A systematic review of postcoital bleeding and risk of cervical cancer.

BACKGROUND Postcoital bleeding may be a symptom of cervical cancer. Guidance to aid a GP in determining whom to investigate or refer exists but recommendations vary. Women need to be involved in decisions about their care and this involves communicating risk and an exploration of the implications of the risk. Risk estimates of postcoital bleeding for cervical cancer are not available. AIM To provide an estimate of the positive predictive values of postcoital bleeding for cervical cancer to aid decision making in primary care about whom to investigate for cervical cancer. DESIGN OF STUDY A systematic review. SETTING Community, primary and secondary care. METHOD Six electronic databases were searched from the beginning of each of their time frames. Inclusion criteria were that the study was published in English and reported or contained enough data to calculate the prevalence or incidence of postcoital bleeding within the study population. No studies were excluded on issues of methodological quality. RESULTS The search strategy identified 910 unique articles. The point prevalence of postcoital bleeding in the community ranged from 0.7 to 9% among women. One study reported an annual cumulative incidence of 6% of menstruating women. The prevalence of postcoital bleeding in women with cervical cancer ranged from 0.7 to 39%. Calculation of risk that a woman in the community developing postcoital bleeding has cervical cancer ranges from 1 in 44,000 at age 20-24 years to 1 in 2 400 aged 45-54 years. There was no information allowing the direct calculation of risk in women presenting to primary care. CONCLUSION The evidence base for management strategies of postcoital bleeding and calculations of risk for cervical cancer in women with postcoital bleeding are poor. Recommendations for clinical practice are made on the current evidence.