Hepatitis C transmission and HIV post-exposure prophylaxis after needle- and syringe-sharing in Australian prisons

In prison A, the index patient (Index patient A) disclosed to nursing staff in November 2000 that he was infected with HIV and had shared two needles and syringes in the previous 5 days, while incarcerated. The inmate did not identify his sharing partners and shared again the following day.

In prison B, the index patient (Index patient B) was found to be HIV sero-positive after presenting for voluntary testing in November 2000. He disclosed sharing a needle and syringe with other inmates in the previous three weeks while incarcerated and identified several sharing partners in the prison.

Inmates in the two prisons were contacted through peer outreach and contact tracing, and were invited to attend the prison clinic if they had shared needles or syringes between 5 and 29 November 2000 (the risk period). All inmates who reported sharing injecting equipment during this period were considered potentially exposed to HIV, HBV and HCV infection, although it was usually not known if they had shared with an index patient.

Attending inmates were assessed for risk behaviour, presence of HIV, HBV and HCV antibodies, HBV vaccination status, and need for HBV immunoglobulin and HIV PEP, before referral for methadone assessment. Specialist medical staff assessed inmates for PEP. Those who were HBV antibody-negative were eligible for HBV vaccination and immunoglobulin, and those who were HIV-negative were eligible for PEP. Clinic staff routinely recorded these data in a structured medical record.

The following baseline data were systematically collected from inmates' medical records by a single researcher (B G O'S): injecting drug history; needle and syringe sharing in the risk period; other risk behaviours; HIV, HBV and HCV antibody results at the time of initial clinic presentation; and PEP prescription.

Inmates were excluded from the study if there was no documented evidence that they shared injecting equipment in the risk period.

Inmates were followed up for up to 14 months. Serological and risk data were systematically collected from medical records by the same researcher (B G O'S). In addition, data on evidence of clinical HIV, HBV or HCV seroconversion illness, and compliance and side effects of PEP were collected.

Inmates were referred for a single follow-up serological test between March and December 2001 if they were still incarcerated. Those who had been released were sent letters requesting that they seek free follow-up testing at a local clinic, with the offer of $30 for travel costs, and had their records flagged for testing in the event of re-incarceration.

Index patient A was positive for anti-HIV antibodies and for HBV infection (positive for HBV surface antigen; HBV DNA level of 33.506 pg/mL; and negative for HBV e antigen and for IgM to HBV core antigen). He was also positive for anti-HCV antibodies, with serum HCV RNA detected in February 2000 but not retested in November 2000. He had been diagnosed with HIV infection in the community in April 1999. In November 2000, he had an HIV viral load of 22 000 RNA copies/mL, and CD4 T-lymphocyte count of 0.09 x 109/L (reference range, 0.38–1.39 x 109/L) and was not compliant with antiretro-viral therapy.

Index patient B was negative for anti-HIV antibodies in early September 2000 while incarcerated. He was released in mid-September and was reported to have shared needles and syringes with an HIV infected person in the community during this period. He was re-incarcerated in mid-October 2000 and had a serological profile suggesting recent HIV-1 infection: HIV plasma load of 64 900 RNA copies/mL, and CD4 T-lymphocyte count of 0.62 x 109/L. This patient was also anti-HBV-positive but not infectious (positive for antibodies to HBV core and surface antigens; not tested for HBV surface antigen [HBs]) and was infected with HCV (anti-HCV-positive; not tested for HCV RNA).

Of the 63 inmates with follow-up serological results, four had undergone HCV seroconversion (14% of the 29 inmates susceptible to HCV in the cohort), but none had undergone HIV or HBV seroconversion. One of those who seroconverted to HCV was in prison A during the risk period (1% of the 91 inmates potentially exposed), and the other three were in prison B (23% of the 13 potentially exposed).

Characteristics of those who seroconverted are shown in Box 3. Seroconversion occurred within 3–14 months of the potential exposure. The three inmates in prison B who seroconverted to HCV reported sharing needles and syringes with each other, as well as with other HCV-infected inmates during the risk period. All three were continuously incarcerated.

Of the 104 potentially exposed inmates, 56 (54%) were eligible for HIV PEP (62% of all potentially exposed in prison A; none in prison B) as their exposure had occurred within the 72 hours before they attended the clinic. Of these, 46 were offered PEP (82% of those eligible), and 34 elected to receive it (74%). They were prescribed zidovudine and lamivudine. The main reasons for inmates electing not to take PEP included the belief that they had used a new syringe, had not shared with the index patients, or had cleaned the injecting equipment with bleach. Reasons for not offering PEP were not recorded.

The median period of taking PEP was 18 days, and compliance with therapy was reported as complete for eight inmates (24%), moderate for 22 (65%), and poor (many doses missed) for four (12%). Of the 26 who did not complete the full course, 11 did not give a reason. However, PEP was commonly ceased when inmates were transferred or released. No serious adverse effects were reported.

Among inmates susceptible to HBV at baseline, 24 (83%) received HBV vaccination or immunoglobulin and had protective antibodies (serum anti-HBs levels > 10 L/mL) at follow-up.