Quasi Experiment of Prenatal Probiotics Against Group B Streptococcus Colonization

Methods

A non-blinded, open-label, quasi-experimental design was approved by the Institutional Review Boards of three institutions (two universities and the clinical research site). The study setting was a large Certified Nurse-Midwife (subsequently referred to as midwife) clinical practice serving a culturally and economically diverse urban population in the Midwest region of the USA.

Participants Women self-selected to participate in screening and informed consent procedures done by the first two authors. The following inclusion criteria were used at study enrollment: low risk (no obstetric, fetal, medical or genetic risk factors), adult (≥18 years of age), pregnant at 28 ± 2 weeks gestation, able to speak and write English, and expressing willingness to participate in the study intervention (oral probiotic) and data collection (vaginal and rectal swabs, questionnaires).

A convenience sample of 20 healthy pregnant participants was sought. Following informed consent, the first 10 subjects were assigned to the experimental group. Nonrandom assignment was used to assure that the 10 experimental group participants were enrolled while remaining within funding, and time constraints. After enrolling the experimental group, the next ten women were assigned to the control group.

Intervention

The study intervention consisted of one capsule of Florajen3 (previously described) orally each day. Florajen3 meets the internationally established probiotic criteria (FAO/WHO, 2001). The probiotics contained in Florajen3 are non-spore-forming, lactose- and hydrogen-peroxide-producing bacteria. The manufacturer recommends refrigeration to maintain maximum potency. As part of the preparation for this study, a sample of Florajen3 was left unrefrigerated for a period of 6 weeks and appropriate colony counts were sustained. Therefore for the purpose of the study, the participants in the experimental group were allowed to leave the probiotic unrefrigerated in an effort to improve daily compliance. Experimental group study participants were made aware of the probiotic brand used (open label) as the intervention.

Further, the current state of knowledge about prenatal probiotics, including potentially rare side effects, were discussed at length using a 10-page informed consent document. The Florajen3 was placed in a study bottle equipped with an electronic cap monitoring system (MEMS ®AARDEX), designed to record each time the bottle was opened by the participant. Each opening of the cap bottle equipped with this system is recorded on a microchip. The number of openings are retrievable via computer software when the bottle caps are returned. Researchers, midwives, and participants were aware of group assignment. The control group participants did not receive a placebo.

Procedures

Midwife prenatal care providers and staff were oriented to the study data collection procedures. Study data were collected at three points during routine prenatal visits with each participant's midwife (28±2 weeks, 32±2 weeks, and 36±2 weeks gestation). The study was considered completed at the 36±2 weeks gestation visit. All participants received compensation at the completion of 2 study visits; 25 USD at 32±2 weeks and 75 USD at 36±2 weeks gestation. Probiotic group participants were offered a supply of probiotics for the remainder of pregnancy.

The midwife prenatal provider collected demographic information (birth date, race, and gestational age), and the vaginal and rectal swabs for GBS and wet mount, using accepted criteria to identify the vaginal pH and the presence of normal flora and/or pathogens. The wet mounts were examined microscopically by each midwife using standardized wet mount diagnostic procedures with normal saline and potassium hydroxide and documented in a standard format. Participants completed a brief questionnaire at each study visit that asked them to describe yogurt ingestion, sexual activity (frequency and type), and vaginal cleansing practices (if used) in the past week; these were considered potential confounding variables. The participants were also asked if these practices were typical for them, and if not, how they were atypical. Women in the experimental group were asked to report any side effects that they attributed to the probiotics by notifying their midwife immediately, if necessary, and then also noting the information on the study questionnaire. As part of routine prenatal care, the CDC-required GBS vaginal to rectal swab was collected by each participant's midwife and sent to the hospital laboratory between the 35 and 37 week visit. This was the only result that determined the need for IAP.

Researchers oversaw data collection and sample packaging. Participants were asked to return their probiotic bottles at each study visit and the Florajen3 capsules were replaced with a fresh supply. This gave the researchers the opportunity to notify the laboratory to record the date for later evaluation of the electronic cap monitoring results and conduct pill counts as an additional means to monitor women's responses to study participation, as well as probiotic adherence at each study visit. The vaginal and rectal swabs were placed on ice, packaged according to accepted procedures for human specimen handling, and shipped overnight to the laboratory of the final author, located 80 miles from the study site, for processing and analysis.

Laboratory Analysis Upon receipt at the laboratory, the vaginal and rectal swabs were processed for quantitative colony counts of GBS. Polymerase chain reaction (PCR) was performed to determine whether bacteria isolated were the same as those in Florajen3. If positive for PCR, pulsed field gel electrophoresis (PFGE) would be performed to confirm the strain identity.

These results were for research purposes only and were not available to the practicing midwives.

The routine CDC-recommended prenatal GBS vaginal to rectal swabs were analyzed at the hospital laboratory, using a culture based method in which a threshold for GBS positive results is considered greater than 102 colony forming units (CFU) per swab. These results determined the need for IAP and were available to the midwives for management during labor.

Statistical Analysis

Demographics, the study variables, and laboratory data were entered into and analyzed with a statistical software package. Descriptive statistics, Confidence Intervals, Chi-squares, T-tests, and correlations were used for analysis. Data were compared between subjects at each study visit and between groups at each data collection point. The qualitative prenatal GBS culture results (positive or negative) from the hospital laboratory were compared to the quantitative vaginal and rectal GBS study results, adherence data, and confounding variables (yogurt ingestion, sexual activity, and vaginal cleansing practices).