Page 168 - Screening for Cervical Cancer: Systematic Evidence Review

Chapter III: Results


Table 6. Performance of ThinPrep in a Prospective Cohort*
Number
ThinPrep with Actual Estimated
‡
ll
§
Cytology Final Diagnosis Sens Spec PPV NPV ¶
†
#
Threshold Diagnosis (%) “Equivocal” (%) (%) (%) (%) +LR -LR**
≥ ASCUS †† ≥ LSIL ‡‡ 323 (3.7) With Normal 87.9 90.5 95.9 99.5 9.3 0.13
≥ ASCUS ≥ LSIL 1019 (11.8) With LSIL 55.4 93.0 51.6 94.0 7.9 0.48
≥ LSIL ≥ LSIL 323 (3.7) With Normal 79.6 97.7 57.8 99.2 34.6 0.21

≥ LSIL ≥ LSIL 1019 (11.8) With LSIL 42.7 99.9 97.8 92.9 42.7 0.57
§§
≥ HSIL ≥ HSIL 137 (1.6) With < HSIL 67.2 99.3 61.3 99.5 98.4 0.33

≥ LSIL ≥ HSIL 137 (1.6) With < HSIL 89.8 96.1 25.8 99.8 23.1 0.11

51
*Data summarized from Hutchinson et al., 1999.
†Category in which final case diagnoses of “equivocal” were assigned for the calculation of test
characteristics of ThinPrep; 696 of 8,636 subjects (8.1%) had a final diagnosis of equivocal
“conferred with various combinations of results, including a single cytologic diagnosis of LSIL by any
51
method, and isolated positive cervigram, or equivocal results based on the review of all available data.”
(50)
‡ Sens, sensitivity
§ Spec, specificity
ll PPV, positive predictive value
¶ NPV, negative predictive value
# +LR, positive likelihood ratio;
** -LR, negative likelihood ratio.
†† ASCUS, atypical squamous cells of uncertain significance
‡‡ LSIL, low-grade squamous intraepithelial lesion
§§ HSIL, high-grade squamous intraepithelial lesion

69