks into the portal circulation from the gut to enter the systemic circulation. 30-32 Another study found that if pyrogen-free water was circulated through new extracorporeal circuits, small amounts of material (3 to 6 pg) would give a positive LAL endotoxin test.!" It was speculated that cellulosic material in the oxygenator and the cardiotomy reservoir was the source of this LAL-reactive material and not gram-negative bacilli. Similar results have been reported during the use of artificial kidneys.P:34 Both of these reasons could explain the greater endotoxin concentrations of blood collected in cell conservation devices. None of the study patients had any signs or symptoms of endotoxemia, consistent with the low concentrations of endotoxin found in the blood samples. Endotoxin in concentrations of 1000 pgjkg of body weight is generally necessary to produce pyrogenic reactions in human beings, and doses of 2000 pgjkg of body weight administered intravenously produce granulocytosis in addition to a febrile response. 33, 35 The concentrations of endotoxin found in our samples were far below these threshold values. In conclusion, our study shows that bacterial and endotoxin contamination of blood from cell conservation devices used for autologous transfusion during cardiac operations is frequent. Bacterial contamination may be caused by airborne contaminants that either fall or are pulled into the operative field by air currents created by suction tubes of the cell conservation device or the manipulation of the salvaged blood before reinfusion, as hypothesized by Bennett!' and others. 24, 36 However, despite bacterial and endotoxin contamination of blood recovered by the Cell Saver System during operations and its subsequent reinfusion, no episodes of sepsis or endotoxic shock were detected in any of our patients; transient bacteremia was rarely detected. The sources of microbial contamination could not be identified but, because the organisms isolated were ubiquitous skin commensals, contamination probably occurred during recovery of the blood from the open incision. The presence of endotoxin in intraoperative and postoperative blood may be due to intestinal leakage of endotoxin into the portal circulation. Endotoxin contamination of blood from the Cell Saver System may also be a result of intestinal endotoxin leak-
Volume 103 Number 3 March 1992
age, but it also may be due to LAL-reactive material in the extracorporeal circuit. Additional studies are needed to determine definitely if endotoxin or bacterial contamination of blood collected by a cell conservation apparatus during operations and its subsequent reinfusion plays a role in postoperative pyrexia and other commonly recognized complications of cardiac operations. Although blood from the Cell Saver System had a high potential for bacterial and endotoxin contamination in this study, its use did not appear to elicit adverse effects in the 38 patients studied. REFERENCES 1. Graves EJ. 1989. National Hospital Discharge Survey: Annual summary, 1987. Hyattsville, Md. National Center for Health Statistics. Vital Health Stat 13(99). 2. National Blood Resource Education Program Expert Panel. The use of autologous blood. JAMA 1990;263:414-7. 3. Cordell AR, Lavender SW. An appraisal of blood salvage techniques in vascular and cardiac operations. Ann Thorac Surg 1980;31 :421-5. 4. Emminizer S, Klopp EH, Hauer JM. Autotransfusion: current status. Heart Lung 1981;10:83-7. 5. Johnson RG, Rosenkrantz KR, Preston RA, Hopkins C, Daggett WM. The efficacy of postoperative autotransfusionin patients undergoing cardiac operations. Ann Thorac Surg 1983;36:173-9. 6. Schaff HV, Hauer JM, Brawley RK. Autotransfusion in cardiac surgical patients after operation. Surgery 1978; 84:713-8. 7. Schaff HV, Hauer J, Gardner TJ, et al. Routine use of autotransfusion following cardiac surgery: experience in 700 patients. Ann Thorac Surg 1979;27:493-9. 8. Weniger J, von der Emde J, Schricker KT, Blechschmidt J. Retransfusion von Drainageblut nach herzchirurgischen Eingriffen. Langenbecks Arch Chir 1980;351:229-41. 9. Weniger J, Shanahan R. Reduction of bank blood requirements in cardiac surgery. Thorac Cardiovasc Surg 1982; 30:142-6. 10. Thurer RL, Lytle BW, Cosgrove DM, Loop FD. Autotransfusion following cardiac operations: a randomized, prospective study. Ann Thorac Surg 1979;27:500-7. 11. Bennett JG. Autotransfusion of drained mediastinal blood. Thorac Cardiovasc Surg 1982;30:28-30. 12. Carter RF, McArdle B, Morritt GM. Autologous transfusion of mediastinal drainage blood. Anaesthesia 1981; 36:54-9. 13. Adan A, Brutel de la Riviere A, Haas F, van Zalk A, de Nooij E. Autotransfusion of drained mediastinal blood after cardiac surgery: a reappraisal. Thorac Cardiovasc Surg 1988;36:10-4. 14. Andersen LW, Baek L, Degn H, Lehd J, Krasnik M, Rasmussen JP. Presence of circulating endotoxins during cardiac operations. J THoRAc CARDIOVASC SURG 1987; 93:115-9.
Salvaged blood used in autologous transfusions
587
15. Schwieger 1M, Gallagher CJ, Finlayson DC, Daly WL, Maher KL. Incidence of Cell-Saver contamination during cardiopulmonary bypass. Ann Thorac Surg 1989;48: 51-3. 16. Novitsky TJ, Remillard JF, Loy N. Design criteria and evaluation of the LAL-4000 for kinetic turbidimetric LAL assay. In: Watson SW, Levin J, Novitsky TJ, eds. Detection of bacterial endotoxins with the Limulus amebocyte lysate assay. New York: Alan R Liss, 1987:189-96. 17. Remillard JF, Gould MC, Roslansky PF, Novitsky TJ. Quantitation of endotoxin in products using LAL kinetic turbidimetric assay. In: Watson SW, Levin J, Novitsky TJ, eds. Detection of bacterial endotoxins with the Limulus amebocytelysateassay. New York: Alan R Liss, 1987:197210. 18. Levin J, Tomasulo PA, Oser RS. Detection of endotoxin in human blood and demonstration of an inhibitor. J Lab Clin Med 1970;75:903-11. 19. Novitsky TJ, Roslansky PF, Siber GR, Warren HS. Turbidimetric method for quantifying serum inhibition of the Limulus amoebocyte lysate. J Clin Microbiol 1985;21: 211-6. 20. Sturk A, Janssen ME, Muylaert FR, Joop K, Thomas LLM, Cate JW. Endotoxin testing in blood. In: Watson SW, Levin J, Novitsky TJ, eds. Detection of bacterial endotoxins with the Limulus amebocyte lysate assay. New York: Alan R Liss, 1987:371-85. 21. Cosgrove DM, Thurer RL, Lytle BW, et al. Blood conservation during myocardial revascularization. Ann Thorac Surg 1979;28:184-9. 22. Schaff HV, Hauer JM, Bell WR, et al. Autotransfusion of shed mediastinal blood after cardiac surgery. J THORAC CARDIOVASC SURG 1978;75:632-41. 23. Kluge RM, Calia FM, Mclaughlin JS. Sources of contamination in open heart surgery. JAMA 1974;230:1415-8. 24. Blakemore WS, McGarrity GJ, Thurer RJ, Wallace HW, MacVaugh H III, Coriell LL. Infection by airborne bacteria with cardiopulmonary bypass. Surgery 1971;70:830-8. 25. Gallis HA. Microbial ecology and normal flora of the human body. In: Joklik WK, Willet HP, Amos DB, eds. Zinsser's Microbiology. 17th ed. East Norwalk, Conn., Appleton & Lange, 1980;524-6. 26. Branson D. How to culture for a specificorganism. In: Procedure manual for clinical bacteriology. Springfield: Charles C Thomas, 1982:168-320. 27. Kohman LJ, Coleman MJ, Parker FB Jr. Bacteremia and sternal infection after coronary artery bypass grafting. Ann Thorac Surg 1990;49:454-7. 28. Loop FD, Lytle BW, Cosgrove DM, et al. Sternal wound complications after isolated coronary artery bypass grafting: early and late mortality, morbidity, and cost of care. Ann Thorac Surg 1990;49:179-87. 29. Rocke DA, Gaffin SL, Wells MT, Koen Y, Brocke-Utine JG. Endotoxernia associated with cardiopulmonary bypass. J THORAC CARDIOVASC SURG 1987;93:832-7. 30. Freeman R, Gould FK. Rises in antibody to enteric gram
The Journal of
5 8 8 Bland et al.
negative bacilli after open heart surgery: a possible mechanism for postoperative pyrexia. Thorax 1985;40:53841. 31. Subramanian V, Mcleod J, Gans H. Effect of extracorporeal circulation on reticuloendothelial function. I. Experimental evidence for impaired reticuloendothelial function following cardiopulmonary bypass in rats. Surgery 1968; 64:775-84. 32. Utley JR. The immune response to cardiopulmonary bypass. In: Utley JR, ed. Pathophysiology and techniques of cardiopulmonary bypass. Baltimore: Williams & Wilkins, 1982:142.
Thoracic and Cardiovascular Surgery
33. Petersen NJ, Carson LA, Favero MS. Bacterial endotoxin in new and reused hemodialyzers: a potential cause of endotoxemia. Trans Am Soc Artif Intern Organs 1981; 27:155-60. 34. Pearson FC, Bohon J, Lee W, et al. Characterization of Limulus amebocyte lysate-reactive material from hollow fiber dialyzers. Appl Environ Microbiol 1984;48:1189-96. 35. Wolff MS. Biologicaleffectsof bacterial endotoxinsin man. J Infect Dis 1973;129:S259-64. 36. Gage AA, Dean DC, Schimert G, Minsley N. Aspergillus infection after cardiac surgery. Arch Surg 1970;101:384-7.
Lee A. Bland, MA, MPH,a Margarita E. Villarino, MD, MPH,b Matthew J. Arduino, MS, DrPH,a Sigrid K. McAllister, BS,a Steven M. Gordon, MD,b Charles T. Uyeda, PhD,c Carol Valdon, MS, RN,d Diana Potts, MS, RN,d William R. Jarvis, MD,b and Martin S. Favero, Phl)," Atlanta, Ga., and Palo Alto, Calif.
From the Nosocomial Infections Laboratory Branch" and the Epidemiology Branch," Hospital Infections Program, Centers for Disease Control, Public Health Service, U.S. Department of Health and Human Services, Atlanta, Ga., and the Microbiology Laboratory" and Infection Control/' Palo Alto Veterans Administration Medical Center, Palo Alto, Calif. Received for publication July 5, 1990. Accepted for publication Dec. 12, 1990. Address for reprints: Lee A. Bland, MA, MPH, Nosocomial Infections Laboratory Branch, Hospital Infections Program, COl, Centers for Disease Control, Atlanta, GA 30333.
12/1/28363
582
Since the 1970s the number of patients undergoing cardiac operations with cardiopulmonary bypass (CPB) has rapidly increased. Currently more than 350,000 cardiac operations are performed annually in the United States. I Volume replacement during heart operations often requires the use of banked blood components. Fear of transfusion-related diseases (e.g., human immunodeficiency virus, hepatitis C virus, cytomegalovirus, and Yersinia enterocolitica) and reactions (isosensitization or hemolytic and febrile reactions) from heterologous blood transfusions have led to research on artificial blood and increased use of autologous transfusion.
Volume 103 Number 3 March 1992
Salvaged blood used in autologous transfusions
583
Some investigators have had success with autologous blood transfusion both clinically and in reducing blood banking expenses, whereas others have found autologous blood transfusion safe but with little to no reduction in the use of blood bank components.v'" Still others have questioned the suitability of using salvaged blood as an alternative to donated blood. 1I, 12 Adan and coworkers 13 found that salvaged autologous blood for reinfusion contained a considerable amount of small-sized debris; they concluded that the dynamics of the coagulation process in the mediastinum do not favor autologous blood transfusion and that it should not be performed on a routine basis. Another study reported the detection of circulating endotoxin in the extracorporeal circuit during CPB, but no intraoperative or postoperative complications were noted in the 10 patients studied.!" The endotoxin contamination was determined to be a result of the CPB procedure and the intraoperatively administered fluids. Schwieger and colleagues'> found no risk of infection in cardiac surgical patients from blood collected in a cell conservation apparatus. We recently investigated an outbreak of massive nonsurgical postoperative bleeding after cardiac operations in which salvaged autologous blood was routinely transfused. As a result of this outbreak, we conducted a prospective study of cardiac operations to determine the prevalence of microbial and endotoxin contamination of the patients' blood collected during preoperative, intraoperative, and postoperative periods and of blood aspirated from the open chest cavity during the operation and collected by the Cell Saver System (Haemonetics Cell Saver System, Haemonetics Corp., Braintree, Mass.* Use of autologous blood transfusion was assessed as to its association with bacteremia, postoperative bleeding, and endotoxic shock.
Bacterial culture. Bacterial cultures were performed by inoculating 3 to 5 ml of sample into Bactec (Johnson Laboratories, Towson, Md.) aerobic 16B and anaerobic 17D blood culture bottlescontaining resin.These bottles were incubated at 35° C for 7 days and read on the Bactec 460 instrument as follows: The aerobic bottles were tested twice during the first 24 hours of incubation and onceon days 2, 3, 4,5, and 7; anaerobic bottles were tested once on days 2, 3, 4,5, and 7. Bottles registering thresholdgrowth indicesgreater than or equal to 20 were subjected to Gram stain and subcultured to a tube of enriched thioglycolate. Gram stain-"positive" bottles were subsequently subcultured to chocolate,phenylethanol,and MacConkey agar plates. Coagulase tests on staphylococci were performed with reconstituted rabbit plasma (BBL Microbiology Systems, Cockeysville, Md.) by the slide and tube methods. Certain staphylococci were speciated by using the gram-positive identification cards on the AutoMicrobic System (Vitek Systems, Hazelwood, Mo.) or by the Staph-Ident (Analytab Products, Plainview, N.Y.) method. Gram-negativebacilliwere identified by the gram-negative identification cards (Vitek Systems) on the AutoMicrobic System and anaerobes by the An-Ident (Analytab Products, Inc.) methods. Endotoxin analysis. Detection and quantification of endotoxin were done by the Limulus amebocyte lysate (LAL) turbidimetric assay method (LAL-5000; Associatesof Cape Cod, Woods Hole, Mass.) (Associates of Cape Cod, LAL Update 5:2, 1987).16,17 An Escherichia coli 0113 control standard endotoxin (lot 36; Associates of Cape Cod) was used for all standard curves, positivecontrol procedures, and product inhibition determinations.lv 19 Endotoxin-free polypropylene microcentrifuge tubes, containing I % heparin sodiuminjection(USP) as an anticoagulant, were used for collectionof blood samples. Blood samples were centrifuged for 3 minutes, and the plasma was removed. The plasma wasdiluted I: I0 with steriledistilledwater for irrigation (Abbott Laboratories, Chicago, Ill.) and heated at 70° C for 10 minutes to remove plasma endotoxin test inhibitors.Pj" After being heated, the samples were again diluted 1:10 with sterile distilledwater (Abbott) to make a final dilution of I: I00. CPB priming fluidwas collectedin endotoxin-freepolystyrenetubes. The priming fluid was diluted I:10 with sterile distilled water (Abbott) before analysis.
Materials and methods
Results
BetweenSeptember 1988and February 1989,bloodsamples from all patients undergoing cardiac operations were collected and assayed for bacterial and endotoxin contamination. The following samples were collectedfrom each patient: (I) preoperative blood drawn through an arterial line in the operating room before the operation and before prophylactic antibiotics were given; (2) blood drawn intraoperatively through an arterial lineimmediatelyafter the discontinuationof CPB; (3) blood collected percutaneously in the surgical intensive care unit I hour postoperatively; and (4) blood from the reservoirbag, collected by the Cell Saver System during the operation, immediately before reinfusion. In addition, samplesof Ringer's lactate solution used to prime the extracorporeal circuit of the CPB equipment were collectedfor bacterial and endotoxin analysis.
Thirty-eight patients were enrolled in the study. Patients were supported by CPB for a mean time of 125 minutes (range 20 to 243 minutes), and the mean length of operation was 299 minutes (range 155 to 470 minutes). Bacteria were isolated from 34.8% (46/132) of all collected blood samples, and 13.7% (17/124) had detectable endotoxin (i.e., > 10 pg/rnl) (Table I). Bacterial cultures were positive in 10.8% (4/37) of the preoperative samples, 30.6% (11/36) of the intraoperative samples, 3.6% (1/28) of the postoperative samples, and 96.8% (30/31) of the blood samples taken from the Cell Saver System. The predominant microorganisms isolated were coagulase-negative staphylococci, which included Sta-
"Use of trade names and commercial sources is for identification only and does not imply endorsement by the Public Health Service or the U.S. Department of Health and Human Services.
phylococcus epidermidis, S. capitis, S. warnerii, S. haemolyticus, S. hominis, S. simulans, and S. saccharolyticus (Table II). Several of the cultures were polymicrobic,
584
The Journal of Thoracic and Cardiovascular
Bland et al.
Surgery
Table I. Bacterial and endotoxin assay ofpatient and Cell Saver System blood and priming fluid during cardiac operations
Blood Preoperative Intraoperative Cell Saver System Postoperative All blood Priming fluid Total
Bacterial cultures
Endotoxin assay
No. positive/No. of tests (%)
No. positive/No. of tests (%)
4/37 (10.8) 11/36 (30.6) 30/31 (96.8) 1/28 (3.6) 46/132 (34.8) 0/38 (0) 46/169 (27.2)
and many consisted of more than one type of coagulasenegative staphylococci, coagulase-negative staphylococci and diphtheroids, or coagulase-negative staphylococci and gram-negative bacilli. Of the two gram-negative bacilli polymicrobic cultures isolated from the Cell Saver System, one was positive for Enterobacter agglomerans and the other for Pseudomonas aeruginosa. All bacterial cultures of the CPR priming fluid were negative. Endotoxin was detected in 10.7% of all samples collected (Table I). None of the patients' blood samples had detectable plasma endotoxin before the operation. Endotoxin was detected in five of 35 (14.8%) of the intraoperative samples (15 to 26 pg/rnl), five of 25 (20%) of the postoperative samples (12 to 44 pg/rnl), and seven of 29 (24.1 %) of the blood samples from the Cell Saver System (10 to 765 pg/rnl) (Table 11). Endotoxin was not detected in the CPB priming fluid. No adverse clinical signs or symptoms, including postoperative bleeding or endotoxemia, were detected postoperatively in any of the 38 study patients despite the frequent isolation of bacteria and detection of endotoxin, especially in blood from the Cell Saver System. Discussion
Recovery and reinfusion of blood aspirated from the open chest cavity during cardiac operations has many benefits, including reduction in the risk of infectious disease transmission from donor blood to the host patient, avoidance of isosensitive, hemolytic, and febrile reactions, and possible reduction in blood bank services.21• 22 One recovery method involves the use of a cell conservation apparatus, which washes and removes excess water, heparin, serum hemoglobin, bacteria, potassium, and cellular debris from the recovered blood cells while maintaining blood sterility.' However, several studies have isolated bacteria from blood retrieved from a cell conservation
0/35 (0) 5/35 (14.3) 7/29 (24.1) 5/25 (20) 17/124 (13.7) 0/35 (0) 17/159 (10.7)
apparatus or other autologous blood collection de-
vices.U: 15.22
Blood collected from the reservoir bag of the Cell Saver System during operations rarely was found to be sterile in this study. Ninety-seven percent of the samples taken of the blood aspirated from the open chest and recovered by the Cell Saver System were culture positive. Kluge, Calia, and Mclaughlin-' found that up to 60% of cultures taken from the open chest cavity from patients undergoing cardiac operations were contaminated with microorganisms. The most common organisms isolated were diphtheroids and coagulase-negative staphylococci. Other studies report 20% or less of salvaged blood yields positive cultures. I I. 22 Kluge, Calia, and Mclaughlin-' suggested that the frequent contamination of the perioperative site during heart operations was probably due to airborne contaminants. Blakemore and coworkers-" found that during cardiac operations the greatest density of airborne microbes occurred directly above the operative site and that coagulase-negative staphylococci were the predominant organisms isolated. Other organisms isolated in their study included Enterobacter sp., Mima sp., P. aeruginosa, Serratia sp., Aspergillus sp., and Candida sp. The chief source of these organisms was the operating room personnel. The number of isolates increased with the number of persons in the operating room, the degree of traffic, and the number of times the door to the operating room was opened. Other investigators suggest that organisms isolated from drained mediastinal blood result from the manipulation of the drained blood before reinfusion or specimen contamination by the laboratory.P II In our study the majority of organisms isolated from blood collected by the Cell Saver System were common skin commensals; 35 of 44 (79.5%) were coagulase-negative staphylococci, and nine of 44 (20.5%) were aerobic or anaerobic diphtheroids. Such microorganisms are part of the normal skin flora and commonly are found in the
Volume 103 Number 3 March 1992
Salvaged blood used in autologous transfusions
585
Table II. Bacterial isolates and endotoxin (pgjml) concentrations in patient and Cell Saver System blood and priming fluid during cardiac operations Patient No.
Preop. blood
lntraop. blood
Bact.
Bact.
1 2 3 4 5 6 7 8
Endo.
Cell Saver System blood Endo.
CNS
CNSt
*
P. acnes
P. acnes
ND:j:
9
CNS (3 strains) CNS ND
S. epidermidis S. epidermidis
10
26
CNS ND
ND
ND
S. capitis S. epidermidis
CNS
S. capitis CNS, P. acnes S. haemolyticus S. epidermidis S. saccharolyticus S. simulans S. epidermidis S. epidermidis (2 strains) S. epidermidis P. acnes
S. epidermidis
ND
S. warnerii, S. epidermidis S. epidermidis ND
25 26 27 28 29 30 31 32 33 34 35 36 37 38
ND
S. hominis S. hominis, S. epidermidis CNS, E. agglomerans Diphtheroids, S. haemolyticus, S. capitis S. epidermidis S. epidermidis, P. aeruginosa
Diphtheroids CNS
11
12 13 14 15 16 17 18 19 20 21 22 23 24
Bact.
ND
S. capitis, S. epidermidis, S. warnerii
ND CNS, P. acnes
CNS CNS
Diptheroids ND
S. epidermidis
S. epidermidis, S. warnerii
17
CNS ND ND
20 15
S. epidermidis, diphtheroids S. epidermidis
18
S. epidermidis S. epidermidis
ND
ND ND
Postop. blood Endo.
Bact.
14 18
Endo.
Prime fluid
Bact. Endo.
25
ND
ND S. epidermidis
19
ND
ND
ND ND ND ND ND ND
ND 21 ND ND ND ND ND ND
ND
ND
ND
ND
ND 14 765 ND ND 205 10
81
ND ND
ND ND ND ND
ND ND ND ND 44 12 ND ND
*Blank entry indicates a negative result. tCoagulase-negative staphylococci (unspeciated). iAssay not done.
air and on environmental surfaces, even in the operating room.23-25 Such ubiquitous organisms are also common laboratory isolates frequently considered contaminants.l? 26 However, Kohman, Coleman, and Parker.F' in a recent study on bacteremia after coronary artery bypass grafting, found that coagulase-negative staphylococci accounted for 55% of the positive postoperative blood cultures. Gram-negative bacilli were isolated in
only two of the samples taken from the Cell Saver System. We could not determine if the samples from the Cell Saver System were contaminated during the surgical procedures by endogenous or exogenous sources or if the samples were contaminated during collection and subsequent laboratory manipulation. However, because blood samples obtained through the arterial lines preoperative-
5 8 6 Bland et al.
Iy, intraoperatively, and postoperatively were assayed in the same manner but did not have the same high contamination rate, the samples from the Cell Saver System probably were not contaminated during collection or laboratory manipulation but during the actual recovery of the blood from the open chest cavity. The fact that all samples of CPB priming fluid were culture negative would also support this hypothesis. One patient had three successive blood cultures (preoperative, intraoperative, and Cell Saver System) positive for coagulase-negative staphylococci and suggestive of bacteremia, but, unfortunately, blood for culture was not drawn immediately postoperatively on this patient. The patient received antimicrobial prophylaxis and did not exhibit any clinical sign of bacteremia postoperatively, however. Blood cultures taken 6 days later were negative. No other patient had successive positive cultures between preoperative and intraoperative samples or intraoperative and postoperative samples. Intraoperative cultures obtained at the end of CPB were positive in 30.6% of the cases studied. No correlation was found between positive intraoperative cultures and the duration of CPB support or the duration of the surgical procedure. Although several patients had positive blood cultures, none had adverse clinical signs or symptoms. Similar findings have been reported in other studies. 10,22 This may be due to the relatively low frequency of reported sternal wound complications after cardiac operations. In a recent study by Loop and coworkers'" of 6504 patients who underwent isolated coronary artery bypass grafting, 1.1% had sternal wound complications. They reported even lower wound infection rates for valve operations (0.4%), and when coronary artery bypass grafting was combined with other cardiac procedures wound infection rates were only 1.0%. It is not surprising that such complications were not seen in our study involving only 38 patients. Plasma endotoxin was not detected in preoperative blood samples or samples of CPB priming fluid. A similar study by Andersen and colleagues 14 of 10 cardiac operations did not find detectable endotoxin in patients preoperatively but did find that 30% of the CPB priming fluid (Ringer's lactate solution) samples had measurable endotoxin concentrations (mean 11 pg/rnl). In another study of nine patients having heart operations, it was found that, the mean preoperative plasma endotoxin concentration was 128 ± 32 pgjml, with concentration increasing during the surgical procedure to 428 ± 60 pgjml and finally decreasing to baseline levels postoperatively.-? A similar trend was noticed in our study, in which endotoxin concentrations, although generally low,
The Journal of Thoracic and Cardiovascular Surgery
increased during the procedure and decreased by the end of the procedure and postoperative recovery. The greatest percentage of positive assays and the highest plasma endotoxin levels (765,205, and 81 pgjml) were in blood samples taken from the Cell Saver System. It has been suggested that, during cardiac operations, depression of the reticuloendothelial system allows endotoxin that leaks into the portal circulation from the gut to enter the systemic circulation. 30-32 Another study found that if pyrogen-free water was circulated through new extracorporeal circuits, small amounts of material (3 to 6 pg) would give a positive LAL endotoxin test.!" It was speculated that cellulosic material in the oxygenator and the cardiotomy reservoir was the source of this LAL-reactive material and not gram-negative bacilli. Similar results have been reported during the use of artificial kidneys.P:34 Both of these reasons could explain the greater endotoxin concentrations of blood collected in cell conservation devices. None of the study patients had any signs or symptoms of endotoxemia, consistent with the low concentrations of endotoxin found in the blood samples. Endotoxin in concentrations of 1000 pgjkg of body weight is generally necessary to produce pyrogenic reactions in human beings, and doses of 2000 pgjkg of body weight administered intravenously produce granulocytosis in addition to a febrile response. 33, 35 The concentrations of endotoxin found in our samples were far below these threshold values. In conclusion, our study shows that bacterial and endotoxin contamination of blood from cell conservation devices used for autologous transfusion during cardiac operations is frequent. Bacterial contamination may be caused by airborne contaminants that either fall or are pulled into the operative field by air currents created by suction tubes of the cell conservation device or the manipulation of the salvaged blood before reinfusion, as hypothesized by Bennett!' and others. 24, 36 However, despite bacterial and endotoxin contamination of blood recovered by the Cell Saver System during operations and its subsequent reinfusion, no episodes of sepsis or endotoxic shock were detected in any of our patients; transient bacteremia was rarely detected. The sources of microbial contamination could not be identified but, because the organisms isolated were ubiquitous skin commensals, contamination probably occurred during recovery of the blood from the open incision. The presence of endotoxin in intraoperative and postoperative blood may be due to intestinal leakage of endotoxin into the portal circulation. Endotoxin contamination of blood from the Cell Saver System may also be a result of intestinal endotoxin leak-
Volume 103 Number 3 March 1992
age, but it also may be due to LAL-reactive material in the extracorporeal circuit. Additional studies are needed to determine definitely if endotoxin or bacterial contamination of blood collected by a cell conservation apparatus during operations and its subsequent reinfusion plays a role in postoperative pyrexia and other commonly recognized complications of cardiac operations. Although blood from the Cell Saver System had a high potential for bacterial and endotoxin contamination in this study, its use did not appear to elicit adverse effects in the 38 patients studied. REFERENCES 1. Graves EJ. 1989. National Hospital Discharge Survey: Annual summary, 1987. Hyattsville, Md. National Center for Health Statistics. Vital Health Stat 13(99). 2. National Blood Resource Education Program Expert Panel. The use of autologous blood. JAMA 1990;263:414-7. 3. Cordell AR, Lavender SW. An appraisal of blood salvage techniques in vascular and cardiac operations. Ann Thorac Surg 1980;31 :421-5. 4. Emminizer S, Klopp EH, Hauer JM. Autotransfusion: current status. Heart Lung 1981;10:83-7. 5. Johnson RG, Rosenkrantz KR, Preston RA, Hopkins C, Daggett WM. The efficacy of postoperative autotransfusionin patients undergoing cardiac operations. Ann Thorac Surg 1983;36:173-9. 6. Schaff HV, Hauer JM, Brawley RK. Autotransfusion in cardiac surgical patients after operation. Surgery 1978; 84:713-8. 7. Schaff HV, Hauer J, Gardner TJ, et al. Routine use of autotransfusion following cardiac surgery: experience in 700 patients. Ann Thorac Surg 1979;27:493-9. 8. Weniger J, von der Emde J, Schricker KT, Blechschmidt J. Retransfusion von Drainageblut nach herzchirurgischen Eingriffen. Langenbecks Arch Chir 1980;351:229-41. 9. Weniger J, Shanahan R. Reduction of bank blood requirements in cardiac surgery. Thorac Cardiovasc Surg 1982; 30:142-6. 10. Thurer RL, Lytle BW, Cosgrove DM, Loop FD. Autotransfusion following cardiac operations: a randomized, prospective study. Ann Thorac Surg 1979;27:500-7. 11. Bennett JG. Autotransfusion of drained mediastinal blood. Thorac Cardiovasc Surg 1982;30:28-30. 12. Carter RF, McArdle B, Morritt GM. Autologous transfusion of mediastinal drainage blood. Anaesthesia 1981; 36:54-9. 13. Adan A, Brutel de la Riviere A, Haas F, van Zalk A, de Nooij E. Autotransfusion of drained mediastinal blood after cardiac surgery: a reappraisal. Thorac Cardiovasc Surg 1988;36:10-4. 14. Andersen LW, Baek L, Degn H, Lehd J, Krasnik M, Rasmussen JP. Presence of circulating endotoxins during cardiac operations. J THoRAc CARDIOVASC SURG 1987; 93:115-9.
Salvaged blood used in autologous transfusions
587
15. Schwieger 1M, Gallagher CJ, Finlayson DC, Daly WL, Maher KL. Incidence of Cell-Saver contamination during cardiopulmonary bypass. Ann Thorac Surg 1989;48: 51-3. 16. Novitsky TJ, Remillard JF, Loy N. Design criteria and evaluation of the LAL-4000 for kinetic turbidimetric LAL assay. In: Watson SW, Levin J, Novitsky TJ, eds. Detection of bacterial endotoxins with the Limulus amebocyte lysate assay. New York: Alan R Liss, 1987:189-96. 17. Remillard JF, Gould MC, Roslansky PF, Novitsky TJ. Quantitation of endotoxin in products using LAL kinetic turbidimetric assay. In: Watson SW, Levin J, Novitsky TJ, eds. Detection of bacterial endotoxins with the Limulus amebocytelysateassay. New York: Alan R Liss, 1987:197210. 18. Levin J, Tomasulo PA, Oser RS. Detection of endotoxin in human blood and demonstration of an inhibitor. J Lab Clin Med 1970;75:903-11. 19. Novitsky TJ, Roslansky PF, Siber GR, Warren HS. Turbidimetric method for quantifying serum inhibition of the Limulus amoebocyte lysate. J Clin Microbiol 1985;21: 211-6. 20. Sturk A, Janssen ME, Muylaert FR, Joop K, Thomas LLM, Cate JW. Endotoxin testing in blood. In: Watson SW, Levin J, Novitsky TJ, eds. Detection of bacterial endotoxins with the Limulus amebocyte lysate assay. New York: Alan R Liss, 1987:371-85. 21. Cosgrove DM, Thurer RL, Lytle BW, et al. Blood conservation during myocardial revascularization. Ann Thorac Surg 1979;28:184-9. 22. Schaff HV, Hauer JM, Bell WR, et al. Autotransfusion of shed mediastinal blood after cardiac surgery. J THORAC CARDIOVASC SURG 1978;75:632-41. 23. Kluge RM, Calia FM, Mclaughlin JS. Sources of contamination in open heart surgery. JAMA 1974;230:1415-8. 24. Blakemore WS, McGarrity GJ, Thurer RJ, Wallace HW, MacVaugh H III, Coriell LL. Infection by airborne bacteria with cardiopulmonary bypass. Surgery 1971;70:830-8. 25. Gallis HA. Microbial ecology and normal flora of the human body. In: Joklik WK, Willet HP, Amos DB, eds. Zinsser's Microbiology. 17th ed. East Norwalk, Conn., Appleton & Lange, 1980;524-6. 26. Branson D. How to culture for a specificorganism. In: Procedure manual for clinical bacteriology. Springfield: Charles C Thomas, 1982:168-320. 27. Kohman LJ, Coleman MJ, Parker FB Jr. Bacteremia and sternal infection after coronary artery bypass grafting. Ann Thorac Surg 1990;49:454-7. 28. Loop FD, Lytle BW, Cosgrove DM, et al. Sternal wound complications after isolated coronary artery bypass grafting: early and late mortality, morbidity, and cost of care. Ann Thorac Surg 1990;49:179-87. 29. Rocke DA, Gaffin SL, Wells MT, Koen Y, Brocke-Utine JG. Endotoxernia associated with cardiopulmonary bypass. J THORAC CARDIOVASC SURG 1987;93:832-7. 30. Freeman R, Gould FK. Rises in antibody to enteric gram
The Journal of
5 8 8 Bland et al.
negative bacilli after open heart surgery: a possible mechanism for postoperative pyrexia. Thorax 1985;40:53841. 31. Subramanian V, Mcleod J, Gans H. Effect of extracorporeal circulation on reticuloendothelial function. I. Experimental evidence for impaired reticuloendothelial function following cardiopulmonary bypass in rats. Surgery 1968; 64:775-84. 32. Utley JR. The immune response to cardiopulmonary bypass. In: Utley JR, ed. Pathophysiology and techniques of cardiopulmonary bypass. Baltimore: Williams & Wilkins, 1982:142.
Thoracic and Cardiovascular Surgery
33. Petersen NJ, Carson LA, Favero MS. Bacterial endotoxin in new and reused hemodialyzers: a potential cause of endotoxemia. Trans Am Soc Artif Intern Organs 1981; 27:155-60. 34. Pearson FC, Bohon J, Lee W, et al. Characterization of Limulus amebocyte lysate-reactive material from hollow fiber dialyzers. Appl Environ Microbiol 1984;48:1189-96. 35. Wolff MS. Biologicaleffectsof bacterial endotoxinsin man. J Infect Dis 1973;129:S259-64. 36. Gage AA, Dean DC, Schimert G, Minsley N. Aspergillus infection after cardiac surgery. Arch Surg 1970;101:384-7.