Improving Shoulder Dystocia Management and Outcomes with a Targeted Quality Assurance Program

Abstract

Background Several investigators have achieved remarkable success in transferring shoulder dystocia management skills mastered with simulation training to clinical practice. However, other investigators have not demonstrated similar benefits, raising questions about the comparative effectiveness of specific simulation schemes, instructional content, and additional quality assurance measures between successful and unsuccessful interventions. After our initial review revealed gaps in following shoulder dystocia management algorithms, documentation and timely follow-up of injured neonates, we developed and implemented five interventions, three educational and two systems-level, aimed at improving shoulder dystocia management. Objective To describe the clinical impact of a systematic program of quality improvement on outcomes of vaginal births complicated by shoulder dystocia. Setting An urban tertiary academic medical center that trains 36 obstetrics/gynecology residents (9 per year) and provides comprehensive obstetrical services for approximately 2,000 deliveries annually. Study Design We use SQUIRE 2.0 (Standards for Quality Improvement Reporting Excellence) to (1) describe our core instructional content and simulation-based practice, emphasizing specific proscriptive and prescriptive recommendations and their evidence basis, and (2) to report an interrupted time series assessment of the clinical impact of our systematic quality improvement program targeting shoulder dystocia-associated brachial plexus injury. Results Compared with baseline (June 1993 to December 2004), the incidence of shoulder dystocia among vaginally delivered infants with birth weight ≥ 2,500 g at Johns Hopkins Hospital (January 2014 to December 2015) increased from 2.6 to 4.6% (X 2 = 29.8; df = 1; p < 0.0001); in addition, documentation improved, direct fetal manipulation increased, while use of episiotomy for the management of shoulder dystocia decreased. While preintervention only 65% of brachial plexus injury were associated with shoulder dystocia, 100% of neonatal brachial plexus injuries were associated with shoulder dystocia postintervention (80/122 [65%] vs. 7/7 [100%], X 2 = 3.66; df = 1; p = 0.055), a trend reflecting simultaneous increased recognition of impacted shoulders and improved overall management of shoulder dystocia. Most importantly, the incidence of brachial plexus injury among shoulder-dystocia-complicated vaginal deliveries has decreased from a baseline of 31.6 to 6.3% (X 2 = 27.9; df = 1; p < 0.0001), and the absolute brachial plexus injury rate declined from 8.2 to 2.9 per 1,000 vaginal births ≥ 2,500 g, a reduction of 64.5% (X 2 = 7.3; df = 1; p = 0.007). Conclusion A systematic program of quality assurance with specific proscriptive and prescriptive instructional content and management recommendations is associated with improved recognition, management, and clinical outcomes of shoulder dystocia.