Anesthesia type—spinal versus general—did not affect outcomes in a pragmatic trial of 1,600 patients, mean age 78, researchers reported in a presentation at Anesthesiology 2021, the annual meeting of the American Society of Anesthesiologists.
The REGAIN investigators, led by Mark D. Neuman, MD, of the Department of Anesthesiology and Critical Care, University of Pennsylvania Perelman School of Medicine in Philadelphia, designed a pragmatic, randomized superiority trial to determine what effect anesthesia choice would have on a patient’s ability to walk 10 feet independently or with a walker 60 days after surgery.
They also evaluated the effect of anesthesia on delirium, days to discharge, or death at 60 days.
Most of the previous randomized trials were “conducted more than 30 years ago and do not reflect current practice, had small numbers of participants, or where not designed to assess outcomes beyond hospital stay,” Neuman and colleagues wrote in The New England Journal of Medicine, which published the study concurrently with its presentation at ASA.
By contrast, the REGAIN investigators focused on outcome at 60 days.
They enrolled 1,600 patients at 46 participating hospitals in the U.S. and Canada. In a modified intention to treat analysis, the composite primary outcome—death or inability to walk at least 10 feet at 60 days—occurred in 18.5% of the spinal anesthesia group versus 18% in the general anesthesia group, which worked out to no measurable difference in relative risk for spinal compared to general anesthesia (RR 1.03, 95% CI 0.84-1.27 P=0.83).
Additional findings:
- Inability to walk independently at 60 days reported in 15.2% of the spinal group versus 14.4% of the general group (RR 1.06, 95% CI 0.82-1.36).
- Death within 60 days 3.9% in the spinal anesthesia group versus 4.1% in the general (RR 0.97; 95% CI, 0.59 to 1.57).
- Delirium rate was 20.5% in the spinal anesthesia arm versus 19.7% in the general (RR 1.04; 95% CI, 0.84 to 1.30).
“The median total anesthesia time was 132 minutes (interquartile range, 102 to 165) in the spinal anesthesia group and 131 minutes (interquartile range, 103 to 165) in the general anesthesia group,” they wrote.
Looking specifically at severe adverse events, the overall rate was low but differed by group: “Death during hospitalization occurred in 5 of 782 patients assigned to spinal anesthesia (0.6%) and in 13 of 790 patients assigned to general anesthesia (1.6%). Acute kidney injury occurred in 32 of 709 patients (4.5%) assigned to spinal anesthesia, and admission to a critical care unit occurred in 18 of 783 (2.3%); the corresponding numbers among the patients assigned to general anesthesia were 55 of 726 (7.6%) and 29 of 793 (3.7%).”
A paper published earlier this year in JAMA noted that as many as 65% of patients age 65 or older experience delirium following noncardiac surgery and an estimated 10% develop long-term cognitive decline. BreakingMED asked Neuman if the REGAIN investigators looked at cognitive decline at 60 days or later time points.
“We did collect these longer-term data, but they are not included in this paper,” Neuman wrote in an email reply. “We plan to analyze them in a separate manuscript.”
Patients were enrolled between Feb. 12, 2016, and Feb 18, 2021. In all, 22,022 patients were screened and 1,60o patients randomly assigned to spinal (n=795) or general anesthesia (n=805). The mean age was 78, a third were men, and 7.6% were Black.
All participants had clinically or radiographically diagnosed femoral neck, intertrochanteric, or subtrochanteric hip fracture requiring surgical repair. Participants were excluded if they were not able to walk 10 feet without assistance, had a periprosthetic fracture, or if they required a concurrent procedure than could not be done under spinal anesthesia.
They also excluded patients at risk for malignant hyperthermia and patients who “had contraindications to spinal anesthesia (coagulopathy, use of anticoagulant or antiplatelet medications, critical or severe aortic stenosis, a high risk of infection at the spinal needle insertion site, or elevated intracranial pressure).” Patients who were judged to have delirium prior to procedure were not excluded if consent could be obtained from a proxy or the patient.
Neuman and colleagues said they included death in the composite outcome measure to eliminate survivor bias.
“Of the 795 patients who were assigned to the spinal anesthesia group, 119 (15.0%) instead received general anesthesia. Reasons for administration of general anesthesia were an inability to place a spinal block (52 patients), clinician selection of general anesthesia (29 patients), patient or proxy request (18 patients), crossover to general anesthesia after spinal block placement (e.g., due to block failure or intraoperative events; 12 patients), and communication issues (e.g., due to case rescheduling or shift changes; 7 patients); no reason was provided in 1 instance… Of the 502 patients with available data on the maximum depth of sedation during spinal anesthesia, 431 (85.9%) had an OAAS score between 5 (lighter sedation) and 2 (deeper sedation), and 71 (14.1%) had a deeper level of sedation,” they wrote.
The crossover rate from general to spinal anesthesia was smaller: just 28 patients (3.5%) of the 805 patients randomized to general anesthesia crossed over. The reasons for crossover included “clinician selection of spinal anesthesia (15 patients), patient or proxy request (7 patients), and communication issues (i.e., as a result of case rescheduling or shift changes; 4 patients); in 2 cases, no reason was provided.”
The authors said the study was limited by “a considerable amount of missing outcome data; however, the sensitivity analyses that account for missing data were similar to those in the primary analysis.” Another limitation was the lower than anticipated occurrence of the primary outcome, which suggests that the patients enrolled were healthier than anticipated by the trial design.
In the U.S. the use of spinal anesthesia for hip surgery increased by 50% over the 10 years from 2007 through 2017 “potentially reflecting a belief that spinal anesthesia is superior to general anesthesia. Our finding of similar outcomes at 60 days with either technique suggests that anesthesia choices for hip-fracture surgery may be based on patient preference rather than on anticipated differences in clinical outcomes,” the concluded.
-
In a pragrmatic, randomized trial, spinal anesthesia for hip-fracture surgery was not superior to general anesthesia.
-
This analysis did not address the longer term congnitive outcomes for spinal anesthesia compared to general anesthesia, so further studies are needed to determine what—if any—cognitive differences as associated with anesthesia choice.
Peggy Peck, Editor-in-Chief, BreakingMED™
Neuman had no conflicting interests to disclose.
Cat ID: 859
Topic ID: 659,859,438,499,791,859,730,192,925,858
