Sign In
Individual Sign In
Create an Account
Institutional Sign In
OpenAthens Shibboleth
From the Centers for Disease Control and Prevention
March 21, 2007

West Nile Virus Transmission Through Blood Transfusion—South Dakota, 2006

JAMA. 2007;297(11):1186-1188. doi:10.1001/jama.297.11.1186

MMWR. 2007;56:76-79

West Nile virus (WNV) transmission through blood transfusion was first reported in 2002,1,2 prompting rapid implementation of nationwide screening of blood donations for WNV by 2003.3,4 Screening strategies were developed using minipool nucleic acid-amplification testing (MP-NAT) based on six or 16 pooled donor samples. To improve sensitivity of WNV detection, blood-collection agencies (BCAs) later implemented enhanced screening by individual donation NAT (ID-NAT), most often used when a given trigger threshold of positive MP-NAT results is reached during the WNV transmission season.5,6 This approach has been effective, resulting in the detection and interdiction of approximately 1,400 potentially infectious blood donations during 2003-2005 and a reduction in recognized transfusion-transmission events.7 A total of 23 confirmed WNV transfusion-transmitted cases were reported in 2002, before screening was implemented; six probable or confirmed cases were detected in 2003 after MP-NAT screening was initiated, one was detected in 2004, and none were detected in 2005.7 This report describes the first WNV transfusion-transmission cases detected since the initiation of enhanced screening strategies using ID-NAT triggering. In 2006, two immunosuppressed patients had onset of West Nile neuroinvasive disease (WNND) after receiving blood products from a single infected donor despite a negative MP-NAT result at the time of donation. Although risk for transmission has been substantially reduced as a result of routine MP-NAT and triggered ID-NAT screening, clinicians should be reminded that transfusion-transmitted WNV infections can still occur, and that immunosuppressed patients are more likely to have onset of WNND.

In September 2006, the South Dakota Department of Health (SDDH) was notified of WNND in a man aged 82 years with end-stage renal disease who had received a kidney transplant on August 25, 2006. Four days after the transplant surgery, the patient received a transfusion of 2 units of packed red blood cells (PRBC) for anemia. Ten days after surgery, the patient was discharged to a long-term–care facility and continued to receive immunosuppressive therapy, including 750 mg of mycophenolate mofetil twice daily, 125 mg of cyclosporin twice daily, and 20 mg of prednisone daily. Twenty-one days after surgery, he had onset of fever, lethargy, and a peri-incisional hematoma, prompting his readmission to the hospital. The patient was treated empirically with broad-spectrum antimicrobial and antifungal agents. Two days after readmission (i.e., 23 days after transplant and 19 days after PRBC transfusion), his mental status deteriorated rapidly. The next day, his cerebrospinal fluid (CSF) had four white blood cells (WBC)/mm3, 46 red blood cells (RBC)/mm3; a protein level of 58 mg/dL, and a glucose level of 67 mg/dL. Anti-WNV immunoglobulin M (IgM) antibody was detected in both serum and CSF by IgM antibody-capture enzyme-linked immunosorbent assay (MAC-ELISA) performed at SDDH. When the patient was discharged to a long-term–care facility (36 days after his transplant surgery), his fever had resolved, and his mental status had improved.

Because the patient had been hospitalized during the 2 weeks before onset of his WNV-related illness, WNV transmission by organ transplantation or blood transfusion was considered more likely than transmission by mosquito bite. The kidney donor's premortem serum was negative for both anti-WNV IgM and WNV RNA by MAC-ELISA and reverse transcription polymerase chain reaction (RT-PCR). One other kidney transplant recipient from the same organ donor had no symptoms of WNV disease, and serum from this recipient was negative for both anti-WNV IgM and WNV RNA. Traceback investigation revealed that the patient with WNND had received blood products from six different donors during the 8 weeks before symptom onset. No donor samples from the time of donation were available for testing. However, all donors consented to have serum collected and tested for anti-WNV IgM. One donor, the source of 1 PRBC unit transfused into the patient with WNND 4 days after transplant, was IgM positive.

The implicated blood donor was a man from a rural area of South Dakota where substantial WNV activity in birds, mosquitoes, and humans occurred during the 2006 transmission season. He had not traveled outside of South Dakota during the month before his last donation on August 4, 2006. He did not report any symptoms consistent with WNV disease during the 2 weeks before this donation or during the 3 subsequent months. Because the BCA that collected the donation did not conduct routine screening for WNV, a sample of the donor's blood was sent for screening at an out-of-state BCA, where the MP-NAT test result for six pooled samples, including his donation, was negative. The out-of-state BCA had a policy of triggering ID-NAT after two WNV-positive MP-NAT results and more than one positive in 500 results during a rolling 7-day period. Two positive MP-NAT results had been detected by the testing BCA during the month before this donation; however, the positive results occurred more than 7 days apart and therefore did not trigger ID-NAT testing.

After identification of the IgM-positive donor, the platelet and fresh frozen plasma (FFP) co-components from his whole blood donation were traced. The platelet unit had been discarded without being transfused. The FFP unit had been transfused on August 10, 2006, into a man aged 60 years who had received a kidney transplant in 2001 for end-stage renal disease attributed to insulin-dependent diabetes mellitus. On the same day as the transfusion, he had undergone surgical repair of a spinal fracture caused by a fall. He received a transfusion of 15 blood products, including 6 units of FFP, one of which was from the blood donor described in this report. One week after surgery, he was discharged to a rehabilitation facility, where he continued to receive immunosuppressive therapy, including 4 mg tacrolimus twice daily and 500 mg mycophenolate mofetil three times daily. Eleven days after the surgery, he had onset of fever and was treated empirically with antimicrobial and antifungal agents. Fifteen days after surgery, he had onset of tremors, encephalopathy, and acute left arm paralysis unexplained by his previous injury but consistent with WNV-associated myelitis. The patient's CSF had four WBC/mm3, zero RBC/mm3, a protein level of 171 mg/dL, and a glucose level of 52 mg/dL. Anti-WNV IgM was detected in the CSF by MAC-ELISA at SDDH. The patient's fever, tremors, and encephalopathy resolved, but his left arm paralysis persisted at the time of transfer to an out-of-state hospital 5 days after symptom onset (20 days after surgery). Three months later, the patient remained in a long-term–care facility.

Reported by:

L Kightlinger, PhD, South Dakota Dept of Health. SM Brend, MPH, Iowa Dept of Health. J Gorlin, MD, Memorial Blood Centers, St. Paul; MM Kemperman, Minnesota Dept of Health. MJ Kuehnert, MD, National Center for Preparedness, Detection, and Control of Infectious Diseases (proposed); JJ Sejvar, MD, GL Campbell, MD, National Center for Zoonotic, Vector-Borne, and Enteric Diseases (proposed); EC Farnon, MD, KD Ellingson, PhD, EIS officers, CDC.

CDC Editorial Note:

This report describes two cases of probable transfusion-transmitted WNV from a common blood donor despite a negative MP-NAT result at the time of donation. The source of infection cannot be proven because blood samples or co-components from the implicated donation were unavailable for testing; however, evidence of WNND in two recipients of blood products from a common donor with serologic evidence of recent infection makes WNV transfusion-transmission probable. Because these two transfusion recipients were hospitalized for at least 2 weeks each before onset of WNND, neither patient was likely to have acquired infection from a mosquito bite. Furthermore, for the patient who underwent transplant surgery on August 25, transmission through the transplanted kidney is unlikely, given that neither the organ donor nor the other organ recipient had evidence of WNV infection.

Nationwide blood screening for WNV has been successful in preventing transfusion-transmitted WNV.3 However, as with all blood donation screening, infections can be transmitted to transfusion recipients on rare occasions despite negative donor test results. Although WNV transmission by blood transfusion is rare, the cases described in this report underscore the importance of clinical recognition, effective WNV blood screening strategies, and investigation coordination.

Transfusion-transmitted WNND might be difficult to recognize, but physicians should consider the disease as a possible diagnosis, particularly when unexplained neurologic complications occur in immunosuppressed patients after transfusion. Both patients described in this report were kidney transplant recipients who were immunosuppressed when they had onset of WNND after receiving blood product transfusions. Although WNND occurs in less than 1% of WNV infections overall (the majority of which are mosquito-borne), transplant patients who acquire WNV infections have an estimated forty-fold greater risk for developing WNND compared with the general population.8

The results of this investigation highlight the potential for false-negative MP-NAT results and the need to evaluate strategies for triggering ID-NAT donor screening; however, they also underscore the rarity of WNV transfusion-transmission events. Since ID-NAT triggering was fully implemented after the start of the 2004 transmission season, no transfusion-transmitted cases had been detected until the cases described in this report. Most false-negative MP-NAT results are caused by low-level viremic donor samples in which WNV is undetected by MP-NAT but is potentially identifiable by the more sensitive ID-NAT. Criteria for triggering ID-NAT differ among BCAs, but most are based on the number of positive MP-NAT results or a threshold rate for all positive results reached during a rolling 7-day period.5 Certain BCAs collect blood and perform NAT screening on-site; however, BCAs without the ability to screen for WNV send donor samples to remote (sometimes out-of-state) BCAs for testing. BCAs performing the testing determine when to trigger ID-NAT upon reviewing their own results.

To enhance the sensitivity of ID-NAT triggering, BCAs are considering the feasibility and utility of more standardized criteria for ID-NAT triggering and methods for enhanced communication among BCAs so that knowledge of positive screening results can be shared. BCAs face many challenges in WNV screening, including seasonal epidemics that are geographically unpredictable, limited resources for ID-NAT, and coordination of blood collection and testing that might be performed by multiple BCAs in a given geographic area. An additional tool for sharing of donor screening results might be useful to enhance ID-NAT triggering. The WNV Biovigilance Network,* currently being piloted by AABB (formerly known as the American Association of Blood Banks) to aggregate WNV blood donor screening results, is a model for successful collaboration. However, timeliness of reporting must be addressed to adapt the network for use in decisions regarding ID-NAT triggering.

Public health investigations involving patients with recent transplantation or blood transfusion are complex and often involve multiple states and local jurisdictions. Coordination among state and local health departments, clinicians, BCAs, hospital blood banks, transplant centers, and CDC often is required. Prompt reporting of suspected cases to local and state health departments, with assistance from CDC, will promote timely traceback investigations that can identify additional cases and prevent further transmission.

*Information available at

CDC.  West Nile virus activity—United States, October 10-16, 2002, and update on West Nile virus infections in recipients of blood transfusions.  MMWR. 2002;51:929-931PubMed
Pealer LN, Marfin AA, Petersen LR.  et al.  Transmission of West Nile virus through blood transfusion in the United States in 2002.  N Engl J Med. 2003;349:1236-1245PubMedArticle
Stramer SL, Fang CT, Foster GA.  et al.  West Nile virus among blood donors in the United States, 2003 and 2004.  N Engl J Med. 2005;353:451-459PubMedArticle
Busch MP, Caglioti S, Robertson EF.  et al.  Screening the blood supply for West Nile virus RNA by nucleic acid amplification testing.  N Engl J Med. 2005;353:460-467PubMedArticle
Custer B, Busch MP, Marfin AA, Petersen LR. The cost-effectiveness of screening the U.S. blood supply for West Nile virus.  Ann Intern Med. 2005;143:486-492PubMedArticle
Custer B, Tomasulo PA, Murphy EL.  et al.  Triggers for switching from minipool testing by nucleic acid technology to individual-donation nucleic acid testing for West Nile virus: analysis of 2003 data to inform 2004 decision making.  Transfusion. 2004;44:1547-1554PubMedArticle
Montgomery SP, Brown JA, Kuehnert M.  et al.  Transfusion-associated transmission of West Nile virus, United States 2003-2005.  Transfusion. 2006;46:2038-2046PubMedArticle
Kumar D, Drebot MA, Wong SJ.  et al.  A seroprevalence study of West Nile virus infection in solid organ transplant recipients.  Am J Transplant. 2004;4:1883-1888PubMedArticle