Context.-Cervical cancer affects 12 000 women in the United States annually. However, despite its prevalence, there remains no good methodology to detect its recurrence.Objective.-To identify the role of cervicovaginal highrisk human papilloma virus (hr-HPV) testing in predicting cervical cancer recurrence.Design.-This is a retrospective study of patients who underwent hr-HPV testing as part of their routine surveillance for cervical cancer. Standard statistical analyses, including v 2 test and multivariable logistic regression, were performed with IBM SPSS 19.0.Results.-A total of 133 patients were identified, of whom 107 (80%) had squamous cell carcinoma. Ninety patients (68%) had bulky disease and were treated primarily with chemoradiation and brachytherapy. Of patients whose disease recurred, 5 patients (42%) had tested positive for hr-HPV during their surveillance period, compared to 13 patients (11%) for whom disease did not recur (relative risk: 3.88, P ¼ .002). On multivariate logistic regression, hr-HPV status remained significantly predictive of disease recurrence (odds ratio: 12.3, P ¼ .02, 95% confidence interval: 1.5-99.6). Using 2 3 2 table analysis, we found that while cervicovaginal cytology has limited specificity (5.7%) in predicting recurrence, the combination of cytology with hr-HPV testing increases the specificity of testing to 89.3%.Conclusions.-Persistence of hr-HPV is a risk factor for disease recurrence. High-risk-HPV testing is not routinely used during surveillance for cervical cancer, but this study suggests that large, prospective trials investigating the role of hr-HPV testing in cervical cancer surveillance are needed.