Page 64 - Screening for Cervical Cancer: Systematic Evidence Review
P. 64

Chapter III.  Results



               management of patients with low grade cytologic abnormalities who will at the time of

               colposcopy have a much lower prevalence of high grade lesions.


                       The indications for the colposcopy, including indications established for the purposes of

               research, vary across studies; this in turn also influences the spectrum of abnormalities identified.


               The study cohort in Iceland had a prevalence of 54.5% of high-grade lesions compared to 15.4%

               among German women referred for colposcopy.  Similarly, among studies that enrolled patients


               based on cytology findings as an entry criterion (rather than referral for colposcopy), the criteria

                                                          66
                                      68
               vary from ASCUS only  to CIN 1 or CIN 2.   This means that aggregating test performance
               characteristics across studies is problematic.  The test characteristics of high-risk HPV testing for

               detecting HSIL as a triage tool are summarized in Table 11.


                       Three strong studies selected the degree of Pap test abnormality a priori and then

               obtained histologic verification of HPV test performance.  The group with the lowest risk of

               high-grade lesion at the time of evaluation was that assembled by Manos and colleagues within


                                                                                  68
               the Kaiser Permanente Medical Care Program in Northern California.   They enumerated all
               women at 4 medical centers who had a Pap test during a 9-month period and invited all women


               with ASCUS to participate if they were not pregnant and did not have a history of treatment for

               cervical neoplasia in the prior six months.  Sixty-one percent of patients with ASCUS enrolled in


               the study; participants were demographically similar to non-participants.  In this setting the

               sensitivity of Hybrid Capture II testing for high-risk HPV to detect HSIL was 89.2%; specificity


               was 64.1%; positive predictive value, 15.1%; negative predictive value 98.8%.  For detection of

               LSIL and more severe lesions, the sensitivity was 76.3%; specificity, 69.5%; positive predictive


               value, 37.8%; and negative predictive value, 92.4%.









                                                             64
   59   60   61   62   63   64   65   66   67   68   69