Aug 20, 2004 -- This modern medical mystery story occurred last year in Georgia. Three patients who all received dialysis on the same machine -- one after another -- were infected with the West Nile virus. Did they become infected at the dialysis center or somewhere else? Two lived in the same neighborhood and the third lived about a mile away. While there definitely was a high potential for mosquito exposure where they lived, no West Nile-positive mosquitoes were found.
This cluster of dialysis patients with West Nile infections suggests possible transmission of the virus in the dialysis center. West Nile virus transmission through blood transfusion, organ transplantation, in utero and possibly through breast milk has been described. But in this situation where the dialysis center was a common denominator, the epidemiologic investigation looked at possible cross-contamination of supplies, equipment, medications and procedures. No source or explanation for the three cases of West Nile Virus has been found. It remains on the books as a modern medical mystery.
- West Nile Encephalitis (article)
- West Nile, Avoid Mosquito Bites to Avoid Infection (Health e-Tool)
- How to Choose Use Insect Repellents Safely (Health e-Tool)
- Dialysis (article)
Possible Dialysis-Related West Nile Virus Transmission --- Georgia, 2003
In October 2003, the Georgia Division of Public Health (DPH) was notified of two patients from the same county with confirmed West Nile virus (WNV) disease who had received hemodialysis on the same day and on the same dialysis machine. The two dialysis patients (patients A and C) had the only confirmed cases of human WNV disease reported in their county in 2003. Review of the dialysis center's records indicated that another patient (patient B) had received dialysis on the same machine between these two patients on the same day. This report summarizes results of the epidemiologic investigation, which suggested that WNV might have been transmitted at the dialysis center. Hemodialysis centers should adhere strictly to established infection-control procedures to avoid WNV transmission through dialysis.
Patient A. The first patient, who received dialysis on the machine (machine A) in late August, was a man aged 77 years with a history of hypertension and end-stage renal disease (ESRD). Eight days after dialysis, patient A was hospitalized with a 48-hour history of fever, chills, confusion, and anorexia. Blood cultures were negative. Serologic testing of serum revealed IgM and IgG antibodies to WNV by enzyme-linked immunosorbent assay (ELISA) and a higher neutralizing antibody titer to WNV (1:1,280) than to St. Louis encephalitis virus (SLEV) (1:320). Patient A had not received a blood transfusion <30 days before symptom onset. After a 9-day hospitalization, he was afebrile at discharge.
Patient B. The second patient, who received dialysis on machine A between patients A and C, was a woman aged 71 years with a history of type 2 diabetes, ESRD, and hypertension. Dialysis center and hospital records and patient interview revealed no symptoms of illness during late August or early September, and patient B had not received a blood transfusion in July, August, or September. In addition, she had never received a flavivirus vaccination (which might elicit cross-reactive antibody to serologic tests for WNV) or traveled outside the United States. A serum sample obtained 42 days after dialysis was uninterpretable for IgM antibody to WNV (i.e., because of high background reactivity), negative for IgM to SLEV, and positive for IgG to both WNV and SLEV by ELISA; the neutralizing antibody titers were 1:160 to WNV and 1:10 to SLEV. A second specimen taken from this patient 84 days after dialysis was negative for IgM antibody to WNV and SLEV by ELISA, positive for IgG to both WNV and SLEV by ELISA, and had neutralizing antibody titers of 1:320 to WNV and 1:20 to SLEV.
Patient C. The third and last patient to receive dialysis on machine A on the same day in late August was a man aged 60 years with a history of type 2 diabetes, hypertension, alcoholism, recent onset of ESRD, and prostate cancer. Nineteen days after his dialysis procedure, patient C was admitted to a local hospital with fever, chills, altered mental status, and cachexia. After admission, he had seizures and was intubated and placed on a ventilator. Analysis of cerebrospinal fluid (CSF) indicated a mild pleocytosis (67 white blood cells [62% polynuclear cells, 38% mononuclear cells] and five red blood cells/mm3) and an elevated protein level (122 mg/dL). Computerized tomography scans of the patient's brain on the second and tenth days of hospitalization revealed bilateral lacunar infarcts, white matter changes, and cortical and subcortical atrophy. Serologic tests of serum were positive for IgM and IgG antibodies to WNV by ELISA. The neutralizing antibody titer was higher to WNV (1:1,280) than to SLEV (1:20). Patient C had not received a blood transfusion <30 days before symptom onset. Twenty days after admission, he had a high fever and respiratory failure and died.
DPH and the local health department investigated practices and procedures at the dialysis center. No breakdowns in disinfection procedures for the dialysis machine or dialyzers and no breaches in infection-control practices were revealed. All bloodline attachments to the dialysis machine were disposable and discarded after each dialysis session. Patient blood samples were withdrawn from the bloodline for testing on a monthly basis, unless otherwise directed by the physician. Medications were bottled in multiple-dose units but were drawn by using a needle in a separate medication room and injected into the patients' bloodlines with a syringe. Both the needle and syringe were then discarded. No single medication was administered to all three patients on the day of their dialyses. However, patients A and B had received a common medication, and patients A and C also had received a common medication, although most likely from separate vials. Blood samples from three other patients who had received dialysis on machine A on the previous day and on the following day were all negative for IgM and IgG antibodies to WNV.
Patients A and C resided in the same neighborhood, 0.2 miles apart, and patient B resided approximately 1 mile away from this neighborhood. An environmental assessment around the homes of patients A and C and in the neighborhood where they resided revealed a high potential for mosquito exposure, including lack of window screens, barrels of stagnant water, and wooded areas between homes. Mosquito surveillance of the area in mid-October indicated that Culex quinquefasciatus mosquitoes were the most abundant mosquito species; however, no WNV-positive mosquitoes were identified, as would be expected from mosquito collections obtained so late in the transmission season. Pesticide spraying to kill adult mosquitoes in the neighborhood was conducted two days after surveillance. Three neighbors of patients A and C submitted blood samples for testing for WNV; all samples were negative for IgM and IgG antibodies to WNV.
Reported by: CE Smith, MD, JM Jenkins, D Staib, PJ Newell, MD, KJ Mertz, MD, S Lance-Parker, DVM, RM Kelly, PhD, KA Bryant, MPH, Georgia Div of Public Health; EB Hayes, MD, GL Campbell, MD, Div of Vector-Borne Infectious Diseases, National Center for Infectious Diseases; A Srinivasan, MD, D Jernigan, MD, M Arduino, MD, Div of Healthcare Quality Promotion; K Abe, PhD, EIS Officer, CDC.
Source: Morbidity and Mortality Weekly Report, published by the CDC, August 20, 2004 / Vol. 53 / No. 32.