High dose nonsteroidal anti-inflammatory drugs compromise spinal fusion.

Purpose: Although nonsteroidal anti-inflammatory drugs (NSAIDs) provide benefit to patients following spinal fusion surgery, their routine administration has remained controversial due to concerns about possible deleterious effects on bone healing. The goal of this retrospective study was to assess the incidence of non-union following the perioperative administration of ketorolac, celecoxib, or rofecoxib.

Methods: We retrospectively analyzed the data of 434 patients receiving perioperative ketorolac (20-240 mg.day(-1)), celecoxib (200-600 mg.day(-1)), rofecoxib (50 mg.day(-1)), or no NSAIDs in the five days following spinal fusion surgery.

Results: There were no significant differences in the incidence of non-union among the groups that received no NSAIDs (11/130; 8.5%), celecoxib 5/60; 8.3%), or rofecoxib (9/124; 7.3%). In contrast, 23/120 of patients (19.2%) that received ketorolac had a higher incidence (P < 0.001) of non-union compared to non-NSAID users. However, only 3/50 patients (6%) receiving low-dose ketorolac (< or = 110 mg.day(-1)) resulted in non-union which was not significantly different from non-NSAID users. Patients administered higher doses of ketorolac (120-240 mg.day(-1)) resulted in a higher incidence (P < 0.0001) of non-union (20/70; 29%) compared to non-NSAID users. For those patients developing non-union, there was a higher incidence comparing smokers vs non-smokers (P < 0.0001) and one level fusion vs two level fusions (P < 0.001).

Conclusions: This study revealed that the short-term perioperative administration of celecoxib, rofecoxib, or low-dose ketorolac (< or = 110 mg.day(-1)) had no significant deleterious effect on non-union. In contrast, higher doses of ketorolac (120-240 mg.day(-1)), history of smoking, and two level vertebral fusions resulted in a significant increase in the incidence of non-union following spinal fusion surgery.