A 47-year-old woman was admitted to the hospital for complex spinal surgery. The surgery went well without complications, and postoperatively she was transferred to a general surgical ward. Shortly thereafter, she spiked a fever, became tachycardic, hypotensive, and hypoxic, and developed a red rash across her chest. She was reintubated (placed back on the mechanical ventilator), given an infusion of dopamine to maintain adequate blood pressure, and transferred to the intensive care unit (ICU). She was found to have severe septic shock and developed multiorgan system failure. On initial evaluation, the clinicians were puzzled and confused because there was no clear cause for her septic shock.

On the same day, a 76-year-old man with coronary artery disease and a prosthetic aortic valve was admitted for spinal surgery. The procedure went well, and he was stable and transferred to a general surgical ward postoperatively. Later that evening, he developed tachycardia, hypotension, and hypoxia, requiring reintubation and transfer to the ICU. He was found to have sepsis and, despite extensive diagnostic testing, the clinical team could not identify a clear cause for his decompensation.

Given the similarity in clinical course, the hospital investigated the two cases. Upon detailed review, the blood bank discovered that both patients had received intraoperative platelet transfusions from the same batch of platelets. With further testing, it was determined that the entire batch of platelets was contaminated with Staphylococcus aureus, a virulent and aggressive bacteria often found in hospitals.

The 47-year-old woman remained critically ill for many days and had Staphylococcus aureus in her bloodstream for more than a week despite antibiotic therapy. She had a long and complicated hospitalization, but she was ultimately discharged in stable condition. The bacterium was never cultured from the blood of the 76-year-old man, but he remained febrile in the ICU for many days. Given his prosthetic valve, an echocardiogram was obtained that showed possible bacterial endocarditis (infection of his heart valve with the bacteria). In addition to a prolonged hospitalization, he required 6 weeks of intravenous antibiotics as a result of the contaminated platelet transfusion.

by Roslyn Yomtovian, MD

Frequency of Transfusion-Transmitted Viral Agents

Despite the ongoing quest for the holy grail of blood transfusion medicine—a zero-risk blood supply (1)—numerous challenges to safety remain.(2-4) It was assumed for many years that with the development of sophisticated testing for viral agents, in particular nucleic amplification testing (NAT) for HIV, hepatitis C virus (HCV), and more recently West Nile virus (WNV), blood safety would be assured. In fact, such testing has dramatically reduced the risk of viral transmission. Today, it is mathematically estimated (estimates are needed since the actual number is so low) that between 1 in one and 1 in two million units of transfused blood units are contaminated with HIV or HCV.(5-7) These are remarkable accomplishments. Prior to routine HIV screening of blood donors 23 years ago, and prior to routine HCV screening 18 years ago, the incidence of transfusion-transmitted HIV in San Francisco was estimated to be as high as 1 in 100 (8), and the incidence of transfusion-transmitted HCV, nationwide, was estimated to be 1 in 100.(9) The risk of WNV today is estimated to be less than 1:350,000, but the incidence is seasonal and higher in endemic areas.(10) The incidence of transfusion-transmitted hepatitis B virus (HBV), to which NAT has not yet been routinely applied, is approximately 1:200,000.(5,6) Comparatively, the risk of receiving ABO-incompatible blood (the wrong blood type) is approximately 1:38,000; approximately half of these transfusions will result in a transfusion reaction.(11) It is clear that concerted efforts to limit viral contamination of blood products have been very successful.

Emergence of Platelet Bacterial Contamination as a Transfusion Risk

Although bacterial contamination of blood has been a well-recognized and documented problem in transfusion therapy for more than 60 years (12), with the reduction in the risk of transfusion-related viral transmission, bacterial contamination has become the leading transfusion-transmitted infectious risk. Until recently, bacterial contamination has been underemphasized, in part because the risks of HIV and HCV were more dramatic, and because the rigorous processing requirements for blood and blood components gave the public the sense that the risk of other contaminants was vanishingly low.(13) In addition, even when it did occur, bacterial contamination has frequently gone unrecognized (described in more detail below).

Bacterial contamination, while associated with all blood components (packed red blood cells, fresh frozen plasma, etc.), is of particular concern with platelets. This is because, unlike other blood components that are stored refrigerated or frozen (which inhibits bacterial growth), platelets are stored at room temperature for up to 5 days (and in some instances 7 days). If the platelet unit is contaminated, bacteria can flourish and grow quickly in the warm, nutrient-rich platelet storage bag. And, because platelets are generally administered to patients who are prone to sepsis (a type of severe infection)—those with hematologic/oncologic disorders—it is common to attribute new symptoms of infection to the underlying condition rather than to the platelet transfusion.

Frequency and Microbiology of Platelet Bacterial Contamination

Platelet bacterial contamination is estimated to occur in as many as 1 in 2000 platelet units. The clinical significance of platelet bacterial contamination is highly variable—reactions range from no obvious clinical effects to severe and life-threatening. The frequency of these outcomes, based on 46 contaminated platelet products transfused at a single facility, is estimated in Table 1.(14) The severity of the clinical reaction is significantly correlated with the pathogenicity and quantity of the organisms transfused. In general, gram-negative organisms and some gram-positive organisms (Bacillus cereus, Streptococcus bovis, and Staphylococcus aureus) are more virulent species, while coagulase-negative staphylococci and viridans group streptococci are less virulent species. Propionibacterium acnes, a frequent skin contaminant, was not isolated by this facility since anaerobic culture methods were not used in this study. This extremely slow-growing gram-positive bacillary organism has very low virulence when associated with platelet bacterial contamination.

In an ongoing surveillance program of platelet bacterial contamination at our facility covering more than 15 years (14), during which bacteria were isolated from 54 platelet products, the most common isolates, as in most studies, were coagulase-negative staphylococci (Table 2). Fortunately, despite being the most frequent isolate, coagulase-negative staphylococci are infrequently implicated in severe transfusion reactions or transfusion-associated fatalities.(14,15) Platelet units contaminated with Staphylococcus aureus or endotoxin-producing gram-negative organisms can cause more serious transfusion reactions and occasional fatalities as these bacteria are more pathogenic.(14-16) In a review of transfusion-associated fatalities reported to the U.S. Food and Drug Administration (FDA) from 1995–2004, 85 of 665 reported deaths (13%) were attributed to bacterial contamination of blood products (primarily platelets), making bacterial contamination of blood products the third leading reported cause of death from transfusion therapy.(16) By comparison, the two leading causes of transfusion-related deaths reported to the FDA for fiscal years 2005 and 2006 were transfusion-related acute lung injury (TRALI) at 51% and non-ABO hemolytic transfusion reactions at 20%.(17) Most fatalities from bacterial contamination, a total of 58/85 (68%), were associated with gram-negative organisms.(16) The most common gram-negative organism causing death was Klebsiella pneumoniae (12/58); this was followed by Escherichia coli (9/58) and Serratia marcescens (5/58). The other culprit bacteria were a wide array of organisms.(16)

Etiology of Platelet Bacterial Contamination

How do platelets become contaminated? Platelets for transfusion are obtained from donors through standard venipuncture and aseptic blood-drawing technique. The primary source of bacteria is believed to be skin-associated microorganisms that frequently reside beneath the skin surface in sebaceous glands and hair follicles and therefore escape standard skin-cleansing procedures. Bacterial contamination also results from lack of careful aseptic procedures in prepping the donor skin for phlebotomy. (For a discussion of contamination in drawing blood, see this AHRQ WebM&M commentary.) In addition, bacterial contamination may result from subclinical donor bacteremia or faulty manufacturing processes of the platelet storage bag.(13) As the bacteria often contaminate the bag at the time of phlebotomy from a particular donor, there is a higher risk of bacterial contamination when blood is pooled from multiple donors. Accordingly, single-donor pools of platelets are generally safer than those from multiple donors.

Detection and Reduction of Platelet Bacterial Contamination

An effort to reduce platelet bacterial contamination was mandated by the College of American Pathologists (CAP) in 1994 and by the American Association of Blood Banks (AABB) in 2004. In addition to urging facilities to ensure aseptic technique when obtaining donor platelets, the CAP requires that a method be in place to detect the presence of bacteria in platelet components. The AABB requires that methods be in place to both reduce and detect bacteria in platelet components. Methods that have been implemented to reduce or detect bacteria in platelets are listed in Table 3. The most common of these techniques, applied in combination, are use of single-donor apheresis platelets with sample diversion during donor collection (to limit bacterial contamination) and application of bacterial culture following collection (to detect bacterial contamination). As described earlier, one way to decrease the risk of contamination (which is often employed in transfusions to high-risk recipients such as immunosuppressed patients) and facilitate screening for contaminants (Table 3) is to use single-donor platelet units (if available). A particularly hopeful technology is pathogen inactivation technology (Cerus Intercept System and the Navigant Mirasol Technology). Both technologies are currently being studied in the United States, with the former used in several countries (Table 3); some observers believe that pathogen inactivation technology might completely eliminate the risk of bacterial contamination.

Clinical Management of Suspected Platelet Bacterial Contamination

It is important for clinicians to recognize that, despite all present efforts to reduce and detect bacteria in platelets, any patient receiving platelet products is at risk for receipt of a contaminated unit. Therefore, patients receiving platelets should be carefully monitored for signs and symptoms of a transfusion reaction—fever, tachycardia, chills, rigors, hypotension, skin flushing, or a sense of "impending doom"—within 4 hours after the receipt of the transfusion. Many different types of transfusion reactions (separate from bacterial contamination) can lead to the development of fever and systemic symptoms, and differentiating minor from life-threatening reactions can be difficult. One guiding principle relates to the temperature rise: prospective data reveal that a 2°C rise in temperature is associated with a 40% chance that the cause is bacterial contamination.(18)

Clinicians should also be aware that gram-positive organisms, including Staphylococcus aureus, the culprit in this case, may cause delayed symptoms—some cases might not present for up to 24 hours following the platelet transfusion.(13) Awareness of the delayed response may be particularly important for those patients who are transfused platelets as outpatients and sent home soon thereafter. These patients should be instructed to check for signs and symptoms suggestive of bacterial contamination (as described above) and told to call their health care provider immediately if they occur.

As soon as a patient is suspected of receiving a bacterially contaminated platelet unit, the transfusion must be discontinued immediately and blood cultures (aerobic and anaerobic) drawn. The blood bank must be notified and given the bag containing the residual platelets with its tubing properly sealed off to prevent leakage and environmental contamination. The blood bank will perform a stat gram stain on a sample drawn from the implicated platelet bag and set up appropriate cultures. With gram-negative organisms, there is a strong correlation between the quantity transfused and the clinical outcome—relating to the quantity of the endotoxin administered—and therefore promptness in discontinuing a potentially bacterially contaminated unit is of the utmost importance. Clinical patient management will depend on the extent of the clinical findings but should begin without delay. This includes appropriate antibiotic coverage and aggressive management for any early signs of sepsis. Procurement of an infectious disease consultation should be considered. Federal statutes require that any death found to be associated with the administration of a bacterially contaminated blood product must be reported to the FDA within 7 days.(19) Typically, the blood bank, in concert with the clinical service, organizes the report for submission to the FDA.

Conclusion

The clinical case provided at the outset of this commentary provides an excellent real-life example of platelet bacterial contamination. Blood centers often divide apheresis platelet units into two (and sometimes three) portions to be used in two (or three) separate patients. Because of this, it is critically important, when a recipient of a contaminated platelet unit is identified, to notify the hospital blood bank/transfusion service immediately so that the recipient(s) of the other platelet portion(s), if any, can be identified as soon as possible. In the case at hand, it was fortuitous that both platelet recipients were in the same facility at approximately the same time. While both recipients developed sepsis, the onset occurred after the platelet transfusion had been concluded. This delay in onset of sepsis is a typical pattern for gram-positive organisms, which contributes to the underdiagnosis and underrecognition of this problem. While procedures are in place to prevent and detect bacteria in platelet units, these are imperfect, and new procedures remain under development (Table 3). In the meantime, platelet bacterial contamination is an uncommon but inevitable risk of platelet transfusion therapy.

Take-Home Points

• Because of the nature of donor phlebotomy and platelet storage, bacterial contamination of platelets is currently an unusual but inevitable risk of platelet transfusion therapy.
• Clinicians and nurses caring for patients undergoing platelet transfusion therapy should monitor patients for signs and symptoms that suggest the possibility of a septic reaction and be prepared to discontinue the transfusion and begin treatment immediately.
• High-risk immunocompromised patients should receive, if possible, platelets from single donors, as they are at lower risk for bacterial contamination than pooled units.
• Because reactions associated with platelet bacterial contamination may be delayed for up to 24 hours or even longer (especially in those contaminated with gram-positive organisms), a patient receiving platelets as an outpatient should be instructed to check for signs and symptoms suggestive of bacterial contamination and to call the health care provider immediately if they occur.
• As with any blood component, platelets should be used only when judged to be clinically judicious and efficacious.
  

Roslyn Yomtovian, MD

VA National Quality Scholar

Louis Stokes Cleveland Department of Veterans Affairs Medical Center

Adjunct Professor, Department of Pathology
Case Western Reserve University School of Medicine

1. Zuck TF. Greetings—A final look back with comments about a policy of a zero-risk blood supply. Transfusion. 1987;27:447-448. [go to PubMed]

2. Stramer SL. Current risks of transfusion-transmitted agents: a review. Arch Pathol Lab Med. 2007;131:702-707. [go to PubMed]

3. Eder AF, Chambers LA. Noninfectious complications of blood transfusion. Arch Pathol Lab Med. 2007;131:708-718. [go to PubMed]

4. Dodd RY. Current risk for transfusion transmitted infections. Curr Opin Hematol. 2007;14:671-676. [go to PubMed]

5. Dodd RY, Notari EP IV, Stramer SL. Current prevalence and incidence of infectious disease markers and estimated window-period risk in the American Red Cross blood donor population. Transfusion. 2002;42:975-979. [go to PubMed]

6. Busch MP, Kleinman SH, Nemo GJ. Current and emerging infectious risks of blood transfusions. JAMA. 2003;289:959-962. [go to PubMed]

7. Stramer SL, Glynn SA, Kleinman SH, et al; National Heart, Lung, and Blood Institute Nucleic Acid Test Study Group. Detection of HIV-1 and HCV infections among antibody-negative blood donors by nucleic acid-amplification testing. N Engl J Med. 2004;351:760-768. [go to PubMed]

8. Busch MP, Young MJ, Samson SM, et al; for Transfusion Safety Study Group. Risk of human immunodeficiency virus (HIV) transmission by blood transfusions before the implementation of HIV-1 antibody screening. Transfusion. 1991;31:4-11. [go to PubMed]

9. Gresens CJ, Holland PV. The disappearance of transfusion-transmitted hepatitis C virus infections in the United States. Clin Liver Dis. 2001;5:1105-1113. [go to PubMed]

10. Petersen LR, Epstein JS. Problem solved? West Nile virus and transfusion safety. N Engl J Med. 2005;353:516-517. [go to PubMed]

11. Linden JV, Wagner K, Voytovich AE, Sheehan J. Transfusion errors in New York state: an analysis of 10 years' experience. Transfusion. 2000;40:1207-1213. [go to PubMed]

12. Yomtovian R. Bacterial contamination of blood: lessons from the past and road map for the future. Transfusion. 2004;44:450-460. [go to PubMed]

13. Yomtovian R, Palavecino E. Bacterial Contamination of Blood Products—History and Epidemiology. In: Brecher ME, ed. Bacterial Contamination of Blood Products. Bethesda, MD: American Association of Blood Banks; 2003.

14. Jacobs MR, Good CE, Lazarus HM, Yomtovian RA. Relationship between bacterial load, species virulence, and transfusion reaction with transfusion of bacterially contaminated platelets. Clin Infect Dis. 2008;46:1214-1220. [go to PubMed]

15. Kuehnert MJ, Roth VR, Haley NR, et al. Transfusion-transmitted bacterial infection in the United States, 1998 through 2000. Transfusion. 2001;41:1493-1499. [go to PubMed]

16. Niu MT, Knippen M, Simmons L, Holness LG. Transfusion-transmitted Klebsiella pneumoniae fatalities, 1995 to 2004. Transfus Med Rev. 2006;20:149-157. [go to PubMed]

17. U.S. Food and Drug Administration, Center for Biologics Evaluation and Research. Fatalities Reported to FDA Following Blood Collection and Transfusion: Annual Summary for Fiscal Years 2005 and 2006. Bethesda, MD: U.S. Food and Drug Administration. Available at: http://www.fda.gov/cber/blood/fatal0506.pdf.

18. Chiu EK, Yuen KY, Lie AK, et al. A prospective study of symptomatic bacteremia following platelet transfusion and of its management. Transfusion. 1994;34:950-954. [go to PubMed]

19. U.S. Food and Drug Administration, Center for Biologics Evaluation and Research. Guidance for Industry: Notifying FDA of Fatalities Related to Blood Collection or Transfusion. Bethesda, MD: U.S. Food and Drug Administration; 2003. Available at: http://www.fda.gov/cber/gdlns/bldfatal.htm.

20. American Association of Blood Banks. Association Bulletin #03-12 Further Guidance on Methods to Detect Bacterial Contamination of Platelet Components. Bethesda, MD: American Association of Blood Banks; 2003.

Table 1. Clinical Outcomes Associated with Transfused Bacterially Contaminated Platelet Products (N = 46 transfused platelet units).(14)

Table 2. Bacterial Isolates from 15-Year+ Surveillance Program (N = 54 cultured platelet units).(14)*

*54 platelet units were contaminated, but only 46 were transfused, as noted in Table 1.
**These isolates collectively comprise the group of coagulase-negative staphylococci.
***This isolate was in a platelet unit co-contaminated with viridans group streptococcus.

Table 3. Methods to Interdict Platelet Bacterial Contamination.(20)