Since its emergence in the 1970s, group B streptococcus (GBS) is still the most common pathogen responsible for sepsis in neonates during the first week (0–6 days, ie, early onset) of life in developed countries.1 Prevention strategies of early-onset GBS disease in newborns rely on reducing or eliminating vertical transmission of GBS to the newborn with systemic administration of intrapartum antibiotic prophylaxis2–4 to GBS colonized mothers or to mothers with specific risk factors if their GBS colonization status is unknown. Until recently, the only method for maternal GBS carriage detection was bacterial culture, which is typically performed at 35–37 weeks of gestation. However, vaginal GBS colonization can be intermittent during pregnancy5; at least 10% of women who were screened negative using late antenatal culture screening were found positive at the time of delivery.6 In the era of intrapartum antibiotic prophylaxis, most remaining early-onset GBS disease cases occur in newborns whose mothers had negative results of culture screening for GBS colonization at 35–37 weeks of gestation.6–9 After the molecular GBS test (Xpert GBS, Cepheid) became available, our team evaluated its analytical performances (Sensitivity 98.5%, Specificity 99.6%) and confirmed the poor positive predictive value of antenatal culture screening (58.3%) compared with intrapartum polymerase chain reaction (PCR) screening.10 Since January 2010, the 40 midwives of our institution perform exclusively intrapartum GBS molecular screening around-the-clock on the GeneXpert system located in the delivery suite for all term deliveries instead of antenatal culture screening.11 A key success factor of the adoption of intrapartum screening was the empowerment of the midwives in the management of PCR processing without wasting time and the assurance that the result would be available to take appropriate action.
The aim of the present study was to assess outcomes and costs associated with around-the clock point-of-care intrapartum GBS PCR screening in the delivery suite.
This single institution, before-after uncontrolled study was conducted at a 680-bed general hospital, the Paris-Saint-Joseph hospital, France. The maternity ward manages around 3,000 deliveries per year at or beyond 32 weeks of gestation.
Intrapartum PCR screening was implemented in January 2010. The study periods included 4 years before and 6 years after the intervention commencing in 2006 and concluding in 2015. Data used for this study were routinely collected at the hospital level. We were limited to four years before intervention because of the change of the computer software of the maternity ward. Women were informed at the time of admission that de-identified patient-level data could be used for research purposed and given the opportunity to refuse. Paris-Saint Joseph hospital Institutional Review Board therefore exempted the study from approval.
During the antenatal screening period a vaginal swab was performed at 35–37 weeks of gestation and cultured.4 Intrapartum antibiotic prophylaxis was administered to women at the onset of labor or rupture of membranes if their antenatal screening was positive, if they had GBS bacteriuria during the current pregnancy, or if they had a history of delivering a neonate who experienced early-onset neonatal GBS disease. If the GBS status was unknown at the time of delivery, intrapartum antibiotic prophylaxis was administered to women with any of the following risk factors: membrane rupture over 12 hours, intrapartum fever equal to or higher than 38°C, and preterm delivery before 37 weeks of gestation.4
During the intrapartum PCR screening period, midwives performed GBS PCR screening around-the-clock, for all term deliveries. The single use cartridge of the Xpert GBS molecular test contains the sample and all necessary reagents and is processed in a completely automated fashion in <50 minutes, on the GeneXpert system (Cepheid) located in the delivery suite. At the onset of labor or after rupture of membranes, midwives swabbed the lower vagina using a double swab. One swab is immediately used for the Xpert GBS PCR test in the delivery suite. The leftover swab is sent to the laboratory for culture, to be available in the event of a nondiagnostic PCR result. Intrapartum antibiotic prophylaxis (5 million international units penicillin G followed by 2.5 million international units every 4 hours until delivery) is then administered at the onset of true labor to women who screened positive, or to those presenting with the same risk factors as above in the event of nondiagnostic PCR results. Prophylaxis was also administered to women with amniotic membrane rupture ≥18 hours following negative PCR testing.12
During both study periods, intrapartum antibiotic prophylaxis was not administered for cesarean deliveries performed before labor onset in women with intact amniotic membranes. In the case of penicillin allergy, clindamycin was used for susceptible GBS strains during the antenatal culture screening period or vancomycin otherwise; during the intrapartum PCR screening period vancomycin was the sole alternative because the Xpert GBS PCR test does not detect clindamycin GBS resistance.12
All cases of early-onset neonatal GBS disease were retrieved from the microbiology laboratory and linked to medical charts and discharge information. Records of newborns with GBS infection were reviewed by a neonatologist for diagnosis, clinical signs, laboratory evaluations, and the number of days of antibiotic therapy. Diagnosis of proven early-onset GBS disease was defined by positive GBS culture results of blood or cerebrospinal fluid or both, in the presence of clinical signs or biological abnormalities consistent with sepsis. Probable early-onset GBS disease was defined by positive GBS culture results of gastric fluid aspiration, ear or rectum culture, clinical signs, biological abnormalities consistent with sepsis, and negative cultures of blood and cerebrospinal fluid.13
During the intrapartum screening period, records of women who were screened positive or with invalid or error PCR results were also analyzed to check compliance with intrapartum antibiotic prophylaxis recommendations of the European Consensus of intrapartum GBS screening.12 During both study periods, we recorded length of stay, antibiotics used for the prophylaxis, and the number of doses.
We calculated the cumulated health care costs of mothers and newborns (healthy and early-onset GBS disease) during the index admission from the perspective of the National French Health Insurance, based on diagnosis related group (DRG).14 The French health care system fully covers all pregnancy-related health care expenditures. The costs of antenatal culture and intrapartum PCR screenings tests were estimated using reimbursement rates of the National French Health Insurance. The cost of intrapartum antibiotic prophylaxis was included in the DRG cost. The unit of analysis was the mother and not the newborn, because in case of twin birth we assumed that prevention would benefit both twins. All costs were calculated using 2018 prices or tariffs for out of hospital care and the latest available hospital cost schedule for hospital admissions.
We estimated the total and yearly costs of screening and of treating early-onset GBS disease cases and we calculated the cost of avoiding one early-onset GBS disease by dividing the difference in costs during the pre and post PCR periods by the difference in the total number of cases diagnosed.
De-identified data was extracted from the microbiology and administrative database into a Microsoft Excel database which was then imported to SAS software for analysis. The rate of early-onset GBS disease was calculated as the number of confirmed and probable early-onset GBS disease cases per 1,000 live births. Continuous variables were reported as mean±SD; we used Student t test to compare variables with normal distribution and Mann-Whitney U test for nonnormal distributions and discrete variables. For dichotomous variables (comparison of proportions), we used Pearson χ2 test. All tests were two-tailed and carried out with a significance level alpha =.05. Analyses were done on SAS version 9.1 (SAS Institute, Cary, NC).
Euros were converted into U.S. dollars using the OECD purchasing parity index (1 U.S.$=0.75€ (https://stats.oecd.org/Index.aspx?DataSetCode=SNA_TABLE4).
A total of 11,226 deliveries were recorded in 2006–2009 (antenatal culture screening) and 18,835 in 2010–2015 (intrapartum PCR screening), corresponding to 11,818 and 18,980 live births, respectively (Table 1). Term deliveries represented respectively 93.0% and 93.3% of all deliveries. There was no significant difference in the before and after cohorts in terms of gestational age, duration of labor, or time from rupture of membranes to delivery.
During the intrapartum PCR screening period, 14.5% of term deliveries screened positive compared with 12.2% during the antenatal culture period (P<.001) and 91.8% women who were screened positive received intrapartum antibiotic prophylaxis compared with 89% during the antenatal culture period (P=.55). During the antenatal culture period, 395 (3.8%) of term deliveries did not undergo GBS testing compared with 23 (0.1%) in the intrapartum PCR period (P<.001). In both periods, Penicillin G was the chosen treatment in 94.8%, and approximately 50% of treated women received at least two doses during both periods.
In the intrapartum screening period, 5.2% of the treated women received one dose of vancomycin because of a history of penicillin allergy. Two hundred and six women with a positive PCR result were not treated (8.8%), mainly because of very rapid labor (<30 minutes) (n=163) or inadvertently (n=43). For the 1,050 (6.5%) women with unknown GBS status at delivery because of a nondiagnostic PCR result, 260 (24.8%) had a duration of membrane rupture before delivery of at least 18 hours and should have received intrapartum antibiotic prophylaxis; missed intrapartum antibiotic prophylaxis occurred in 58 (22%).
The number of proven (bacteremia and meningitis) and probable early-onset GBS disease cases was 60% lower in the intrapartum PCR period, 3.8/1,000 (95% CI 2.3–7.4) compared with 0.9/1,000 (95% CI 0.6–1.5) (P<.001). Between the two periods, the rate of proven early-onset GBS disease cases decreased from 1.01/1,000 to 0.21/1,000 (P=.026) and probable early-onset GBS disease cases from 2.8/1,000 to 0.73/1,000 (P<.001) with a risk ratio of 0.25, 95% CI 0.14–0.43 for both proven and probable cases. The proportion of newborns with GBS infection in mothers who screened positive decreased from 0.86% to 0.38% (P=.043), and from 1.40% to 0.28% (P=.009) in those with unknown GBS status at delivery. More importantly, in women who screened negative, the proportion of early-onset GBS disease fell from 0.36% to 0.04% (P<.001) with a 78.5% decrease in the number of early-onset GBS disease cases during the intrapartum PCR screening period compared with the antenatal culture screening period (Table 2).
Total days of hospital and antibiotic therapy for early-onset GBS disease declined by 64% and 60%, respectively, proportionate to the reduction in infections with no significant difference for average length of stay or antibiotic duration preintervention and postintervention. The average length of stay for delivery declined one day between the two periods owing to changes in postnatal practices of newborns' surveillance (Table 3). The average cost per delivery (including all vaginal or cesarean deliveries and the costs of tests) was $2,691 in the antenatal screening period and $2,381 in the intrapartum screening period. The yearly cost of delivery and treatment of newborns with GBS infection was reduced from $41,875±6,823 to $11,945±10,303 after the implementation of intrapartum PCR screening (Table 4). The added cost of PCR compared with culture was partly offset by the reduction in early-onset GBS disease and the remaining extra cost represented on average $49 per newborn. The estimated extra cost of PCR to avoid one additional case of early-onset GBS disease was $5,819 (95% CI 2,909–10,183).
We found that intrapartum PCR screening compared with antenatal culture screening was associated with a significant decrease in the rate of proven and probable early-onset neonatal GBS disease cases. The decrease in early-onset neonatal GBS disease was associated with a threefold reduction in the total numbers of days of hospital and antibiotics for early-onset GBS disease.
The lack of a control group in our pre–postintervention comparison was a limitation. The 3.8/1,000 incidence of probable and confirmed cases of early-onset GBS disease was close to the figure of 4.75/1,000 found in a regional survey in France.15 An Australian study reported a decreasing incidence of neonatal sepsis of 0.4/1,000 per year, from 1.20/1,000 live births in 2002 to 0.83 in 2012 but no change in the incidence of early-onset GBS disease. Our starting point, however, was higher at 1.01/1,000 of proven early-onset GBS disease compared with the 0.43/1,000 live births reported in Australia and New Zealand and therefore some of the observed reduction in our study could be explained by external factors.16 However, given the French intrapartum antibiotic prophylaxis coverage, the average risk of early-onset GBS disease was predicted to be 3.8/1,000 (95% CI 0–9) which corresponds to the results of the antenatal screening period in our study.17 Since intrapartum antibiotic prophylaxis coverage did not change over the 10-year period covered by our study, we can assume that some of the observed change can be attributed to our intrapartum screening policy.
When comparing the costs of screening tests, we found that PCR screening did not affect the overall costs of delivery which have been steadily declining over time. When we restricted the analysis to the sole costs and savings attributable to PCR screening, we found a cost increase of $49 per newborn with an estimated extra cost of $5,819 to avoid one additional case of early-onset GBS disease based upon our current yearly incidence. The current DRG tariff has so far permitted this cost to be absorbed. However, a PCR test cost of $26–33 including the cost of the GeneXpert system, would be fully balanced by the early-onset GBS disease avoided. Although we compared only short-term health care costs of both screening strategies, a decision-analytic model developed by Haberland et al18 found intrapartum PCR screening to be potentially cost saving when taking into account long-term health care costs of possible disabilities and benefits of more healthy infants.
Limitations of this study are the lack of a prospective control group and population selection. Women delivering at Saint-Joseph hospital are mostly well informed and well monitored during their pregnancy. A randomized-controlled multicenter study is probably needed to evaluate the cost-effectiveness of this prevention strategy and demonstrate a better efficacy in populations where poorly followed women are of unknown GBS status at presentation for delivery. In term newborns, however, using infection rate as an endpoint is problematic given the sample size needed.
Intrapartum GBS PCR screening in our large cohort of mothers was associated with a significant decrease of both proven early-onset GBS disease and antibiotic use in newborns. The additional PCR costs were offset in part by the reduction in early-onset GBS disease treatment costs. The possibility of intrapartum PCR screening being cost saving is limited by the current incidence of early-onset GBS disease in developed countries, which is always below 1%, and by the current price of the test.
1. Baker CJ. The spectrum of perinatal group B streptococcal disease. Vaccine 2013;31(Suppl 4):D3–6.
2. Boyer KM, Gotoff SP. Prevention of early-onset neonatal disease with selective intrapartum chemoprophylaxis. N Engl J Med 1986;314:1665–9.
3. Verani JR, McGee L, Schrag SJ. Prevention of perinatal group B streptococcal disease—revised guidelines from CDC. MMWR Recomm Rep 2010;59:1–36.
4. Agence Nationale d'Accréditation et d'Evaluation en Santé. Antenatal prevention of the risk of early neonatal bacterial infection/clinical practice guidelines. Available at: https://www.has-sante.fr/portail/upload/docs/application/pdf/prevention_antenatale_du_risque_infectieux_bacterien_-_rec.pdf
. Retrieved August 28, 2018.
5. Goodman JR, Berg RL, Gribble RK, Meier PR, Fee SC, Mitchell PD. Longitudinal study of group B streptococcus carriage in pregnancy. Infect Dis Obstet Gynecol 1997;5:237–43.
6. Puopolo KM, Madoff LC, Eichenwald EC. Early-onset group B streptococcal disease in the era of maternal screening. Pediatrics 2005;115:1240–46.
7. Van Dyke MK, Phares CR, Lynfield R, Thomas AR, Craig AS, Mohle-Boetani J, et al. Evaluation of universal antenatal screening for group B streptococcus. N Engl J Med 2009;360:2626–36.
8. Pulver LS, Hopfenbeck MM, Young PC, Stoddard GJ, Korgenski K, Daly J, et al. Continued early onset group B streptococcal infections in the era of intrapartum prophylaxis. J Perinatol 2009;29:20–5.
9. Lin FY, Weisman LE, Azimi P, Young AE, Chang K, Cielo M, et al. Assessment of intrapartum antibiotic prophylaxis for the prevention of early-onset group B Streptococcal disease. Pediatr Infect Dis J 2011;30:759–63.
10. El Helali N, Nguyen JC, Ly A, Giovangrandi Y, Trinquart L. Diagnostic accuracy of a rapid real-time polymerase chain reaction assay for universal intrapartum group B streptococcus screening. Clin Infect Dis 2009;49:417–23.
11. El Helali N, Giovangrandi Y, Guyot K, Chevet K, Gutmann L, Durand-Zaleski I. Cost and effectiveness of intrapartum group B streptococcus polymerase chain reaction screening for term deliveries. Obstet Gynecol 2012;119:822–29.
12. Di Renzo GC, Melin P, Berardi A, Blennow M, Carbonell-Estrany X, Donzelli GP, et al. Intrapartum GBS screening and antibiotic prophylaxis: a European consensus conference. J Matern Fetal Neonatal Med 2015;28:766–82.
13. Diagnostic et traitement curatif de l'infection bactérienne précoce du nouveau-né: recommandations pour la pratique Clinique. Available at: https://www.has-sante.fr/portail/upload/docs/application/pdf/recos-_inn-_mel_2006.pdf
. Retrieved August 28, 2018.
14. Agence Technique de l'Information sur l'Hospitalisation. Echelle nationale de coûts par GHM (en euros), Référentiel national de coûts. Available at: www.atih.sante.fr
. Retrieved May 22, 2018.
15. Kuhn P, Dheu C, Bolender C, Chognot D, Keller L, Demil H, et al. Incidence and distribution of pathogens in early-onset neonatal sepsis in the era of antenatal antibiotics. Paediatr Perinat Epidemiol 2010;24:479–87.
16. Singh T, Barnes EH, Isaacs D; Australian Study Group for Neonatal Infections. Early-onset neonatal infections in Australia and New Zealand, 2002-2012. Arch Dis Child Fetal Neonatal Ed 2018 Mar 27. pii: fetalneonatal-2017-314671.
17. Russell NJ, Seale AC, O'Sullivan C, Le Doare K, Heath PT, Lawn JE, et al. Risk of early-onset neonatal group B streptococcal disease with maternal colonization worldwide: systematic review and meta-analyses. Clin Infect Dis 2017;65(Suppl 2):S152–9.
18. Haberland CA, Benitz WE, Sanders GD, Pietzsch JB, Yamada S, Nguyen L, et al. Perinatal screening for group B streptococci: cost-benefit analysis of rapid polymerase chain reaction. Pediatrics 2002;110:471–80.