Glucocorticoids have broad inhibitory effects on specific immune responses mediated by T- and B-cell lymphocytes as well as potent suppressive effects on the effector functions of monocytes and neutrophils. Although these attributes coupled with their rapid onset of action render steroids extremely valuable in suppressing undesired inflammatory processes, corticosteroids have similar broad effects on the function of cells comprising other organ systems. The immunocompromised status and catabolic consequences associated with use of corticosteroids limit their long-term use in high doses and dictate the need for careful surveillance and preventive interventions to avoid undesired complications.

Use of high doses of corticosteroids first and foremost requires vigilance for the development of intercurrent infections. Patients with either rheumatoid arthritis or systemic lupus erythematosus (SLE) have an intrinsic susceptibility to infection, and the administration of glucocorticoids enhances the risk of infection. In addition to typical bacterial organisms, infections with mycobacteria, cryptococci, listeria, and nocardia have been associated with corticosteroid therapy. The combination of steroid use with cytotoxic agents, such as cyclophosphamide, has been associated with higher risk of infection with Pneumocystis jirovecii pneumonia, most notably among patients with lymphopenia. Unless life- or organ-threatening disease complications dictate otherwise, in the setting of serious intercurrent infection, doses of corticosteroids should be attenuated to that required to avoid adrenal crisis.

Given the significant catabolic effects of glucocorticoids on muscle, skin, and bone, patients taking moderate or high doses of steroids for prolonged intervals require periodic assessment for the evolution of steroid myopathy or
development of steroid-induced osteoporosis. Since steroid myopathy most commonly affects the proximal hip-girdle musculature, assessment of hip-girdle strength by having the patient squat or arise from a chair unassisted are simple maneuvers that can be employed during clinic visits. Corticosteroid-induced muscle wasting and weakness may be difficult to distinguish from inflammatory muscle diseases for which they are prescribed. Muscle tenderness and elevation in creatine kinase favor the presence of active myositis. On muscle biopsy, loss of type I and type II fibers as well as vacuolar changes may be observed in steroid-induced myopathy or myositis.

Periodic assessment for osteoporosis is now a standard of care for patients on chronic corticosteroids. The employment of alternate-day dosing regimens does not appear to confer protection from steroid-induced osteopenia. Exogenous administration of calcium and vitamin D may suffice to protect patients from steroid-induced osteopenia. Glucocorticoid-induced suppression of adrenal dehyrdoepiandrosterone (DHEA) production may render women at increased risk for the catabolic effect of steroids on bone, but a role for DHEA administration in the prevention of bone complications has not yet been confirmed. Bisphosphonates (alendronate, risedronate, ibandronate, and zoledronic acid) and have emerged as proven therapies for the prevention and treatment of glucocorticoid-induced osteoporosis (1). Periodic assessment of bone density at 1- to 2-year intervals is recommended to assess the efficacy of these interventions in patients on chronic steroid therapy.

The predictable metabolic consequences of steroids include salt and water retention as well as variable degrees of insulin resistance with hyperglycemia. The mineralocorticoid effects of steroids warrant expectant observation for the development of hypertension or heart failure exacerbations in patients who have or are at risk for these cardiovascular disorders. Long-term metabolic consequences of corticosteroid use in patients with rheumatic disease may include accelerated atherogenesis. Attention to other cardiovascular risk factors including assessment for and treatment of hypercholesterolemia may slow the progression of atherogenesis and lower the risk for vascular events in patients who require long-term steroid use for management of rheumatic disease manifestations.

Other complications of corticosteroid therapy are less predictable but nonetheless require vigilance for their occurrence so as to avoid unfavorable outcomes. Corticosteroids may have untoward effects on the central nervous system, including emotional irritability, difficulty in concentration, depression, confusion, or psychosis. High-dose corticosteroids therapy has been implicated as a possible inducer of pancreatitis. Since pancreatitis may be a manifestation of lupus, the occurrence of pancreatitis in patients with lupus receiving glucocorticoid therapy may result in a therapeutic dilemma. Osteonecrosis is a recognized complication of high-dose steroid use. In patients with lupus, other disease-related factors may account for the development of osteonecrosis, but the incidence appears to correlate with the cumulative steroid dose. Since routine radiographs typically fail to reveal the presence of osteonecrosis during its early stages, patients on high doses of steroids who develop otherwise unexplained pain in the shoulders, hips, knees, or ankles should be evaluated with magnetic resonance imaging to rule out the presence of osteonecrosis.

Nonsteroidal anti-inflammatory drugs constitute the most frequently prescribed class of medication used in the treatment of patients with rheumatic disorders. A rapid onset of action and their combined analgesic or anti-inflammatory attributes render NSAIDs very useful in the management of rheumatic disease. Although a number of cellular effects distinct from those related to prostaglandin
production have been described for various NSAIDs, the major therapeutic effect of NSAIDs relates to their ability to inhibit cyclooxygenase-mediated synthesis of prostaglandins, affecting vascular permeability and hyperalgesia. However, prostaglandins generated by cyclooxygenase also play an important role in hemostasis, in maintaining the integrity of the intestinal mucosa, and in regulating renal blood flow. These physiologic effects of prostaglandins account for the majority of NSAID side effects and toxicity, most notably bleeding, intestinal ulceration, azotemia, and retention of salt and water.

Certain toxic effects of a given NSAID may be governed by its specificity for the respective isoforms of cyclooxygenase, COX-1 and COX-2. COX-1 is expressed constitutively in most organ systems and is the isoform primarily responsible for synthesis of prostaglandins maintaining the integrity of the gastrointestinal (GI) mucosa and the hemostatic function of platelets. COX-2 is primarily induced and expressed in response to cytokines at sites of tissue injury and inflammation and is not expressed in platelets. Traditional nonselective NSAIDs inhibit both COX-1 and COX-2, whereas celecoxib selectively inhibits COX-2, substantially sparing activity of COX-1.

Monitoring of patients taking NSAIDs, particularly those not selective for COX-2, entails careful attention to symptoms referable to the GI tract and the possibility of bleeding complications. As the majority of NSAID-induced ulcerations are silent, periodic assessment of the hematocrit and red cell indices are prudent in patients taking NSAIDs for extended durations. Although there are no published studies to provide guidelines for how frequently such monitoring should occur, risk factors for NSAID-induced GI bleeding and perforation are now well recognized (Table 28.1) (2,3), and the presence of these risk factors in a given patient should guide the frequency of blood count or hemoccult monitoring.

Both COX-1 and COX-2 are constitutively expressed in the kidney and generate prostaglandins (PGE₂ and PGI₂) that regulate renal blood flow under conditions of volume contraction and/or decreased effective arterial blood volume. PGE₂ and PGI₂ furthermore stimulate secretion of renin with attendant release of aldosterone and potassium secretion. Accordingly, diminution in GFR with salt
and water retention and/or hyperkalemia may occur as a consequence of treatment with either nonselective or COX-2 selective NSAIDs. Patients with preexisting renal disease or diminished effective arterial blood volume (congestive heart failure, cirrhosis, and renal vascular disease) are at particular risk for effects of NSAIDs on glomerular perfusion. Effects of NSAIDs on GFR may cause significant complications in patients with diabetes with type IV renal tubular acidosis (hyporeninemic hypoaldosteronism), as the attendant inhibition of renin release accompanied by diminution of salt load to distal nephrons may precipitate significant hyperkalemia. Careful monitoring for fluid retention and elevations of creatinine or potassium should be undertaken in these at-risk patient populations within several days of instituting treatment with an NSAID.

In addition to their potential effects on glomerular perfusion and renin secretion, NSAIDs may induce idiosyncratic, drug-specific complications of interstitial nephritis. While this complication may occur with any NSAID, interstitial nephritis has been reported most commonly in patients receiving fenoprofen. Although an appropriate frequency of monitoring renal function in patients taking NSAIDs has not been established by relevant outcome studies, at least semiannual assessment of creatinine and urinalysis is prudent for patients on long-term NSAID therapy to minimize the risk of permanent kidney damage from drug-induced interstitial nephritis.

Use of one or more disease-modifying antirheumatic drugs (DMARDs) is now the standard of care for patients with active rheumatoid arthritis. Many DMARDs, including methotrexate, hydroxychloroquine, azathioprine, cyclosporine, and mycophenolate are used to manage manifestations of diseases other than rheumatoid arthritis, including lupus and polymyositis. Use of DMARDs entails titration of the dose to achieve the desired clinical benefit without inducing toxicity. Selection and successful use of a DMARD or DMARD combination for a given patient rests upon multiple clinical considerations, including stage and activity of the disease, patient comorbidities, concurrent medication use,
and the known side-effect profiles of the respective DMARDs. Monitoring for DMARD toxicity and side effects is therefore critical to the appropriate use of these drugs.