In the United States, the obesity “epidemic” has become a serious public health concern, affecting greater numbers of citizens1 and resulting in increased morbidity2,3 and health costs.4 Affected adults have increased predisposition to developing degenerative orthopedic conditions,5 which can contribute to increased functional limitation in obese patients.6 In particular, obesity is known to be associated with an increased risk of low back pain7,8 and increasing obesity rates have been suggested to contribute to increased rates of degenerative spine pathology.9–11 As a result, more obese patients are undergoing elective spine surgery.12
Paralleling the trend toward increased spine surgical procedures in obese patients is an increased rate of spine surgery in elderly patients. Lumbar spine procedures in particular are increasingly being used in elderly patients to treat degenerative lumbar conditions. For example, lumbar fusion rates have increased from 0.3 per 1000 Medicare beneficiaries in 1992 to 1.1 per 1000 in 2003.13 The impact of advanced age on both medical and surgical complications is well studied, demonstrating that elderly patients are much more susceptible to experiencing procedural morbidity as compared to younger cohorts.14,15 However, with the recent rise in obesity, the obese elderly population represents a new sector of the spine surgery population that is poorly studied with regards to short-term spine surgery outcomes. In particular, elderly patients with morbid obesity (BMI ≥40) will be more commonly encountered in the coming years.
Given the projected increase in elderly, obese patients in the next several years, it is of utmost importance that complication rates in this patient population be analyzed for the purposes of patient education and appropriate risk stratification. The primary aim of the present study is to determine how obesity, specifically morbid obesity, modifies 90-day complication rates and 30-day readmission rates following 1- to 2-level, primary, elective lumbar spinal fusion surgery for degenerative pathology in an elderly population. A secondary goal was to determine how these conditions increase length of stay (LOS) and in-hospital costs.
Medicare data from the PearlDiver Patient Records Database (www.pearldiverinc.com; PearlDiver Inc, Fort Wayne, Indiana) was queried. This privately managed repository of the full 100% sample of Medicare data (2005 – 2012, over 30 million individual patient records) was queried for the purpose of academic research. Data was remotely accessed from a password-protected server maintained by PearlDiver. A waiver of institutional review board (IRB) approval was received for this study as all data were previously acquired and deidentified.
Patients who were 65 to 84 years of age and underwent posterior lumbar or lumbosacral spinal fusion surgery (International Classification of Diseases, Ninth Revision, Clinical Modification diagnosis and treatment (ICD-9-CM) ICD-9-CM: 81.07) were identified. This cohort of patients was then further narrowed to only include patients treated with 1- to 2-level spinal fusion surgery (ICD-9-CM: 81.62) with a diagnosis of degenerative lumbar spine condition on the same day of surgery (ICD-9-CM: 722.10, 722.51, 722.52, 722.73, 724.02, 724.03, 738.4). Exclusion criteria included presence of previous trauma, infection, or metastasis of the spine; revision fusion surgery; three-level or greater surgeries (ICD-9-CM: 81.63, 81.64); concomitant cervical, thoracic, anterior lumbar fusion, or posterior/transforaminal lumbar interbody fusion (ICD-9-CM: 81.01–81.06, 81.08); or anterior lumbar spine fusion during the 3 months before or after the index procedure, to exclude staged procedures. The query resulted in 64,813 65- to 84-year-old patients who underwent elective one- to two-level posterior lumbar fusion.
From this overall cohort, three specific subgroups were selected based on reported body mass index (BMI) or obesity diagnoses. The first group consisted of all patients with a diagnosis of morbid obesity (BMI ≥40; ICD-9 CM: V85.41-.45, 278.01) within 3 months before the index procedure (n = 2594). The second group contained all patients with a diagnosis of obesity (BMI ≥30; ICD-9 CM: V85.30-.45, 278.00, 278.01) within 3 months before the index procedure, who were not included in the first cohort (n = 5534), creating a cohort of patients with BMI 30 to 39.9, also referred to as the “non-morbid obesity” cohort. The final group, which served as a cohort of control patients, consisted of all patients who did not have a coded diagnosis of obesity at any point during their full set of records (n = 48,210).
Patients in each of the three cohorts were queried for demographics (sex, age) and comorbid diagnoses: diabetes mellitus, congestive heart failure, chronic pulmonary disease, chronic kidney disease, and smoking history.16 This information was utilized to match each of the two study cohorts to individualized control groups. Matching was performed in a strict one-to-one fashion, wherein for every patient in one of the study cohorts, one patient from the full control cohort was selected for inclusion in the study cohort's unique control cohort. The selected patient was chosen at random from the set of control patients with the same combination of comorbidities, sex, and age, the latter of which was dichotomized as either younger than 75 years or 75 years and older. As such, four total cohorts were created for comparative purposes: two study cohorts and two control cohorts.
Ninety-day complication rates were assessed. Major medical complications included respiratory failure, pulmonary embolism, acute renal failure, myocardial infarction, and cerebrovascular accident. Minor medical complications included deep venous thrombosis, urinary tract infection, and ileus. Wound infection and dehiscence rates were also assessed.
From a socioeconomic perspective, several additional parameters were determined and compared between cohorts. Thirty-day readmission rates owing to any reason were calculated. Length of stay and in-hospital costs were also determined.
Pearson χ2 analysis was used to compare proportions of patients in each cohort with given demographics or comorbid conditions and also to compare complication and readmission rates. Student t test was utilized to compare in-hospital costs. Median length of stay was compared between cohorts using the Wilcoxon rank-sum test. For all analyses, significance was set at P < 0.001.
Patient Demographics and Comorbidities
All assessed demographics and comorbidities are summarized in Table 1. Both obesity cohorts had significantly higher proportions of younger and female patients, when compared with the overall control cohort. Proportions of patients with each assessed comorbidity were significantly higher in each obesity cohort, as compared with the control cohort. Notably, 63% of patients in the morbid obesity cohort had diabetes mellitus, 48% had chronic pulmonary disease, and 40% had evidence of smoking history. Analogous proportions in the obesity (BMI 30–39.9) cohort were 51%, 42%, and 39%.
Ninety-Day Medical Complications
Both nonmorbid obesity and morbid obesity patients demonstrated an increased predisposition for experiencing complications during the 90 days following the index procedure (Tables 2 and 3). Obese and morbidly obese patients had significantly higher rates of acute renal failure and urinary tract infection when compared with matched controls. Uniquely, morbidly obese patients had a significantly higher rate of respiratory failure than matched controls (OR 1.91, 95% CI: 1.52–2.39, P < 0.0001). Of note, both morbidly obese and obese patients had significantly higher odds of experiencing any one major medical complication (OR 1.79, 95% CI: 1.54–2.08, P < 0.0001 and OR 1.32 95%, CI: 1.18–1.47, P < 0.0001, respectively).
Surgical complications were elevated in both pathologic cohorts. Wound infection (OR 3.71, 95% CI: 2.69–5.12, P < 0.0001 and OR 2.22, 95% CI: 1.76–2.81, P < 0.0001) and dehiscence (OR 3.80, 95% CI: 2.45–5.87, P < 0.0001 and OR 2.59, 95% CI: 1.84–3.63, P < 0.0001) occurred with greater than two times increased odds in the morbidly obese and obese cohorts, respectively.
Thirty-Day Readmission Rates, Length of Stay, and In-Hospital Costs
Thirty-day readmissions for any cause were significantly increased in both the obese (7.8 vs. 4.9%; OR: 1.62, 95% CI: 1.39–1.90, P < 0.0001) and morbidly obese (10.1 vs. 5.2%; OR: 2.06, 95% CI: 1.66–2.55, P < 0.0001) cohorts compared with matched controls. For obese and morbidly obese cohorts, median LOS was 3 and 4 days, respectively. Though control populations had a similar median LOS of 3 days, when considering the distribution of length of stay data, length of stay was increased for both obese (P < 0.0001) and morbidly obese (P < 0.0001) cohorts. In-hospital costs were increased by almost $8000 in morbidly obese patients as compared with controls (P < 0.0001). Data concerning length of stay, readmission rates, and in-hospital costs are summarized in Table 4.
The obese, elderly population represents a poorly studied sector of society with increased complication rates following elective, one- to two-level lumbar spinal fusion surgery. In the present study, patients who were morbidly obese were found to experience increased odds of developing any major medical complication (OR 1.8), urinary tract infection (OR 1.4), wound infection (OR 3.7), and wound breakdown (OR 3.8) within 90 days, in addition to significantly increased all-cause 30-day readmissions. Patients with nonmorbid obesity had similarly increased complication rates, though at lower ORs. In-hospital costs and length of stay were significantly increased in both pathologic cohorts compared with controls.
Increased complication rates following lumbar spine surgery in obese patients have been demonstrated in many previous studies. In particular, increases in wound and major medical complications have been observed.17–21 In a cohort of over 800 patients undergoing cervical or lumbar fusion for degenerative spine pathology, Higgins et al reported 2.8 and 2.5 times increased rates of wound and major medical complications, respectively, in obese patients. In the same study, morbidly obese patients were found to have 15 times increased risk of both major medical and wound complications, as compared with nonobese patients.19 Similar findings were illustrated by De la Garza-Ramos et al17 who found that obesity was independently correlated with an increased risk of both developing any postoperative complication (risk ratio [RR]: 2.1) and developing a surgical site infection (RR: 3.1). Similar significant increases in wound-related complications or infection have been demonstrated in other studies.18,20 The present study utilized the Medicare database to study a different target population with similar results. The increased odds of wound complications (2.2–2.6 times increased odds in obese patients and 3.7–3.8 times increased odds in morbidly obese patients) and any major medical complication (1.3 and 1.8 times increased odds) reported in the current study fall within the range of values presented in other reports. However, it should be noted that our study focused on a uniform surgical population, only selecting those patients treated with short fusions (one to two level) for elective reasons. As such, it is possible that we have underestimated the full potential of obesity to increase complication rates owing to the baseline low risk of complications associated with short, elective fusions.
From a socioeconomic perspective, possibly the most significant finding from this study involved comparison of 30-day readmission rates in the two obese cohorts. Obese and morbidly obese patients were found to have more than 1.5 and 2 times increased odds, respectively, of having a readmission during the 30 days following the index procedure, as compared with matched controls. This finding is notable because of the recent decision by the Centers for Medicare and Medicaid Services (CMS) to target joint arthroplasty for hospital readmission rate assessment,22–24 suggesting that other common elective orthopedic procedures, such as one- to two-level lumbar spine fusion could soon be similarly targeted. Other studies have demonstrated a similarly increased risk of readmission associated with obesity in patients treated with elective lumbar surgery.25,26 Our finding that both morbid and nonmorbid obesity significantly increase in-hospital costs has been observed in previous research. Higgins et al19 similarly found that morbid obesity results in an over $9,000 increase in the cost of care when considering various spine surgeries.
Though data from the present study indicate that both obese and morbidly obese patients are at much greater risk of postoperative complications as compared with nonobese elderly patients with similar demographic characteristics and comorbid diagnoses, data from this study do not provide information regarding other clinical outcomes, including functionality and disability. It is conceivable that amelioration of back pain in this patient population may lead to healthier lifestyles and increased activity level that could improve the overall health of the patient. In fact, Djurasovic et al demonstrated that at 2 years of follow-up, obese patients had similar improvement in both Oswestry Disability Index and Short Form 36 Physical Component Summary Scores as compared with nonobese patients treated with lumbar fusion.18 Ultimately, it is of utmost importance that each patient and each unique comorbidity profile be thoroughly considered when approaching patients with significant obesity. A future topic of interest may be how extreme weight loss in morbidly obese patients can modify short-term surgical outcomes. Unfortunately, our current data do not inform how weight loss can affect individual outcomes in obese patients undergoing elective spine surgery.
This study has several advantages stemming from use of a large, national insurance database. Adequately sized cohorts allowed for comparisons with adequate statistical power to demonstrate clinically significant findings. However, with administrative databases, there are many limitations. First, comorbidity coding, particularly by ICD-9 diagnosis codes, is known to be unreliable because of poor coding practices.27,28 To combat this issue, we only selected patients for each of the pathological cohorts who had a given diagnosis of obesity within the 3-month period preceding surgery. Additionally, when selecting a control cohort, we specifically only chose patients who did not have a coded diagnosis of obesity at any time during their full set of records, covering potentially an 8-year span. In the full cohort of Medicare patients 65- to 84-years’ old, we found that only 10.6% were coded as obese at some point during the full scope of potentially 8 years of records, which possibly limits the validity of this study, given that national estimates are above 30% within this age demographic.29 Second, data from our Medicare database is only available in aggregate form, preventing multivariate analysis and potentially increasing the risk of confounders. To avoid this potential limitation, we used a strict matching algorithm that specifically selected control cohorts with the exact same set of patients as the pathological cohorts, with respect to demographics and comorbid diagnoses. One final limitation is the generalizability of the present study. Our study only focuses on 65- to 84-year-old patients, limiting conclusions about elderly patients 85 years of age or older.
Patients with obesity are at significantly increased risk of major medical complications, urinary tract infection, wound complications, and 30-day readmissions. Patients who are specifically morbidly obese have even higher risks and associated costs. Morbidly obese patients should be appropriately counseled of these risks and should be carefully selected to reduce postoperative morbidity.
- Morbidly obese lumbar fusion patients have over 3.5 times increased odds of developing wound infection or breakdown than matched nonobese controls.
- Morbidly obese patients have significantly increased rates of 30-day readmission and length of stay.
- This patient cohort incurs on average almost $8000 more in-hospital charges than nonobese controls.
1. Yang L, Colditz GA. Prevalence of Overweight and Obesity in the United States, 2007-2012. JAMA Intern Med
2. Mihalko WM, Bergin PF, Kelly FB, et al. Obesity, Orthopaedics, and Outcomes. J Am Acad Orthopaed Surg
3. Werner BC, Evans CL, Carothers JT, et al. Primary total knee arthroplasty in super-obese patients: dramatically higher postoperative complication rates even compared to revision surgery. J Arthroplasty
4. Tsai AG, Williamson DF, Glick HA. Direct medical cost of overweight and obesity in the USA: a quantitative systematic review. Obes Rev
5. Thijssen E, van Caam A, van der Kraan PM. Obesity and osteoarthritis, more than just wear and tear: pivotal roles for inflamed adipose tissue and dyslipidaemia in obesity-induced osteoarthritis. Rheumatology (Oxford, England)
6. Friedmann JM, Elasy T, Jensen GL. The relationship between body mass index and self-reported functional limitation among older adults: a gender difference. J Am Geriatr Soc
7. Deyo RA, Bass JE. Lifestyle and low-back pain. The influence of smoking and obesity. Spine
8. Leboeuf-Yde C, Kyvik KO, Bruun NH. Low back pain and lifestyle. Part II—obesity. Information from a population-based sample of 29,424 twin subjects. Spine
9. McClendon J Jr, Smith TR, Thompson SE, et al. The impact of body mass index on hospital stay and complications after spinal fusion. Neurosurgery
10. Gandhi R, Woo KM, Zywiel MG, et al. Metabolic syndrome increases the prevalence of spine osteoarthritis. Orthopaed Surg
11. Liuke M, Solovieva S, Lamminen A, et al. Disc degeneration of the lumbar spine in relation to overweight. Int J Obes
12. Marawar S, Girardi FP, Sama AA, et al. National trends in anterior cervical fusion procedures. Spine
13. Weinstein JN, Lurie JD, Olson PR, et al. United States’ trends and regional variations in lumbar spine surgery
: 1992-2003. Spine
14. Carreon LY, Puno RM, Dimar JR. Perioperative complications of posterior lumbar decompression and arthrodesis in older adults. J Bone Joint Surg Am
15. Smith JS, Shaffrey CI, Glassman SD, et al. Risk-benefit assessment of surgery for adult scoliosis: an analysis based on patient age. Spine
16. Quan H, Sundararajan V, Halfon P, et al. Coding algorithms for defining comorbidities in ICD-9-CM and ICD-10 administrative data. Med Care
17. De la Garza-Ramos R, Bydon M, Abt NB, et al. The impact of obesity on short- and long-term outcomes after lumbar fusion. Spine
18. Djurasovic M, Bratcher KR, Glassman SD, et al. The effect of obesity on clinical outcomes after lumbar fusion. Spine
19. Higgins DM, Mallory GW, Planchard RF, et al. Understanding the Impact of Obesity on Short-term Outcomes and In-hospital Costs After Instrumented Spinal Fusion. Neurosurgery
20. Pereira BJ, de Holanda CV, Ribeiro CA, et al. Impact of body mass index in spinal surgery for degenerative lumbar spine disease. Clin Neurol Neurosurg
21. Buerba RA, Fu MC, Gruskay JA, et al. Obese Class III patients at significantly greater risk of multiple complications after lumbar surgery: an analysis of 10,387 patients in the ACS NSQIP database. Spine J
22. Centers for Medicare & Medicaid Services. Hospital Readmissions Reduction Program (HRRP): Fiscal Year (FY) 2015 Fact Sheet]. 7/1/2014. Available at: https://www.cms.gov/Medicare/Quality-Initiatives-Patient-Assessment-Instruments/Value-Based-Programs/HRRP/HRRP-2015-Fact-Sheet-.pdf
. Accessed December 29, 2015.
24. Healthcare Cost and Utilization Project. Statistical Brief #172: Conditions With the Largest Number of Adult Hospital Readmissions by Payer, 2011]. April 2014. Available at: http://www.hcup-us.ahrq.gov/reports/statbriefs/sb172-Conditions-Readmissions-Payer.pdf
. Accessed December 29, 2015.
25. Basques BA, Varthi AG, Golinvaux NS, et al. Patient characteristics associated with increased postoperative length of stay and readmission after elective laminectomy for lumbar spinal stenosis. Spine
26. Seicean A, Alan N, Seicean S, et al. Impact of increased body mass index on outcomes of elective spinal surgery. Spine
27. Gologorsky Y, Knightly JJ, Lu Y, et al. Improving discharge data fidelity for use in large administrative databases. Neurosurg Focus
28. Samuel AM, Lukasiewicz AM, Webb ML, et al. ICD-9 diagnosis codes have poor sensitivity for identification of preexisting comorbidities in traumatic fracture patients: A study of the National Trauma Data Bank. J Trauma Acute Care Surg
29. Ogden CL, Carroll CM, Fryar CD. Prevalence of obesity among adults and youth: United States, 2011-2014]. NCHS Data Brief