Materials and Methods

This was a double-blind, randomized, multicenter, placebo-controlled study of the delayed-release combination of doxylamine succinate (10 mg) and pyridoxine hydrochloride (10 mg) (Diclectin) in the treatment of NVP. Women were recruited in 2008-2009 in 3 university medical centers: The University of Texas in Galveston, TX, University of Pittsburgh, Pittsburgh, PA, and Georgetown University, Washington, DC. All subjects were pregnant, at least 18 years of age, in the gestational age range of 7-14 weeks, had a NVP, had a pregnancy unique quantification of emesis (PUQE) score ≥ 6 ( Table 1 ), and had not responded to conservative management consisting of dietary/lifestyle advice according to the 2004 American College of Obstetrics and Gynecology (ACOG) practice bulletin. Women treated with other antiemetics, having chronic medical conditions, or those who could not communicate in either English or Spanish, were excluded. After normal physical examination and laboratory tests (hemoglobin and blood count, liver function tests, electrolytes, amylase), and after confirming in utero singleton pregnancies by ultrasound, women were randomly assigned to receive Diclectin or a similar-appearing placebo. Interactive voice response system (IVRS) provided study sites the ability to randomly assign a subject, report a study clinic visit, early termination of a subject, report receipt of drug, manage drug supply, replace lost or damaged drug, report an unscheduled visit, report study completion, report a compassionate visit, and report compassionate use subject discontinuation.

In recent studies on the effect of 500 mg ginger or 10 mg vitamin B6 on “nausea score” and on number of vomiting episodes, a sample size of 64 per group showed significant differences at power of 90% and alpha of .001. Therefore, for this study, 280 patients (140 patients per treatment group) were enrolled to achieve 200 evaluable patients.

Two tablets of study drug (Diclectin or placebo) were administered at bedtime on day 1. If symptoms of nausea and vomiting persisted into the afternoon hours of day 2 (ie, PUQE score >3), the subject was directed to take her usual dose of 2 tablets at bedtime and an additional tablet the next morning on day 3. Based on assessment in the clinic on day 4 (± 1 day), the subject might have been directed to take an additional (fourth) tablet in the midafternoon to control evening symptoms. Therefore, the minimum assigned dose of study medication was 2 tablets daily at bedtime, increasing when indicated to the maximal dosage of 4 tablets per day according to the timing, duration, severity, and frequency of the symptoms experienced by the subject. This was a 15-day study, which included 14 days on which study drug was administered. Telephone contact was made on days 2, 6, 12, and 14 to assess subject diary information, adverse events (AEs), concomitant medications, and compliance with the study medication. Patients returned to the clinic in the morning prior to their morning dose on day 4 (± 1 day), day 8 (± 1 day), and on day 15 (± 1 day; end of study visit) to collect diary report and complete all study-related data.

Subjects completed the PUQE score and the study diary once daily every morning before study dose at approximately the same time each day. Subjects completed the global assessment of well being scale of the PUQE on days 1, 8, and 14 at the same time that the PUQE score was completed.

AEs and concomitant medications were recorded at all visits and follow-up phone calls. A follow-up phone call was conducted 30 days after last dosing to capture any serious AEs for patients completing the treatment period or early termination. At the end of the 2-week trial, patients were offered compassionate use of the product (Diclectin or placebo), which they had received. If compassionate use of medication was commenced after day 15, drug accountability and dispensing was conducted at clinic visits and AEs were reported during this compassionate use period. All AEs and concomitant medications that were reported during the first 30 days of compassionate use treatment were recorded on the case report form. After the first 30 days of compassionate use treatment only serious AEs were recorded.

The primary effectiveness endpoint consisted of change from baseline in the 2 domains of the PUQE score. The symptom domain of the PUQE score incorporates the number of daily vomiting episodes, number of daily retching, and length of daily nausea in hours, for an overall score of symptoms rated from 3 (no symptoms) to 15 (most severe) ( Table 1 ). The quality of life domain of the PUQE score incorporates patients’ report of their present well-being from zero (worst possible) to 10 (best possible). Secondary effectiveness criteria included the day-by-day area under the curve for change in PUQE from baseline, time loss from employment, and the number of women in each arm who continued with (blinded) compassionate use of her medication (Diclectin or placebo). We also recorded the number of patients in each group who reported concurrent use of alternate therapy for NVP. These therapies included management strategies such as nutritional modifications, teas, aroma therapy, massage, and yoga.

The frequency and severity of all adverse effects was tabulated by treatment group, system organ class, preferred term severity, and relationship to study drug. In addition, laboratory tests were conducted on days 1 and 15 (± 1 day).

Intent to treat (ITT) group differences for continuous demographic variables at baseline were examined by analysis of variance (ANOVA) for unpaired results or by Mann-Whitney U test (based on the normality of data distribution). Categorical variables such as race and ethnicity were examined by Pearson’s χ ² test or Fisher’s exact test. Analysis included all randomized patients who had received at least 1 dose of study medication.

Baseline variables (eg, ethnic origin, race, maternal age, gestational age at start of symptoms, gestational age at enrollment, and PUQE score at enrollment) were evaluated to identify potential differences between treatment groups that were not eliminated by randomization. If there were any significant baseline differences, post hoc analyses would have been conducted to discern any effect of such differences between the treatment groups on the primary and secondary endpoints.

PUQE scores, based on the complete data, were evaluated using an ANCOVA model, with change from baseline to day 15 (± 1 day) as the response variable, baseline PUQE score was the covariate, and the treatment group and study center were the fixed effects.

Secondary analyses were used to compare the 2 treatment groups using ANCOVA where change from baseline to day 15 (± 1 day) was the response variable, the baseline value was the covariate, and the treatment group and study center were the fixed effects. Compliance with medication and time loss from employment were analyzed using an ANOVA model where period total was the response variable and the treatment and study center were the fixed effects.

AEs experienced by the subjects which occurred on or after day 1 (after the patient signed the informed consent form) through day 15, were compared between groups using Pearson’s χ ² test or Fisher’s exact test if more appropriate.