Canadian Task Force on Preventive Health Care

Full Text Review

Screening for Asymptomatic Bacteriuria in Pregnancy

Prepared by Lindsay E. Nicolle, MD, Department of Medicine, University of Manitoba

These recommendations were finalized by the Task Force in October 1993

Up Contents

UpOverview

Asymptomatic bacteriuria is common in women and increases in prevalence with age and/or sexual activity. The impact of asymptomatic bacteriuria on pregnancy outcome has been a focus of controversy since the development of quantitative urine culture technique in the mid 1950s allowed clear differentiation of women with bacteriuria from those without. While it is generally accepted that asymptomatic bacteriuria is detrimental to pregnancy, data available to support this contention is limited. Randomized controlled trials and cohort studies have shown that the detection and treatment of asymptomatic bacteriuria can decrease the occurrence of acute pyleonephritis later in pregnancy and decrease the occurrence of intrauterine growth retardation. Asymptomatic bacteriuria in children and in the elderly are addressed in Chapters 21 and 81 respectively.

Up Burden of Suffering

The prevalence of asymptomatic bacteriuria in pregnancy varies from 4-7% (range 2-11%) and is similar to that observed in non-pregnant women.<1,2> The prevalence is higher among individuals in lower socioeconomic classes, and those with a past history of asymptomatic urinary infection. Increased frequency of screening during pregnancy identifies more cases. Approximately 1-2% of women who are not bacteriuric at initial screening early in pregnancy will develop bacteriuria later in the pregnancy.

There is a high incidence of pyelonephritis occurring later in pregnancy, usually at the end of the second trimester or beginning of the third trimester, in women with asymptomatic bacteriuria identified and not treated early in pregnancy. Initial studies reported 20% to 27% of women with asymptomatic bacteriuria developed pyelonephritis<3,4> compared to 0.4% to 1.2% of those without bacteriuria. A more recent study reported 13% of untreated women with asymptomatic bacteriuria developed pyelonephritis, compared with 0.4% of those with negative screening cultures.<5> Pyelonephritis in pregnancy generally requires hospitalization for treatment and, as with any severe febrile illness in later pregnancy, may be associated with premature labour and delivery. In the pre-antibiotic era 23-54% of women with acute pyelonephritis had premature births, and 6% of all premature births were reported to be due to pyelonephritis.<2> In the post-antibiotic era, Gilstrap et al<6> reported 25% of women with intrapartum pyelonephritis delivered low birth weight infants compared to 15% of controls and suggested this was evidence for acute pyelonephritis in pregnancy being associated with pre-term labour.

The association of asymptomatic bacteriuria with other complications of pregnancy including stillbirth, intrauterine growth retardation, and preterm labour in the absence of acute pyelonephritis has been more controversial.<1> However, a review of birth certificate data for the state of Washington for 1980 and 1981 reported that women with urinary infection (UTI) associated pregnancy had a fetal mortality rate 2.4 times greater, low birth rate 2.04 times greater, and prematurity 2.4 times greater than those without urinary infection.<7> Naeye<8> reported observations from the Collaborative Perinatal Study of the U.S. National Institute of Neurological and Communicative Disorders and Stroke which followed 60,000 pregnancies between 1959 and 1966. The perinatal mortality rate was twice as high in women with UTI within 15 days of delivery. The low birth weight associated with urinary infection was due to a contribution of both preterm delivery and fetal growth retardation. Both of these studies included both symptomatic and asymptomatic infections.

In a meta-analysis of 19 studies, Romero et al reported that women with asymptomatic bacteriuria had a 54% higher risk of a low birth weight infant and twice the risk of a pre-term infant compared with non-bacteriuric women.<9> The mechanism by which asymptomatic bacteriuria promotes preterm labour is not clear, but subclinical amnionitis or phospholipid A2 production by bacteria have been proposed.<1> One study reported that 40% of women with bacteriuria at delivery had post-partum endometritis compared with 2.2% of women without bacteriuria.<10> There is no compelling evidence of an association of asymptomatic bacteriuria with hypertension in pregnancy or of long-term renal damage associated with asymptomatic bacteriuria of pregnancy in the antibiotic era.<1>

Up Maneuver

The gold standard for screening for asymptomatic bacteriuria is growing bacterial cultures of urine samples from women in early pregnancy (12-16 weeks gestation).

Non-culture methods are not, generally, reliable for the identification of bacteriuria in asymptomatic populations, including pregnant women.<1,11> Routine urinalysis is imprecise for identification of pyuria and other pyuria-based methods, particularly leukocyte-esterase dipstick, are subject to false negatives in bacteriuria without pyuria and false positives with contamination from vaginal secretions. Currently only culture methods can be considered acceptable in pregnancy. The routine use of the standard semi-quantitative culture may be expensive and problems with contamination with vaginal and periurethral flora at collection and overgrowth of organisms when culture is delayed may lead to erroneous results. A dipslide method is substantially less costly, allows immediate inoculation after specimen collection, requires no special equipment for incubation as it may be left at room temperature, and will identify specimens with significant bacteriuria which should be fully analyzed by the laboratory.<1> For women with asymptomatic bacteriuria two consecutive positive specimens are necessary for diagnosis. Thus, for women with a positive dipslide, a second specimen should be forwarded to the laboratory for quantitative culture, identification of organisms, and antimicrobial susceptibility testing.

The optimal timing to obtain specimens to identify women with asymptomatic bacteriuria in pregnancy has been investigated in a number of studies. A specimen obtained at 12-16 weeks will identify 80% of women who will ultimately have asymptomatic bacteriuria during pregnancy.<12> Repeated screening on a monthly basis will identify bacteriuria in only an additional 1-2% of patients. A single urine specimen obtained from 12-16 weeks of pregnancy will identify most women with asymptomatic bacteriuria.

Up Effectiveness of Prevention and Treatment

Prevention of Acute Pyelonephritis

Studies have consistently reported a decrease in acute pyelonephritis later in pregnancy from 20-30% to 2-4% for women who have been identified with asymptomatic bacteriuria in early pregnancy and treated.<1,3-5> One study, reported in letter form, found no difference in pyelonephritis in treated (3.0%) and untreated (2.5%) women with asymptomatic bacteriuria, but this is at variance with all other reports.<13> However, as 0.5%-1.0% of women with initial negative screening urine cultures will develop pyelonephritis subsequently in pregnancy, only 50-80% of the burden of pyelonephritis in pregnancy is prevented through screening and treatment. One study suggested that screening 4,470 pregnant women with treatment of asymptomatic bacteriuria prevented only 6 cases of pyelonephritis and was likely not cost effective, but no formal cost analysis was provided.<5> Wadland and Plante<14> reported that screening with a single urine specimen in early pregnancy was cost effective as long as the population prevalence of bacteriuria was over 2%.

Prevention of Pre-term Delivery and Low Birth Weight

Early studies of the treatment of asymptomatic bacteriuria generally did not report decreased low birth weight or premature delivery with therapy of asymptomatic bacteriuria.<1> The meta-analysis of Romero et al,<9> however, of 8 clinical trials, reported a relative risk of 0.56, (95% confidence interval 0.43-0.73) for antibiotic treatment of asymptomatic bacteriuria reducing low birth weight. The mechanism by which treatment of asymptomatic bacteriuria prevents low birth weight is not clear. It has been suggested that antibiotic therapy, in fact, eradicates organisms from the cervix and vagina which may be associated with subclinical chorioamnionitis, rather than a direct effect of treatment of urinary infection.<9>

Up Recommendations of Others

In 1989, the U.S. Preventive Services Task Force<15> recommended periodic testing for asymptomatic bacteriuria for pregnant women. The optimal frequency for urine testing was unknown and therefore left to clinical discretion. Since urine culture is a more accurate screening test then dipstick urinalysis, it was recommended for detecting asymptomatic bacteriuria during pregnancy. This recommendation is currently under review.

Up Conclusions and Recommendations

Identification and treatment of asymptomatic bacteriuria will lead to a 10-fold decrease in the occurrence of acute pyelonephritis later in pregnancy in women with asymptomatic bacteriuria. Treatment of asymptomatic bacteriuria will also decrease pre-term delivery and low birth weight. A single screening test using a culture method at 12-16 weeks of pregnancy will identify 80% of women with asymptomatic bacteriuria of pregnancy. Screening for and treatment of asymptomatic bacteriuria is likely cost-effective. There is good evidence to recommend screening for asymptomatic bacteriuria in pregnancy (A Recommendation).

Up Unanswered Questions (Research Agenda)

1. Further studies of optimal screening methods, both nonculture and culture methods, with cost analyses are needed. These should be developed as randomized prospective trials comparing the different methods with endpoints of pregnancy outcomes as well as identification of bacteriuria.

2. The efficacy of screening only targeted high-risk groups such as those with a past history of urinary infection or of lower socioeconomic status compared to screening all pregnant women should be compared.

3. The optimal frequency of screening for recurrent bacteriuria in subjects initially identified with bacteriuria and treated should be determined.

Up Evidence

The literature was identified with a MEDLINE search to March 1993 using the following MESH headings: urinary tract infections, pregnancy, human, case reports.

This review was initiated in June 1993 and recommendations were finalized by the Task Force in October 1993.

 Full Citation

Link to Structured Abstract of this review

Link to Summary Table of this review

Link to Selected References list of this review

Top of Page

Home PageCTFPHC Home Page

Reprinted in modified format by the Canadian Task Force on Preventive Health Care
with permission.
For any technical issues please contact: webmaster@ctfphc.org
Original Copyright © 1994 Minister of Supply and Services Canada.
Last modified March 27, 1998.