Effects of cyclooxygenase inhibitors on fracture healing and the reversibility of effects after short term treatment
Date
2005
DOI
Authors
Al-Ghawas, Mohammad Nabil
Version
OA Version
Citation
Abstract
Introduction: Previous studies have suggested that both COX-2-specific (coxibs) and nonspecific anti-inflammatory drugs (NSAIDs) inhibit fracture healing. Because drug specificity, dose, and duration may influence fracture healing, and because patients with fractures require analgesia for only short periods of time after initial stabilization, we tested the hypothesis that the negative effects of these drugs on fracture healing would be reversible with short-term treatment.
Materials and Methods: Two hundred fifty male Sprague Dawley rats underwent standard, closed mid diaphyseal femoral fracture. Immediately post fractures, animals were divided in groups for histomorphometry, mechanical testing, and PGE2 assay. Combinations of placebo, non-specific (ketorolac), COX 2 specific NSAID (valdecoxib) treatment were given at different doses and for different durations. For statistical analysis ANOVA, Fisher’s Exact Test, and multivariate analysis of variance at P[less than]0.05 were used.
Results: In rats treated for 7 days, showed no significant differences in % of nonunions after 21 days (16% of ketorolac, 22% of valdecoxib, and 8% of vehicle-treated animals, p=0.47), nor after 35 days (5% of ketorolac , 0% of valdecoxib, and 9% of vehicle-treated, p= 1.0). In rats treated for 21 days, there were differences in % of nonunions after 21 days (13% of ketorolac, 36% of valdecoxib, and 4% of vehicle-treated,p =0.01), but the differences disappeared by 35 days (none in the ketorolac and valdecoxib groups, 5% in vehicle-treated, p=1.0). Torsional testing showed statistically significant diminishment in both stiffness and strength at 35 days post fracture for those animals treated for 21 days with the COX2 specific NSAID but there was no observable difference in mechanical strength in those animals treated for only 7 days. Histomoxphometry showed a strong correlation between non-union rate and percent cartilage in the calluses which is consistent with diminished mechanical stiffness of the calluses. Histological assessment further showed diminished marrow cellular contents in both 2 1 day non and specific NSAID treated tissues which was recovered if the drug was withdrawn at seven days. Analysis of PGE2 levels in the calluses showed that the non-specifc NSAID had 2 to 3 fold lower levels than the specific coxib and exhibited an increasing inhibition at progressively higher doses while the COX-2 specific NSAID did not show any dose dependency of inhibition over a log difference in oral doses.
Conclusions: These data suggest COX2 specific NSAIDs delay fracture healing and the magnitude of the effect as measured by non-union in this study was related to the duration of treatment. Treated animals recover their healing capacity upon discontinuation of treatment. The analysis of PGE2 1evels in the callus suggests that the effects of these drugs may not simply be related to their actions on PGE2 1evels but that the specific COX2 effectS may be related to other phamacological actions that are not clearly defined. Extraplation of these findings to a clinical setting suggests that management of fracture-associated pain with inhibitors of COX-2 should neither impair nor delay healing as Iong as the duration of treatment is consistent with current standards of care.
Description
PLEASE NOTE: This work is protected by copyright. Downloading is restricted to the BU community: please click Download and log in with a valid BU account to access. If you are the author of this work and would like to make it publicly available, please contact open-help@bu.edu.
Thesis (MSD)--Boston University, Goldman School of Dental Medicine, 2005 (Restorative Sciences and Biomaterials).
Includes bibliographical references: leaves 66-72.
Thesis (MSD)--Boston University, Goldman School of Dental Medicine, 2005 (Restorative Sciences and Biomaterials).
Includes bibliographical references: leaves 66-72.
License
This work is protected by copyright. Downloading is restricted to the BU community. If you are the author of this work and would like to make it publicly available, please contact open-help@bu.edu.