THE CONTRIBUTION of infection to untoward outcomes of pregnancy has been of considerable interest since the classic studies of Kass,1 in which bacteriuria was associated with low birth weight in untreated but not in treated bacteriuric pregnant women. Subsequent studies evaluated the role of single infectious agents such as Ureaplasma urealyticum.2 More recently, it has become evident that many of these single agents were part of the complex microbial flora that was associated with adverse outcomes of pregnancy.3
Bacterial vaginosis is not only associated with adverse pregnancy outcomes, but at least in certain high-risk populations successful treatment of bacterial vaginosis can reduce the incidence of preterm delivery.4 It follows that any study of the etiology of low birth weight and other untoward outcomes of pregnancy must control for the presence of bacterial vaginosis.
There are many other factors that can potentially influence the outcome of pregnancy and thus, act as confounders in any study that attempts to determine causes of low birth weight, premature delivery, and other adverse outcomes of pregnancy. Such factors include race, marital status, education, cigarette smoking, parity,5 and use of illicit drugs.6
The issue of trichomoniasis during pregnancy has attracted the attention of physicians for more than 70 years. In 1923, Wilhelm Liss found that febrile puerperal morbidity occurred twice as often in women who had trichomoniasis as in controls.7 Similarly, Gragert8 in 1923 and Bland et al9 in 1931 reported an association between trichomoniasis and febrile puerperal morbidity. Febrile puerperal morbidity was defined as a temperature of 100.4 °F or higher on 2 postpartum days.
In this issue, Cotch and associates, using data from the Vaginal Infections and Prematurity (VIP) Study, report a multicenter study in which 13,816 pregnant women were followed up prospectively. They determined the prevalence of T. vaginalis at enrollment and performed multivariate analysis for many other possible factors such as age, race, geographic location, smoking, marital status, education, income, gravidity, weight, other microorganisms present, and metronidazole use. Using logistic regression analysis, they found that women infected with T. vaginalis were 30% more likely than uninfected women to deliver preterm or to have low birth weight infants. They were also 40% more likely to have an infant who was delivered preterm and at low birth weight.
The clear-cut association demonstrated by Cotch et al between trichomoniasis, preterm delivery, and low birth weight raises important questions about the management of trichomoniasis during pregnancy. Should all pregnant women be examined for trichomoniasis as part of routine prenatal care? Should all infected women be treated? Can we recommend metronidazole for widespread use during pregnancy?
The magnitude of these questions becomes evident if one considers that the worldwide estimates of the number of infections with T. vaginalis are as high as 180,000,000. Most infections occur in persons of reproductive age, and there is a higher prevalence in pregnant women.10 The prevalence of trichomoniasis in women of reproductive age ranges from 2% among private patients to 56% in women attending sexually transmitted disease (STD) clinics. Thomason11 reported that 42% of symptomatic women attending an STD clinic in Milwaukee were infected with T. vaginalis. In unselected pregnant women attending an antenatal clinic in South Africa, trichomoniasis was present in 65%.12
At this stage of our knowledge, do we have enough information to recommend screening of all pregnant women for trichomoniasis and to recommend treatment for any pregnant women found to be infected? The association between trichomoniasis low birth weight and premature delivery reported herein by Cotch and coworkers is fairly convincing, but it is only an association. A treatment study such as that currently being conducted by The National Institute of Child Health and Human Development will be needed to determine whether the adverse outcomes associated with trichomoniasis can be prevented by treating the infected pregnant women.
However, Trichomonas vaginalis is not part of the normal vaginal flora. There is no evidence that its presence does any good for the pregnant woman or her offspring. Accordingly, a fairly strong case can be made for initiating screening for and treatment of pregnant women with trichomoniasis while awaiting the results of the aforementioned treatment trial.
Although metronidazole is usually considered to be contraindicated during the first trimester of pregnancy and a drug to be avoided if at all possible later in pregnancy, the evidence supporting its proscription is scanty. Metronidazole is mutagenic in bacteria,13,14 and case reports describe mid-line facial defects in infants born to women exposed to metronidazole.15,16 On the other hand, a study published in 1965 17 found no toxicity to the fetus after treating 375 pregnant women. Most of these women were treated during the second or third trimester. A year later, an Air Force study of 206 women who received metronidazole while pregnant found no overall increase in the incidence of congenital abnormalities in their infants, although a slightly higher than expected incidence of birth defects in babies exposed to the drug during the first trimester was detected, using historical controls.18 The authors noted that there was no apparent cause and effect relationship nor any discernible pattern, but in cautionary note, they recommended withholding metronidazole during the first trimester until more data were available. Similarly, a retrospective cohort study of 1,387 patients19 who received oral metronidazole while pregnant and 1,387 controls found no evidence of increased birth defects. A meta-analysis published in 1995 evaluated 32 studies, 7 of which met the criteria of including at least 10 patients exposed to metronidazole during the first trimester of pregnancy, inclusion of either a group of pregnant women not exposed to metronidazole, or a group of women exposed only during the third trimester, and a report of the number of malformations observed in live-born infants in each of these groups. Their conclusion was that metronidazole does not appear to be associated with an increased teratogenic risk.20
Although evidence for the teratogenicity of metronidazole is diminishing, the evidence describing a relationship between trichomoniasis and preterm delivery and low birth weight is increasing. It may be that withholding metronidazole from pregnant women in order to protect them against its largely theoretical adverse consequences exposes patients to the not-so-theoretical risk of adverse pregnancy outcomes. Continued avoidance of metronidazole during pregnancy may do more harm than good.
1. Elder HA, Santamarina BAG, Smith S, et al. The natural history of asymptomatic bacteriuria during pregnancy: the effect of tetracycline on the clinical course and the outcome of pregnancy. Am J Obstet Gynecol 1971; 111:441-445.
2. McCormack WM, Rosner B, Lee Y-H, Munoz A, Charles D, Kass EH. Effect on birth weight of erythromycin treatment of pregnant women. Obstet Gynecol 1987; 62:202-207.
3. Hillier SL, Nugent RP, Eschenbach DA, et al. Association between bacterial vaginosis and preterm delivery of low birth weight infants. N Engl J Med 1995; 333:1737-1742.
4. Hauth JC, Goldenberg RL, Andrews WW, DuBard MB, Cooper RL. Reduced incidence of preterm delivery with metronidazole and erythromycin in women with bacterial vaginosis. N Engl J Med 1995; 333:1732-1736.
5. Kleinman JC, Kessel SS. Racial differences in low birth weight: trends and risk factors. N Engl J Med 1987; 317:749-753.
6. Shiono PH, Klebonoff MA, Nugent RP. The impact of cocaine and marijuana use on low birth weight and preterm birth: a multicenter study. Am J Obstet Gynecol 1995; 172:19-27.
7. Liss W. Der EinfluB der Trichomonadenkolpitis auf die Wochenbettsmorbiditar. Monatsschrift fur Guburtshulfe und Gynakologie 1923; 64:31-36.
8. Gragert O. Wochenbettsmorbiditat bei ant partum nichthehan-delten und ant partum benhandelten Fallen von Trichomonadenkolpitis. Monatsschrift fur Geburtshulfe und Gynakologie 1923; 63:37-41.
9. Bland PB, Goldstein L, Wenrich DH. Vaginal trichomoniasis in the pregnant woman. A clinical and morphologic study. JAMA 1931; 96:157-163.
10. Brown MT. Trichomoniasis. The Practitioner 1972; 209:639-644.
11. Thomason JL, Gelbart SM. Trichomonas vaginalis. Obstet Gynecol 1989; 74:536-541.
12. Hoosen AA, Ross SM, Mulla MJ, Patel M. The incidence of selected vaginal infections among pregnant urban blacks. S Afr Med J 1981; 59:827-839.
13. Voogd CE, Vand Der Stel JJ, Jacobs JA. The mutagenic action of nitromidazoles. Mutation Res 1979; 66:207-221.
14. Rosenkraz Jr. HS, Speck WT, Stambaugh JE. Mutagenicity of metronidazole: Structure-activity relationships. Mutation Res 1976; 38:203-206.
15. Greenberg F. Possible Metronidazole teratogenicity and clefting. Am J Med Gen 1985; 22:825.
16. Cantu JM, Garcia-Cruz D. Midline facial defect as a teratogenic effect of metronidazole. Birth Defects 1982; 18:85-88.
17. Robinson SC, Mirchanani G. Trichomonas vaginalis. Further observations on metronidazole (Flagyl) including infant follow-up. Am J Obstet Gynecol 1965; 93:502-505.
18. Peterson WF, Stauch JE. Metronidazole in Pregnancy. Am J Obstet Gynecol 1966; 94:343-349.
19. Piper JM, Mitchel EF, Ray WA. Prenatal use of metronidazole and birth defects: no association. Obstet Gynecol 1993; 82:348-352.
20. Burtin P, Taddio A, Ariburnu O, Einarson TR, Koren G. Safety of metronidazole in pregnancy: a meta-analysis. Am J Obstet Gynecol 1995; 172:525-529.