Staphylococcus epidemidis bacteremia from transfusion of contaminated platelets: application of bacterial DNA analysis R.R. MUDER,Y.C. YEE, J.D. RIHS, AND M. BUNKER Septicemia is a rare complication of platelet transfusion. A case is reported of transfusion-associated septicemia in a 66-year-old man who received a 1O-unit pool of platelets. Durin transfusion, h e experienced rigors, wheezing, dyspnea, and fever. A total of four %lood cultures drawn 10 and 36 hours after discontinuation of the transfusion grew Stephylococcus epidermidis. Culture of the residual platelet pool yielded S. epidermidis with a colony count of lo5 organisms per rnL. Strain identity of all four blood isolates and the platelet pool isolate was confirmed by gel electrophoresis of EcoRl and Hindlll restriction digests of whole-cell DNA. There have been 31 prior reported cases of platelet transfusion-associated septjcemia, of which 9 have been caused by coagulase-negative staphylococci. Systemic reactions to platelet transfusions should prompt consideration of transfusion-associated bacteremia as the cause. TRANSFUSION 1992;32:771-774.
without incident. Approximately 90 minutes after the initiation of a 10-unit transfusion of pooled donor platelets, he abruptly experienced rigors, wheezing, dyspnea, and fever to 104.6"F (measurement taken orally). The platelet transfusion was discontinued after a total transfusion of 200 mL. The patient was treated with acetaminophen, diphenhydramine, and albuteral and the application of a cooling blanket. Sixteen hours later, he was afebrile and asymptomatic. Cultures of four of four blood samples obtained by venipuncture of separate peripheral sites grew S. epidermidis within 24 hours of their drawing. Two samples were drawn 10 hours after the cessation of the platelet transfusion, and the other two were drawn 36 hours after the transfusion; the patient was afebrile at that time. The isolates were resistant to penicillin but susceptible to oxacillin, cephalothin, vancomycin, and clindamycin. We initiated vancomycin administration, 1 g intravenously every 12 hours, 8 hours after the final blood cultures were drawn. This was changed to nafcillin, 1 g every 4 hours, when the results of the susceptibilities of the blood isolates became known. Therapy was administered for a total of 10 days, during which time the patient remained afebrile. He was discharged in stable condition.
SEPTICEMIA IS A RARE but potentially fatal complication of the transfusion of blood components.'.* Because platelets are stored at 20 to 24°C for up to 5 days, contaminating bacteria initially present at very low levels may multiply to extremely high level^,^ which creates the potential for overwhelming septicemia. We report a case of septicemia due t o the transfusion of platelets contaminated with Staphylococcus epidermidis. Using DNA typing techniques, we were able to demonstrate that the contaminated platelet pool was the source of infection, and that the patient had persistent bacteremia with the organism for 36 hours. Case Report A 66-year-old man was admitted for blood transfusion. The patient had a long history of myelodysplastic disorder and had required repeated transfusions of red cells and platelets. He had a history of coronary artery disease, hypertension, dietcontrolled adult onset diabetes, peptic ulcer disease, and carotid artery stenosis requiring endarterectomy. The patient's medications included prednisone (20 mg twice daily), digoxin, isosorbide, captopril, folic acid, thiamine, and ranitidine. On examination he was in no distress. He was afebrile, with a pulse of 88 and a blood pressure of 132 over 50. The remainder of the examination was unremarkable. His hemogloHis white cell bin concentration was 7.7 g per dL (77 6). count was 1600 per mL (1.6 x 109/L),with 36 percent (0.36) segmented neutrophils, 8 percent (0.08) bands, 6 percent (0.06) metamyelocytes, and 50 percent (0.50) lymphocytes, and his platelet count was 10,000 per mL (10 x 109/L). On the evening of admission, a new peripheral cannula was inserted, and the patient received 2 units of packed red cells
Materials and Methods Culture After the transfusion was discontinued, the pooled platelets were stored at 4°C for 10 hours before culture. We aspirated samples from the bag of residual pooled platelets by sterile technique and plated them on 5-percent sheep's blood agar plates incubated at 35 and 30°C in room air, chocolate agar plates incubated at 37°C in 5-percent C02, and reducible anaerobic blood agar plates incubated at 35°C anaerobically. We inoculated additional aliquots into thioglycolate broth and incubated it at 30 and 35°C. Speciation of staphylococcal isolates was performed with the bacterial identification system (VITEK, Hazelwood, MO).
From the Medical Service, Laboratory Service, and Special Pathogens Laboratory, Veterans Affairs Medical Center. and the Departments of Medicine and Pathology, University of Pittsburgh School of Medicine, Pittsburgh, Pennsylvania. Received for publication November 13, 1991; revision received March 6, 1992, and accepted March 25, 1992.
DNA analysis We inoculated a single colony of each S. epidemidis isolate into 5 mL of trypticase soy broth and incubated the tubes for
771
772
TRANSFUSION Vol. 32, No. 8-1992
MUDER ET AL.
FIG. 1. Gel electrophoresis of DNA fragments of staphylococcal isolates after digestion with EcoRI (A)and Hind111 (B).Lane 1 is A DNA (a molecular size marker); Lane 2 is the platelet pool isolate; Lanes 3 through 6 are isolates obtained from four separate blood cultures; Lane 7 is the isolate from the patient’s skin.
18 hours at 37°C. We centrifuged bacterial cells from 1.5 mL of suspension at 9500 x g for 2 minutes, resuspended them in 567 pL of buffer (10 mM [lo mmol/L] Tris, pH 7.9, 1 mh4 [l mmolL] EDTA), and lysed them with 20 p L of lysostaphin for 30 minutes at 37°C. After lysis, whole-cell DNA was extracted and purified4 and incubated with either EcoRI or H i d 1 1 (Toyobo. Ltd., Tokyo, Japan) for 18 hours in accordance with the manufacturer’s recommendations. Electrophoresis of digested DNA was carried out in 0.7-percent agarose in Trisborate buffer for 20 hours at 1volt per cm. Gels were stained for 1 to 1.5 hours in 0.5 pg per mL of ethidium bromide and photographed (665 film, Polaroid Corp., Cambridge, MA).
Results Review of blood bank records showed that the platelet pool consisted of 10 individual donor units collected 5 days before transfusion. The units were stored at room temperature and pooled shortly before administration. Gram’s stain of the residual pooled platelets showed no organisms, but culture yielded S. epidennidis at an approximate concentration of lo5 organ-
isms per mL. The bags containing the individual units had been discarded and were not cultured. Culture of the patient’s antecubital fossa yielded S. horninis; no staphylococci were isolated from his anterior nares. The S. epidennidis isolated from the residual pooled platelets and from each of the four blood isolates had identical restriction patterns after digestion with EcoRI (Fig. 1A) and Hind111 (Fig. 1B).The staphylococcal isolate from the patient’s skin was distinct from the platelet and blood isolates.
Discussion Platelet concentrates stored at room temperature maintain adequate viability and function for up to 7 days.5 With extended storage, small numbers of bacteria inadvertently introduced at the time of collection or during processing could multiply to high numbers. Microbiologic surveys6-’0 have shown that 0 to 6.3 percent of individual platelet units are contaminated.
TRMSFUSION 1992-vOi. 32. N ~ a.
773
PLATELET TRANSFUSION BACTEREMIA
Contamination of fresh platelet pools ranges from 7.4 to 10.9 percent; the proportion may rise as high as 38 percent after 3 days of s t ~ r a g e . ~S.* ' ~epidennidis and diphtheroid bacilli, components of the normal skin flora, are the most frequently isolated bacteria, which indicates that contamination at the time of collection is the most likely source. Clinically recognized septicemia due to platelet transfusion is extremely rare.2 Bacterial counts are typically quite low in contaminated units,*-1° and the vast majority are transfused without adverse effect. However, the inoculation of platelet concentrates with a single colony of S. epidennidis can result in logarithmic phase growth with colony counts of lo3 per 0.1 mL at 3 days and lo7 per 0.1 mL at 6 days.' Duration of platelet storage appears to be an important determinant of the risk of sepsis. The transfusion of multidonor platelets stored for 5 days is associated with an incidence of sepsis five times greater than that seen with the transfusion of platelets stored for 4 days or less.2 To date, 32 cases (including the present one) of platelet-associated bacteremia have been rep~rted;'-'~~-'~ seven of the episodes have been fatal. Given the frequency of isolation of S. epidennidis from platelet concentrates, it is interesting to note that only nine prior cases of plateletassociated septicemia have been reported with coagulase-negative staphylococci as the cause. S. epidennidis, although of limited pathogenicity in the normal host, may cause sustained bacteremia in neutropenic patient~.~'The relatively low number of reported cases associated with platelet transfusion could be due to the frequency of concomitant antibiotic therapy in these patients or to frequent dismissal as a blood culture contaminant. Our patient's primary physicians initially dismissed the first two sets of blood cultures as contaminants and were unconvinced of the significance of the results until the residual platelets and the third and fourth sets of blood cultures were reported as positive. Subsequently, we were able to confirm by DNA typing that all four blood isolates were identical to the S. epidennidis contaminating the platelet pool and were different from the staphylococci isolated from the patient's skin. Restriction endonuclease analysis of S. epidennidis appears to be a reproducible method of demonstrating clonal identity among isolates. It discriminates between strains more effectively than biotyping or phage typing and is more stable than plasmid content or antibiotic susceptibility profiles.'* This case underscores the potential difficulties in differentiating transfusion-associated septicemia from noninfectious transfusion reactions. The isolation from the blood of organisms typically regarded as blood culture contaminants, such as coagulase-negative staphylococci, should be regarded as potentially significant. Our observations demonstrate the utility of molecular typing in confirming the source of transfusion-associated septi-
cemia and in differentiating a true pathogen from a skin contaminant. One other case of platelet-associatedS. epidennidis bacteremia has been confirmed by field-inversion gel electrophoresis and ribotyping.16 Molecular typing could be especially valuable in investigating common-source outbreaks of septicemia in which multiple patients are involved. Bacteria isolated from patients' blood or from blood components in suspected cases of transfusion-associated septicemia should be saved for further typing in case epidemiologic investigation suggests the need for it.
Acknowledgment The authors thank Victor L. Yu, MD, for review of the manuscript.
References 1. Morduchowicz G, Pitlik SD, Huminer D, et al. Transfusion reactions due to bacterial contamination of blood and blood products. Rev Infect Dis 1991;13:307-14. 2. Morrow JF, Braine HG, Kickler TS, Ness PM, Dick JD, Fuller AK. Septic reactions to platelet transfusions. A persistent problem. JAMA 1991;266:555-8. 3. Braine HG, Kickler TS, Charache P, et al. Bacterial sepsis secondary to platelet transfusion: an adverse effect of extended storage at room temperature. Transfusion 1986;26:391-3. 4. Ausbel FM, Brent R, Kingston RE, et al. Short protocols in molecular biology. New York: Wiley, 1989. 5. Simon TL, Nelson El, Carmen R, Murphy S. Extension of platclet concentrate storage. Transfusion 1983;23:207-12. 6. Silver H, Sonnenwirth AC, Beisser LD. Bacteriologic study of platelet concentrates prepared and stored without refrigeration. Transfusion 1970;10:315-6. 7. Wrenn HE, Speicher CE. Platelet concentrates: sterility of 400 single units stored at room temperature. Transfusion 1974;14171-2. 8. Cunningham M, Cash JD. Bacterial contamination of platelet concentrates stored at 20°C. J Clin Pathol 1973;26:401-4. 9. Buchholz DH, Young VM, Friedman NR, Reilly JA, Mardiney MR Jr. Bacterial proliferation in platelet products stored at room temperature. Transfusion-induced Enterobacter sepsis. N Engl J Med 1971;285:429-33. 10. Buchholz DH, Young VM, Friedman NR, Reilly JA, Mardiney MR Jr. Detection and quantitation of bacteria in platelet products stored at ambient temperature. Transfusion 1973;13:268-75. 11. Rhame FS, Root RK, Macbwry JD, Dadisman TA, Bennett JV. Salmonella septicemia from platelet transfusions. Study of an outbreak traced to a hematogenous carrier of Salmonella choleraesuir. Ann Intern Med 1973;78:633-41. 12. Blajchman MA, Thornley JH. Richardson H, Elder D, Spiak C, Racher J. Platelet transfusion-induced Serrafia marcescem sepsis due to vacuum tube contamination. Transfusion 1979;19:39-44. 13. Van Lierde S, Fleisher GR, Plotkin SA, C a m p s JM. A case of platelet transfusion-relatedSerratia marcescem sepsis. Pediatr Infect Dis 1985;4:293-5. 14. &now PM, Weiss LM, Weil D, Rosen NR. Escherichia coli sepsis from contaminated platelet transfusion. Arch Intern Med 1986;146:321-4. 15. Heal JM, Jones ME, Forey J, Chaudhry A. Stricof RL. Fatal Salmonella septicemia afkr platelet transfusion. Transfusion 1987;27:2-5. 16. Shayegani M, Parsons LM. Waring AL. et al. Molecular relatedness of Staphylococcus epidemidis isolates obtained during a platelet transfusion-associated episode of sepsis. J Clin Microbiol 1991;29:2768-73.
774
MUDER ET AL.
17. Wade JC, Schimpff SC, Newman KA, Wiernik PH. Stuphylococcus epidemidis: an increasing cause of infection in patients with granulocytopenia. Ann Intern Med 1982;97:503-8. 18. Renaud F, Freney J, Etienne J, et al. Restriction endonuclease analysis of Sraphylococcus epidemidis DNA may be a useful epidemiological marker. J Clin Microbiol 1988;26:1729-34.
TRANSFUSION
Vol. 32. No. 8-1992
Robert R. Muder, MD, Chief, Infection Control, Infectious Disease university ~~i~~ c, Pittsburgh, PA 15240. [Reprint requests] Ying C. Yee, MS, Microbiologist, Special Pathogens Laboratory, VA Medical Center. John D. Rihs, BS, Supervisor, Microbiology Laboratory, VA Medical Center. Mark Bunker, MD, Chief, Blood Bank, VA Medical Center.
section, VA Medical
without incident. Approximately 90 minutes after the initiation of a 10-unit transfusion of pooled donor platelets, he abruptly experienced rigors, wheezing, dyspnea, and fever to 104.6"F (measurement taken orally). The platelet transfusion was discontinued after a total transfusion of 200 mL. The patient was treated with acetaminophen, diphenhydramine, and albuteral and the application of a cooling blanket. Sixteen hours later, he was afebrile and asymptomatic. Cultures of four of four blood samples obtained by venipuncture of separate peripheral sites grew S. epidermidis within 24 hours of their drawing. Two samples were drawn 10 hours after the cessation of the platelet transfusion, and the other two were drawn 36 hours after the transfusion; the patient was afebrile at that time. The isolates were resistant to penicillin but susceptible to oxacillin, cephalothin, vancomycin, and clindamycin. We initiated vancomycin administration, 1 g intravenously every 12 hours, 8 hours after the final blood cultures were drawn. This was changed to nafcillin, 1 g every 4 hours, when the results of the susceptibilities of the blood isolates became known. Therapy was administered for a total of 10 days, during which time the patient remained afebrile. He was discharged in stable condition.
SEPTICEMIA IS A RARE but potentially fatal complication of the transfusion of blood components.'.* Because platelets are stored at 20 to 24°C for up to 5 days, contaminating bacteria initially present at very low levels may multiply to extremely high level^,^ which creates the potential for overwhelming septicemia. We report a case of septicemia due t o the transfusion of platelets contaminated with Staphylococcus epidermidis. Using DNA typing techniques, we were able to demonstrate that the contaminated platelet pool was the source of infection, and that the patient had persistent bacteremia with the organism for 36 hours. Case Report A 66-year-old man was admitted for blood transfusion. The patient had a long history of myelodysplastic disorder and had required repeated transfusions of red cells and platelets. He had a history of coronary artery disease, hypertension, dietcontrolled adult onset diabetes, peptic ulcer disease, and carotid artery stenosis requiring endarterectomy. The patient's medications included prednisone (20 mg twice daily), digoxin, isosorbide, captopril, folic acid, thiamine, and ranitidine. On examination he was in no distress. He was afebrile, with a pulse of 88 and a blood pressure of 132 over 50. The remainder of the examination was unremarkable. His hemogloHis white cell bin concentration was 7.7 g per dL (77 6). count was 1600 per mL (1.6 x 109/L),with 36 percent (0.36) segmented neutrophils, 8 percent (0.08) bands, 6 percent (0.06) metamyelocytes, and 50 percent (0.50) lymphocytes, and his platelet count was 10,000 per mL (10 x 109/L). On the evening of admission, a new peripheral cannula was inserted, and the patient received 2 units of packed red cells
Materials and Methods Culture After the transfusion was discontinued, the pooled platelets were stored at 4°C for 10 hours before culture. We aspirated samples from the bag of residual pooled platelets by sterile technique and plated them on 5-percent sheep's blood agar plates incubated at 35 and 30°C in room air, chocolate agar plates incubated at 37°C in 5-percent C02, and reducible anaerobic blood agar plates incubated at 35°C anaerobically. We inoculated additional aliquots into thioglycolate broth and incubated it at 30 and 35°C. Speciation of staphylococcal isolates was performed with the bacterial identification system (VITEK, Hazelwood, MO).
From the Medical Service, Laboratory Service, and Special Pathogens Laboratory, Veterans Affairs Medical Center. and the Departments of Medicine and Pathology, University of Pittsburgh School of Medicine, Pittsburgh, Pennsylvania. Received for publication November 13, 1991; revision received March 6, 1992, and accepted March 25, 1992.
DNA analysis We inoculated a single colony of each S. epidemidis isolate into 5 mL of trypticase soy broth and incubated the tubes for
771
772
TRANSFUSION Vol. 32, No. 8-1992
MUDER ET AL.
FIG. 1. Gel electrophoresis of DNA fragments of staphylococcal isolates after digestion with EcoRI (A)and Hind111 (B).Lane 1 is A DNA (a molecular size marker); Lane 2 is the platelet pool isolate; Lanes 3 through 6 are isolates obtained from four separate blood cultures; Lane 7 is the isolate from the patient’s skin.
18 hours at 37°C. We centrifuged bacterial cells from 1.5 mL of suspension at 9500 x g for 2 minutes, resuspended them in 567 pL of buffer (10 mM [lo mmol/L] Tris, pH 7.9, 1 mh4 [l mmolL] EDTA), and lysed them with 20 p L of lysostaphin for 30 minutes at 37°C. After lysis, whole-cell DNA was extracted and purified4 and incubated with either EcoRI or H i d 1 1 (Toyobo. Ltd., Tokyo, Japan) for 18 hours in accordance with the manufacturer’s recommendations. Electrophoresis of digested DNA was carried out in 0.7-percent agarose in Trisborate buffer for 20 hours at 1volt per cm. Gels were stained for 1 to 1.5 hours in 0.5 pg per mL of ethidium bromide and photographed (665 film, Polaroid Corp., Cambridge, MA).
Results Review of blood bank records showed that the platelet pool consisted of 10 individual donor units collected 5 days before transfusion. The units were stored at room temperature and pooled shortly before administration. Gram’s stain of the residual pooled platelets showed no organisms, but culture yielded S. epidennidis at an approximate concentration of lo5 organ-
isms per mL. The bags containing the individual units had been discarded and were not cultured. Culture of the patient’s antecubital fossa yielded S. horninis; no staphylococci were isolated from his anterior nares. The S. epidennidis isolated from the residual pooled platelets and from each of the four blood isolates had identical restriction patterns after digestion with EcoRI (Fig. 1A) and Hind111 (Fig. 1B).The staphylococcal isolate from the patient’s skin was distinct from the platelet and blood isolates.
Discussion Platelet concentrates stored at room temperature maintain adequate viability and function for up to 7 days.5 With extended storage, small numbers of bacteria inadvertently introduced at the time of collection or during processing could multiply to high numbers. Microbiologic surveys6-’0 have shown that 0 to 6.3 percent of individual platelet units are contaminated.
TRMSFUSION 1992-vOi. 32. N ~ a.
773
PLATELET TRANSFUSION BACTEREMIA
Contamination of fresh platelet pools ranges from 7.4 to 10.9 percent; the proportion may rise as high as 38 percent after 3 days of s t ~ r a g e . ~S.* ' ~epidennidis and diphtheroid bacilli, components of the normal skin flora, are the most frequently isolated bacteria, which indicates that contamination at the time of collection is the most likely source. Clinically recognized septicemia due to platelet transfusion is extremely rare.2 Bacterial counts are typically quite low in contaminated units,*-1° and the vast majority are transfused without adverse effect. However, the inoculation of platelet concentrates with a single colony of S. epidennidis can result in logarithmic phase growth with colony counts of lo3 per 0.1 mL at 3 days and lo7 per 0.1 mL at 6 days.' Duration of platelet storage appears to be an important determinant of the risk of sepsis. The transfusion of multidonor platelets stored for 5 days is associated with an incidence of sepsis five times greater than that seen with the transfusion of platelets stored for 4 days or less.2 To date, 32 cases (including the present one) of platelet-associated bacteremia have been rep~rted;'-'~~-'~ seven of the episodes have been fatal. Given the frequency of isolation of S. epidennidis from platelet concentrates, it is interesting to note that only nine prior cases of plateletassociated septicemia have been reported with coagulase-negative staphylococci as the cause. S. epidennidis, although of limited pathogenicity in the normal host, may cause sustained bacteremia in neutropenic patient~.~'The relatively low number of reported cases associated with platelet transfusion could be due to the frequency of concomitant antibiotic therapy in these patients or to frequent dismissal as a blood culture contaminant. Our patient's primary physicians initially dismissed the first two sets of blood cultures as contaminants and were unconvinced of the significance of the results until the residual platelets and the third and fourth sets of blood cultures were reported as positive. Subsequently, we were able to confirm by DNA typing that all four blood isolates were identical to the S. epidennidis contaminating the platelet pool and were different from the staphylococci isolated from the patient's skin. Restriction endonuclease analysis of S. epidennidis appears to be a reproducible method of demonstrating clonal identity among isolates. It discriminates between strains more effectively than biotyping or phage typing and is more stable than plasmid content or antibiotic susceptibility profiles.'* This case underscores the potential difficulties in differentiating transfusion-associated septicemia from noninfectious transfusion reactions. The isolation from the blood of organisms typically regarded as blood culture contaminants, such as coagulase-negative staphylococci, should be regarded as potentially significant. Our observations demonstrate the utility of molecular typing in confirming the source of transfusion-associated septi-
cemia and in differentiating a true pathogen from a skin contaminant. One other case of platelet-associatedS. epidennidis bacteremia has been confirmed by field-inversion gel electrophoresis and ribotyping.16 Molecular typing could be especially valuable in investigating common-source outbreaks of septicemia in which multiple patients are involved. Bacteria isolated from patients' blood or from blood components in suspected cases of transfusion-associated septicemia should be saved for further typing in case epidemiologic investigation suggests the need for it.
Acknowledgment The authors thank Victor L. Yu, MD, for review of the manuscript.
References 1. Morduchowicz G, Pitlik SD, Huminer D, et al. Transfusion reactions due to bacterial contamination of blood and blood products. Rev Infect Dis 1991;13:307-14. 2. Morrow JF, Braine HG, Kickler TS, Ness PM, Dick JD, Fuller AK. Septic reactions to platelet transfusions. A persistent problem. JAMA 1991;266:555-8. 3. Braine HG, Kickler TS, Charache P, et al. Bacterial sepsis secondary to platelet transfusion: an adverse effect of extended storage at room temperature. Transfusion 1986;26:391-3. 4. Ausbel FM, Brent R, Kingston RE, et al. Short protocols in molecular biology. New York: Wiley, 1989. 5. Simon TL, Nelson El, Carmen R, Murphy S. Extension of platclet concentrate storage. Transfusion 1983;23:207-12. 6. Silver H, Sonnenwirth AC, Beisser LD. Bacteriologic study of platelet concentrates prepared and stored without refrigeration. Transfusion 1970;10:315-6. 7. Wrenn HE, Speicher CE. Platelet concentrates: sterility of 400 single units stored at room temperature. Transfusion 1974;14171-2. 8. Cunningham M, Cash JD. Bacterial contamination of platelet concentrates stored at 20°C. J Clin Pathol 1973;26:401-4. 9. Buchholz DH, Young VM, Friedman NR, Reilly JA, Mardiney MR Jr. Bacterial proliferation in platelet products stored at room temperature. Transfusion-induced Enterobacter sepsis. N Engl J Med 1971;285:429-33. 10. Buchholz DH, Young VM, Friedman NR, Reilly JA, Mardiney MR Jr. Detection and quantitation of bacteria in platelet products stored at ambient temperature. Transfusion 1973;13:268-75. 11. Rhame FS, Root RK, Macbwry JD, Dadisman TA, Bennett JV. Salmonella septicemia from platelet transfusions. Study of an outbreak traced to a hematogenous carrier of Salmonella choleraesuir. Ann Intern Med 1973;78:633-41. 12. Blajchman MA, Thornley JH. Richardson H, Elder D, Spiak C, Racher J. Platelet transfusion-induced Serrafia marcescem sepsis due to vacuum tube contamination. Transfusion 1979;19:39-44. 13. Van Lierde S, Fleisher GR, Plotkin SA, C a m p s JM. A case of platelet transfusion-relatedSerratia marcescem sepsis. Pediatr Infect Dis 1985;4:293-5. 14. &now PM, Weiss LM, Weil D, Rosen NR. Escherichia coli sepsis from contaminated platelet transfusion. Arch Intern Med 1986;146:321-4. 15. Heal JM, Jones ME, Forey J, Chaudhry A. Stricof RL. Fatal Salmonella septicemia afkr platelet transfusion. Transfusion 1987;27:2-5. 16. Shayegani M, Parsons LM. Waring AL. et al. Molecular relatedness of Staphylococcus epidemidis isolates obtained during a platelet transfusion-associated episode of sepsis. J Clin Microbiol 1991;29:2768-73.
774
MUDER ET AL.
17. Wade JC, Schimpff SC, Newman KA, Wiernik PH. Stuphylococcus epidemidis: an increasing cause of infection in patients with granulocytopenia. Ann Intern Med 1982;97:503-8. 18. Renaud F, Freney J, Etienne J, et al. Restriction endonuclease analysis of Sraphylococcus epidemidis DNA may be a useful epidemiological marker. J Clin Microbiol 1988;26:1729-34.
TRANSFUSION
Vol. 32. No. 8-1992
Robert R. Muder, MD, Chief, Infection Control, Infectious Disease university ~~i~~ c, Pittsburgh, PA 15240. [Reprint requests] Ying C. Yee, MS, Microbiologist, Special Pathogens Laboratory, VA Medical Center. John D. Rihs, BS, Supervisor, Microbiology Laboratory, VA Medical Center. Mark Bunker, MD, Chief, Blood Bank, VA Medical Center.
section, VA Medical
