Abstract
Introduction: There is limited evidence to guide peri-operative steroid regimen in patients receiving chronic glucocorticoid (GC) therapy. Many patients who undergo minor surgical procedures receive stress-dose steroids. While hemodynamic instability and hypotension are well-recognized risks of inadequate peri-operative GCs, there may be harms associated with using doses higher than necessary. Whether surgical outcomes differ according to peri-operative steroid dose is not known. We hypothesized that patients who had greater GC exposure have less hypotension, but higher rates of hyperglycemia and post-operative complications. Methods: This retrospective study investigated the relationships between peri-operative GC use and post-operative complications following total hip/knee joint replacement (arthroplasty) in patients with rheumatoid arthritis (RA). All GCs were converted to prednisone equivalents; GC exposure was assessed by number of doses and total cumulative dose during the hospitalization. Complications (infection, thromboembolism and cardiovascular events) were determined by chart review. Results: Of 432 patients with RA included, half (54%) underwent knee arthroplasty. Mean age was 64±12 years, 78% were women. Thirty percent of patients were on home GCs (mean dose 7±4mg/day). Median cumulative GC dose during hospitalization was 37mg [IQR 27, 57]. Compared to patients who only received one peri-operative dose of steroids, those who received multiple doses had a greater risk of post-operative complications (OR 3.319 (95% CI 1.03, 12.62; p<0.05] and hyperglycemia, glucose >180 mg/dl, [OR 1.812(0.99, 3.32; p<0.05]. They did not have an increase in hypotension or need for pressors. Among patients who received steroids while in the hospital (90%), there was a small but significant dose response relationship with hyperglycemia (r=0.16; p<0.01). Higher cumulative dose was also associated with higher risk of complications; for every 10 mg increase in cumulative dose, the risk of complications increased by 15% (p<0.01). Conclusions: Among RA patients undergoing arthroplasty, we did not find that lower doses of GCs were related to more hypotension. However, patients with higher GC exposure were more likely to have hyperglycemia and post-operative complications. Our results suggest that use of peri-operative GC is not without risk, and the lowest doses possible should be considered. Further studies are needed to confirm these findings and to define the optimal dosing strategies for patients receiving peri-operative GCs.