The effect of anesthesia type on major lower extremity amputation in functionally impaired elderly patients.

OBJECTIVE: Patients undergoing major lower extremity amputations are at risk for a wide variety of perioperative complications. Elderly patients with any functional impairment have been shown to be at high risk for these adverse events. Our goal was to determine the association between the type of anesthesia-general anesthesia (GA) and regional/spinal anesthesia (RA)-on perioperative outcomes after lower extremity amputation in these elderly and functionally impaired patients.

METHODS: The American College of Surgeons National Surgical Quality Improvement Program (ACS-NSQIP) data set (2005-2012) was queried to identify all patients aged ≥75 years with partial or total functional impairment who underwent major lower extremity amputations. Propensity matching and multivariate analysis were performed to isolate the effect of anesthesia type.

RESULTS: We identified 3260 patients (50% male), 2558 GA patients and 702 RA patients, who were a mean age of 82 years. Anatomic distribution was 59% above-the-knee and 41% below-the-knee amputations. Patients undergoing GA were more likely to have impaired sensorium (9% vs 6%; P = .035), be on anticoagulation or have a bleeding disorder (33% vs 17%; P < .001), have had a previous operation ≤30 days (16% vs 10%; P < .001), and be operated on by a nonvascular surgeon (16% vs 12%; P = .033). GA was associated with shorter anesthesia time to surgery (36 ± 48 vs 42 ± 49 minutes; P < .001) but a similar operative time (66 ± 33 vs 64 ± 33 minutes; P = .292) compared with RA. After propensity matching, rates of 30-day mortality (14% vs 12%; P = .135), postoperative myocardial infarction/cardiac arrest (2.9% vs 3.1%; P = .756), pulmonary complications (7.3% vs 6.7%; P = .632), stroke (0.7% vs 0.9%; P = .694), urinary tract infections (6.7% vs 6.5%; P = .887), and wound complications (7.6% vs 7.6%; P = .999) were similar in patients undergoing GA and RA, respectively. Median length of stay was similar in both groups (5 vs 5.5 days; P = .309). Multivariable analyses confirmed that anesthesia type did not significantly affect morbidity and mortality.

CONCLUSIONS: The mode of anesthesia, GA vs RA, did not have significant effect on perioperative outcomes after major lower extremity amputation in the functionally impaired geriatric population. These findings provide an evidence base that will allow surgeons, anesthesiologists, and patients to make an informed decision about anesthesia type for their procedure.