A prospective, randomised, double-blind, controlled clinical trial (RCT) was conducted to compare the efficacy and clinical safety of the induction of labour using the combination of isosorbide dinitrate-oxytocin (experimental arm, X = 1) compared to misoprostol-oxytocin (standard arm, X = 0). The main result of the study was the success rate of foetal expulsion within 15 hours, while the average administration induction interval defined the secondary result. A total of 60 women with pregnancies greater than 20 weeks gestation were referred for foetal evacuation. We defined late intrauterine foetal death (IUFD) as babies without signs of life in the uterus after 20 complete weeks of pregnancy. The approval to conduct the study was obtained from the institutional board ethical committees of both hospitals. All participants were informed about the objectives of the study, and informed consent was required. Patients with the following characteristics were included: closed cervix without evidence of cervical dilation or baseline uterine activity, A Bishop score of \<5, having intact membranes, gestation greater than or equal to 20 weeks established by the date of menstruation or by fetometry and ultrasound-confirmed late IUFD. This study did not include the medical management of multiple pregnancies, IUFD after late foeticide or the management of specific medical conditions associated with an increase in the risk of IUFD or patients with a history of hypertension, along with women with a history of unexplained antepartum haemorrhage, pelvic dystocia or any another counter-indications where medications were used. Patients were assigned through a computational random number generator to receive isosorbide dinitrate or misoprostol. Allocation to groups was predetermined and arranged in numerical order. All participants were given a vaginal exam by the same person, who was blinded to the treatment allocation. The medications were administered in the posterior fornix, and cervical activity was evaluated at baseline and every 3 hours to monitor any change based on the Bishop score.
If the cervical conditions did not change after the treatment application, participants received a new dose, without exceeding 4 doses, to facilitate cervical ripening. Once a Bishop score of over 7 was reached, oxytocin was infused in a balanced electrolyte solution beginning with an infusion rate of 2 milli-International Units per minute (mIU/min) and doubling the dose every 15 minutes. The labour induction time from the first application of medication to the expulsion of the foetus was determined with a digital stopwatch. Each woman's vital signs were verified to determine that she was in stable condition and demonstrated haemodynamic stability as a requirement for the application of a new dose. During this time, the data and medical information were collected on paper and were later entered into a computational database. The participants remained at rest during the evaluation of adverse effects, such as headache, abdominal pain, pelvic pain, lower back pain, nausea, dizziness, and vomiting. The lack of cervical activity after 4 doses of medication was considered treatment failure.
A base solution of isosorbide dinitrate and misoprostol was prepared to obtain a concentration of 20 micrograms per milliliter (mcg/mL) in 10% lactose solution. Serial dilutions were performed using known concentrations in mobile phase buffer solution (150 g ammonium acetate and 11.5 mL glacial acetic acid to 1 L water), methanol and water in a proportion of 350:100:550, respectively. The concentration of both drugs was determined using a ultraviolet (UV) light spectrophotometer reading at lambda 220 nm and a reference filter of lambda 278 nm. The mean absorbance for each set of standards, controls and samples was calculated by a standard plot curve. Computer-based curve-fitting statistical software was employed. Peak absorption and area under the curve were taken into account to determine stability of the pharmacological integration. Known samples of isosorbide dinitrate and misoprostol were added to 100% glycerin to prepare the gel solution. The reagents had final concentrations of either 80 mg of isosorbide dinitrate in 1.5 ml of gel solution or 100 micrograms of misoprostol in the same presentation.
Both solutions were packed in syringes that had the same appearance for the purpose of keeping the physician, patient and researcher blinded. The pharmacist was the only participant who knew the contents of the syringes. Four syringes were placed in an opaque and sealed envelope consecutively numbered with a unique study number. The envelopes were opened by the physician who applied the contents of the syringe, repeating the administration every 3 hours according to the prior selection and random allocation of patients. The entire "stock" of reagents remained stable for a period of 20 days and was maintained at room temperature (18-25° C) before using.
The calculation of the sample size was based on the Cox regression model using the risk quotient (log Hazard Ratio) and a 95% confidence interval (95% CI).17 To reach a power (superiority of the clinical trial) of 80% and an α value of 0.05, with a standard deviation of 0.5 and under the assumption that 25% of all women will fail to induce labour, the estimated sample size required for the trial was 60 patients (30 in each arm).
All continuous variables were summarised using histograms and the measures of central tendency before conducting the statistical analyses. The statistical analyses were performed using the Chi-Square test, Fisher's exact test and the Mantel-Haenszel test, and Student's t test when appropriate. The Shapiro-Wilk test was used to verify the normality of the data. To model the time-event of labour induction after IUFD, Cox regression was utilised, controlling for the covariates maternal age, foetal size and weight, and weeks of gestation. The proportional hazards assumption was evaluated using the two graphed lines, which corresponded to the survival curves (absence of crossing) and the log-log plots in the figure. Respiratory and cardiac rate, temperature (measured orally) and systolic and diastolic blood pressures (recorded using sphygmomanometry) were determined at baseline and every 3 hours until the trial ended. Relative risks (RRs) and 95% CIs were estimated utilising bivariate analysis, including non-serious adverse events among the treatment groups. Attributable risk was calculated to determine the probability of total risk of each adverse effect observed with the use of isosorbide dinitrate or misoprostol. All of the tests were two-tailed, and results that had values of p \< 0.05 were considered statistically significant.
