to the editor: Drs. Hainer and Gibson provided an excellent review of the assessment and treatment of vaginal discharge in symptomatic patients. However, we disagree with the statement that treatment of trichomoniasis “should not be based on a Papanicolaou (Pap) smear finding of trichomonads.” On the contrary, recent research supports the practice of treating all trichomoniasis infections diagnosed on Pap smears.

A 2003 study demonstrated a specificity of 99.4 percent and a sensitivity of 61.4 percent for the diagnosis of trichomoniasis, using liquid-based cytology Pap smears.¹ This article concludes, “The presence of [Trichomonas] vaginalis organisms, as stated by the liquid-based Pap smear pathology report, is accurate and warrants treatment without further testing.” A similar 2009 study of liquid-based cytology for T. vaginalis found a 98 percent sensitivity (95% confidence interval [CI], 94 to 100 percent) and a 96 percent specificity (95% CI, 94 to 98 percent).² The authors also recommend treatment for trichomoniasis when organisms are detected in a Pap smear.

Because this sexually transmitted infection may be a surprise finding for physicians and patients, counseling needs to be sensitive to the social needs of the patient. Physicians should also offer testing for additional sexually transmitted infections in patients with trichomoniasis detected on a Pap smear.

in reply: Drs. Wadhwa and Stanford raise the important question of the reliability of diagnosing trichomoniasis using Papanicolaou (Pap) smear cervical cytology. A meta-analysis of the performance of the Pap smear in the diagnosis of trichomoniasis found that assuming a prevalence of 20 percent, with a likelihood ratio for a positive Pap smear of 19, the probability that a woman with a positive test has trichomoniasis (the positive predictive value) is 83 percent.¹ The studies cited by Wadhwa and Stanford to support treatment of trichomoniasis diagnosed on Pap smears were conducted at a government referral center for sexually transmitted infections in Hong Kong, where the prevalence of trichomoniasis by culture was 22 percent,² and at an urban women's health center in Cleveland, Ohio, where the prevalence of trichomoniasis by culture was 21.6 percent.³

In primary care settings, however, the prevalence of trichomoniasis in patients with symptomatic vaginitis ranges from 6 to 17 percent. Using a prevalence of 10 percent, the positive predictive value for a positive Pap smear is 68 percent. At a prevalence of 1 percent, trichomoniasis has a positive predictive value of only 16 percent. Therefore, the conventional Pap smear does not appear to be a useful tool in the diagnosis of Trichomonas vaginalis in most primary care settings. However, liquid-based cytology appears to be more reliable.^4,5

The Centers for Disease Control and Prevention's Sexually Transmitted Diseases Treatment Guidelines, 2010 concludes: “While the sensitivity of a Pap test for T. vaginalis diagnosis is poor, use of a liquid-based testing has demonstrated enhanced sensitivity; however, false-positive tests can occur, and confirmatory testing might be needed in some circumstances.”⁶