A clinical decision analysis of cryotherapy compared with expectant management for cervical dysplasia

Screening for precursors of cervical cancer with colposcopic examination for women with abnormal Papanicolaou (Pap) smears identifies those with cervical dysplasia. Though the majority of mild dysplasias (CIN I) will regress, many are treated with cryotherapy.

We used decision analysis to compare immediate cryotherapy with expectant management (following with another Pap smear or colposcopy, with treatment reserved for progression or a duration of 2 years). The decision tree included the possibility of more invasive surgical procedures if the cryotherapy was ineffective or if the dysplasia progressed in extent of cervical involvement or in grade. Probabilities were derived from literature review and expert judgment. The analysis considered the disutility of the follow-up examinations, cryotherapy, and the more invasive procedures, using expert assessment.

Using the baseline assumptions, expectant management led to a better outcome for most patients (57%), who recover with no procedure. However, more patients treated with expectant management required surgical procedures (loop electrosurgical excisional procedure, conization, or, rarely, hysterectomy) than did those treated with immediate cryotherapy. In the expected disutility analysis, expectant management was better than immediate cryotherapy. Sensitivity analysis showed that three factors had the potential to change the recommendation of the analysis: (1) the probability the dysplasia will regress, (2) the disutility of the process of expectant management, and (3) the disutility of invasive procedures compared with cryotherapy.

The analysis indicated that expectant management is preferable to immediate cryotherapy for women with histologically proven mild cervical dysplasia. However, this conclusion depended on assumptions about three factors for which there is insufficient evidence in the literature. More research is needed.