Introduction:

Optimal anesthesia is an essential component of the enhanced recovery pathway after surgery (ERAS) as it has the potential to facilitate earlier mobilization, expedite the return of bowel function--leading to shorter length of stay, and accelerated convalescence. Neuroaxial (i.e. spinal, regional, epidural) anesthesia has also been associated with lower 30-day readmission rates. Furthermore, a reduction in hospital mortality has also been reported, in particular due to decreased deep venous thromboembolism. Prior studies have promoted the combination of spinal and general anesthesia as providing a more predictable analgesic effect, lesser hemodynamic and respiratory side effects in patients undergoing cardiac, vascular, orthopedic, pelvic and abdominal surgery. Indeed, local anesthetics interferes with blood clotting and prevent hypercoagulability, which can reduce incidence of post-operative DVT. Based on the above-mentioned considerations, the National Institute of Health and Care Excellence (NICE) in the United Kingdom recommends including neuroaxial anesthesia as part of venous thromboembolism (VTE) prevention in patients undergoing non-cardiac surgery.

Radical nephrectomy is a commonly performed urologic surgery in the United States. Given the unique benefits provided by neuroaxial anesthesia in orthopedic and cardiac literature, we aimed to evaluate the benefits of adjunct anesthesia in the open nephrectomy cohort.  Specifically, the objective of the present study was to identify differences in intra- and post- operative complications, length of stay and readmission rates between patients managed with general anesthesia alone as compared to general + neuroaxial ( spinal, regional and epidural) anesthesia. 

Methods:

The NSQIP database includes over 150 perioperative data elements, which were collected from over 600 hospitals. Patients were included in the study if they were 18 years or older at the time of surgery between 2014 and 2017. CPT codes were used to identify patients having undergone open nephrectomy. Patients were further subdivided based on anesthesia modality: general anesthesia alone vs in combination with neuroaxial anesthesia.

Data elements that were collected included age, race/ethnicity, body mass index (BMI), smoking status, diabetes mellitus, hypertension on medications, dyspnea, chronic obstructive pulmonary disease, bleeding disorders, steroid use, greater than 10% weight loss within 6 months before surgery, preoperative hematocrit, American Society of Anesthesiology (ASA) physical status, blood transfusion within 72 hours prior to surgery, congestive heart failure, and functional status. Functional status was defined as dependent (partial or total) versus independent.

Patients were excluded if they had preoperative ascites, disseminated cancer or sepsis, or on mechanical ventilation at the time of surgery. Patients with missing values (preoperative hematocrit, height, weight, and those with unknown or unassigned ASA status), older than 90 years old, or undergoing emergent nephrdectomy were excluded. Patients with any additional concurrent general surgery procedures (i.e cholecystectomy, appendectomy, colectomy, etc) were also excluded. 

Intraoperative variables that were evaluated in the analysis included procedure type, total operating time, and wound classification, defined using the National Healthcare Safety Network.

Postoperative variables included postoperative length of hospital stay, total length of stay, complications, readmission, procedure related readmissions and reoperation rates.

Results:

2,346 out of 3633 patients were included. Before propensity score matching, patients in the two groups were unevenly matched for BMI (p=0.031), race (p<0.001), and hypertension (p=0.033).  After 1:1 propensity score matching, 1090 patients were evenly distributed in each category, with no demographic differences between the two groups. The differences previously seen prior to matching were no longer significant. The remainder of the descriptive statistics were equally distributed among the groups.

Compared to GA alone after multivariable logistic regression, adjuvant neuroaxial anesthesia showed increased odds ratio of prolonged post-operative stay [aOR: 1.107, 95% CI: 1.042-1.176, p= 0.001] (Table 3C) after adjusting for age, dyspnea, CHF, COPD, ASA status, ARF, dialysis, operative time, and preoperative hematocrit value.  The addition of neuroaxial anesthesia was not associated with decreased procedure related readmission rate (aOR 0.966, 0.537-1.736, p=0.909), or a decrease in the complication rate (aOR 0.9, 0.659-1.227, p=0.505) when compared to general anesthesia alone. (Table 3A & 3B) 

Overall, the complication rate was 23.9 %, with no differences in complications between GA alone vs adjuvant neuroaxial anesthesia group, p=0.434. The most common complication after open nephrectomy in the GA cohort was a urinary tract infection, seen in 11(2%) patients. Pneumonia was more common in the NA group, 15 (2.8%)--although these difference were not significant. Both groups experienced a high rate of intra and postoperative transfusion requirements 101(18.5%) and 94(17.2%), p=0.635, as well as all-readmissions 25(4.6%) vs 24(4.4%), p=0.096, albeit the differences again were not significant in either of the postoperative outcomes.

Conclusion:

Using the 2014-2017 NSQIP database, we were able to demonstrate no difference in procedure related readmissions and complications in the neuroaxial anesthesia group. Furthermore, patients who received neuroaxial anesthesia experienced a longer postoperative stay. Future prospective trials with increased focus on postoperative opioid consumption and early mobilization are encouraged. 

Funding: N/A