Journal Club: The Impact of Body Mass Index on Hospital Stay and Complications After Spinal Fusion SIGNIFICANCE/CONTEXT AND IMPORTANCE OF THE STUDY M cClendon et al 1 investigate a widely acknowledged and increasingly preva- lent health concern, obesity, in a novel context: its effects on the postoperative course of patients undergoing extensive (5 or more levels) spine surgery for deformity. The effect of obesity on surgical morbidity has garnered much interest and is the subject of several recently published articles. 2-4 Many neurosurgeons consider that obese patients carry a higher surgical risk by virtue of their comorbid conditions, their body habitus and its effect on the technical aspects of surgery, and the effect of obesity on wound healing. The ability to quantify surgical risk and complications according to body mass index (BMI) would facilitate preoperative risk discussions with pa- tients, help with resource allocation, and could potentially alter the timing of elective procedures— all critical considerations in today’s health care universe. ORIGINALITY OF THE WORK There is an extensive body of literature looking at obesity and its association with spinal surgery complications and outcomes. This article extends these investigations by focusing on a unique subset of spine patients, namely those undergoing fusion of 5 or more levels for spinal deformity. To our knowledge, this is the first article to look at the effect of body mass index (BMI) on elective spinal surgeries with 5 or more levels of fusion for deformity. APPROPRIATENESS OF THE STUDY DESIGN OR EXPERIMENTAL APPROACH The authors ask a prognostic question (the association of increased BMI with poor outcomes from 5 or greater levels elective spinal surgery for deformity) by using a retrospective cohort study design. Data were generated by review of inpa- tient and outpatient charts from appropriate patients seen at the authors’ institution between 2007 and 2010. Unfortunately, several aspects of this study design potentially weaken the reli- ability of the authors’ conclusions. First, the authors did not identify a priori primary or secondary outcome measures or adhere to a consistent definition of their independent variable (which is, variously, BMI category, “ideal vs not ideal,” and “obese vs not obese”). As a result, there is a substantial risk of detecting an association when one really does not exist (ie, a type I error). For example, Table 2 shows that, for each BMI category, 22 different outcomes were assessed. Within each BMI category, therefore, the risk of finding at least 1 spurious association (at the .05 level of statistical significance) is 67.6% (1 2 [1 2 0.05] 22 ). Taking all 5 BMI categories together, the risk approaches 100%. Table 3, which presents even more individual comparisons, conveys an even higher risk of type I error. Second, patients in the various BMI groups differed with respect to a number of prognosti- cally important risk factors. Some of these (for example, smoking) seem not to have been controlled for in the authors’ multivariate anal- ysis, and others were excluded from the analysis because of the authors’ relatively stringent threshold for inclusion (P # .05 in univariate analysis). A less stringent threshold (0.1 or 0.2 is often used in studies of this type) might have been helpful. In addition, some known risk factors in spinal surgery such as diabetes mellitus, 5 which the authors mentioned in their opening paragraph, were not included in their analysis. Finally, the abundance of unequally distributed potential known confounders raises the concern that additional important but unknown confounders are also asymmetrically present within the different BMI categories. This problem cannot be addressed by multivariate analysis or any other statistical technique. Russell Payne, MD Einar Bogason, MD Brian Anderson, MD Nicholas Brandmeir, MD Ephraim Church, MD Jonathon Cooke, MD* Gareth Davies, MD Namath Hussain, MD Akshal Patel, MD Pratik Rohatgi, MD Emily Sieg, MD Omar Zalatimo, MD Endrit Ziu, MD, PhD Justin Davanzo, MD Department of Neurosurgery, Penn State Hershey Medical Center, Hershey, Pennsylvania *This author is a military service member. This work was prepared as part of his official duties. Title 17, USC, §105 provides that, “Copyright protection under this title is not available for any work of the US Government.” Title 17, USC, §101 defines a US Government work as a work prepared by a military service member or employee of the US Government as part of that person’s official duties. The views expressed in this presentation are those of the author and do not necessarily reflect the official policy or position of the Department of the Navy, Department of Defense, or the US government. Correspondence: Russell Payne, MD, Department of Neurosurgery, Penn State Hershey Medical Center, 500 University Dr, Hershey, PA 17033. E-mail: rpayne@hmc.psu.edu Copyright © 2014 by the Congress of Neurological Surgeons. JOURNAL CLUB JOURNAL CLUB NEUROSURGERY VOLUME 75 | NUMBER 5 | NOVEMBER 2014 | 599 Copyright © Congress of Neurological Surgeons. Unauthorized reproduction of this article is prohibited Downloaded from https://academic.oup.com/neurosurgery/article-abstract/75/5/599/2447752 by guest on 27 May 2020