JMIR Publications

ABSTRACT

Background:

An estimated 18 million people are living with Rheumatoid Arthritis (RA) in the world (1). The disease comes with significant morbidity for patients, including the increased risk of fractures compared to the general population due to the chronic inflammation associated with RA, which can lead to reduced bone mineral density (2). Presently, disease severity and progression have been helped by the rapid evolution of anti-rheumatic medications. These medications are broadly categorized into two main types: non-steroidal anti-inflammatory drugs (NSAIDs) and disease-modifying antirheumatic drugs (DMARDs) (3). Patients with RA have an increased risk of post-operative complications of orthopaedic surgery because of its chronic impact on bone as well as the use of immunomodulatory medications that may interfere with bone healing (4). Drugs with immunosuppressive action can lead to potential complications with both wound healing and bone union during elective osteotomy. This is particularly important in foot and ankle surgery, where corrective osteotomies are commonly performed, and the risk of wound breakdown is high. Decisions to continue or suspend taking these medications need to be based on evidence weighing the potential for post-operative complications versus potentiating a disease flare. There is an abundance of literature that highlights the adverse effects of NSAIDs on bone healing and fracture union, but there is little robust evidence surrounding the use of DMARDs in orthopaedic surgery (3, 5). Patients requiring surgery in trauma or elective settings will often be on one or more of these medications. Therefore, it is vital to understand the effect of DMARDs on postoperative outcomes to improve their recovery and rehabilitation, and if required, to suspend the medications in the perioperative period (6). Current guidance published by the American College of Rheumatology has been formulated for elective hip and knee surgery with a focus on preventing wound complications, typically restarting DMARDs at the 14-day mark when the wound has healed. These guidelines do not consider the time to union of bone, which is typically 6 weeks. The currently available research has been in vivo or in vitro studies, with little to few studies assessing the clinical implications of DMARDs on bone healing in the rheumatoid patient.

Objective:

This literature review aims to synthesise and evaluate current evidence on the impact of DMARDs on bone healing.

Methods:

A literature search was conducted on PubMed, Embase, and Medline. An initial search was conducted looking at the effect of DMARDs or anti-rheumatic medications on bone healing in foot and ankle osteotomies. We used the keywords ‘DMARDs OR anti-rheumatic medications’, ‘foot and ankle surgery OR osteotomy’, AND ‘Bone healing OR bone union’. It yielded only four original papers for review after removing duplicates, case reports, conference abstracts, and non-English Language material. Due to the limited data in this field, we expanded our search and question to look at the effects of DMARDs on bone union in elective and trauma patients. The keywords were subsequently refined to ‘rheumatic disease OR rheumatoid arthritis’ AND ‘anti-rheumatic medication OR disease-modifying anti-rheumatic medications OR DMARDs’ AND ‘fracture healing OR bony union OR malunion or non-union’. As there was still a limited number of original studies on this theme, we decided to include any study design apart from case studies. These were excluded due to their potential bias and limited generalisability. We also only included papers written in the English language and within the last 50 years. We selected papers that looked specifically at either DMARDs or methotrexate and their effect on bone healing, fracture union, or bone metabolism. The search resulted in 80 papers for review. After applying the above inclusion and exclusion criteria with two independent reviewers, a total of 9 papers were included for a narrative analysis.

Results:

The effect of methotrexate (MTX) on bone appears to be dose-dependent. Satoh et al showed that new bone formation in a fracture gap in rats did not differ significantly between low-dose MTX and control groups (7). However, there was a marked reduction in bone formation in the high-dose MTX group, particularly periosteal bone formation de novo in the fracture gap site in the first week. The study showed no difference between the three groups for intramedullary bone formation or chondroid tissue formation. A key limitation of this study was that it only looked at bone formation rather than bone strength or mineral density. Several other animal studies support the finding that high-dose MTX has a greater adverse effect on bone metabolism than low-dose MTX (8, 9, 10). Pountos et al’s systematic review analysed 70 papers of in vivo and animal studies on the effect of MTX on fracture healing (11). The review gave rise to contradictory evidence. Some in vitro studies concluded that MTX reduces mitochondrial activity, bone cell metabolism, and turnover. Other studies showed no effect on osteoblast proliferation, which is a crucial step in bone healing (3). Some studies also showed there was a reduction in biochemical markers of osteogenesis, such as ALP (alkaline phosphatase), while in others, ALP increased (12). In clinical studies, the impact of DMARDs on bone healing has been studied for patients undergoing elective spinal surgery (13). One study looked at bone fusion rates after craniovertebral junction surgery and found that those who continued DMARDs showed higher radiographic fusion outcomes than those who discontinued (92.8% vs 75%, P value = 0.276). However, the study was not statistically significant due to its small sample size of 30 patients in total (14). Guadiani et al studied revision spinal surgery rates for patients using DMARDs and TNF-alpha inhibitors compared to a control group not on either medication. The reoperation rate within 1 year was 19% for the TNF-alpha inhibitor group and 11% for the DMARD group compared to 6% for the control group. According to the Cox proportional hazard model they used the TNF-alpha group had a 3.1-fold increased risk compared to the control group (95% CI 1.4-7.0), while the DMARD group showed a 2.2-fold increase (95% CI: 0.96-5.3). The reasons for revision surgery were due to infection (40%) or other causes (60%), such as failure to fuse in the DMARDs group, while in the TNF-alpha inhibitor group, it was 47% for infection and 53% for other causes (15). This implies there is a higher rate of infection for the TNF-alpha inhibitor cohort. The authors concluded that continuing DMARDs, especially TNF-alpha inhibitors, 90 days before surgery, appeared to have a higher rate of revision spinal surgery than those who discontinued. In 2017, the American College of Rheumatology and American Association of Hip and Knee Surgery (ACR/AAHK) performed an extensive meta-analysis of the literature around the use of DMARDs in orthopaedic surgery (10). They advised that conventional DMARDs, which include methotrexate, leflunomide, hydroxychloroquine, and sulfasalazine, can be continued in orthopaedic surgery as they did not lead to adverse post-operative outcomes. However, they recommended holding biologics for two weeks before surgery as there was an increased risk of poor wound healing. The effect on bone healing itself was not studied in this review. There is, in fact, very limited data on the effect of biologic DMARDs on bone healing, but in an in vivo study, they have shown an inhibitory effect on osteoblast proliferation (3). This is particularly true of TNF-alpha inhibitors such as infliximab, which showed a reduction in overall osteoblast cell numbers. This suggests it could interfere with the bone repair and remodelling (3). Furthermore, the 2021 critical analysis review by Saunders et al on the perioperative management of antirheumatic drugs in foot and ankle surgery also concluded that conventional DMARDS are generally safe to use throughout the perioperative period, while biologics should be held typically before surgery (16).

Conclusions:

Our narrative review has highlighted an important literature gap within the field of DMARDs and bone healing, whether in a traumatic or elective setting. Much of the original research is in vivo or animal studies, and although they show statistically significant results, they cannot accurately predict human outcomes due to significant differences in physiology and biology (3,8,9,12). Clinical studies are even fewer, and the ones conducted so far have included small study populations. Moreover, they are antiquated and often do not examine the latest anti-rheumatic drugs. For instance, an important study we included, Elia et al, included only 30 patients, which reduced the statistical power of the results (14). All the clinical studies we have included so far are for elective procedures such as spinal surgery or foot and ankle surgery (13,14, 16, 17). To our knowledge, there are currently no randomised controlled trials that study the effect of DMARDs on bone healing, either in a trauma or elective setting. Nevertheless, there is a greater number of publications available to consider for the effect of DMARDs on wound healing in orthopaedic surgery. This is of important consequence as surgical site infections, especially when involving the bone, can lead to impaired fracture healing, causing malunion or non-union (18). Current evidence suggests MTX has no adverse effect on wound healing in orthopaedic surgery and can be safely continued pre- and postoperatively (10,11). However, biologics are recommended to be held perioperatively due to the increased risk of surgical site infections and impaired wound healing. The current guidance is to schedule surgery at the end of their dosing cycle (10). Some hospital trusts have advised only restarting when most of the wound has healed (19). Considering a wider evidence base, biologics have shown an increased risk of serious infection, so there is certainly a research gap to explore on how these medications impact patient outcomes in orthopaedic surgery (20). Any decision to stop anti-rheumatic medications in the preoperative period should be a carefully considered decision, with patients fully informed as to the risks and benefits of stopping such therapy. Patients on DMARDs tend to have more severe disease, and withholding them may result in disease flares, which can cause significant morbidity. Flares may lead to joint swelling, stiffness, pain, and increased cardiovascular risk (21). This can ultimately impair rehabilitation following major surgery, which predisposes the patient to further post-operative complications such as venous thromboembolism, hospital-acquired infections, or reduced functional baseline from a prolonged hospital stay (22). Grennan et al found that those who discontinued MTX two weeks before and after surgery showed a higher rate of flare-ups than those who continued their medication. Patients who continued MTX before surgery had even fewer post-operative complications than the control group that was not on any MTX (23). The 2017 American College of Rheumatology study also concluded that continuing glucocorticoids and DMARDs perioperatively for hip and knee arthroplasty resulted in better function, a greater range of motion, and improved post-operative pain (10). Therefore, we advise that the decisions around anti-rheumatic medications in patients undergoing orthopaedic surgery should be determined on an individual basis, with consideration given to their disease severity, functional baseline, and risk factors for poor bone healing, as we currently do not have enough evidence to suggest that they should be held. Our literature review, however, has some limitations. Firstly, we used specific terminology to capture the effect of anti-rheumatic medications on bone healing, so we may have missed articles that contain this information, which did not include our keywords. Secondly, there is such little data available for our topic that the papers we have selected for review have small study populations or no controls. None of the studies we included showed any randomisation. Therefore, results were interpreted with caution as there is a potential for bias and reduced generalisability. Finally, many papers we included were animal studies, so their findings cannot be applied directly to humans. In conclusion, the effect of DMARDs on bone union remains largely unstudied, especially considering human studies and large randomised controlled trials. Our literature review has identified that MTX may be safe to continue before orthopaedic surgery, as it does not appear to affect bone union at low doses that are used in RA. However, biologics should be withheld as there is evidence to suggest they can cause an increased risk of infection or wound breakdown. The effect of biologics specifically on bone healing, has not been studied to our knowledge. Given that millions of patients suffer from rheumatoid arthritis, and many will at one point undergo a joint procedure, it is important to further understand the clinical impact of DMARDs on bone so we can recommend evidence-based guidance. Until then, we advise a multi-disciplinary approach in determining which anti-rheumatic medications to withhold before any orthopaedic surgery.