Background and methods: From June 1989 through March 1990, 26 patients, of whom 23 had diabetes, contracted acute hepatitis B virus (HBV) infection in a hospital in California. All 26 patients and one HBV carrier (also a diabetic) had been admitted to a single medical ward during the six months before the case patients became infected with HBV. To determine the source of the infection, we conducted a retrospective cohort study of the 72 patients with diabetes who had been admitted to the ward from January through December 1989 and a case-control study comparing the 3 nondiabetic patients who contracted hepatitis with 20 nondiabetic controls.
Results: The retrospective cohort study of all the patients with diabetes who were admitted to the ward during 1989 found that those who underwent capillary blood sampling by finger stick with a spring-loaded lancet device were more likely to contract HBV infection than those who did not have finger sticks (attack rate, 42 percent vs. 0 percent; P = 0.08). In addition, a dose-response relation was observed between the number of finger sticks received and the frequency of hepatitis B (P = 0.002). The case-control study found that all 3 of the nondiabetic patients who contracted hepatitis underwent finger-stick blood sampling with the device, as compared with none of the 20 nondiabetic controls (P = 0.0006). A review of nursing procedures indicated that the platform of the device was not routinely changed after each use; this finding suggested that contamination of the platform by HBV-infected blood was the mechanism of percutaneous transmission of HBV.
Conclusions: Proper use of finger-stick devices as well as strict adherence to universal precautions to avoid contamination by blood are required to decrease the possibility of transmission of blood-borne pathogens among hospitalized patients.