Nosocomial Hepatitis B Virus Infection Associated with Reusable Fingerstick Blood Sampling Devices -- Ohio and New York City, 1996

Fingerstick devices are widely used for capillary-blood sampling for glucose monitoring in patients with diabetes. In 1996, outbreaks of hepatitis B virus (HBV) infection occurred among patients with diabetes in an Ohio nursing home and in a New York City hospital. In response to these outbreaks, nursing-home and hospital personnel, state and local public health officials, and CDC conducted epidemiologic investigations. This report summarizes the investigations, which suggest that, in both outbreaks, HBV transmission was associated with use of spring-loaded fingerstick devices on multiple patients.

In the Ohio outbreak investigation, acute HBV infection was defined as a positive serologic test result for IgM antibody to hepatitis B core antigen (IgM anti-HBc) during June 1995-April 1996, and in the New York outbreak investigation, was defined as a positive serologic test result for IgM anti-HBc or seroconversion from hepatitis B surface antigen (HBsAg)-negative to HBsAg-positive during January-October 1996. Chronic HBV infection was defined as a positive serologic result for HBsAg and total anti-HBc and a negative result for IgM anti-HBc. Persons were considered immune to HBV if their test results were positive for antibody to HBsAg (anti-HBs) as a result of vaccination or positive for anti-HBc and/or anti-HBs as a result of natural infection. Persons were considered susceptible to HBV if their test results were negative for HBsAg, anti-HBc, and anti-HBs.

Nursing Home A, Ohio

From January through March 1996, acute hepatitis B was diagnosed in four residents of an Ohio nursing home. In March 1996, the local health department requested that the Ohio Department of Health (ODH) investigate these cases; after completing its assessment, ODH requested assistance from CDC. To determine the extent of transmission, in May 1996 CDC conducted a serosurvey of current and discharged residents who had resided at the nursing home during June 1995-April 1996. Serologic results were available for 74 (94%) of 79 current and former residents (17 other residents had died during this period). Nine (12%) persons had acute HBV infection, two (3%) had chronic infection, five (7%) were immune, and 58 (78%) were susceptible. The attack rate was 13% (nine of 67). HBsAg subtyping was performed on blood samples obtained from eight HBsAg-positive residents (seven with acute infection and one with chronic infection); all were subtype adw2, a common subtype among HBsAg-positive persons in the United States (1).

All 11 residents with acute or chronic infection had diabetes and received insulin injections on a routine or supplemental basis. Six susceptible residents also had diabetes, but only one had received insulin on a routine or supplemental basis. To determine risk factors for HBV infection, medical charts of residents who had acute infection or were susceptible were reviewed for history of medications, use of ancillary medical services, and frequency and types of percutaneous exposures, including injections and invasive procedures. Infection-control practices at the nursing home were assessed through interviews with personnel and direct observations of nursing procedures.

All acute infections occurred among persons with diabetes (relative risk {RR}=infinite; 95% confidence interval {CI}=10.1-infinite). Routine (daily) receipt of insulin (RR=19.0; 95% CI=4.5-730.0) and receipt of supplemental insulin (RR=50.8; 95% CI=7.1-360.5) also were highly associated with infection. Among persons who routinely underwent fingerstick capillary sampling, the attack rate was 53%, compared with zero attack rate among persons who did not require finger sticks (RR=infinite; 95% CI=13.2-infinite). The risk for infection was greater among persons who underwent greater than or equal to 60 finger sticks per month than among those who underwent less than 60 (RR=15.0; 95% CI=3.5-64.8). The mean number of venous blood draws during June 1995-April 1996 was approximately three times greater for case-patients than for noncase-patients (23 versus six, p less than 0.01). Because all case-residents underwent both fingerstick capillary sampling and phlebotomy, a possible association of venous blood draws with HBV infection could not be examined independently; however, examination of the dates of blood draws for case-patients did not indicate clustering in time.

Neither age, race, sex, room location, nor length of stay were associated with HBV infection, and there was no significant relation between HBV infection and history of blood transfusions; dialysis; medical procedures; intramuscular injections; medication with noninsulin subcutaneous drugs; or visits to the emergency department, podiatrist, dentist, or optometrist.

When the first case was identified in January 1996, the facility routinely used three Monojector{Registered} * (Sherwood-Davis and Geck {St. Louis, Missouri}) lancet devices for all residents who required finger sticks. The Monojector{Registered} is a pen-like device with a lancet and an end cap that rests on the patient's finger during blood sampling. Nursing personnel had routinely changed lancets between residents, but after the initial supply of end caps for each device had been used, end caps were no longer changed. After recognition of this outbreak, nursing home staff began using individual fingerstick devices for each resident who required finger sticks. In November 1996, a repeat HBV serosurvey was conducted of all susceptible residents, and no new HBV infections were detected.

Hospital B, New York City

In May 1996, acute hepatitis B was diagnosed in three inpatients in a hospital in New York City. All three patients had diabetes mellitus and had been hospitalized in December 1995 on the same medical ward as a patient with diabetes mellitus and acute hepatitis B (the suspected source case). Attempts were made to contact other patients hospitalized on the medical ward during the 19-day period in December when the suspected source case had been hospitalized. Of the 68 patients identified, 21 (31%) had died, and seven (10%) could not be contacted. Serologic results for the remaining 40 patients indicated that none had acute HBV infection: 20 were susceptible, and 20 were immune.

The median age of the three case-patients was 64 years; all denied commonly recognized risk factors for HBV infection. All three case-patients had had fingerstick capillary blood sampling, compared with none of the 20 susceptible patients (odds ratio=undefined, p less than 0.001). Receipt of insulin also was associated with infection; however, only two of the three case-patients had received insulin. No other common hospital procedure or medication was associated with infection. The HBsAg subtype identified for the three case-patients and the suspected source patient was adw2.

Finger sticks had been performed with the Glucolet{Registered} 2 (distributed by Miles {Elkhart, Indiana}), a pen-like device with a disposable lancet and end cap assembly. The lancet assembly was changed after each finger stick, but the pen-like lancet-holding device was used for multiple patients. The nursing staff typically performed finger sticks by starting at one end of the ward and moving from room to room sampling and recording blood glucose levels for patients with diabetes. The case-patients were hospitalized close to the suspected source case; one was a roommate, one was in an adjacent room, and one was in a room diagonally across the hall. Improper use of the fingerstick device (e.g., reusing disposable parts) was not reported by the nursing staff; however, nurses reported observing that hands were not always washed nor gloves changed between patients and that used lancet caps were placed in the same box as unused lancet caps.

Based on a review of serologic records at the hospital, 11 additional patients were identified with possible nosocomially acquired acute HBV infection diagnosed during January-October 1996. These 11 patients had no known risk factors for HBV infection, but all had been hospitalized at least once during the 6 months before diagnosis of their HBV infection. Ten of the 11 patients had received finger sticks while hospitalized; the remaining patient had diabetes mellitus but did not have fingerstick procedures documented in his hospital record. Eight of the 11 patients had been hospitalized on the same two wards. HBsAg subtyping was performed for seven of the eight patients, and all were identified as subtype adw4, a rare subtype present in less than 1% of HBsAg-positive persons in the United States (1).

In May 1996, the hospital instituted use of a completely disposable, nonreusable fingerstick device for capillary blood sampling. In addition, infection-control practices were reemphasized among nursing staff, including handwashing and changing of gloves after contact with each patient. Since implementation of these changes, no new cases of nosocomially acquired HBV infection have been identified.

Reported by: A Purdy, Allen County Health Dept, Lima; F Smith, MD, E Salehi, MPH, TJ Halpin, MD, State Epidemiologist, Ohio Dept of Health. J Quale, MD, D Landman, MD, E Atwood, V Ditore, Dept of Veterans Affairs Medical Center, Brooklyn; D Ackman, MD, PF Smith, MD, State Epidemiologist, New York State Dept of Health. Div of Cancer Prevention and Control, National Center for Chronic Disease Prevention and Health Promotion; Div of Applied Public Health Training (proposed), Epidemiology Program Office; Hepatitis Br, Div of Viral and Rickettsial Diseases, National Center for Infectious Diseases, CDC.

Editorial Note

Editorial Note: Outbreaks of HBV infection associated with use of spring-loaded fingerstick devices on multiple patients have been reported both in the United States and elsewhere (2-4). Various designs of these capillary-blood sampling devices are cleared by the Food and Drug Administration (FDA), but all function similarly. In general, the device is stabilized on the patient's finger by a platform, tip, or end cap containing a hole through which the lancet punctures the skin. In the only previously reported outbreak associated with fingerstick devices in the United States, 26 patients at a hospital in California acquired HBV infection (2,3). The platform, a disposable component of the fingerstick device, was not routinely changed between patients and became contaminated, which was the probable mechanism of HBV transmission in this outbreak. After the California outbreak, CDC and FDA issued recommendations for the safe use of spring-loaded fingerstick devices, including, optimally, using a separate device for each patient (2,5).

Both of the hepatitis B outbreaks described in this report probably were associated with use of fingerstick devices with a pen-like design, consisting of both reusable and disposable components. The device used in the Ohio nursing home has a separate lancet and end cap. Because the end cap rests on the finger during blood sampling, it can be contaminated with blood after the lancet pierces the skin. The package insert for the device indicates that both the lancet and end cap should be replaced after each use. The finding that the end caps were not routinely changed between residents suggests that HBV transmission occurred through exposure of subsequent patients to residual blood on the end caps.

The device used in the New York hospital has a combined lancet and end cap assembly that was changed as a unit after each use. The exact mechanism of HBV transmission resulting from use of this device is unclear but may have occurred by blood contamination of nurses' gloves or unused lancet caps. The pen-like lancet-holding device, which was shared and not cleaned between patients, also may have served as a vehicle for transmission. The package insert for the device recommends disinfecting the lancet-holding device only after visible contamination with blood.

The results of these investigations reemphasize the need to restrict use of fingerstick capillary-blood sampling devices to individual patients (2) and to discard used parts appropriately. In addition, when medical procedures are performed on multiple patients, gloves should be changed after contact with each patient (6).

References

1.      Swenson PD, Riess JT, Krueger LE. Determination of HBsAg subtypes in different high risk populations using monoclonal antibodies. J Virol Methods 1991;33:27-38.

2.      CDC. Nosocomial transmission of hepatitis B virus associated with a spring-loaded finger stick device -- California. MMWR 1990;39:610-3.

3.      Polish LB, Shapiro CN, Bauer F, et al. Nosocomial transmission of hepatitis B virus associated with the use of a spring-loaded finger-stick device. N Engl J Med 1992;326:721-5.

4.      Douvin C, Simon D, Zinelabidine H, Wirquin V, Perlemuter L, Dhumeaux D. An outbreak of hepatitis B in an endocrinology unit traced to a capillary-blood sampling device {Letter}. N Engl J Med 1990;322:57-8.

5.      Food and Drug Administration. FDA safety alert: hepatitis B transmission via spring-loaded lancet devices. Rockville, Maryland: US Department of Health and Human Services, Public Health Service, Food and Drug Administration, August 28, 1990.

6.      CDC. Update: universal precautions for prevention of transmission of human immunodeficiency virus, hepatitis B virus, and other bloodborne pathogens in health-care settings. MMWR 1988;37:377-82,387-8.

o        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.