ATLANTA – Patients with rheumatoid arthritis were more at risk of postoperative infection because of a high Charlson Comorbidity Index or longer surgery time than because of perioperative use of antirheumatic medications, according to a presentation at the annual meeting of the American College of Rheumatology.
, of the University of Minnesota in Minneapolis and colleagues performed a retrospective cohort study of 154 patients with seropositive RA who were in the Fairview Health System between Jan. 2010 and Dec. 2017 and underwent either orthopedic or major organ surgery. The patients were classified based on their use of disease-modifying antirheumatic drugs (DMARDs) and biologics alone or in combination, with patients divided into “no DMARD or biologic,” “DMARD but no biologic” and “biologic with or without DMARD” groups.
The question of whether to discontinue antirheumatic medications before surgery is still controversial, with conflicting evidence across studies, Dr. Shmagel said in her presentation. A study by Giles and colleagues found 10 of 91 patients (11%) RA who underwent an orthopedic surgical procedure developed a postoperative infection, with patients receiving tumor necrosis factor (TNF) inhibitors more likely to develop an infection, compared with patients who were not receiving TNF inhibitors (Arthritis Care Res. 2006.).
However, other studies have challenged that idea, and a 2018 study from Goodman and colleagues raised the issue of whether patients stopping biologics prior to surgery are at increased risk of flares. Of 120 RA patients in their study who underwent total hip or total knee arthroplasty, 75% of patients flared at 6 weeks after surgery. While patients who halted biologics before surgery were more likely to flare, stopping biologics did not predict flaring after surgery (J Rheumatol. 2018.).
“It’s not entirely clear whether these theories are related to what we do with antirheumatic medications, but we felt that it was pertinent to further study this question.” Dr. Shmagel said.
Dr. Shmagel and colleagues examined the 30-day infection rate of RA patients postoperatively, with 30-day readmission and 30-day mortality rates as secondary outcomes. Patient-associated factors such as age, gender, race, body mass index, smoking status, Charlson Comorbidity Index, income, and use of corticosteroids were analyzed as covariates in addition to factors involving surgery such as expected surgery time, perioperative antibiotic use, and whether the procedure was elective or emergency surgery.
A majority of the patients in the study across all groups were white women about 63 years old with a body mass index above 30 kg/m2 and almost all undergoing electing surgery compared with emergency surgery. While patients in each group were similar with regard to Charlson Comorbidity Index, expected length of surgery, and percentage of patients undergoing elective surgery, patients in the biologic with or without DMARD group had a significantly lower median income level compared with those in the other two groups (P = .01).
Overall, there were 244 surgeries in 154 patients, with 117 surgeries in the group not receiving biologics or DMARDs, 95 surgeries in the group receiving DMARDs but no biologics, and 32 surgeries in the biologics with or without DMARD group. In the DMARD but no biologics group, most patients were receiving methotrexate (45%) or hydroxychloroquine (44%), while the most common biologics in the biologics with or without DMARD group were infliximab (25%), tocilizumab (19%), abatacept (16%), etanercept (13%), rituximab (9%), and tofacitinib (9%).
There was an 11% overall rate of infection, with a similar rate of infection across all groups (P = .09). While there was a higher rate of surgical site infections among patients in the biologics with or without DMARD group (9%) and a higher percentage of urinary tract infections in the no DMARD and no biologics group (4%), the results were not statistically significant. When the rate of infections was examined by type of surgery, there were no significant differences between infections from musculoskeletal surgery (P = .7) and major organ surgery (P = .8).
The overall 30-day readmission rate was 12%, but there were no statistically significant differences between groups. Although there were five deaths in the study, four deaths were in the group not receiving DMARDs or biologics, and one death was in the biologic with or without DMARD group.
Higher Charlson Comorbidity Index did predict infection risk, with an odds ratio of 1.37 per 1-point increase in the index (95% confidence interval, 1.10-1.70). Length of surgery also increased the risk of infection, with an OR of 1.16 per 15-minute increase in surgery time (95% CI, 1.09-1.23).
Dr. Shmagel noted that the retrospective nature of the study and the midwestern cohort may mean the results are not generalizable to other populations and that larger randomized trials should be considered. “Certainly, a larger study with more events would be needed,” she said.
This study was funded by the University of Minnesota. Dr. Shmagel reported no relevant conflicts of interest.
SOURCE: Kerski M et al. Arthritis Rheumatol. 2019;71 (suppl 10), Abstract 1805.