Table 5.
Characteristics of the Included Studies II: Effects of NSAIDs on Chondrocyte Differentiation and Allograft and Autograft Bone Incorporation.
| Year of the Study | Study | In Vitro | In Vivo | Species | N | Follow-up, Min-Max | Outcome Measures | NSAIDs | NSAIDs Administration | Frequency of Administration | Source of Cells | Effects |
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 2011 | Pountos et al. | 1 | 0 | Human MSCs | 10 | 21 days | Functional and quantitative assays, immunohistochemistry | Ketorolac, diclofenac, parecoxib, meloxicam | Drugs were administered into serum-free chondrogenic medium | Constantly in the medium in their stated plasma concentration | Human trabecular bone, bone marrow aspirated from superior iliac crest | No effect on MSC proliferation or its potential to
differentiate into osteogenic lineage. Use of a therapeutic concentration of diclofenac or ketorolac decreased significantly (P < 0.005) sGAG content by 45% and 55%, respectively, while parecoxib and meloxicam, inhibited sGAG to a lesser degree, 22% and 27%, respectively (P < 0.005). Pellet chondrogenesis was associated with increased COX-1 expression levels, but not COX-2, and COX-1 specific drugs suppressed MSC prostaglandin E2 (PGE-2) more than COX-2 specific inhibitors. |
| 2016 | Caron et al. | 1 | 0 | ATDC5, primary hBMSCs, ex vivo periosteal agarose cultures of New Zealand white rabbits, human mature chondrocytes | 8 | Indomethacin administration started at the onset of
differentiation (t = 0): Follow-up with
ATDC5 cells at 10 and 14 days. Follow-up with hBMSCs at 7-28
days. Follow-up with periosteal agarose culture at 21
days. Indomethacin administration beginning at varying days into differentiation onward: Follow-up at 14 days after administration of treatment. |
Bicinchoninic acid (BCA) protein assay (immunoblotting), histochemistry, quantitative reverse transcription polymerase chain reaction | Indomethacin | Drugs were added to the differentiation medium consisting of supplemented proliferation medium | N/A | Human bone marrow stem cells obtained from iliac crest bone marrow aspirate. Periosteal agarose culture harvested from the proximal tibia of skeletally mature rabbits. ATDC5 cell line obtained from mouse embryonic carcinoma cells. Human chondrocytes obtained from unaffected regions of OA cartilage from total knee arthroplasty. | Decreased gene and protein expression of chondrogenic and hypertrophic markers (measured by RT-qPCR and immunoblotting) as well as decreased glycosaminoglycan content (by Alcian blue histochemistry) was observed in all indomethacin concentration. Higher concentration of the drug resulted in more changes. When mature chondrocytes were treated with indomethacin, elevation in collagen type 2 mRNA expression was observed. Similarly, when ATDC5 cells and bone marrow stem cells were predifferentiated to obtain a chondrocyte phenotype and indomethacin was added from this time point onward, low concentrations of indomethacin also resulted in increased chondrogenic differentiation. |
| 2017 | Janssen et al. | 0 | 1 | Skeletally immature New Zealand white rabbits | 12 | 25 days | Standardized enzyme immunoassay analysis, Lane and Sandhu radiological score, micro-finite element analyses (FEA), Heiple histologic fracture score, gene expression analysis, RT-qPCR using SYBR assay | Celecoxib | Administered orally | Daily | Growth plate of the distal femur, tibia, ulna of New Zealand white rabbits | Negative effect on the initiation of the chondrogenic phase
of endochondral ossification. Celecoxib inhibited in
vivo PGE-2 synthesis by 80%. COX-2 inhibition
impaired and delayed the bone fracture healing process
(Lane-Sandhu radiologic fracture score for the control group
was 8.8 compared with a score of 5.7 for the treated group).
Mineralization front was significantly less advanced in the
celecoxib-treated group. Significant decrease in the
hypertrophic zone and total growth plate thickness in the
celecoxib-treated group, but no significant difference was
observed in the proliferative zone. Lack of cartilaginous PEO formation suggested an inhibited chondrogenic differentiation. |
| 2006 | O’Keefe et al. | 0 | 1 | C57BL/6 mice | 61 | X-ray follow-up weekly, semiquantitative histomorphometric analysis follow-up at 5 weeks postsurgery | Histological and histomorphometric analysis | Celecoxib, ketorolac | Orally for celecoxib, intramuscular injection for ketorolac | Daily | Mouse femoral bone graft | 60% decrease in bone formation of treated groups compared
with controls (P < 0.05). Differences
between bone formation in celecoxib and ketorolac were
insignificant. A lower percentage of histologic bony
bridging (55% for celecoxib, 61% for ketorolac) was observed
in both NSAID-treated groups (P < 0.05)
compared with control (82%). Stopping administration did not allow for a full recovery in bone formation; a 45% reduction in formation remained. |
| 2008 | van der Heide et al. | 0 | 1 | Dutch milk goats | 27 | 6 weeks | Histological and histomorphometric analysis: (fibrous tissue ingrowth and bone ingrowth) | Ketoprofen, meloxicam | Subcutaneously | Daily | Bone chambers (from each side in the cortical bone of the proximal medial tibia) with surrounding cortex | Normal ingrowth pattern was observed in the majority of the
bone chambers (138 specimens), with only 17 of the total
bone chambers having no ingrowth or only fatty tissue and no
fibrous tissue or bone. No significant differences in bone growth (P = 0.5) or fibrous tissue ingrowth (P = 0.6) were observed between the different medication groups. No significant differences in bone ingrowth were found with regard to whether allograft, autograft, or rinsed allograft was used. |
| 2008 | Antoniolli et al. | 0 | 1 | Wistar rats | 120 | 7, 14, and 30 days | Histological and immunohistochemistry | Diclofenac sodium, meloxicam | Intramuscular | Immediate postoperative period and for 6 consecutive days | Rat femurs | NSAIDs delayed the repair of both groups of animals,
including those with bovine devitalized matrix and
autogenous bone graft, while dexamethasone did not (null
hypothesis rejection at 5%). Inflammatory reaction was greatest in animals with bovine devitalized matrix as their bone defect filling, regardless of whether they were treated with NSAIDs or not. All groups, including those treated with NSAIDs and those not treated with NSAIDs, had a significant increase in bone neoformation at all follow-up times. Collagen formation was significantly increased at all time points. |
| 2016 | Soreide et al. | 0 | 1 | Humans | 4144 (treated with NSAIDs) | 2, 5, and 10 years postoperative | Risk of revision and survival, Knee Injury and Osteoarthritis Outcome Score (KOOS), and patient-reported quality of life (QOL) | Diclofenac (91.5% of patients), ketorolac (3.1%), celecoxib (2.4%), other (1.0%) | Not reported | Not reported | N/A | No significant difference in overall graft survival between
the groups who received NSAIDs and those who did not.
Beneficial significant effect on risk of revision in
patients with BPTB grafts (95% CI). Reduced risk of an Osteoarthritis Outcome Score (KOOS)–quality of life (QOL) subscale score <44 at 2-year follow-up was observed for patients who received NSAIDs and statistically significant higher mean scores for all subscales at 2-year follow-up. NSAIDs did not increase risk of revision or increase risk of worse patient-reported knee function. |
BPTB = bone–patellar tendon–bone; hBMSC = human bone marrow stem cells; MSCs = mesenchymal stem cells; N/A, not applicable; NSAIDs = nonsteroidal anti-inflammatory drugs; OA = osteoarthritis; PEO = periosteal endochondral ossification; RT-qPCR = real time quantitative polymerase chain reaction; sGAG = sulfated glycosaminoglycan.