ABSTRACT
OBJECTIVES: The aims of the study were to determine whether a Gram stain of cervical mucus can accurately rule out infection with Neisseria gonorrhoeae or Chlamydia trachomatis and to compare a diagnostic test that is based on the polymerase chain reaction with a deoxyribonucleic acid probe in the detection of these organisms. STUDY DESIGN: Gravid patients were screened for N gonorrhoeae and C trachomatis with a deoxyribonucleic acid probe, Gram stain, and analysis with the polymerase chain reaction. A normal, noninfected sample was defined by <10 polymorphonuclear leukocytes per high-power field on the Gram stain. Standard statistical methods were used to compare results of the Gram stain and the deoxyribonucleic acid probe, as well as to compare results of deoxyribonucleic acid probe hybridization and polymerase chain reaction analysis. A P value of <.05 was considered statistically significant. RESULTS: Patient enrollment totaled 519. The prevalence of infection as determined by deoxyribonucleic acid probe hybridization was 1.4% for N gonorrhoeae (7/518) and 6.8% for C trachomatis (35/518). The cervical Gram stain predicted the absence of infection in 17% (90/518) of patients, with a negative predictive value of 99% for N gonorrhoeae and 97% for C trachomatis. African American race, age <20 years, and unmarried status were all predictors of the presence of C trachomatis or N gonorrhoeae cervicitis. For the patients who lacked these risk factors (n = 74), the Gram stain had 100% negative predictive value. Analysis with the polymerase chain reaction detected 8 additional patients with C trachomatis and 105 additional patients with N gonorrhoeae, in comparison with deoxyribonucleic acid probe hybridization. CONCLUSION: The cervical Gram stain can accurately predict the absence of N gonorrhoeae and C trachomatis in gravid women. Analysis with the polymerase chain reaction indicates that N gonorrhoeae and C trachomatis are significantly more prevalent in this population than previously reported.
