Comparison of the POP-Q examination, transvaginal ultrasound, and direct anatomic measurement of cervical length.

INTRODUCTION AND HYPOTHESIS: Our objective was to determine the relationship between the Pelvic Organ Prolapse Quantification (POP-Q) examination for determining cervical length (CL) and CL at hysterectomy. Secondary objectives were to define cervical elongation using both measures in a urogynecologic population, determine the relationship between POP-Q estimate and CL on ultrasound (US) and examine the interobserver reliability of each mode of measurement. METHODS: This was a prospective cohort study of women scheduled for hysterectomy at the Los Angeles County + University of Southern California (LAC + USC) medical center. CLs were measured by POP-Q and at the time of hysterectomy. Transvaginal US CLs were determined when available. Exam CL (eCL) was compared with anatomic (aCL) and US (uCL) CL. Repeat measures of eCL, uCL, and aCL were all compared for interobserver reliability. RESULTS: The study enrolled 151 women. Median eCL was 3.0 cm (0.5-9.0) (n = 149); average uCL was 2.3 cm ± 0.7 (n = 108), average aCL 2.8 cm ± 1.1 (n = 87); eCL correlated fairly with aCL (r = 0.3, p = 0.005, n = 88) but poorly with uCL (r = -0.13, p = 0.18, n = 105); uCL correlated poorly with aCL (r = 0.19, p = 0.14, n = 64). Interobserver reliability for eCL and aCL were good to excellent (eCL α=0.881; aCL α=0.889) but for uCL adequate (α=0.699). The 97.5 percentile cutoff for aCL was 5.0 cm and for eCL 8.0 cm. CONCLUSIONS: The POP-Q examination estimate of CL correlates fairly with aCL at the time of hysterectomy; uCL does not appear to correlate with aCL or eCL. Cervical elongation may be defined as an anatomic length of 5.0 cm or a POP-Q estimate of 8.0 cm.