Effect of preoperative low-normal cervical length on perinatal outcome after laparotomy-assisted fetoscopic spina bifida repair.

OBJECTIVE: To determine if preoperative cervical length in the low-normal range increases the risk of adverse perinatal outcome in patients undergoing fetoscopic spina bifida repair. METHODS: This was a retrospective cohort study of patients who underwent fetal spina bifida repair between September 2014 and May 2022 at a single center. Cervical length was measured on transvaginal ultrasound during the week before surgery. Eligibility for laparotomy-assisted fetoscopic spina bifida repair was as per the criteria of the Management of Myelomeningocele Study, although maternal body mass index (BMI) up to 40 kg/m2 was allowed. Laparotomy-assisted fetoscopic spina bifida repair was performed, with carbon dioxide insufflation via two 12-French ports in the exteriorized uterus. All patients received the same peri- and postoperative tocolysis regimen, including magnesium sulfate, nifedipine and indomethacin. Postoperative follow-up ultrasound scans were performed either weekly (< 32 weeks' gestation) or twice a week (≥ 32 weeks). Perinatal outcome was compared between patients with a preoperative cervical length of 25-30 mm vs those with a cervical length > 30 mm. Logistic regression analyses and generalized linear mixed regression analyses were used to predict delivery at less than 30, 34 and 37 weeks' gestation. RESULTS: The study included 99 patients with a preoperative cervical length > 30 mm and 12 patients with a cervix 25-30 mm in length. One further case which underwent spina bifida repair was excluded because cervical length was measured > 1 week before surgery. No differences in maternal demographics, gestational age (GA) at surgery, duration of surgery or duration of carbon dioxide uterine insufflation were observed between groups. Cases with low-normal cervical length had an earlier GA at delivery (median (range), 35.2 (25.1-39.7) weeks vs 38.2 (26.0-40.9) weeks; P = 0.01), higher rates of delivery at < 34 weeks (41.7% vs 10.2%; P = 0.01) and < 30 weeks (25.0% vs 1.0%; P < 0.01) and a higher rate of preterm prelabor rupture of membranes (PPROM) (58.3% vs 26.3%; P = 0.04) at an earlier GA (mean ± SD, 29.3 ± 4.0 weeks vs 33.0 ± 2.4 weeks; P = 0.05) compared to those with a normal cervical length. Neonates of cases with low-normal cervical length had a longer stay in the neonatal intensive care unit (20 (7-162) days vs 9 (3-253) days; P = 0.02) and higher rates of respiratory distress syndrome (50.0% vs 14.4%; P < 0.01), sepsis (16.7% vs 1.0%; P = 0.03), necrotizing enterocolitis (16.7% vs 0%; P = 0.01) and retinopathy (33.3% vs 1.0%; P < 0.01). There was an association between preoperative cervical length and risk of delivery at < 30 weeks which was significant only for patients with a maternal BMI < 25 kg/m2 (odds ratio, 0.37 (95% CI, 0.07-0.81); P = 0.02). CONCLUSIONS: Low-normal cervical length (25-30 mm) as measured before in-utero laparotomy-assisted fetoscopic spina bifida repair may increase the risk of adverse perinatal outcomes, including PPROM and preterm birth, leading to higher rates of neonatal complications. These data warrant further research and are of critical relevance for clinical teams considering the eligibility of patients for in-utero spina bifida repair. Based on this evidence, patients with a low-normal cervical length should be aware of their increased risk for adverse perinatal outcome. © 2022 International Society of Ultrasound in Obstetrics and Gynecology.

Assessing the performance of hospital support services.

A simple to operate, nontechnical quality assurance program was developed to evaluate the support services provided at this Veterans Administration hospital. The program has received unanimous support from the entire hospital staff and promises to be a highly effective, and practical, program.

A Monte Carlo computer model to investigate patient scheduling.

A Monte Carlo computer model was developed to investigate various types of patient appointment schedules for a single channel queue. Input parameters included the incidence of no-shows, the rate of unscheduled walk-ins, the frequency distribution of physician examination times, and the string of appointment times. The results included the frequency distributions of physician utilization rates and the total waiting time of all patients. Five hospital clinics were simulated. Even increment schedules yielded the best trade-off between physician utilization and patient waiting.