Objective We sought to examine the impact of the loop electrosurgical excision procedure (LEEP) on the rate and magnitude of HIV-1 genital shedding among women undergoing treatment for cervical intraepithelial neoplasia 2/3 (CIN2/3). Design Prospective cohort study. Population Women infected with HIV-1 undergoing LEEP for CIN2/3 in Kisumu, Kenya. Methods Participants underwent specimen collection for HIV-1 RNA prior to LEEP and at 1, 2, 4, 6, 10, and 14 weeks post-LEEP. HIV-1 viral load was measured in cervical and plasma specimens using commercial real-time polymerase chain reaction (PCR) assays, to a lower limit of detection of 40 copies per specimen. Main outcome measures Presence and magnitude of HIV-1 RNA (copies per specimen or cps) in post-LEEP specimens, compared with baseline. Results Among women on highly active antiretroviral therapy (HAART), we found a statistically significant increase in cervical HIV-1 RNA concentration at week 2, with a mean increase of 0.43 log10 cps (95% CI 0.03–0.82) from baseline. Similarly, among women not receiving HAART, we found a statistically significant increase in HIV-1 shedding at week 2 (1.26 log10cps, 95% CI 0.79–1.74). No other statistically significant increase in concentration or detection of cervical HIV-1 RNA at any of the remaining study visits were noted. Conclusions In women infected with HIV undergoing LEEP, an increase in genital HIV shedding was observed at 2 but not at 4 weeks post-procedure. The current recommendation for women to abstain from vaginal intercourse for 4 weeks seems adequate to reduce the theoretical increased risk of HIV transmission following LEEP.