correlates and risk factors for PROM in Mysore, India. WBCs in vaginal fluid, leucocytes in urine, UTI and infection with E. coli, S. aureus, C. albicans and BV ...
October 2006 of Medical Microbiology, (2006) 24 (4):283-5 Indian Journal
283
Brief Communication
THE CLINICAL AND MICROBIOLOGICAL CORRELATES OF PREMATURE RUPTURE OF MEMBRANES *C Karat, P Madhivanan, K Krupp, S Poornima, NV Jayanthi, JS Suguna, E Mathai
Abstract Prematurity is the cause of 85% of neonatal morbidity and mortality. Premature rupture of the membranes (PROM) is associated with 30-40% of preterm deliveries. A case-control study conducted between July 2002 and 2003 examined the correlates and risk factors for PROM in Mysore, India. WBCs in vaginal fluid, leucocytes in urine, UTI and infection with E. coli, S. aureus, C. albicans and BV were significantly associated with PROM. BV, E. coli and WBCs in vaginal fluid were independent risk factors. Screening and treatment of BV and E. coli infection in pregnancy may reduce the risk of PROM.
m o fr d nsage, gestational age and parity, Preterm premature rupture of the fetal membranes (PROM) and 150 controls a matched for is associated with 30-40% of premature births and is an was conductedlo betweenio July 2002 and 2003 at Holdsworth t n important cause of perinatal morbidity and mortality. It has Memorial Hospital, Mysore, South India. a w been estimated that 10% of perinatal deaths are directly or c Theoinclusionlicriteria were gestational age > 31 weeks; indirectly attributable to PROM. Limited data are available for d PROM b confirmed by clinical finding of posterior India, but studies from other parts of the world have shown diagnosis of u e vaginal pool, vaginal. pH or ferning; cervical dilation of < 3 the etiology of the condition to be multifactorial. Infection e P ) and clear liquor. Exclusion criteria r cms; singleton pregnancy; has been cited as a major cause of membrane damage. f w m woman in labour; mal-presentations; multiple r included: o o There is growing evidence associating upper genitalfo tract gestation; medical disorders; and a history of cervical c n . infections with PROM. One possible mechanism by which kencirclage. ecervical/ l d ow infections might act is through ascension from the b e for Protection of Human Subjects at C.S.I. nThe Committee vaginal area and replication in the placenta, the decidua and a l M k i Holdsworth Memorial Hospital approved the study. Data were the membranes. Another hypothesis is that a several organisms d y collected from all eligible participants who gave informed that are commonly present in the vaginal flora, including group e v b consent. A structured questionnaire collected data on a that B streptococci, S. aureus and microorganisms cause BV, m d . sociodemographic and medical variables. Collected biological s and tweaken secrete proteases that degrade collagen the fetal e i w specimens included liquor from the posterior fornix of the membranes leading to PROM. FResearchers have postulated s w vagina and midstream urine. Dipstick urinalysis was performed that PROM may be the result Dof directhobacterial winsults that (Multistix 10SG, Bayer Diagnostics). If nitrites and leucocytes necrotize tissue leading toP host-mediated auto ( destruction. found in the urine analysis, that patient’s urine was also eto evaluate the risk factors were t sstudy was i i The purpose of this sent for culture. s presenting with premature h women and correlates forT PROM ina Vaginal fluids were cultured on three primary culture media: rupture of membranes at a tertiary care hospital in Mysore, Key words: Case control study, prematurity, PROM
1
2
3,4
5,6
7,8
9,10
11,12
13
blood, chocolate and McConkey agar. Cultures were incubated at 37oCelsius for 24 hours. Gram stain was done for diagnosis of BV using Nugent’s score.
South India. Materials and Methods A hospital-based case-control study of 150 patients presenting with PROM at greater than 31 weeks of gestation *Corresponding author (email: )
Department of Obstetrics and Gynecology (CK, SP, NVJ, SJS),
Holdsworth Memorial Hospital, Mysore - 570 021, Karnataka, India,
Division of Epidemiology (PM, KK), University of California,
Berkeley, USA and Division of Microbiology (EM), Christian
Medical College, Vellore - 632 014, India
Received : 18-03-06
Accepted : 21-06-06
Data were entered in MS Excel and analyzed using Stata 8.0 (Stata Corporation, College Station, TX). The primary outcome was PROM, analyzed as a binomial variable. In univariate analysis, χ2 test and Fischer’s exact test were used to assess the associations between PROM and categorical variables and t-test and the Wilcoxon rank sum test were used for continuous variables. Likelihood-ratio tests were used to determine the statistical significance of each variable. Associations, measured as odds ratio (OR) and adjusted for relevant confounding factors, were estimated using
www.ijmm.org 283 CMYK
Indian Journal of Medical Microbiology
284
Table 1: Descriptive characteristics of 150 cases and 150 controls Characteristic n (%) Number of antenatal women Registration status* Booked Unbooked Region Urban Rural Socioeconomic status* Upper class Upper middle class Middle class Lower middle class Religion Hindu Muslim Christian Buddhist Occupation Housewife Engineer Teacher White blood cells* Many Few Leucocytes* History of recent coitus History of previous PROM* History of preterm delivery* History of UTI* Bacterial vaginosis* Positive Indeterminate Negative Culture of high vaginal swab* No significant growth S. aureus E. coli Candida albicans Enterobacter Group B streptococcus P. mirabilis
Cases n (%)
Controls
150
150
115 (76.7) 35 (23)
124 (82.7) 26 (17.3)
144 (96) 6 (4)
149 (99.3) 1 (0.6)
22 (14.7) 95 (63.3) 2 (1.3) 31 (20.7)
19 (12.7) 75 (50) 1 (0.6) 55 (36.7)
13 (8.7) 4 (2.6) 4 (2.6) 1 (0.67)
28 (19) 0 0 0
vol. 24, No. 4
Results About 23% of the study cases were unbooked as compared to 17% of the controls. Rates of PROM positively correlated with women residing in rural areas, unbooked status and coming from a lower socio-economic background (Table). Fifty-eight cases (39%), as compared to 22 controls (14%), had WBCs in their vaginal fluid indicating presence of genital tract infection. Overall, 6% of cases had urinary tract infections at the time of admission, as compared to 2% of controls. Close to a third (n=56) of the cases reported having coitus in the past week as compared to 49 controls. The detection of BV and presence of E. coli or S. aureus were significantly associated with occurrence of PROM. Five women with PROM reported a previous history of PROM as compared to only one control. Although four cases of Group B streptococcus and four of P. mirabilis were detected, none of the controls had either infection.
m o fr 98 (65) 90 (60) d analysis Logistic regression showed (Table 2) a significant sincreased 46 (30.7) 55 (37) a n positive association between WBCs in the vaginal o95% CI:ti2.09, o 6 (4) 4 (3) l fluid (OR: 3.67; 6.4), leucocytes the urine (OR: n a prior history of urinaryintract 0 1 (0.6) 6.43; 95%w CI: 1.4, 29.3), infections c (OR: 3.5; 95% CI:li1.05, 14.9), detection of BV (OR: 10.5, 95% o 150 (100) 142 (94.7) d 31.5),upresence b of E. coli (OR: 7.5; 95% CI: 2.8, 20.0), CI: 3.5, 0 1 (0.6) S.eaureus (OR: 2.8;. 95% CI: 1.24, 3.6) and subsequent Pof PROM.) 0 7 (4.7) rf eoccurrence wmultivariate manalysis, BV, E. coli or WBCs detected in 58 (38.8) 22 (14.7) r o o In n fluid.c were all independent risk factors for being fo kvaginal 92 (61) 128 (85) e d diagnosed 12 (8) 2l(1.3) wwith PROM. Interestingly, Candida infection was o b 56 (37) 49 (33) e protective n for PROM (a OR: 0.20; 95% CI: 0.07, 0.5).
la 1 (0.67) M dDiscussion
5 (3) k i 0 (2.00) a 33b(2.00)
y e 10 (6.7)v a d .m To the best of our knowledge, this is the first study s the risk factors and correlates for PROM in India. i (16) te 4 (3)w exploring 24 Interventions for prevention of PROM will have large public F 48 (32) s 9 (6) w o health implications here where less-than-optimal care and D 78 (52) 137 (91) w h nutrition are observed during pregnancy. Studies have shown ( P te that malnourished women having decreased level of host s si 57 (38) 93 (62) i defense factors regularly present in amniotic fluid so infectious 24 (16) agents such as E. coli and S. aureus may play a larger role. Th a 4229 (28) (20) 5 (3.3)
*P
