SAN DIEGO – Functional dependence following elective cervical spine procedures was associated with a significantly increased risk of almost all 30-day complications analyzed, including mortality, a large retrospective analysis of national data demonstrated.
The findings suggest that physicians should “include the patient’s level of functional independence, in addition to more traditional medical comorbidities, in the risk-benefit analysis of surgical decision making,” Dr. Alpesh A. Patel said in an interview in advance of the annual meeting of the Cervical Spine Research Society. “Those individuals with dependence need to be counseled appropriately about their increased risk of complications including mortality.”
Dr. Patel, professor and director of orthopedic spine surgery at Northwestern University Feinberg School of Medicine, Chicago, and his associates retrospectively reviewed the American College of Surgeons National Surgical Quality Improvement Program (ACS NSQIP) data files from 2006 to 2013 and limited their analysis to patients undergoing elective anterior cervical fusions, posterior cervical fusions, cervical laminectomy, cervical laminotomy, cervical discectomy, or corpectomy. They divided patients into one of three groups based on the following preoperative functional status parameters: independent, comprising those not requiring assistance or any equipment for activities of daily living (ADLs); partially dependent, including those with equipment such as prosthetics, equipment, or devices and requiring some assistance from another person for ADLs; and totally dependent, in which patients require total assistance for all ADLs. The researchers used univariate analysis to compare patient demographics, comorbidities, and 30-day postoperative complications among the three groups, followed by multivariate logistic regression to analyze the independent association of functional dependence on 30-day complications when controlling for procedure and comorbidity variances.
Dr. Patel reported findings from 24,357 patients: 23,620 (97.0%) functionally independent, 664 (2.7%) partially dependent, and 73 (0.3%) totally dependent. Dependent patients were significantly older and had higher rates of all comorbidities (P less than .001), with the exception of obesity (P = .214). In addition, 30-day complication rates were higher for all complications (P less than .001) other than neurological (P =.060) and surgical site complications (P =.668). When the researchers controlled for type of procedure and for disparities in patient preoperative variables, multivariate analyses demonstrated that functional dependence was independently associated with sepsis (odds ratio 6.40; P less than .001), pulmonary (OR 4.13; P less than .001), venous thromboembolism (OR 4.27, P less than .001), renal (OR 3.32; P less than .001), and cardiac complications (OR 4.68; P =.001), along with mortality (OR 8.31; P less than .001).
“The very strong association between functional dependence and mortality was quite surprising,” Dr. Patel said. “It was, to the contrary, also surprising to see that, despite wide variance in medical comorbidities and functional status, surgical complications such as infection and neurological injury were similar in all groups.” He characterized the study as “the first large-scale assessment of functional status as a predictor of patient outcomes after cervical spine surgery. It fits in line with other studies utilizing large databases. Big data analysis of outcomes can be used to identify risk factors for complications including death after surgery. Identifying these factors is important if we are going to improve the care we provide. Accurately quantifying the impact of these risk factors is also critical when we risk stratify and compare hospitals and physicians.”
He acknowledged certain limitations of the study, including the fact that it is a retrospective study “with a heterogeneous population of patients, surgeons, hospitals, and procedures. This adds uncertainty to the analysis at the level of the individual patient but does provide generalizability to a broader patient population.”
Dr. Patel reported having no conflicts of interest.