Literature review > Issue 1 > Review Bianchi et al. 

At Cornell Medical School, I walked every day past George Papanicalaou's laboratory to classes. Since Dr. Pap, millions of women have undergone regular periodic gynecologic examinations, including Pap smears. Recently, thin layer preparations of cervical cells have proven easier to interpret than conventional Pap smears, and it soon became clear that specimens collected in vials containing alcohol-based, buffered fixative for preparation of monolayers for cytologic screening could also be tested for human papillomavirus; adjunctive HPV testing of cervical specimens collected in PreserveCyt solution has been approved by the US FDA. Can the residual material also be used - perhaps selectively - to screen for other STIs?

Because Bianchi et al. did not prospectively compare PCR screening for C. trachomatis in specimens collected in PreserveCyt medium for thin layer cytology vs. PCR screening of specimens simultaneously collected from 2-SP medium for the Cobas PCR CT/NG test, we unfortunately cannot really assess the sensitivity and specificity of PCR testing of clinical specimens collected for ThinPrep Pap tests. A second concern with this study is that women with any clinical sign of urogenital inflammation or with atypical cells such as inflammatory cells on the smear, were excluded. Thus, only 2.2% of women were C. trachomatis positive.

In contrast, Anguenot et al. (1) in Geneva, Switzerland collected two cervical samples from each of 590 women specifically selected on the basis of high risk of chlamydial infection (e.g., sexually active adolescents, suspected pelvic inflammatory disease, presence of an STD, or high risk sexual behavior); one sample was collected in AutoCyte's preservation fluid, and the other with swabs placed in Abbott's LCx transport medium. Of 30 (5.1%) C. trachomatis positive specimens by LCR, 28 were positive in both cervical samples, one in LCx medium only, and one in AutoCyte medium only (kappa=0.96). Importantly, the odds ratio for inflammation on Pap smear and detection of C. trachomatis was 2.0 (p=0.07). Although conventional PAP smears showing inflammation have often been rejected as uninterpretable in the past, this has been less of a problem with thin layer cytology, because of better cellular dispersion.

Kiviat et al. (2) first described in some detail the infectious correlates of inflammatory cell changes on Papanicalou-stained cervical cytologic smears. Specifically in randomly selected STD clinic patients, C. trachomatis was isolated by culture from 21% was significantly correlated with increased numbers of histiocytes and with transformed lymphocytes (presence of either had sensitivity of 95%, specificity of 75%, and positive predictive value of 50% for C. trachomatis in this population), as well as with increased polymorphonuclear leukocytes, and with reactive or atypical metaplastic cells. Others have found similar associations among women seeking routine gynecologic care (3).

Currently, the practice of "reflex HPV DNA testing" of residual liquid thin layer cytology specimens (e.g., when thin layer cytology shows atypical squamous cells of uncertain significance) has been recommended as cost effective (4). Should not "reflex" nucleic acid amplification C. trachomatis testing of residual liquid specimens be evaluated when the thin layer cytology shows inflammation? If indeed that approach proves as sensitive as currently approved methods for diagnosis of C. trachomatis infection, and could be combined with routine chlamydia screening of specimens from high risk adolescent and young adult women seeking routine gynecologic examination and PAP smears, the impact on further control of chlamydial infections could be enormous.

References

1. Anguenot JL, de Marval F, Vassilakos P, Auckenthaler R, Ibecheole V, Campana A. Combined screening for Chlamydia trachomatis and squamous intra-epithelial lesions using a single liquid-based cervical sample. Hum Reprod 2001;16:2206-10.

2. Kiviat NB, Paavonen JA, Brockway J et al. Cytologic manifestations of cervical and vaginal infections. I. Epithelial and inflammatory cellular changes. JAMA 1985;253:989-996.

3. Freund KM, Buttlar CA, Giampaolo C, Phillips RS, Aronson MD, Moskowitz MA. The use of cervical cytology to identify women at risk for chlamydial infection. Am J Prev Med 1992;8:292-7.

4. Kim JJ, Wright TC, Goldie SJ. Cost-effectiveness of alternative triage strategies for atypical squamous cells of undetermined significance. JAMA 2002;287:2428-9.

about SDI | newsletters | grants | publications | literature reviews

WHO Home - WHO Search - TDR Home - SDI Home - SDI Contact us
(c) WHO/OMS 2001