The Impact of Obesity on Risk Factors for Adverse Outcomes in Patients Undergoing Elective Posterior Lumbar Spine Fusion

Deeptee Jain, MD; Wesley Durand, BS; Jeremy D. Shaw, MD; Shane Burch, MD; Vedat Deviren, MD; Sigurd Berven, MD


Spine. 2021;46(7):457-463. 

In This Article


The main goal of this study was to determine whether risk factors for postoperative outcomes have different effects in obese vs nonobese patients. Simple multivariate regression as most frequently performed in orthopedic literature assumes a linearity between variables, that is, that the impact of independent variables is the same across various patient groups. In this study, we attempted to examine the impact of independent variables in obese versus nonobese patients by studying the interaction variable.

Not surprisingly, we confirmed previous studies' findings that obese patients have higher rates of medical complications as compared to nonobese patients after spine surgery.[3,4,16–18] Interestingly, in general, we demonstrated that many comorbidities like cerebrovascular disease, history of MI, CHF, chronic pulmonary disease, and renal disease, while still absolutely more than one, had lower odds ratios for increasing risk for medical complications in obese patients versus nonobese patients. This finding was contrary to our hypothesis. It is known that obese patients have higher rates of medical comorbidities; thus, preoperative optimization is especially crucial in this population.[19] That being said, it may be interpreted that obesity itself is the dominating independent risk factor, as opposed to the other comorbidities that come along with obesity. Although medical optimization in conjunction with multidisciplinary evaluation has been shown to reduce complication rates after spine surgery, in obese patients, perhaps weight loss should be a primary target to mitigate this risk.[20]

Ninety-day readmission rates in obese patients undergoing elective posterior LSF in this study were high, 15.6%, and significantly higher than nonobese patients at 12.2%. These rates fit between those previously reported in the general population, with 30-day readmission rates of 5.1% and 90-day readmission rates of 24.8% in other studies.[11,21] This finding is consistent with previous findings that BMI is a risk factor for readmission,[22,23] which follows given the known increased risk of complications in this patient population.[17,18] Similarly as with medical complications, we surprisingly found that the OR of history of MI and renal disease was significantly lower in obese patients than in nonobese patients, again emphasizing the importance of obesity as very important independent risk factor.

Interestingly, we found that TLIF/PLIF was protective against 90-day revision in the nonobese patient, but not in the obese patient. This may suggest that obese patients are more prone to surgical complications in undergoing TLIF/PLIF in the short term. One explanation could include technical difficulties leading to hardware malpositioning, which may occur due to limited direct visualization or fluoroscopic visualization intraoperatively.[24]

We found that Medicare/Medicaid insurance was significantly associated with higher major complication, readmission, reoperation, and infection rates but not different between obese and nonobese patients, consistent with previous studies.[11,25] Interestingly, it is not surprising that impact of having Medicare/Medicaid insurance among obese patients is similar compared to the general population; according to the Center for Disease Control, obesity and socioeconomic status are not strongly associated.[26] Medicare/Medicaid status likely reflects poorer economic backgrounds; the exact reasons for worse outcomes are unclear, but may include limited resources and education, language and cultural barriers, all preventing good follow-up care.[27,32] The impact of insurance on readmission rates highlights the need to treat spine fusion patients with a team based approach; although these may not be modifiable risk factors, there should be members dedicated to addressing socioeconomic issues and access to postoperative care before surgery.[20]

Limitations of this study include those often suffered by administrative claims databases; most notably they are subject to underreporting or errors in coding.[28] One such is example in this study is that patient selection of obese was based solely on coding, as opposed to BMI, which may contribute to a selection bias. Although it has been reported that 80% of patients undergoing spine surgery are either overweight or obese, in the present study only about 12% of patients were obese. Databases using administrative claims are notorious for undercoding obesity. Furthermore, it has been demonstrated that obese patients are more likely to be coded for complications, further skewing the results.[29] Second, we defined reoperation rates as any return to the operating room for the lumbar spine, but this does not explore the surgical indications for reoperation, which may include hardware failure, neurologic symptoms, infection, but may also include a planned staged second surgery. Finally, SID only captures inpatient visits, and thus any complications treated in the outpatient setting were not included in the present study.

In conclusion, in this retrospective case–control study using an administrative claims database, we demonstrated that many medical comorbidities have less impact in obese patients than nonobese patients in predicting adverse outcomes despite increased rates of adverse outcomes in obese patients. These findings overall reflect the importance of obesity as an independent risk factor. Despite the prevalence of obesity in our society and specifically in the spine surgery population, this is an understudied group, and these results have important implications for future studies targeting preoperative factors to improve outcome.