Real-World Outcomes of Glaucoma Filtration Surgery Using Electronic Health Records: An Informatics Study

Precis: Utilizing an automated pipeline for data extraction from electronic health records provides real-world information on the success of various glaucoma procedures, with tube shunt implantation associated with increased failure rates compared with trabeculectomy. Background: We aimed to evaluate the long-term survival of glaucoma surgeries using an automated pipeline for extraction of outcomes from electronic health records. Methods: A retrospective observational study from a single academic center. Patients undergoing trabeculectomy, Ex-PRESS shunt, Baerveldt, and Ahmed tube shunt insertion from 2009 to 2018 were identified from electronic health record procedure codes. Patient characteristics were identified from structured and unstructured fields using a previously validated natural language processing pipeline. Results: Five hundred twelve patients underwent 711 glaucoma surgeries: 287 trabeculectomies, 47 Ex-PRESS shunts, 274 Baerveldt and 103 Ahmed tube implantations. The Median follow-up was 359 days. The mean baseline IOP was 24.4 mm Hg (SD 10.9), and 73.1% were on >= 3 medications. Compared with trabeculectomy, tube shunt surgery had a higher risk of failure (Baerveldt: Hazard Ratio (HR) 1.44, 95% CI 1.02 to 2.02; Ahmed: HR 2.01, 95% CI 1.28 to 3.17). Previous glaucoma surgery was associated with increased failure (>= 2 previous surgeries: HR 2.74, 95% CI 1.62 to 4.64), as were fewer baseline medications (<3 medications: HR 2.96, 95% CI 2.12 to 4.13) and male sex (HR 1.40, 95% CI 1.03 to 1.90). At 1 year, tube shunt patients had a 2.53 mm Hg (P=0.002) higher IOP compared with trabeculectomy patients. Conclusions: Baerveldt and Ahmed tube shunt implantation was associated with increased failure compared with trabeculectomy. Fewer baseline medications, previous glaucoma surgeries, and male sex were also risk factors for failure. These results demonstrate the utility of applying an informatics pipeline to electronic health records to investigate key clinical questions using real-world evidence.