Histological diagnosis and evaluation of the Swede score colposcopic system in a large cohort of pregnant women with atypical cervical cytology or cervical malignancy signs.

Objective: We evaluated the distribution of histological diagnoses in pregnant women with atypical cytology or cervical malignancy signs, as well as the usefulness of the Swede score colposcopic scoring system to reduce the need for diagnostic cervical biopsy.

Design: Prospective clinical study.

Setting And Population: The study comprised 261 pregnant women undergoing colposcopic investigation because of atypical cervical cytology, dysplastic biopsy changes, recurrent non-obstetric bleeding or pathological appearance of the cervix.

Methods: Five colposcopic variables (acetowhiteness, margins plus surface, vessel patterns, lesion size and iodine staining) were scored with 0, 1 or 2 points. Colposcopically directed biopsies or loop electrosurgical excision biopsies were taken from all lesions. Histology was compared with the colposcopic score. Sensitivity and specificity were calculated for each variable, and the combination of all five variables, with high-grade lesions (i.e. cervical intraepithelial neoplasia (CIN2, CIN3 or adenocarcinoma-in-situ (AIS)) as endpoints.

Main Outcome Measures: Colposcopic score (Swede score) and histology (CIN1, 2, 3; AIS; cancer).

Results: The specimens consisted of normal tissue in 19.5% of cases, low-grade lesions (i.e. CIN1, koilocytosis, glandular dysplasia of lower grade than AIS) in 26.1%, high grade lesions in 52.9% and cancer in 1.5%. All high grade lesions and cancers had total Swede scores of ≥ 5 and ≥ 8, respectively. Vessel patterns, lesion size and margins plus surface were most important for high grade lesion detection.

Conclusion: The Swede score seems to be a useful tool in evaluating atypical cervical cytology in pregnant women and may reduce the need for diagnostic biopsies.