Stopping Biologics Before Joint Surgery: Is Earlier Better? Save

Withholding intravenous abatacept (Orencia) for longer than a month before hip or knee arthroplasty among patients with rheumatoid arthritis (RA) did not lower the risk for postoperative infection, a retrospective claims-based analysis found.

Compared with stopping the biologic less than 4 weeks prior to surgery, stopping 4 to 8 weeks before the procedure was not associated with a greater risk of 30-day hospitalized infection (OR 0.93, 95% CI 0.65-1.34). Nor was stopping more than 8 weeks before (OR 1.13, 95% CI 0.73-1.77), according to Jeffrey R. Curtis, MD, of the University of Alabama at Birmingham, and colleagues.

There also was no difference in rates of prosthetic joint infections at 1 year for patients stopping the abatacept 4 to 8 weeks before arthroplasty (HR 1.29, 95% CI 0.62-2.69) or more than 8 weeks before (HR 1.20, 95% CI 0.48-3.05), the researchers reported online in Arthritis Care & Research.

Patients with RA have an elevated risk of infection, relating to their disease activity, immunosuppressive treatments, and comorbidities. The American College of Rheumatology has recommended continuing treatment with conventional disease-modifying antirheumatic drugs through the perioperative period when RA patients undergo elective knee or hip arthroplasty -- procedures that are common among these patients.

The guidelines also recommend that biologics be withheld for one dosing interval before surgery, although the data supporting this recommendation are limited.

However, stopping biologic treatment can place the patient at risk for disease flare, and the optimal time of stopping remains uncertain.

Curtis and colleagues previously found that stopping infliximab (Remicade) before surgery did not decrease the risk of postoperative infection, but the effects of stopping on other types of biologics with different dosing intervals and mechanisms of action remain uncertain.

Accordingly, they analyzed data from Medicare claims and the Truven MarketScan database for the years 2006 to 2015, identifying patients with RA being treated with once-monthly intravenous abatacept who underwent knee or hip arthroplasty (Patients on subcutaneous abatacept were not included; the timing of intravenous infusions can be more precisely established).

Covariates included demographics, comorbidities, medication use, and overall healthcare utilization during the year before surgery.

The analysis included 1,939 surgeries in 1,780 patients. The majority of patients were women enrolled in Medicare, and mean age was 67.

Time of stopping abatacept was less than 4 weeks, or within one dosing interval, in 37.8%, 4 to 8 weeks in 44.5%, and more than 8 weeks in 17.8%. Those who stopped within 4 weeks or 4 to 8 weeks before surgery were similar, but those who stopped more than 8 weeks before were less likely to be receiving methotrexate; to have had other prior biologic therapies; more likely to have received opioids or antibiotics in the previous 3 months; and more likely to have been hospitalized within the past year.

In the 30 days after surgery, there were 175 (9%) infections, with the most common being urinary tract, skin and soft tissue, and pneumonia.

Urinary tract infections (UTI) were expected to predominate for reasons including the need for catheterization, so the researchers also considered non-UTI separately, and found no difference in rates according to abatacept stop time. There also were no differences in 30-day readmission rates.

There were, however, differences in risks of having a prolonged hospital stay (>4 days for primary surgery and >5 days for revision surgery), with odds ratios of 1.74 (95% CI 1.17-2.58) for patients whose stop time was 4 to 8 weeks before surgery and 2.26 (95% CI 1.41-3.62) for those with more than 8 weeks. This may reflect disease flares or patient characteristics such as risks of complications, the researchers noted. 

In a sensitivity analysis that evaluated stop times in 2-week intervals, those who stopped within 2 weeks of the surgery had a numerically higher risk for hospitalized infection (OR 1.63, 95% CI 0.91-2.91), non-UTI (OR 1.45, 95% CI 0.70-2.99) and 30-day readmission rates (OR 1.23, 95% CI 0.60-2.51). These findings were not statistically significant, but the authors noted that they were unable to rule out the possibility of "clinically important" infection risks with the 2-week stop time.

Finally, compared with no glucocorticoid use, receiving a glucocorticoid dose above 7.5 mg/day in the 3 months before surgery was associated with a greater risk for hospitalized infection (OR 2.19, 95% CI 1.28-3.77). In addition, patients who stopped more than 8 weeks before surgery were more likely to fill a prescription for a glucocorticoid within 3 months after the procedure.

The researchers concluded that stopping intravenous abatacept for one dosing interval did not decrease the risk of 30-day hospitalized infection or 1 year prosthetic joint infection.

"The questionable benefits of [withholding] therapy should be balanced against the known risks of disease flares, especially because glucocorticoids are associated with adverse postoperative outcomes," they observed.

A study limitation was the possibility of residual confounding by indication.

The study was funded by Bristol-Myers Squibb (BMS). Several co-authors are company employees.

The authors disclosed relevant relationships with BMS, AbbVie, Corrona, Galapagos, Roche, Gilead, UCB, Eli Lilly, GlaxoSmithKline, and Pfizer.