Premium
Trends in lobectomy/amygdalohippocampectomy over time and the impact of hospital surgical volume on hospitalization outcomes: A population‐based study
Author(s) -
Kwon ChurlSu,
Blank Leah,
Mu Lan,
Jetté Nathalie
Publication year - 2020
Publication title -
epilepsia
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 2.687
H-Index - 191
eISSN - 1528-1167
pISSN - 0013-9580
DOI - 10.1111/epi.16664
Subject(s) - medicine , epilepsy surgery , confidence interval , odds ratio , epilepsy , population , logistic regression , surgery , anesthesia , environmental health , psychiatry
Abstract Objective Despite national guidelines supporting surgical referral in drug‐resistant epilepsy, it is hypothesized that surgery is underutilized. We investigated the volumes of lobectomy/amygdalohippocampectomy surgeries over time and examined differences in outcomes between (1) high‐volume (HV), middle‐volume (MV), and low‐volume (LV) hospitals and (2) Level 4 Centers versus non–Level 4 Centers. Methods The 2003‐2014 National Inpatient Sample (the largest all‐payer hospitalization database, representative of the US population) was utilized. Epilepsy was identified using a previously validated International Classification of Diseases, Ninth Revision, Clinical Modification (ICD‐9‐CM) case definition and surgeries using ICD‐9‐CM procedure codes. A hospital was considered a Level 4 Center if it performed intracranial electroencephalographic (EEG) monitoring. Tumor surgeries were excluded. Linear regression was used to perform trend tests. Weighted multivariate logistic regression was used to summarize association of surgery with outcomes. Results A total of 4,487 lobectomy/amygdalohippocampectomy surgeries were performed in children and adults with epilepsy. Lobectomy/amygdalohippocampectomy surgeries significantly decreased over time (slope: −0.24, P < .001). This declining surgical trend was greater for all resective/disconnective surgery (slope: −0.45, P < .001), and greatest when compared to all types of epilepsy surgery, for example, resection/disconnection/radiosurgery/laser interstitial thermal therapy/vagus nerve stimulation/deep brain stimulation/responsive neurostimulation/intracranial EEG (slope: −0.95, P < .001). LV compared to HV hospitals had higher odds of transfer to other facilities (13.60% vs 4.24%, odds ratio [OR] = 2.76, 95% confidence interval [CI] = 1.11‐6.82). LV hospitals had higher odds of surgical complications versus MV (12.69% vs 6.80%, OR = 2.20, 95% CI = 1.01‐5.09). HV hospitals incurred the least total charges. There were no differences in discharge status, adverse events, length of stay, or cost between Level 4 Centers versus non–Level 4 Centers. Significance Lobectomies/amygdalohippocampectomies are decreasing over time, suggesting ongoing underutilization. LV centers are associated with greater complication and transfer rates. Future studies are required to understand the reason for worse outcomes in LV centers and to determine whether a minimum number of surgeries must be performed to meet necessary standards.