Perioperative Management of Immunosuppressive Medications in Rheumatic Disease Patients Undergoing Arthroscopy

Abstract

Purpose of Review

This manuscript reviews relevant prior literature regarding management of immunosuppressants in patients with rheumatic diseases around the time of orthopedic surgery, highlighting important considerations specifically regarding arthroscopy.

Recent Findings

Utilization rates of arthroscopic surgery in patients with rheumatic diseases are on the rise, as immunosuppressive treatment options enable them to lead more active lives and hence experience more injuries. Physicians regularly manage patients’ glucocorticoids and conventional synthetic and biologic disease modifying antirheumatic drugs around the time of orthopedic surgery, aiming to minimize infection risk while optimizing disease control. However, there is a paucity of randomized controlled trial data for orthopedic surgery—and specifically nothing in the literature pertaining to arthroscopic surgery. Recent guidelines for rheumatic disease patients undergoing elective total hip and knee arthroplasty recommend that most immunosuppressive medications should be held perioperatively, citing the high-risk profile of arthroplasty cases and arthroplasty patients.

Summary

While 2017 societal guidelines for perioperative immunosuppression during arthroplasty currently serve as a guide for physicians, they may not be applicable to arthroscopy. The less aggressive arthroscopic surgeries span a broader range of patient ages and risk profiles, indications for surgery, and procedural complexity and associated risks. Given these considerations, the majority of routine arthroscopic patients may not require holding of their immunosuppressive medications in the perioperative period.

Introduction

Rheumatic diseases including rheumatoid arthritis (RA), systemic lupus erythematosus (SLE), juvenile idiopathic arthritis (JIA), and the spondyloarthropathies (psoriatic, ankylosing, reactive, and inflammatory bowel disease (IBD)-related arthritis) affect more than 7 million people of all ages in the USA. Management of these patients has changed significantly over the last 20 years, well beyond the use of glucocorticoids and conventional synthetic disease-modifying antirheumatic drugs (csDMARDs). Immunosuppression has expanded with the advent of targeted biologics (bDMARDs) and janus kinase (JAK) inhibitors. Most patients with RA are receiving one or more csDMARD or a combination of csDMARDs and bDMARD [1], while 22.5% of RA patients live with disease requiring two or three DMARDs or a DMARD and anti-TNF inhibitor [2, 3], with similar use in other rheumatic diseases.

While the widespread use of these medications has reduced inflammatory articular and extraarticular disease manifestations in rheumatic disease patients, increasing their longevity, it has also improved patients’ physical function [4] and risk for activity-induced injuries. Both the chronicity/longevity as well as the increased injuries have augmented the use of arthroscopic surgery in this cohort. As rheumatic disease patients begin to live healthier, more functional lives [5, 6], we expect to see increased rates of utilization across all arthroscopic procedures.

Although guidelines exist for perioperative management of immunosuppressive treatment around the time of total joint arthroplasty [7], there are no such guidelines for arthroscopy. In fact, there are no recent reports on this topic in the medical literature. More physicians will be tasked with difficult medication management decisions in this population. These considerations underscore the need for further research and establishment of evidence-based guidelines on the topic. The purpose of this manuscript is to review any relevant prior literature regarding perioperative management of immunosuppressants in patients with rheumatic diseases, around the time of orthopedic surgery, and to highlight important considerations specifically regarding arthroscopy.

Risks of Surgery in Rheumatic Disease Patients—Lessons from Arthroplasty

There is a robust body of literature on arthroplasty in patients with autoimmune rheumatic diseases [8•]. These patients, who are on average 65 years of age, experience a higher rate of perioperative complications, infections, and revisions compared to their counterparts with osteoarthritis (not necessarily as a result of continuing or holding immunosuppressive treatment). [7, 9, 10]. Patients with any subtype of inflammatory arthritis undergoing total knee arthroplasty (TKA) have significantly more inpatient complications during their immediate post-operative course [11]. With the exception of psoriatic arthritis patients, these patients also have more orthopedic complications [11]. RA patients undergoing TKA have a higher rate of deep periprosthetic infections and revision rate due to infectious causes [12, 13] and higher risk of dislocation following total hip arthroplasty (THA) [9, 13] compared to OA patients. Data from several smaller studies has suggested similar trends in patients with SLE [11, 14]. While studies have also suggested that long term outcomes after arthroplasty in RA and SLE patients are comparable to OA patients [15, 16], data also show that RA patients are at higher risk of poor postoperative pain control, function, and lower quality of life 2 years after arthroplasty [16, 17]. These risks can be exacerbated both by intrinsic disease processes, and are also known risks of immunosuppressive medications like glucocorticoids [18••] and biologics, which are increasingly being investigated [7, 19, 20, 21••, 22]. When coupled with multiple other factors including age and comorbidities such as, obesity, smoking, and diabetes, the risks of infection are additive [23].

In addition to common postoperative complications, many patients experience perioperative disease exacerbations or flares. Although the etiology of flares is multifactorial, cessation of medication in any patient with established rheumatoid arthritis receiving stable, effective long-term DMARD has been shown to significantly worsen risk of disease flare or deterioration compared to a patient who continues treatment. Stressors like surgery can also contribute to the development of flares, which have been reported within six weeks of TKA and THA in as many as 63% of RA patients. Median time to flare was 2 weeks postop, median duration was 4–7 days, and median severity was 7/10 [24, 25••]. While more patients who had flares used biologics, stopping biologics was not an independent predictor of flares and continuing methotrexate was not protective. Higher baseline disease levels were an independent predictor of flares within 6 weeks postoperatively [25••]. When faced with the perfect storm of stressors like surgery and medication cessation, patients with higher baseline disease levels are likely at a higher risk of experiencing flares and resulting poorer postoperative outcomes. This may suggest a subset of patients who would benefit continuing medications throughout pre and post-operative period.

Immunosuppression in the Perioperative Setting with Arthroplasty

Perioperative management of antirheumatic medications poses the unique challenge of balancing the risk of postoperative infection with the risk of flares. While the goal is to minimize risk of both infection and flares, the decision to continue or to hold a particular medication depends, in part, on the type of medication and the known risks of medication.

There is a growing body of literature (Table 1) investigating the risks of perioperative csDMARDS, bDMARDs (biologics), JAK inhibitors, and glucocorticoids during orthopedic surgery, with a majority focusing on arthroplasty. Perioperative use of immunosuppressants is associated with serious risks including but not limited to superficial and deep wound infection, surgical complication, and delayed wound healing. Results of a systematic review suggest an increased risk of infection in RA patients who use or recently used TNF alpha inhibitors during elective surgery (OR=2.47, p<0.0001) [22]. While the risk of hospitalized infection and prosthetic joint infection within a year of surgery is reportedly similar across all biologics (abatacept, etanercept, adalimumab, infliximab, rituximab, and tocilizumab), glucocorticoid use (especially >10g/day) is associated with a greater, dose dependent risk of postoperative infection (RR=2.98, p=0.05) [18••, 33]. Data regarding perioperative use of csDMARDs is conflicting. In a 2019 meta-analysis, Ibrahim et al. found a significantly increased risk of infection in RA patients on MTX therapy compared to placebo (RR = 1.25, p = 0.04) [34]. In their 1996 clinical trial, Carpenter et al reported 4 postoperative infections in a total of 16 RA patients who continued MTX during their total joint arthroplasty procedures [27]. In contrast, a larger 2001 clinical trial with 388 patients found that RA patients who continued MTX perioperatively had a lower frequency of infection and surgical complications (2%) than those patients who either discontinued (15%) or never used MTX (10.5%) [35]. In RA patients taking leflunomide perioperatively, Tanaka et al. reported no significant increase in risk of postoperative infection complications compared to patients who stopped leflunomide within 4 weeks of total joint arthroplasty (6.1% vs. 6.3%) [36]. Fuerst et al recommended stopping leflunomide therapy in RA patients undergoing all types of elective orthopedic surgery after their 2006 clinical trial found that 13 of the 32 patients taking leflunomide alone or in conjunction with a corticosteroid had wound-healing disturbances [28]. Considering the uncertainty and contradictions in some of the literature paired with the increased risk of complication due to combination therapy [37], further investigation is required to elucidate the impact of immunosuppression in all types of orthopedic surgery.

Table 1.

Current recommendations and evidence on perioperative use of immunosuppressive medication

	Target [26]	ACR/AAHKS recommendations [7]	Current Evidence [8•]
csDMARDs (Conventional Synthetic DMARDs)
Methotrexate (MTX), Sulfasalazine, Hydroxychloroquine, Leflunomide (Arava)	AICAR Transformylase and Dihydrofolate Reductase, NF-kB, TH17 cells, DHODH	Continue without interruption perioperatively.	Typically not associated with increased risk, but contradicting data on whether MTX and Leflunomide (Ariva) increase postoperative infection if continued perioperatively [27, 28].
Azathioprine (Imuran)	purine synthesis	Withhold 1 week prior to surgery.	Recommended to monitor renal function when used perioperatively, but direct evidence is limited [29].
Mycophenolate (Cellcept)	IMPDH	Withhold 1 week prior to surgery.	No direct evidence so decision should be made on an individual basis.
bDMARDs (Biologic DMARDs)
Adalimumab (Humira)	Anti-TNF	Schedule surgery 2 weeks after last administered weekly dose or 3 weeks after 2-week dose.	Non-surgical studies indicate serious risk of infection.
Etanercept (Enbrel)	Anti-TNF	Schedule surgery 2 weeks after last administered dose.	Non-surgical studies indicate serious risk of infection.
Golimumab (Simponi)	Anti-TNF	Schedule surgery 5 weeks after last administered 4-week dose or 9 weeks after last administered 8-week dose.	Non-surgical studies indicate serious risk of infection.
Certolizumab (Cimzia)	Anti-TNF	Schedule surgery 3 weeks after last administered 2-week dose or 5 weeks after last administered 4-week dose.	Non-surgical studies indicate serious risk of infection.
Infliximab (Remicade)	Anti-TNF	Schedule surgery 5 weeks after last administered 4-week dose, 7 weeks after last administered 6-week dose, or 9 weeks after last administered 8-week dose.	Non-surgical studies indicate serious risk of infection.
Abatacept (Orencia)	T-Cell Depletor	Schedule surgery 2 weeks after last administered weekly dose or 5 weeks after last administered monthly dose.	Non-surgical studies indicate serious risk of infection.
Rituximab (Rituxan)	B-Cell Depletor	Schedule 7 months after last administered 4–6-month dose.	Non-surgical studies indicate serious risk of infection.
Tocilizumab (Actemra)	IL-6	Schedule surgery 2 weeks after last administered weekly dose or 5 weeks after last administered 4-week dose.	Non-surgical studies indicate serious risk of infection.
Anakinra (Kineret)	IL-1 Receptor	Schedule surgery 2 days after last administered daily dose.	Non-surgical studies indicate serious risk of infection.
Secukinumab (Cosentyx)	IL-17A	Schedule surgery 5 weeks after last administered 4-week dose.	Non-surgical studies indicate serious risk of infection.
Ustekinumab (Stelara)	p40 subunit of IL-12 and IL-23	Schedule surgery 13 weeks after last administered 12-week dose.	Non-surgical studies indicate serious risk of infection.
Belimumab (Benlysta)	BLyS	Schedule surgery 5 weeks after last administered 4-week dose.	Non-surgical studies indicate serious risk of infection.
Ixekizumab (Taltz)	IL-17A	British Association of Dermatologists suggests consulting physician [30]; FDA notes risk of serious infection in non-surgical use [31].	No direct research
JAK (Janus Kinase) Inhibitors
Tofacitinib (Xeljanz)	JAK	Schedule surgery 7 days after last administered dose.	Serious risk of infection.
Upadacitinib (Rinvoq)	JAK	Withhold 1 week prior to surgery [32].	No direct research
Glucocorticoids
Prednisone, Methylprednisolone		Taper to < 20 mg/day.	High risk of serious infection at doses ≥10 mg/day and possibly even lower.

Existing Guidelines

In 2017, the American College of Rheumatology and American Association of Hip and Knee Surgeons (ACR/AAHKS) published guidelines for perioperative management of anti-rheumatic medication in patients with rheumatic diseases undergoing elective total hip and knee arthroplasty [7]. Per these recommendations, csDMARDS (Hydroxychloroquine, sulfasalazine, methotrexate, leflunomide) should be continued without interruption. Azathioprine, cyclosporine mycophenolate mofetil, tacrolimus should be continued without interruption in severe SLE but should be stopped for 7 days before surgery in non-severe SLE. Biologics and targeted therapies including tumor necrosis alpha (TNF) inhibitors, abatacept, rituximab, tocilizumab, secukinumab, ustekinumab, belimumab, and anakinra should be stopped for 1 dosing interval before surgery. Tofacitinib should be stopped for 7 days before surgery. Glucocorticoids should be tapered to <20 mg/day if possible. Guidelines recommend waiting at least 14 days after surgery to restart therapy that was stopped before surgery in order to ensure proper wound healing and no signs of infection before restarting the medications [7].

While the recommendations were supported by low or moderate quality evidence, several studies since their publication have brought some of their core recommendations into question [8•]. Regardless, they are a step forward in establishing standardized management practices.

Unfortunately, there is a paucity of literature and similar guidelines for the less-invasive arthroscopic surgeries [21••]. Existing arthroplasty guidelines may not be applicable to arthroscopy patients for several reasons: 1) arthroscopy patients are often younger and have fewer comorbidities than arthroplasty patients [7]. 2) There is generally more variability of rheumatic disease activity and infection risk in patients undergoing arthroscopy versus arthroplasty. Some patients have acute inflammation and synovitis that need arthroscopic attention by biopsy or synovectomy, others are well controlled and need treatment for acute trauma/injury unrelated to their inflammatory disease. 3) Arthroscopic procedures themselves range from lower risk (e.g., menisectomies) to somewhat higher risk (e.g., ACL reconstructions), unlike joint replacement surgeries which are generally considered a much higher risk for infection [38].

Patient Risk Profiles

Although immunosuppression from medication raises concern for perioperative risk, other factors specific to RA and also from rheumatic diseases in general must be considered. Risk factors for infection include smoking, obesity, age [39], and a history of previous infection [8•, 38]. An elevated Charleson Comorbidity Index [38], frailty, preoperative hyperglycemia, and preoperative anemia are important risk factors for poor postoperative outcomes [7, 8•]. Disease severity and activity are also risk factors for infection and mortality after surgery, as shown in Lin et al.’s study of SLE patients undergoing major surgery and Cordtz et al.’s study of RA patients undergoing hip or knee arthroplasty [40, 41]. These findings suggest that preoperative optimization of rheumatic disease control with immunosuppressive medications may be important not only for prevention of disease flares, but also for surgical outcomes themselves.

In addition, studies have shown that increased hospital volume and surgeon experience in treating patients with autoimmune diseases are strongly associated with better outcomes in these patients [42, 43].

The rheumatology patient on immunosuppression undergoing arthroscopy can fall anywhere along the spectrum of each of these risks. Although arthroscopy is often associated with low risk patients, these patients are not always risk free, often due to other patient factors. For example, sports medicine patients trend young and free of common comorbidities; 50% are between ages 20 and 29 [44]. Lyman et al. found that patients less than 40 years of age who underwent ACL reconstruction were more likely to undergo another ACL reconstruction than were older patients, and they postulated that this was due to higher postoperative activity levels resulting in graft rupture or lower compliance with rehab protocols and activity restrictions [45]. That said, arthroscopic surgeons see a cohort of patients greater than 40 years of age who present for surgery with comorbidities and age related risk factors that increase their risk profile at the onset.

Indications for Arthroscopic Surgery Impacting Immunosuppression Use

Clinical indications for arthroscopy in rheumatic disease patients are diagnostic arthroscopy, therapeutic arthroscopy for synovectomy, and therapeutic arthroscopy for treatment of sports related injuries.

Diagnostic arthroscopy allows for direct visualization of the synovium and targeted sampling that provides more accurate histopathological diagnosis to confirm (or refute) rheumatic disease. There is increased interest among rheumatologists in diagnostic arthroscopy both for clinical and research purposes. Such procedures can also identify other mechanical causes of symptoms that hadn’t been identified by imaging. Like therapeutic arthroscopy described below, complication rates in diagnostic arthroscopy are low (6 of 342 patients in one recent report) [46, 47]. Given the very low risk profile of the procedure, the risk of infection with continued immunosuppression may be low.

Therapeutic arthroscopy can be indicated for disease control by removal of aggressive synovitis, often in patients with one refractory joint while the overall inflammatory burden is well treated by medication. In recent decades, arthroscopic synovectomy has become accepted as a safe and efficacious treatment for refractory joint swelling in RA and PsA patients [48, 49] in spite of the fact that synovectomy patients often have poorly controlled disease and are taking multiple immunosuppressants. These trends suggest that continuing immunosuppression in synovectomy may be associated with a low risk of infection.

Arthroscopy can also be helpful in this rheumatology population for reasons similar to anyone else, for management of sports injuries or other trauma resulting in soft tissue damage. In the general population, epidemiological studies have shown that arthroscopy patients tend to fall in one of two groups—patients less than 45 years of age with traumatic (including sports injury) or overuse injury requiring arthroscopy, and patients greater than 45 years of age with degenerative disorders requiring arthroscopy [50]. Both these groups utilize upper extremity procedures like carpal tunnel release, rotator cuff repair, and shoulder arthroscopy for indications other than rotator cuff repair. Sports injuries, resulting in loss of function or structure due to participation in sports [51], are most common among children and teenagers [52]. While treatment ranges from conservative therapy to surgery and rehabilitation depending on the type of the injury [53], management with arthroscopic surgery has been increasing. Between 1997 and 2006, the frequency of anterior cruciate ligament reconstructions increased by 21.5% in New York State and by as much as 67.8% nationally [45]. Advances in the management of rheumatic diseases have led to patients being healthier and to increased patient participation in sports and other activities [5, 6]. Repercussions of this participation in the autoimmune population include an increased incidence of sports related injuries like rotator cuff tears, ACL tears or ruptures, and meniscal tears [5, 54] and degenerative diseases like carpal tunnel syndrome. In addition, rheumatic diseases occasionally present as sports related injuries [5]. However, anecdotal evidence suggests that many rheumatic disease patients present to sports medicine practices with stable disease well controlled on their established immunosuppressant medication regimen. They are often young, lack significant comorbid conditions, and are considered low risk surgical candidates. These factors suggest that patients may be at lower risk of infection if maintained on their immunosuppressive medications.

Arthroscopic Procedure Risk Profiles

Various arthroscopic procedures carry different inherent levels of infection risk, and should factor in the consideration of perioperative management of immunosuppression. Low risk surgeries like arthroscopic capsulorrhaphy harbor a postoperative infection rate of 0.16% [55], whereas higher risk surgeries like total joint replacements have an infection risk of 3 to 5% [56]. The overall incidence of infection after arthroscopy is reportedly low, with estimates ranging from 0.01% to as high 3.4% [57]. In their 2021 review of guidelines for perioperative management in rheumatic disease patients, Goodman et al. highlighted that the type of surgery may guide the interpretation of the 2017 recommendations. These authors suggested that lower risk procedures, including certain arthroscopic surgeries, may not require patients to hold their immunosuppressive medications [21••].

Arthroscopic procedures span a wide range of complexity and associated infection risk levels (without considering any immunosuppression), ranging from low (e.g., synovectomies and meniscectomies), to high (e.g., ACL reconstruction and meniscal transplantation) [38]. Results of a retrospective review of a large insurance company database showed significant variation in incidence of infection after arthroscopy from lowest to highest in arthroscopic capsulorrhaphy (0.16%), claviculectomy (0.21%), SLAP tear repair (0.25%), and rotator cuff repair (0.29%) [39]. The authors suggest that these differences may be attributable to procedure specific techniques; rotator cuffs require implant placement in the subacromial space, which has direct communication with the skin through the lateral portal [39]. Similarly, in a study of complications resulting from 92,565 knee arthroscopic procedures obtained from the American Board of Orthopedic Surgery database, Salzler et al also reported significant variation in complication rates after knee arthroscopy from lowest to highest in meniscectomy (2.8%), chondroplasty (3.6%), meniscal repair (7.6%), ACL reconstruction (9.0%), and PCL reconstruction (20.1%). In general, increasing the complexity of the cases corresponded with higher overall complication rates, including infections.

Procedural complexity has been also cited as a confounder in several studies that found younger patients to be at higher risk of complications than older patients undergoing arthroscopy [38, 58]. In their study of risk factors for infection after knee arthroscopy, Clement et al. conducted a subgroup analysis among patients undergoing high and low complexity arthroscopic procedures respectively, and reported an association between younger age and deep, superficial, and total infection among patients undergoing low complexity arthroscopic procedures. This association did not hold true with high-complexity procedures. However, 24% of arthroscopies performed in patients less than 50 years of age were high complexity, whereas only 3% of such procedures were performed in patients 50 years of age or older. Similarly, Salzler et al. found that younger patients experienced a higher complication rate than older patients [58]. Although both authors acknowledged that procedure complexity was likely a confounding factor, the results show that arthroscopic procedures are not without risk, even for young patients with low risk profiles.

It is important to consider the impact of arthroscopy beyond the index surgery. Hip arthroscopy for example has an 11.7% rate of conversion to total hip arthroplasty within 2 years, with a higher risk conversion in patients who are older than 40 or have osteoarthritis or obesity at the time of hip arthroscopy [59]. Although more research is needed to determine a rate of conversion in immunosuppressed patients with autoimmune disease, this known risk increases the stakes of index arthroscopy and related perioperative management.

Summary for Management of Immunosuppression for Rheumatic Disease Patients Undergoing Arthroscopic Surgery

Immunosuppression in patients with rheumatic disease patients around the time of arthroscopic surgery requires balancing the risk of infection with the risk of disease flares. As highlighted in the ACR/AAHKS report on immunosuppression and arthroplasty, infection prevention is a priority in high-risk surgeries involving implants and prosthetic material. However, disease flares from holding immunosuppression can also lead to adverse outcomes in post-operative joints, and medications required later to treat such flares can also contribute to infection risk in those instrumented joints. This challenges the current management paradigm of balancing the risks of immunosuppression against the risks of postoperative complications, as immunosuppression may enable positive postoperative outcomes by reducing disease flares in a subset of patients.

While the ACR/AAKHS infection prevention mandate and the associated guidelines for perioperative immunosuppression during arthroplasty currently serve as a guide for physicians, they may not be applicable to arthroscopy for a number of reasons. Arthroscopy patients carry a wider range of ages and risk profiles, their indications for arthroscopic surgery vary broadly, and the arthroscopic procedures themselves have significant variation in complexity and associated risks.

We surmise that the ACR/AAHKS guidelines for holding immunosuppression may apply at least in part to more involved arthroscopic procedures like ACL reconstructions, even if most patients can continue these medications with lower risk arthroscopic procedures. The guidelines to hold immunosuppression may also apply to those patients with higher risk profiles who are undergoing lower risk arthroscopic surgeries. Consideration beyond the index arthroscopy in procedures with high revision or conversion rates may also be warranted in these patients. Given these factors, we hypothesize that the majority of routine arthroscopic patients may perhaps not require holding of their immunosuppressive medications.

Conclusion

In patients undergoing arthroscopy who are on immunosuppressive medication for rheumatic diseases, the multidisciplinary team is faced with the challenge of preventing infection while maintaining adequate disease control. While such infectious risks are better understood in arthroplasty, the high complexity arthroscopy subset, and patients with more risk factors at baseline for infection, may warrant more caution in their immunosuppressive medication regimens around the time of surgery. At our institution, a leading center in both orthopedics and rheumatology, practicing physicians continue csDMARDs, biologic DMARDS, and low-dose glucocorticoids, and stop high dose glucocorticoids for their routine low risk arthroscopy patients. Although outcomes at our center have not been yet studied, anecdotal reports reveal a minimal incidence of postoperative infection and complications. Overall, more research in arthroscopic patients is needed for an evidence base that can guide their perioperative management of immunosuppression.

Author Contribution

All authors contributed to the study conception and design. All authors read and approved the final manuscript.

Data Availability

NA

Declarations

Conflict of Interest

LMJ receives fellowship support from Arthrex, Smith & Nephew, and AANA. JS and KV have no conflicts of interest to report.

Human and Animal Rights and Informed Consent

This article does not contain any studies with human or animal subjects performed by any of the authors.