1151 The Preterm Prediction Study: Failure of midtrimester cervical sialidase level elevation to predict subsequent spontaneous preterm birth William W. Andrews, PhD, MD, Jun Tsao, PhD, Robert L. Goldenberg, MD, John C. Hauth, MD, Brian Mercer, MD, Jay Iams, MD, Paul Meis, MD, Atef Moawad, MD, Anita Das, MS, Peter J. Van Dorsten, MD, Steve N. Caritis, MD, Gary Thurnau, MD, Menachem Miodovnik, MD, James Roberts, MD, and Donald McNellis, MD Bethesda, Maryland, and Birmingham, Alabama OBJECTIVE: Our objective was to determine any associations among midtrimester cervical fluid sialidase activity, bacterial vaginosis, and subsequent spontaneous preterm birth. STUDY DESIGN: In this nested case-control study all patients (n = 126) with spontaneous preterm birth at <35 weeks’ gestation and selected control subjects delivered at ≥37 weeks’ gestation (n = 126, matched for race, parity, and center) were derived from women enrolled in the multicenter National Institute of Child Health and Human Development Preterm Prediction Study. Sialidase activity and presence of bacterial vagi- nosis according to Gram stain were determined in cervical swabs and vaginal smears, respectively, obtained at 22 weeks’ to 24 weeks 6 days’ gestation. RESULTS: The mean ± SD sialidase activities were similar in case patients and control subjects (0.64 ± 1.60 vs 0.41 ± 0.94 nmol · mL –1 · min –1 , P = .21). Neither sialidase activity above the 90th percentile (10.3% vs 9.5%, P = .8) nor sialidase activity above the 95th percentile (7.9% vs 4.8%, P = .3) of control specimens (>1.43 and >2.23 nmol · mL –1 · min –1 , respectively) was associated with spontaneous preterm birth. The frequency of com- binations of bacterial vaginosis and elevated sialidase activity was similar (P ≥ .63 with either cutoff) in case pa- tients and control subjects. Sialidase activity was significantly higher among women with bacterial vaginosis than among those without bacterial vaginosis (1.35 ± 1.87 vs 0.03 ± 0.14 nmol · mL –1 · min –1 , P < .0001 ). CONCLUSIONS: Elevated cervical fluid sialidase activity at 22 to 24 weeks’ gestation did not distinguish women at increased risk for spontaneous preterm birth, nor did it discriminate a subgroup of patients who had bacterial vaginosis associated with spontaneous preterm birth. (Am J Obstet Gynecol 1999;180:1151-4.) Key words: Bacterial vaginosis, genital tract infection, neuraminidase, prematurity, preterm birth, sialidase Considerable data generated during the last several years indicate a strong relationship between subclinical upper genital tract infection and early spontaneous preterm birth. 1, 2 Many of the microorganisms isolated from the upper genital tracts of women with a sponta- neous preterm delivery have been associated with bacter- ial vaginosis, a condition that has consistently been asso- ciated with an increased risk for spontaneous preterm birth. 1-4 Indeed, bacterial vaginosis may be a lower geni- tal tract marker indicating an increased risk for upper genital tract infection. Several recently published randomized clinical trials indicate the efficacy of midtrimester antibiotic treatment in reducing spontaneous preterm birth among symptom- free high-risk women with bacterial vaginosis. 5-7 Deter- mination of other potential markers of asymptomatic genital tract infection is considered crucial to discrimi- nate the women who will be the most appropriate candi- dates for antibiotic intervention designed to prevent spontaneous preterm birth. 2, 8 It has been hypothesized that bacterial colonization of the upper genital tract in women who have a spontaneous preterm birth results from microorganisms that have as- cended from the lower genital tract. 1-3 Sialidase, also known as neuraminidase, is an enzyme produced by certain bacteria that colonize mucosal surfaces, including some of the microorganisms associated with bacterial vaginosis. 9-11 This enzyme, considered a virulence factor for pathogenic From the National Institute for Child Health and Human Development Maternal-Fetal Medicine Units Network and the Department of Microbiology, University of Alabama at Birmingham. A complete list of the members of the Network and their institutional affiliations appears at the end of the article. Received for publication August 13, 1998; revised December 18, 1998; accepted December 31, 1998. Reprint requests: William W. Andrews, PhD, MD, University of Alabama at Birmingham, Department of Obstetrics and Gynecology, 618 South 20th St, Birmingham, AL 35233-7333. 6/1/96883