1232. Systematic Review of ICD-9-CM Procedure Codes for Craniotomy/Craniectomy in Nationwide Inpatient Sample Stroke Hospitalizations
Authors: Hannah Kristen Weiss; Roxanna Garcia; Pouya Nazari, MD; Pedram Golnari, MD; Shyam Prabhakaran; Andrew Naidech; Babak Jahromi (Chicago, IL)
Introduction: Large administrative datasets, such as the Nationwide Inpatient Sample (NIS), can provide substantial longitudinal, geographical, and novel clinical information; yet heterogeneity exists in analyzing this data by corresponding International Classification of Diseases, Version 9 Clinical Modification (ICD-9 CM) codes. We assessed the utilization of ICD-9 CM procedure codes for craniotomy/craniectomy in studies using NIS data for stroke care. Methods: We performed a systematic review of PubMed for all publications that incorporated the use of the NIS to investigate craniotomy and/or craniectomy in acute ischemic stroke and non-traumatic hemorrhagic stroke hospitalizations. Eligible articles were reviewed and ICD-9 procedure codes for craniotomy/craniectomy were extracted from the included studies. Results: A total of 13 articles met final eligibility criteria. A total of 9 studies specified the the ICD-9 procedure code used for craniotomy/craniectomy and 4 studies included craniotomy/craniectomy in their analysis without specifying the applied code. A total of 69.2% of studies were ischemic stroke, 15.4% were hemorrhagic stroke, and 15.4% were mixed. The studies specifying the exact ICD-9 CM procedure code referenced 10 unique procedure codes, with 66.7% (n=6) studies including codes 1.25 and 1.24. Of the studies that specified procedural codes, the majority (77.8%; n=7) utilized two codes for craniotomy/craniectomy. There did not appear to be consistency across studies in selection for ICD-9-CM procedure codes. Conclusion: The two most common procedural codes for craniotomy/craniectomy in NIS analysis publications include 1.24 (“Other Craniotomy”) and 1.25 (“Other Craniectomy”) . There is substantial heterogeneity in selection of procedure codes. Standardizing procedure codes is necessary as its application can alter research findings and outcomes reported from the NIS database.