Background: Vaginal bleeding is the most common symptom of vault relapse in gynecological cancers. This symptom may be overlooked or attributed to other causes, such as atrophy, infection, or post-radiation changes. Colposcopy allows direct magnified visualization of vaginal mucosa and abnormal vasculature though its role in detecting relapse is unclear.
Methods: It is a retrospective observational study of 31 patients who were referred for colposcopy and biopsy with complaints of abnormal vaginal bleeding during follow up of endometrial or cervical cancer.
Results: Among 31 patients, 19 patients were diagnosed to have primary cervical cancer and 12 endometrial cancers. Primary treatment was surgery alone (n=10, 32.2%), chemo-radiation alone (n=4, 12.9%) and both (n=17, 54.8%). Abnormal colposcopic findings were atropic features (n=21, 67.7%), radiation changes (n=8, 25.8%), erosion (n=9, 29.0%), acetowhite areas (n=9, 29.0%), abnormal vessels (n=8, 25.8%) and iodine staining abnormality (n=16, 51.6%). According to Swede scoring system, 77.1% (n=24) were normal or low grade lesions, 22.6% (n=7) were high grade lesions. Total of 6 recurrence cases identified out of which 5 cases had high grade (Swede score >7) and one had low grade (Swede score 5-7) colposcopic features. On taking Swede score cut off of 5 or more, the sensitivity of colposcopy in detecting vault recurrence is 100%, specificity 96%, positive likelihood ratio of 25%and negative likelihood ratio of 0%. Overall the accuracy of colposcopy in detecting relapse was 96.8%.
Conclusions: In our experience colposcopy is worth in detecting the cause of abnormal vaginal bleeding after treatment for endometrial and cervical cancer. Swede score is a good measure to decide on taking biopsy among these patients. Furthermore, larger studies are needed for better clarification.