Postcoital bleeding is a predictor for cervical dysplasia

Background Postcoital bleeding (PCB) is a common gynecological symptom that may cause concern among both patients and physicians. Current literature is inconclusive regarding management recommendations. Objective To identify risk-factors for dysplasia/cancer among patients presenting post-coital bleeding (PCB). Methods Using large health maintenance organization (HMO) database, all women reporting PCB in 2012–2015 were identified. PCB patient records in a single colposcopy center were reviewed. Age, marital status, ethnicity, gravidity, parity, BMI, smoking, PAP smear result (within 1 year of PCB presentation), colposcopy and biopsy results were recorded. Cases were matched by age and socio-economic enumeration area to controls accessing primary care clinics for routine care. Results Yearly incidence of PCB ranged from 400 to 900 per 100,000 women; highest among patients aged 26–30 years. Among the sample of 411 PCB cases with colposcopy, 201 (48.9%) had directed biopsy. Biopsy results included 68 cervicitis (33.8%), 61 koilocytosis/ CIN 1/condyloma (30.3%), 44 normal tissue (21.9%), 25 cervical polyp (12.4%), 2 CIN 2/3 (1%) and 1 carcinoma (0.5%). Positive predictive value for koilocytosis/CIN 1 or higher pathology was 15.6% (64/411) and 0.7% for CIN 2 or higher grade pathology (3/411). In conditional logistic regression, multiparty was a protective factor: OR 0.39 (95% CI 0.22–0.88, P = 0.02), while pathological PAP smear was a related risk-factor: OR 3.3 (95% CI 1.31–8.35, P = 0.01). When compared to controls, PCB patients were significantly (P = 0.04) more likely to present CIN 1 or higher grade pathology (OR 1.82, 95% CI 1.02–3.33). Conclusions Study results indicate that PCB may require colposcopy, especially for nulliparous women with an abnormal PAP smear.