Is the Presence of Trichomonas on a Pap Smear Associated with an Increased Incidence of Bacterial Vaginosis?

Debra S. Heller, MD,1 Svetlana Maslyak, MD,2 and Joan Skurnick, PhD3 1Departments of Pathology and Laboratory Medicine, 2Obstetrics, Gynecology, and Women’s Health, and 3Preventive Medicine and Community Health, UMDNJYNew Jersey Medical School, Newark, NJ

h Abstract acterial vaginosis (BV) is the most common cause of Objective. Bacterial vaginosis (BV) is associated with Bvaginal discharge, estimated to be present in 10% to significant morbidity. Bacterial vaginosis is due to an 20% of white women, and 30% to 50% of African- overgrowth of anaerobic organisms in the vagina. It has American women [1]. Bacterial vaginosis may be the been postulated that the presence of Trichomonas cre- ates an anaerobic environment that favors BV. Thus, cause of significant morbidity and has been associated BV should be more frequent in the presence of Tricho- with premature birth [2], infections after gynecologic monas if Trichomonas is creating a favorable anaerobic surgery [3], and an increased rate of acquisition of environment. HIV after exposure [3]. Bacterial vaginosis results from a Materials and Methods. A retrospective review of Pap shift in the balance of the normal vaginal flora, with test reports was performed for a diagnosis of coccobacilli overgrowth of anaerobic bacteria, including Gardnerella consistent with shift in vaginal flora, that is, the presence of clue cells. Cases were Pap smears with Trichomonas vaginalis. Vaginitis due to Trichomonas vaginalis (TV) is identified. Controls were cases without Trichomonas. associated with an anaerobic environment that favors Results were analyzed using Fisher exact test. the development of BV [4]. If this is true, an increased Results. Four hundred cases were reviewed; 200 with prevalence of BV would be expected in the presence of and 200 without Trichomonas. The incidence of BV was Trichomonas. The purpose of this study was to evaluate significantly higher in the Trichomonas group (46.5%) than in group without Trichomonas (24.5%) (p G .0001.). this association. Conclusion. The presence of Trichomonas on a Pap smear is associated with an increased incidence of BV. The MATERIALS AND METHODS report of Trichomonas on a Pap smear should prompt clinical consideration that BV may be present. h A retrospective review of Pap smear reports was performed on patients who presented to University Hospital, Newark, Key Words: Trichomonas, vaginosis, bacterial, vaginal smears between January 1, 1999, and December 31, 2003. The study was approved by the UMDNJ institutional review board. Cases were Pap smears with Trichomonas identified. Controls were cases without Trichomonas.Caseswere selected sequentially. Bacterial vaginosis is diagnosed on a Correspondence to: Debra S. Heller, MD, Department of Pathology, UH/ Pap smear by the presence of clue cells and coccobacilli, E158, UMDNJ-NJSM, 185 South Orange Avenue, PO Box 1709, Newark NJ 07101. E-mail: [email protected]. which represents a shift in vaginal flora consistent with the clinical diagnosis of BV. Cases and controls were evaluated

Ó 2006, American Society for Colposcopy and Cervical Pathology for the prevalence of this finding. Data were analyzed using Journal of Lower Genital Tract Disease, Volume 10, Number 3, 2006, 137Y139 Fisher exact test.

Copyright @ 2006 American Society for Colposcopy and Cervical Pathology. Unauthorized reproduction of this article is prohibited. 138 & HELLER ET AL.

RESULTS One of the issues that has been cited as a confound- Results are shown in Table 1. Four hundred Pap smears ing factor when evaluating the clinical effects of BV is were reviewed, 200 with Trichomonas and 200 without. how the diagnosis of BV was made. Epidemiological The prevalence of BV on the Pap smear was significantly studies were more likely to use Gram stain, whereas in higher in the Trichomonas group (46.5%) than in group practice, clinical assessment was more commonly used without Trichomonas (24.5%), p G .0001. [2]. The clinical criteria of Amsel et al. [11] require 3 of 4 signs: homogeneous white adherent vaginal discharge, pH of more than 4.5, clue cells on saline wet DISCUSSION smear, and positive Bwhiff^ test on adding KOH. It has Bacterial vaginosis is secondary to anaerobic bacterial been suggested that the presence of clue cells on the overgrowth. Although not considered a sexually trans- Pap smear agrees reasonably well with clinical criteria mitted disease, it is associated with some sexual (J index of 0.87 compared with clue cells on wet smears, behaviors, including early age of sexual intercourse, and J index of 0.94 compared with Gram stain) [12]. orogenital sex, new or multiple partners, sex during Similarly, although the Pap smear is not performed for menses, sex with another woman, and use of an detection of Trichomonas, it has been suggested that intrauterine device [5]. Associations with BV include recognition of Trichomonas by Pap smear is specific, postabortal endometritis, pelvic inflammatory disease, although insensitive [13]. premature rupture of the membranes, chorioamnionitis, Although BVand Trichomonas have significant overlap postpartum endometritis, preterm birth, and neonatal in clinical and laboratory findings (bacterial cultures) [14], sepsis [5]. The presence of BV increases the risk for few studies have explored the association. Franklin and infection after gynecologic surgery [3]. Risk of transmis- Monif [15] found concomitant TV and BV in 15% of sion of HIV to a woman with BV is greater [3]. pregnant women examined by wet smear, concluding that The prevalence of BV is greater in African-American these were frequent coinfections. Demirezen et al. [4] women than in other ethnic groups studied [6], and evaluated the association of TV and BV on Pap smears. In African-American women bear an increased burden of their study, 44.4% of cases with TV also had BV, whereas preterm deliveries, with a high rate seen at our hospital, only 5.5% of TV-negative cases had BV. It has been which serves an inner-city population. It has been postulated that TV increases vaginal pH due to phagocy- postulated that increased BV in black women may be a tosis of lactobacilli [16] and that this is the mechanism of treatable cause of the increased risk for preterm delivery facilitation of the development of BV in the presence of TV and low birth weight in blacks [7]; however, some recent [4]. Moodley et al. [17] have also suggested that TV alters studies have found no benefit from routine BV screening vaginal flora, predisposing to BV. Of note, in the study of and treatment to reduce adverse pregnancy outcome Demirezen et al., the prevalence of BV in the TV-negative [2, 8]. Some authors have suggested that there may be a controls was only 5.5%, as opposed to the 24.5% in our place for screening and treating a subset of high-risk controls, yet we are still seeing a significant difference. patients [2, 9], and one study suggests benefit in screening This may reflect the predominantly African-American and treating low-risk patients [10], but the issue remains population we serve, as BV has been shown to be more controversial. It has been suggested that genetic poly- prevalent in African-American women [6]. morphisms in the cytokine response might alter inflam- Our study lends additional support to the increased matory response and hence alter the risk for BV in frequency of BV in the presence of TV.This study suggests different women [3]. that the diagnosis of BV should be considered when a diagnosis of TV is made on a Pap smear. Table 1. Presence of BV in Pap Smears with and without Trichomonas REFERENCES Positive Negative 1. Sobel JD. What’s new in bacterial vaginosis and Pap smears for BV for BV trichomoniasis? Infect Dis Clin North Am 2005;19:387Y406. With Trichomonas, n = 200 46.5%a 53.5% 2. Guise JM, Mahon SM, Aickin M, Helfand M, Peipert n = 93 n = 107 JF, Westhoff C. Screening for bacterial vaginosis in pregnancy. a Without Trichomonas, n = 200 24.5% 75.5% Y n = 49 n = 151 Am J Prev Med 2001;20(3 suppl):62 72. 3. Larsson PG, Bergstrom M, Forsum U, Jacobsson B, ap G .0001 for significant difference, by Fisher exact test. Strand A, Wolner-Hanssen P. Bacterial vaginosis. Transmission,

Copyright @ 2006 American Society for Colposcopy and Cervical Pathology. Unauthorized reproduction of this article is prohibited. Trichomonas and Bacterial Vaginosis & 139

role in genital tract infection and pregnancy outcome: an Holmes KK. Nonspecific vaginitisVdiagnostic criteria and micro- enigma. APMIS 2005;113:233Y45. bial and epidemiologic associations. Am J Med 1983;74:14Y22. 4. Demirezen S, Kormaz E, Beksa0 M. Association 12. Platz-Christensen JJ, Larsson PG, Sundstrom E, between trichomoniasis and bacterial vaginosis: examination Wiqvist N. Detection of bacterial vaginosis in wet mount, of 600 cervicovaginal smears. Cent Eur J Public Health Papanicolaou stained vaginal smears, and in Gram stained 2005;13:96Y8. smears. Acta Obstet Gynecol Scand 1995;74:67Y70. 5. O’Brien RF. Bacterial vaginosis: many questionsVany 13. Lobo TT, Carvalho SE, Costa PL, Chagas C, Xavier J, answers? Curr Opin Pediatr 2005;17:473Y9. Simoes-Barbosa A. A comparative evaluation of the Papani- 6. Goldenberg RI, Klebanoff MA, Krohn MA, Hillier S, colaou test for the diagnosis of trichomoniasis. Sex Transm Dis Andrews WW. Bacterial colonization of the vagina during preg- 2003;30:694Y9. nancy in four ethnic groups. Vaginal Infections and Prematurity 14. van der Meijden WI, Duivenvoorden HJ, Both-Patoir Study Group. Am J Obstet Gynecol 1996;174:1618Y21. HC, Hazen-Engelsman ME, Drogendijk AC. Clinical and 7. Fry YW. Bacterial vaginosis and other maternal laboratory findings in women with bacterial vaginosis and infections: treatable causes of the black-white gap? Ethn Dis trichomoniasis versus controls. Eur J Obstet Gynecol Reprod 2003;13(suppl 3):108Y11. Biol 1988;28:39Y52. 8. Okun N, Gronau KA, Hannah ME. Antibiotics for 15. Franklin TL, Monif GR. Trichomonas vaginalis and bacterial vaginosis or Trichomonas vaginalis in pregnancy: a bacterial vaginosis. Coexistence in vaginal wet mount pre- systematic review. Obstet Gynecol 2005;105:857Y68. parations from pregnant women. JReprodMed2000; 9. McGregor JA, French JI, Parker R, Draper D, Patterson 45:131Y4. E, Jones W, et al. Prevention of premature birth by screening 16. Rendon-Maldonado JG, Espinosa-Cantellano M, and treatment of common genital tract infections: results of a Gonzalez-Robles A, Martinez-Palomo A. Trichomonas prospective controlled evaluation. Am J Obstet Gynecol 1995; vaginalis: in vitro phagocytosis of lactobacilli, vaginal epithelial 173:157Y67. cells, leukocytes, and erythrocytes. Exp Parasitol 1998; 10. Varma R, Gupta JK. Antibiotic treatment of bacterial 89:241Y50. vaginosis in pregnancy: multiple metaanalysis and dilemmas in 17. Moodley P, Connolly C, Sturm AW. Interrelationships interpretation. Eur J Obstet Gynecol Reprod Biol 2006; among human immunodeficiency virus type I infection, 124:10Y14. bacterial vaginosis, trichomoniasis, and the presence of yeasts. 11. Amsel R, Totten PA, Spiegel CA, Chen KC, Eschenbach D, J Infect Dis 2002;185:69Y73.

ERRATUM

In figure legends for Figures 1, 2, and 3 that appeared in the Home Study Course, which was in the Spring issue of the Journal of Lower Genital Tract Disease (2006;10:123Y125), the legends should have read: Figure provided courtesy of the National Cancer Institute.

Copyright @ 2006 American Society for Colposcopy and Cervical Pathology. Unauthorized reproduction of this article is prohibited.