BACKGROUND: Interruption of a non-nucleoside reverse transcriptase inhibitor (NNRTI)-regimen is often necessary, but must be performed with caution because NNRTIs have a low genetic barrier to resistance. Limited data exist to guide clinical practice on the best interruption strategy to use. METHODS: Patients in the drug-conservation arm of the Strategies for Management of Antiretroviral Therapy (SMART) trial who interrupted a fully suppressive NNRTI-regimen were evaluated. From 2003, SMART recommended interruption of an NNRTI by a staggered interruption, in which the NNRTI was stopped before the NRTIs, or by replacing the NNRTI with another drug before interruption. Simultaneous interruption of all antiretrovirals was discouraged. Resuppression rates 4-8 months after reinitiating NNRTI-therapy were assessed, as was the detection of drug-resistance mutations within 2 months of the treatment interruption in a subset (N = 141). RESULTS: Overall, 601/688 (87.4%) patients who restarted an NNRTI achieved viral resuppression. The adjusted odds ratio (95% confidence interval) for achieving resuppression was 1.94 (1.02-3.69) for patients with a staggered interruption and 3.64 (1.37-9.64) for those with a switched interruption compared with patients with a simultaneous interruption. At least one NNRTI-mutation was detected in the virus of 16.4% patients with simultaneous interruption, 12.5% patients with staggered interruption and 4.2% patients with switched interruption. Fewer patients with detectable mutations (i.e. 69.2%) achieved HIV-RNA of 400 copies/ml or less compared with those in whom no mutations were detected (i.e. 86.7%; P = 0.05). CONCLUSION: In patients who interrupt a suppressive NNRTI-regimen, the choice of interruption strategy may influence resuppression rates when restarting a similar regimen. NNRTI drug-resistance mutations were observed in a relatively high proportion of patients. These data provide additional support for a staggered or switched interruption strategy for NNRTI drugs.