Modified-frailty index does not independently predict complications, hospital length of stay or 30-day readmission rates following posterior lumbar decompression and fusion for spondylolisthesis

Abstract

BACKGROUND CONTEXTFrailty has been associated with inferior surgical outcomes in various fields of spinal surgery. With increasing healthcare costs, hospital length of stay (LOS) and unplanned readmissions have emerged as clinical proxies reflecting overall value of care. However, there is a paucity of data assessing the impact that baseline frailty has on quality of care in patients with spondylolisthesis. PURPOSEThe aim of this study was to investigate the impact that frailty has on LOS, complication rate, and unplanned readmission after posterior lumbar spinal fusion for spondylolisthesis. STUDY DESIGNA retrospective cohort study was performed using the National Surgical Quality Improvement Program (NSQIP) database from 2010 through 2016. PATIENT SAMPLEAll adult (≥18 years old) patients who underwent lumbar spinal decompression and fusion for spondylolisthesis were identified using ICD-9-CM diagnosis and procedural coding systems. We calculated the modified frailty index (mFI) for each patient using 5 dichotomous comorbidities - diabetes mellitus, congestive heart failure, hypertension requiring medication, chronic obstructive pulmonary disease, and dependent functional status. Each comorbidity is assigned 1 point and the points are summed to give a score between 0 and 5. As in previous literature, we defined a score of 0 as “not frail”, 1 as “mild” frailty, and 2 or greater as “moderate to severe” frailty. OUTCOME MEASURESPatient demographics, comorbidities, complications, LOS, readmission, and reoperation were assessed. METHODSA multivariate logistic regression analysis was used to identify independent predictors of adverse events (AEs), extended LOS, complications, and unplanned readmission. RESULTSThere were a total of 5,296 patients identified, of which 2,030 (38.3%) were mFI=0, 2,319 (43.8%) patients mFI=1, and 947 (17.9%) were mFI ≥2. The mFI≥2 cohort was older (p≤.001) and had a greater average BMI (p≤.001). The mFI≥2 cohort had a slightly longer hospital stay (3.7 ± 2.3 days vs. mFI=1: 3.5 ± 2.8 days and mFI=0: 3.2 ± 2.1 days,p≤.001). Both surgical AEs and medical AEs were significantly greater in the mFI≥2 cohort than the other cohorts, (2.6% vs. mFI=1: 1.8% and mFI=0: 1.2%,p=.022) and (6.3% vs. mFI=1: 4.8% and mFI=0: 2.6%,p≤.001), respectively. While there was no significant difference in reoperation rates, the mFI≥2 cohort had greater unplanned 30-day readmission rates (8;4% vs. mFI=5.6: 4.8% and mFI=0: 3.4%,p≤.001). However, on multivariate regression analysis, mFI≥2 was not a significant independent predictor of LOS (p=.285), complications (p=.667), or 30-day unplanned readmission (p=.378). CONCLUSIONSOur study indicates that frailty, as measured by the mFI, does not significantly predict LOS, 30-day adverse events, or 30-day unplanned readmission in patients undergoing lumbar spinal decompression and fusion for spondylolisthesis. Further work is needed to better define variable inputs that make up frailty to optimize surgical outcome prediction tools that impact the value of care.