Efficacy of Intramuscular Penicillin in the Eradication of Group B Streptococcal Colonization at Delivery
Abstract
Objective. Due to rapid deliveries and human error, not all group B streptococcal positive mothers will receive adequate prophylactic antibiotic treatment in labor. We sought to determine if long acting intramuscular penicillin given after a positive culture result would be efficacious in eradicating group B streptococcal colonization at the time of delivery.
Methods. Patients positive for group B streptococci at 35-37 weeks were randomized to receive 2.4 million units of intramuscular benzathine penicillin G suspension (Bicillin L-A) versus no treatment. Study patients were recultured at the time of admission to labor and delivery prior to receiving prophylactic antibiotics according to CDC guidelines.
Results. A total of 53 patients were enrolled. A small but significant decrease in the rate of group B streptococcal colonization was observed in the treatment group (14/27, 52%) versus the control group (20/23, 87%), p = 0.03.
Conclusion. The large number of persistent carriers suggests that 2.4 million units of intramuscular benzathine penicillin G suspension (Bicillin L-A) is insufficient as sole therapy. However, the decline in group B streptococcal carriers might lessen the risk of failed or insufficient intrapartum treatment. Intramuscular benzathine penicillin G suspension (Bicillin L-A) may be useful as an adjunctive treatment for patients at risk for rapid delivery, before adequate intrapartum prophylaxis can be given.
Keywords: IM penicillin, group B streptococcal prophylaxis
Introduction
Recently the Center for Disease Controls (CDC) amended its guidelines to recommend routine screening of all pregnant mothers at 35-37 weeks, and treatment of all group B streptococcus positive mothers with intrapartum prophylactic antibiotics [1]. Coincident with the release of the CDC consensus guidelines in 1996, the rate of early onset group B streptococcus (GBS) infections declined significantly from 1.7 per 1000 in 1993 to 0.4 per 1000 in 2001 [1]. Prevention of vertical transmission has resulted in a dramatic decline in this devastating disease of the newborn.
There is no perfect prevention strategy. Mothers identified as carriers may deliver too precipitously and therefore fail to receive antibiotics altogether, or fail to achieve adequate levels of antibiotics prior to delivery [2]. In addition, human error may result in the failure to administer antibiotics for culture positive mothers. Finally, there will be mothers who are GBS positive at delivery, despite a negative culture at the time of their routine screening at 35-37 weeks. Yancey et al. demonstrated an 87% positive predictive value and a 96% negative predictive value for the identification of colonization at delivery when cultures were obtained at 35-37 weeks [3].
Ideally treatment closer to time of the culture results, and not in labor would help avoid the problems of human error and rapid deliveries. However, such a strategy is ineffective when using standard oral antibiotic therapy, owing to the high rate of recolonization. Yancey et al. demonstrated a 70% rate of colonization when oral antibiotics were given 10-14 days prior to delivery [3]. Several nonrandomized small trials have suggested that intramuscular benzathine penicillin G suspension (Bicillin L-A) may be efficacious owing to the fact that adequate serum levels are obtained for a longer period of time [4-6].
In this study we sought to determine if 2.4 million units of intramuscular benzathine penicillin G suspension (Bicillin L-A) was efficacious for the treatment of GBS positive carriers prior to labor in a prospective randomized fashion.
Methods
A prospective randomized clinical trial was carried out, with a protocol approved by the IRB at the Maine Medical Center. Written informed consent was obtained on all patients. Eligible patients were 18 years and older and GBS positive at 35-37 weeks. Cultures were obtained from rectovaginal swabs using Stuart’s transport medium and Lim’s broth through the Maine Medical Center microbiology lab (Nordx). Patients were recruited from four different sites, including three clinics and one private practice office, from September 1999 through March of 2002. Patients were consecutively approached by the providers participating in this study. Exclusion criteria included penicillin allergy, antibiotic use within the past month, chronic liver or renal disease, and the use of anticonvulsant medications.
GBS positive patients at their next office visit (4-10 days post culture) were recruited and randomized on the same day. Using a table of random numbers, patients were assigned to either a treatment or control group. Treated patients received a single injection of 2.4 million units of intramuscular benzathine penicillin G suspension (Bicillin L-A) into the buttocks. Patients receiving treatment were observed for a minimum of 30 minutes for signs of an allergic response and were advised to call with any abnormal reaction. The control patients received no injections. Prior to any patients receiving antibiotics upon admission for labor, all study patients were recultured using Stuart’s medium and Lim’s broth. Group B streptococcus positive patients on culture at 35-37 weeks subsequently received intravenous antibiotics per CDC guidelines. Semiquantitative colony counts were not performed in women who were colonized with group B streptococci.
We assumed a two-sided alpha of 0.05, a p^sub 0^ of 0.8 (estimated proportion GBS positive in control group), a p^sub 1^ of 0.5 (estimated proportion GBS positive in treatment group) and a power of 80% to calculate that we needed 44 patients. The power analysis was only applied to the primary outcome. The primary outcome variable was the difference in intrapartum and antepartum maternal GBS colonization. Secondary outcome measures included difference in failure rates for deliveries greater than 40 weeks compared to less than 40 weeks, difference in patient weight among successful treatment and treatment failures and the percentages of patients receiving antibiotics 2 hours prior to delivery. Statistics were calculated using GraphPad Quickcalcs (GraphPad Software Inc 2002) and StatView (SAS Institute, Inc., Cary, NC). Fisher’s exact test for categorical variables was used to analyze the main outcome measure and a relative risk for colonization at delivery was computed. Fisher’s exact test was also used to evaluate baseline characteristics of treatment and control groups.
Results
A total of 53 patients were enrolled with positive GBS cultures obtained at 35-36 weeks. Two patients in the control group and one in the treatment group were excluded because cultures were not obtained on admission to labor and delivery. There was no significant difference in the baseline characteristics of treatment and control groups (Table I). The average gestational age at the time intramuscular benzathine penicillin G suspension (Bicillin L- A) was given was 37.5 weeks, while the average gestational age at the time of delivery was 40.3 weeks.
Cultures obtained at the time of delivery from the treatment group showed a small but significant reduction in the proportion of patients with persistent GBS colonization. In those patients who received penicillin, 14 of 27 (52%) were GBS positive at the time of delivery, while 20 of 23 (87%) were GBS positive in the control group (p = 0.0139). The relative risk of colonization at delivery for the intramuscular benzathine penicillin G suspension (Bicillin L- A) group was 0.596 (CI 0.472-0.869).
After excluding 7 patients who underwent a cesarean without labor, 4 of 43 (9%), did not receive antibiotics. The reasons cited were “rapid labor, presented late, and no time for antibiotics”. The average time elapsed from admissions to delivery for patients receiving no antibiotics was 5 hours.
Of those delivering after 40 weeks, 11 of 17 (65%) failed penicillin treatment, while 3 out of 10 (30%) delivering prior to 40 weeks failed treatment (p = 0.12). The patients’ weight in treatment failure and treatment success groups were not statistically different (204 + 44 in success vs. 188 51 in failure, p = 0.43). There were no cases of neonatal group B streptococcal infections in either the control or treatment groups.
Discussion
In this small prospective randomized controlled study we were able to demonstrate a significant reduction in the numbers of GBS carriers at the time of delivery. However the large number of persistent carriers, 52%, suggests that 2.4 million units of intramuscular benzathine penicillin G suspension (Bicillin L-A) is insufficient treatment as sole therapy for GBS carriers. The difference in GBS status on admission to labor and delivery (80% control compared to 52% treatment) suggests that intramuscular benzathine penicillin G suspension (Bicillin L-A), given at the time of the initial positive GBS culture, might potentially lessen the risk of failed or insufficient intrapartum treatment.
Table I. Characteristics of treatment and control groups.
Aside from small numbers, there are two major limitations of this study. First, the dose of 2.4 million units was less than the dose of 4.8 million units as suggested by Bland et al. [6]. Secondly, we did not obtain data in regard to whether theGBS colonization was heavy or mild. One of the main strengths of our study was its prospective randomized design. Due to a relatively small sample size, conclusions can only be drawn with caution, especially when interpreting the secondary outcomes. The length of time required to enroll a sufficient number of patients may reflect a higher rate of patients declining to participate due to the existence of effective intrapartum therapy and a desire to avoid an uncomfortable intramuscular injection.
The rationale of using 2.4 million units versus 4.8 million units was based on limiting patient discomfort and at the same time choosing a dose that would likely provide a bactericidal level in maternal serum for a prolonged period of time. Weeks et al. demonstrated a level of greater than 0.20 ug/ml in all patients 40 or more days post treatment with 4.8 million units of intramuscular benzathine penicillin G suspension [5]. Despite three times the adequate bactericidal level (0.06 ug/ml) at the time of delivery, 28% of the treated patients had persistent GBS. Lewin and Amstey used 2.4 million units, based on work done on pregnant patients with syphilis, and demonstrated 18% persistence of GBS on vaginal cultures only [4]. Likewise, recent work by Bland et al. with 4.8 million units, reported a 25% persistent carrier rate, although none had heavy growth [6]. Bacteriocidal doses of penicillin following 2.4 million units of intramuscular benzathine penicillin G suspension have been reported at least 10 days post administration in pregnant patients [7]. MIC^sub 90^ values were not determined at deliveries for the cultures obtained in this study. None of the previous studies were randomized. In an effort to minimize maternal discomfort (single IM injection), we opted to randomize patients to receive a dose that was likely to produce reasonable levels of antibiotics and had shown beneficial effect in previous studies. It is possible that higher doses of intramuscular benzathine penicillin G suspension (Bicillin L-A) could have achieved a greater reduction in carrier status.
Even with 4.8 million units of intramuscular benzathine penicillin G suspension, the Bland study still reports an unacceptably high persistence of GBS [6]. Our results are consistent with the Bland study in that the risk reduction is not substantial enough to recommend intramuscular benzathine penicillin G suspension (Bicillin L-A) as an alternative to intrapartum prophylaxis. The high rate of failure of antepartum intramuscular benzathine penicillin G suspension (Bicillin L-A) may be due to inadequate antibiotic levels. An alternative but unlikely hypothesis, is that the failures were secondary to resistant GBS selected by the chronic low penicillin exposure. However, as we demonstrated in this study, treatment failures may occur as a result of rapid delivery. Given that the average time from admission to delivery in the patients who failed to receive antibiotics was 5 hours, human error resulting in the delay of administration of antibiotics likely also contributes to treatment failures. If the number of GBS negative patients entering labor and delivery can be increased (from 13 to 48% in our study), intramuscular benzathine penicillin G suspension (Bicillin L- A) might be used as an adjunctive treatment and administered selectively to women who are at risk for rapid delivery before adequate intrapartum prophylaxis can be given.
Acknowledgement
Supported by Maine Medical Center Research funds.
References
1. Schrag S, Gorwitz R, Fultz-Butts K, et al. Prevention of perinatal group B streptococcal disease. Revised guidelines from CDC. MMWR 2002;51:1-22.
2. Pinette MG, Wax JR, Blackstone J, et al. Culture-based group B streptococcal screening. J Reprod Med 2003;48:309-312.
3. Yancey MK, Schuchat A, Brown LK, et al. The accuracy of late antenatal screening cultures in predicting genital group B streptococcal colonization at delivery. Obstet Gynecol 1996;88:811- 815.
4. Lewin EB, Amstey MS. Natural history of group B streptococcus colonization and its therapy during pregnancy. Am J Obstet Gynecol 1981;139:512-515.
5. Weeks JW, Myers SR, Lasher L, et al. Persistence of penicillin G benzathine in pregnant Group B streptococcus carriers. Obstet Gynecol 1997;90:240-243.
6. Bland ML, Vermillion ST, Soper DE. Late third-trimester treatment of rcctovaginal group B streptococci with benzathine penicillin G. Am J Obstet Gynecol 2000;183:372-376.
7. Nathan L, Bawdon RE, Sidawi E, Stettler R et al. Penicillin levels following the administration of benzathine penicillin G in pregnancy. Obstet Gynecol 1993;82:338-342.
MICHAEL G. PINETTE, KATHARINE THAYER, JOSEPH R. WAX, JACQUELYN BLACKSTONE, & ANGELINA CARTIN
Division of Maternal-Fetal Medicine, Department of Obstetrics and Gynecology, Maine Medical Center, Portland, Maine, USA
Correspondence: Michael G. Pinette, MD, MMC Ob/Gyn Associates, 887 Congress Street, Suite 200, Portland, ME 04102, USA. Tel: 207771 5549. Fax: 207771 7834. E-mail: cartia@mmc.org
Copyright CRC Press May 2005