Cardiovascular.

Data from several clinical trials and observational studies in adults have suggested that there is an increased risk of cardiovascular toxicity associated with several NSAIDs and COX-2 inhibitors. Cardiovascular toxicity not only led to the withdrawal of rofecoxib and valdecoxib from the market but also resulted in more restricted, similar product labels in the United States for celecoxib and traditional NSAIDs.

Meaningful data in children are scarce, so pediatric rheumatologists have traditionally relied on adult data. Consideration of the underlying cardiovascular risk of the patient, including the rheumatic disease being treated, is likely to enter into the calculation. In adults with rheumatoid arthritis (RA) and osteoarthritis, COX-2 inhibitors are recommended to be administered with low-dose aspirin in patients with cardiac risk factors. However, a recent multicenter, prospective, observational registry of 274 JIA patients receiving NSAIDs (55 receiving celecoxib) for a total of 410 patient-years (PY) of observation, revealed no difference in adverse events between nonselective NSAIDs and celecoxib. The two reported cardiovascular events were observed in the nonselective NSAID group.

Gastrointestinal.

GI toxicity is common to all NSAIDs. The pathogenesis of gastroduodenal mucosal injury involves multiple mechanisms and ranges from mild epigastric discomfort to symptomatic or asymptomatic peptic ulceration.

The average relative risk of developing a serious GI complication in adult patients exposed to NSAIDs is fivefold to sixfold that of patients not taking NSAIDs. Possible risk factors for GI complications during NSAID therapy include advanced age, past history of GI bleeding or peptic ulcer disease, and cardiovascular disease. Most patients who have a serious GI complication requiring hospitalization have not had prior GI side effects, however. Additional risk factors include longer disease duration, higher NSAID dose, use of more than one NSAID, longer duration of NSAID therapy, concomitant glucocorticoid or anticoagulant use, and serious underlying systemic disorders. Infection with Helicobacter pylori does not seem to play a major role.

The magnitude of this problem in children is poorly documented but has traditionally been thought to be considerably less than in adults, partly because of the absence of the associated risk factors identified in adults. H. pylori has not been reported to be an important pathogen in children with JRA treated with NSAIDs. Studies in children confirm that although mild GI disturbances are frequently associated with NSAID therapy, the number of children who develop clinically significant gastropathy is low. A retrospective study of a cohort of 702 children receiving NSAID therapy for JRA who were monitored for at least 1 year found 5 children (0.7%) with clinically significant gastropathy defined as esophagitis, gastritis, or peptic ulcer disease. The retrospective nature of this study may have resulted in underestimation of the prevalence of NSAID-associated gastropathy. A prospective study of a cohort of 203 children found that although 135 children (66.5%) had documented GI symptoms at some point during NSAID therapy, only 9 (4.4%) had endoscopically detected ulcers or erosions. A prospective study reported 45 children (24 of whom who were symptomatic with abdominal pain) who underwent routine endoscopy (in association with general anesthesia for joint injections). Of these children, 19 (42%) had normal gastric and duodenal mucosa, and 20 had histologically mild gastritis. A clear association was seen between abdominal pain and gastroduodenal pathology, but the severity of gastric inflammation did not correlate with the duration of NSAID therapy. The phase IV safety registry of celecoxib and nonselective NSAIDs revealed no evidence of GI ulcer and one report of gastritis in the nonselective NSAID group.

Studies have shown differences in rates of serious GI complications associated with different NSAIDs. Systematic reviews have found ibuprofen to be associated with the lowest risk; indomethacin, naproxen, sulindac, and aspirin with moderate risk; and tolmetin, ketoprofen, and piroxicam with the highest risk. GI symptoms can be minimized further by ensuring that NSAIDs are always given with food. The utility of antacids and histamine ₂ -receptor antagonists for prophylaxis against serious NSAID-induced GI complications is controversial. Although these medications suppress symptoms, they do not prevent significant GI events such as endoscopically documented gastric ulcers. Asymptomatic patients on acid-reduction therapies seem to be at greater risk for serious GI complications than patients not taking these medications, so their routine use in asymptomatic patients receiving NSAIDs cannot be recommended. Misoprostol, a synthetic prostaglandin E ₁ analogue, has been shown in adults to be effective in prophylaxis and treatment of NSAID-induced gastroduodenal damage, thereby allowing continuation of NSAID therapy while achieving the healing of an ulcer. Studies of misoprostol cotherapy in children are limited, but they also suggest that misoprostol may be effective in the treatment of GI toxicity symptoms in children receiving NSAIDs. Omeprazole, a proton pump inhibitor, has been shown to be superior to ranitidine and misoprostol for the prevention and treatment of NSAID-related gastroduodenal ulcers in adults.

Hepatotoxicity.

Hepatitis with elevation of transaminase levels can occur with any NSAID but has most commonly been reported in children with JRA receiving ASA. In one retrospective study, transaminase levels were increased in 6% of children receiving naproxen. Elevated transaminase levels are rarely of clinical significance and often resolve spontaneously. However, when they are greater than twice the upper limit of normal, or when present for prolonged periods of time without resolution, it may be necessary to reduce the dose or temporarily stop NSAID therapy. Rarely, hepatotoxicity is severe; NSAIDs have been associated with macrophage activation syndrome (MAS). Liver function should be monitored in children taking daily NSAIDs for extended periods, particularly children with systemic JIA.

Renal.

Several types of renal complications have been associated with NSAID therapy, including reversible renal insufficiency and acute renal failure; acute interstitial nephritis; nephrotic syndrome; papillary necrosis; and sodium, potassium, and water retention. Although more common in adults, cases have been described in children. A 4-year prospective study of 226 children with JRA treated with NSAIDs found the prevalence of renal and urinary abnormalities attributable to NSAID therapy to be only 0.4% ; an even lower prevalence of 0.2% was reported in another cohort of 433 children.

Central nervous system effects.

Three general categories of central nervous system (CNS) side effects have been reported in association with NSAID therapy in adults: (1) aseptic meningitis, (2) psychosis, and (3) cognitive dysfunction. The NSAID most commonly reported to cause aseptic meningitis has been ibuprofen; susceptibility seems to be greater in patients with SLE. Indomethacin and sulindac have been reported to induce psychotic symptoms, including paranoid delusions, depersonalization, and hallucinations, in a few patients. More subtle CNS effects, such as cognitive dysfunction and depression, can also occur and are probably underrecognized and underreported. Tinnitus may occur with any NSAID, but particularly with ASA. A prospective study of 203 children with JRA found that CNS symptoms occurred in 55% of patients receiving NSAIDs; the most common symptom was headache, which occurred in about one third of children. Other reported symptoms included fatigue, sleep disturbance, and hyperactivity.

Cutaneous toxicity.

A diverse group of skin reactions, including pruritus, urticaria, morbilliform rashes, erythema multiforme, and phototoxic reactions, have been described. The syndrome of pseudoporphyria that occurs in association with naproxen therapy in children with JRA is a distinctive photodermatitis marked by erythema, vesiculation, and increased skin fragility characterized by easy scarring of sun-exposed skin ( Fig. 12-1 ). In spite of the name, porphyrin metabolism is normal. All findings except scarring resolve with discontinuation of naproxen, but the vesiculation may persist for several months. Children with fair skin and blue eyes are particularly susceptible; one study reported a relative risk of 2.96 if the child had blue-gray eyes and was taking naproxen. In one retrospective and parallel prospective study, young age, JIA itself, duration of therapy, evidence of systemic inflammation, and concurrent antimalarial therapy seemed to be additional risk factors for naproxen-induced pseudoporphyria. It is also rarely reported with other NSAIDs.

A and B, Distant and close-up views of the face of an 8-year-old boy with pseudoporphyria who was taking naproxen. Note a blistered lesion adjacent to a superficial scar. Superficial scars are also visible on the nose.

Effects of coagulation.

NSAIDs decrease platelet adhesiveness by interfering with platelet prostaglandin synthesis. This inhibition is reversible in the case of all NSAIDs except ASA, which irreversibly acetylates and inactivates COX, an effect that persists for the life of the platelet; bleeding time returns to normal only as new platelets are released into the circulation. NSAIDs also displace anticoagulants from protein-binding sites, potentiating their pharmacological effect.

Hypersensitivity and miscellaneous effects.

The precipitation of asthma or anaphylaxis with NSAIDs has been reported in adults as a unique syndrome associated with nasal polyps. Although this syndrome can theoretically be provoked by any NSAID, it has most commonly been reported with ASA or tolmetin. True hypersensitivity to ASA is exceedingly rare in childhood. ASA hypersensitivity occurs in about 0.3% to 0.9% of the general population, in 20% of patients with chronic urticaria, and in 3% to 4% of patients with chronic asthma and nasal polyps.

Hematological toxicity, including aplastic anemia, agranulocytosis, leukopenia, and thrombocytopenia, has been reported but is uncommon. Mild anemia occurs in about 2% to 14% of children and may be due partly to hemodilution, hemolysis, or occult GI blood loss secondary to NSAID therapy.

Salicylism.

Symptoms of salicylism include tinnitus, deafness, nausea, and vomiting. Early on, there is CNS stimulation (hyperkinetic agitation, excitement, maniacal behavior, slurred speech, disorientation, delirium, convulsions). Later, CNS depression (stupor and coma) supervenes. There is a narrow margin between therapeutic and toxic levels. In Kawasaki disease, hypoalbuminemia may predispose children to salicylate toxicity due to increased free levels of drug. The reader is referred to the recommendations of Mofenson and Caraccio for details of the management of severe salicylate poisoning.

Mechanism of action.

MTX ( Fig. 12-2 ) is a folic acid analogue and a potent competitive inhibitor of several enzymes in the folate pathway ( Fig. 12-3 ). MTX is absorbed via the proton-coupled folate transporter ( PCFT / SLC46A1 ) in the gut and enters the cells primarily through the reduced folate carrier ( RFC/SLC19A1 ) and folate receptors (FOLR) 1 and 2. Intracellularly, MTX is bioactivated to a polyglutamated (MTXGlu _n ) form by folylpolyglutamyl synthase (FPGS), which enhances the pharmacological activity and intracellular retention of MTX. The first MTX target to be identified was dihydrofolate reductase (DHFR), the enzyme responsible for reducing dietary folates and dihydrofolate to the biologically active tetrahydrofolate. Tetrahydrofolate is the source of one carbon donors supporting the synthesis of thymidylate, purines, and serine, as well as the remethylation of homocysteine to form methionine and subsequently S- adenosylmethionine (SAM), the one-carbon donor for multiple methyltransferase enzymes. Additionally, MTX inhibits thymidylate synthetase (TYMS) directly and indirectly via depletion of tetrahydrofolate, leading to inhibition of pyrimidine (thymidylate) biosynthesis with a resultant antiproliferative effect. Importantly, MTX targets aminoimidazole carboxamide ribonucleotide (AICAR) transformylase (gene name, ATIC ), which inhibits de novo purine synthesis and promotes the accumulation of extracellular adenosine. Extracellular adenosine is thought to be a large contributor to the site-specific antiinflammatory effects of MTX through inhibition of neutrophil adherence. Pharmacogenomic studies in RA and JIA have provided additional support for the involvement of the purine synthesis and adenosine pathways in mediating MTX response. Glutamate residues are removed from MTXGlu _n by gamma-glutamyl hydrolase (GGH), allowing efflux of MTX from the cell by the ATP-binding cassette (ABC) family of transporters.

Structures of folic acid and methotrexate, with notable structural differences in red.

Intracellular folate pathway. The red dotted lines and squares denote known enzymes inhibited by methotrexate (MTX). MTX acts as a folate antagonist, entering the cells through the reduced folate carrier (SLC19A1). Once intracellular, MTX is bioactivated to methotrexate polyglutamates (MTXGlun) by folylpolyglutamyl synthase (FPGS). No or low glutamation, facilitated by the deglutamating enzyme g-glutamyl hydrolase (GGH), leads to the efflux of MTX by the ATP-binding cassette (ABC) family of transporters. MTX’s initial enzymatic target was identified as dihydrofolate reductase (DHFR), important in the formation of tetrahydrofolate (THF). The list of target genes has been extended to include aminoimidazole carboxamide ribonucleotide (ΑICΑR) transformylase (gene name, ΑTIC ), and thymidylate synthetase (TYMS). Additional endogenous enzymes in the folate pathway include methylenetetrahydrofolate dehydrogenase (MTHFD1), methylenetetrahydrofolate reductase (MTHFR), methionine synthase (MTR), methionine synthase reductase (MTRR), S-adenosylmethionine (SAM), S-adenosylhomocysteine (SAH), glycinamide ribonucleotide transformylase (GART), serine hydroxymethyltransferase (SHMT), and folate hydrolase 1 (FOLH1). Folate isoforms and their polyglutamated states are represented as: tetrahydrofolate (THFGlun), 10-formyl-tetrahydrofolate (10-formyl-THFGlun), 5,10-methenyltetrahydrofolate (5,10-methenyl-THFGlun), 5,10-methylene-tetrahydrofolate (5,10-methylene-THFGlun), and 5-methyl-tetrahydrofolate (5-methyl-THFGlun). The mitochondrial folate pathway produces a formic acid for utilization in de novo purine synthesis. SLC25A32 is a mitochondrial specific folate transporter. The bifunctional methylenetetrahydrofolate dehydrogenase 2 (MTHFD2) and methylenetetrahydrofolate dehydrogenase 1-like (MTHFD1L) in mitochondria replicate the function of cytosolic MTHFD1.

MTX also modulates the function of many of the cells involved in inflammation and affects the production of various cytokines, including the reduction of TNF-α, interferon-γ (IFN-γ), IL-1, IL-6, and IL-8 production, thereby acting as a potent inhibitor of cell-mediated immunity. By reducing the expression of adhesion molecules on endothelial cells, MTX may reduce the permeability of the vascular endothelium. In addition, adenosine inhibits adherence of stimulated neutrophils to endothelial cells, protecting the vascular endothelium from neutrophil-induced damage. MTX may also have more direct effects in inflamed joints by inhibiting the proliferation of synovial cells and synovial collagenase gene expression.

Pharmacology.

There is significant intraindividual and interindividual variability in the absorption and pharmacokinetics of MTX after oral administration. On average, oral bioavailability is about 0.70 (compared with intravenous dosing) and highly variable, ranging from 0.25 to 1.49, with 25% of subjects in one study absorbing less than half their dose. In adults with RA, factors such as age, body weight, creatinine clearance, sex, dose, and fed-versus-fasted state significantly influenced MTX disposition. The bioavailability of MTX has also been shown to be greater in the fasting state in children with JIA. Oral bioavailability is generally about 15% less than after intramuscular or subcutaneous administration, and oral absorption is saturable ( eFig. 12-5 ).

Bioavailability of oral and subcutaneously administered methotrexate. AUC , Area under the curve.

(Adapted and redrawn from C.A. Wallace, New uses of methotrexate, Contemp Pediatr 11: 43–53, 1994.)

After a single dose of MTX, the drug is present in the circulation for a short period before it is redistributed to the tissues ( Fig. 12-4 ). Peak serum levels are reached in approximately 1.5 hours (range 0.25 to 6 hours), with elimination half-life being approximately 7 hours in subjects with normal renal function. Circulating levels diminish rapidly as the drug is distributed into tissue and eliminated. The predominant route of elimination is renal, with more than 80% of the drug eliminated unchanged via glomerular filtration and tubular secretion within 8 to 48 hours. A smaller but significant route of elimination is the biliary tract. The pharmacokinetics of MTX are triphasic. The initial rapid phase represents tissue distribution and renal clearance; the second phase is prolonged because of slow release from tissues, tubular reabsorption, and enterohepatic recirculation; the third phase is flat, reflecting the gradual release of tissue MTX.

Time course of methotrexate (MTX) and 7-hydroxymethotrexate (7-OH-MTX) after an oral dose of 15 mg.

(Redrawn from J.L. Hillson, D.E. Furst, Pharmacology and pharmacokinetics of methotrexate in rheumatic disease: practical issues in treatment and design, Rheum Dis Clin North Am 23: 757–778, 1997.)

Plasma drug levels do not correlate well with clinical effects and are not useful in routine monitoring of MTX therapy. The pharmacokinetics of oral MTX in JIA seem to be age dependent, with more extensive metabolism of MTX in younger children. This difference may account for the observation that children require higher doses of MTX than adults to obtain similar therapeutic effects. As MTX is cleared rapidly from serum, attention has turned to measurement of intracellular concentrations of the therapeutically active polyglutamated forms of MTX (MTXGlu _n ) as more stable and reliable biomarkers of the effect of MTX. Higher concentration of MTXGlu _n have been shown to correlate with drug efficacy in patients with RA, although similar conclusions have not been consistently seen in children with JIA. MTXGlu _n concentrations have been shown to be quite variable in JIA and are associated with drug dose, route, and duration of MTX therapy. Accumulation of long-chain MTXGlu _n have been shown to be higher in children who experience GI side effects with the drug, and although cross-sectional studies have found no association with drug effectiveness, a recent prospective report has shown that responders to MTX had higher concentrations of long-chain MTXGlu _n .

At low doses, MTX is only moderately protein bound (11% to 57%), so the potential for interactions with other protein-bound drugs is small and usually is not clinically significant. Several studies in children have shown an interaction between MTX and NSAIDs that may be clinically significant, particularly in patients with renal dysfunction. The combination of MTX and trimethoprim-sulfamethoxazole should be avoided because it may lead to hematological toxicity through the synergistic effects of these drugs on dihydrofolate reductase within the folate pathway.

The effect of genetic variation within the folate pathway upon drug response has been a focus in adult RA, and several review papers have discussed the vast work for the interested reader. However, there has been a lack of reproducibility of results, likely a consequence of several factors including variability in MTX treatment regimens, outcome measurements, and folate supplementation between studies; small sample sizes; the under characterized role of MTXGlu and folate on drug metabolism; and the unknown functional impact of the genetic polymorphisms. In children, recent work has identified clinical outcomes associated with genetic variation in the folate pathway, specifically in genes involved in the purine synthesis portion of the pathway as well as cellular transporters of MTX and folate ; however, these findings have not yet been replicated in larger cohorts.

Efficacy.

The efficacy of MTX in controlling the signs and symptoms of JIA is now well established. Benefits reported in countless initial retrospective and uncontrolled studies were subsequently confirmed in a randomized, placebo-controlled clinical trial.

There have been various attempts at identifying the clinical predictors of response to MTX in children with JIA, with regards to the specific JIA subtype. There are data to support MTX as an effective therapy in extended oligoarticular JIA more than several other JIA subtypes. Although there have been studies to support its effectiveness in systemic JIA, there are have been others that show no effect, or even worsening in patients on MTX therapy in this subtype of JIA. The utilization of MTX also differs by subtype in clinical practice. In a recent report from the Childhood Arthritis and Rheumatology Research Alliance (CARRA) registry, JIA patients with oligoarticular JIA (53%) and enthesitis-related arthritis (ERA) (63%) were the least likely to ever receive nonbiologic DMARDS such as MTX, compared to RF+ polyarticular (91%) or extended oligoarticular JIA patients (89%). Recently published American College of Rheumatology (ACR) recommendations for treatment of JIA utilize MTX differently by JIA subtype as well. Although in most subtypes a “ step-up”/escalation approach is recommended, in patients with high disease activity, poor prognostic, or systemic features, MTX (or biological therapy for systemic JIA) is recommended to be used earlier, and some advocate using it as first-line therapy. In fact, with newer cytokine-targeting therapies now available, the most recently updated ACR recommendations for the treatment of systemic JIA suggest MTX primarily for the treatment of mild or moderate arthritis in systemic JIA rather than for the treatment of systemic features or MAS where it has been shown to be less effective.

MTX is also used in many other rheumatic disorders, including systemic lupus erythematosus (SLE), some vasculitides, sarcoidosis, systemic sclerosis, localized scleroderma, and uveitis.

Dosage, route of administration, and duration of methotrexate therapy.

Standard effective dosing regimens of MTX in children with JIA are 10 to 15 mg/m ² /week or 0.3 to 0. 6 mg/kg/week ( Table 12-2 ). Improvement is generally seen by about 6 to 8 weeks on effective doses, but may take up to 6 months to see the full effect. Children seem to tolerate much higher doses than adults, and some series have described using 20 to 25 mg/m ² /week or 1. 1 mg/kg/week in children with resistant disease, with relative safety in the short term. Early reports supporting the efficacy of higher dosing regimens (25 to 30 mg/m ² /week) for JIA have been followed with studies that do not support additional gains with higher doses. However, higher dosages of MTX (1 mg/kg/dose up to 40 mg weekly) have been used in other disease processes such as juvenile dermatomyositis and juvenile localized scleroderma.

Many pediatric rheumatologists advocate using parenteral MTX at initiation of treatment to ensure complete absorption and achievement of early disease remission ; the 2011 ACR recommendations for treatment of JIA assumes MTX dosing to be 15 mg/m ² /week administered via the parenteral route. However, there remains variability in clinical practice. F or example, approximately one quarter of the MTX users in the CARRA registry received MTX vial the oral route, and in the German Methotrexate Registry over half of patients (63%) received oral MTX exclusively for the first 6 months. Patients in the German registry reported similar rates of ACR Pediatric 30,50, and 70 response, as well as toxicity between routes of MTX administered. It remains agreed upon that parenteral MTX administration should be considered in children who (1) have a poor clinical response to orally administered MTX (this may be due to poor compliance or to reduced oral bioavailability for various reasons); (2) need dosages greater than about 10 to 15 mg/m ² /week to achieve maximum clinical response (oral MTX absorption is a saturable process, whereas subcutaneous administration is not) ( eFig. 12-5 ) ; or (3) develop significant GI toxicity with orally administered MTX. Studies in adult patients with rheumatoid arthritis suggest that oral absorption of MTX is considerably reduced at doses of 15 mg or more, and MTX should be administered parenterally. Bypassing the enterohepatic circulation may also reduce hepatotoxicity.

The issue of when, how, and by what criteria to consider withdrawing MTX therapy in JIA remains unclear. However, the criteria for “remission” or “relapse” have usually not been well defined or standardized among various studies, and the assessment of outcomes has not been the subject of blind studies. Given these limitations, no firm conclusions can be drawn about the optimal time and mode of MTX discontinuation in children with JIA. MTX withdrawal may result in disease flare in more than 50% of patients; this rate may be even higher in younger children. Cellular biomarkers such as myeloid-related protein (MRP) 8 (S100A8) and MRP 14 (S100A9) heterocomplex (calprotectin, or MRP8/14) secreted by activated phagocytes at local sites of inflammation may be viable biomarkers to determine the appropriate time to discontinue MTX. Levels of MRP8/14 at the time of MTX discontinuation were significantly higher in patients who subsequently developed flares, compared to those who remained in stable remission.

Safety.

Although MTX is associated with many potential toxicities, the documented overall frequency and severity of adverse effects in children with arthritis have been low. Most side effects are mild and reversible and can be treated conservatively. Although the precise mechanism of all MTX-related toxicities is not clearly understood, at least some of MTX’s adverse effects are directly related to its folate antagonism and its cytostatic effects. This relationship is especially evident in tissues with a high cell turnover rate, such as the GI tract and bone marrow, that have a high requirement for purines, thymidine, and methionine, which may explain why supplementation with folic or folinic acid may diminish these symptoms.

Other rare adverse effects

Folate supplementation.

As a potent antifolate drug, the side effects of MTX are also consistent with symptoms of folate deficiency, and it is rational to question how the folate pathway and folic acid supplementation may impact drug efficacy at the expense of minimizing toxicity. Baseline plasma and erythrocyte folate concentrations have been shown to negatively correlate with MTX toxicity scores in RA, and children with historical intolerance to MTX have shown significantly lower cellular folate concentrations in a cross-sectional study. Numerous studies have examined the issue of minimizing MTX toxicities with the use of concurrent folic or folinic acid (leucovorin) supplementation in adults with rheumatoid arthritis. A Cochrane review of all trials on “ low-dose” folic acid (≤7 mg/week) or folinic acid in adults with RA from 1999 through March 2012 revealed a 26% relative risk reduction in the incidence of GI side effects, a 76.9% relative risk reduction in transaminase elevation, and a 60.5% relative risk reduction in MTX withdrawal for any reason, with no observed effect upon efficacy.

However, the effect of folate supplementation upon drug effectiveness is far from clear. Some studies have shown that concurrent folate supplementation may worsen disease activity in psoriasis and RA. A small number of clinical studies that investigated supplemented folic acid in JRA have suggested no substantial effect upon MTX efficacy, but higher doses of folinic acid were associated with disease flares. Baseline variability in the endogenous target folate pathway may also be important for drug outcomes, as preliminary data suggest that initiating MTX in a folate replete state may be associated with improved outcomes on MTX, as enhanced cellular folate uptake may also represent enhanced cellular MTX uptake.

Based on the data from adult studies and the small trial in children with JRA, it seems that daily (1 mg/day) folic acid supplementation confers a beneficial effect in terms of GI and mucosal toxicities associated with low-dose weekly MTX treatment and does not have any significant detrimental effect on disease control. Without firm data to direct otherwise, folic acid supplementation should be considered at least in symptomatic patients. High-dose folinic acid rescue should be reserved for patients with severe, life-threatening toxicity (e.g., aplastic anemia).

Mechanism of action.

The exact mechanism of action of hydroxychloroquine remains unknown, although several physiological effects have been attributed to this drug class that may be pertinent to rheumatic disease. These include the inhibition of neutrophil chemotaxis, nitric oxide production, and phagocytosis. Hydroxychloroquine may also antagonize the action of prostaglandins, interfere with IL-1 release by monocytes ; interfere with production of TNF-α, IL-6, and IFN-γ ; inhibit natural killer activity ; and induce apoptosis. It has antiplatelet and antihyperlipidemic effects extremely important in patients with SLE, and antagonistic effects upon Toll-like receptor 7/9.

Pharmacology.

Hydroxychloroquine is rapidly absorbed from the intestine. Equilibrium concentrations are reached after 2 to 6 months of a constant daily dose, and the half-life exceeds 40 days. Tissue levels are much greater than plasma concentrations, and there is increased affinity of the drug for the liver, pituitary, spleen, kidney, lung, adrenals, and specifically for melanin. Excretion is primarily via the kidney.

Dosing/efficacy.

The recommended dosage for hydroxychloroquine is less than or equal to 6.5 mg/kg/day to a maximum dosage of 400 mg/day ( Table 12-2 ). Early studies have not shown hydroxychloroquine to be an extremely effective disease-modifying agent in JRA ; however, it is used commonly in pediatric SLE and cutaneous LE in children, as well as for juvenile dermatomyositis. Although this medication is commonly used, like many older DMARDs, it has been inadequately studied in children. It is used mostly in combination with other medications, or in mild or well- controlled disease. Data from adults has shown an increased risk of lupus flare once their hydroxychloroquine was withdrawn, and utilizing hydroxychloroquine in addition to MTX and sulfasalazine has been shown to be superior than single or double therapy in adults with RA. Recent data support a protective effect against fetal heart block in neonatal lupus.

Safety.

When used at recommended doses, antimalarials are considered extremely safe. At least four young children have died of respiratory failure after accidental ingestion of large doses (1 to 3 g) of chloroquine, however, as there is no antidote.

GI intolerance occurs in 10% of adults, and skin hyperpigmentation, myasthenia, and muscle weakness have been described. CNS side effects that include headache, light-headedness, tinnitus, insomnia, and anxiety are common. These side effects may be reversible with dose reduction and may remit spontaneously.

The major concerning side effect is retinal toxicity. Retinal toxicity, although rare, can cause blindness, even after the medication has been stopped. Antimalarials accumulate in the pigmented cells of the retina and persist; however, retinitis is sometimes, but not always, reversible. Evidence in adults suggests that retinal toxicity does not occur if the dosage of hydroxychloroquine is maintained at less than 6. 5 mg/kg/day, even for up to 7 years. Routine ophthalmological monitoring can lead to early detection of premaculopathy; vision loss can be prevented if the medication is discontinued. Newly revised recommendations in adults suggest a baseline exam and then annual screening starting after 5 years on therapy. Each examination should include visual acuity, color vision testing, visual field examination, and retinoscopy. In addition, newer objective tests including multifocal electroretinogram, spectral domain optical coherence tomography, and fundus autofluorescence have been shown to be more sensitive than visual fields, and at least one is recommended to be performed, if available, in addition to standard testing. Retinal abnormalities or interference with vision, especially with foveal recognition of red, is an absolute indication for discontinuation of hydroxychloroquine. Use of hydroxychloroquine in children younger than 7 years may be limited by difficulty in obtaining satisfactory evaluation of color vision in this age group, and the standard of care in children remains an annual exam until definitive studies in children suggest increasing the frequency of monitoring.

Hydroxychloroquine crosses the placenta but is considered safe to use during pregnancy. Hydroxychloroquine does appear in breast milk, but the amount ingested per day by a breast-feeding infant would be very low.

Mechanisms of action.

Several mechanisms of action may explain the antiinflammatory effect of sulfasalazine. Bacterial growth is reduced by sulfasalazine and sulfapyridine, and the bacterial antigenic load delivered to the gut-associated lymphoid tissue may be reduced. This mechanism may be important for patients with spondyloarthropathies, in whom bacteria may gain access through inflamed gut mucosa and stimulate the immune system. Sulfasalazine interferes with many enzymes that are important in inflammation in the formation of leukotrienes and prostaglandins, and it is a potent inhibitor of AICAR transformylase, resulting in an accumulation of extracellular adenosine. There are several additional pharmacological effects reported in the literature.

Pharmacology.

Sulfasalazine is poorly absorbed from the GI tract. Peak serum concentrations are reached after 5 days of therapy. The half-life of the drug is 10 hours. Approximately one third of the dose is absorbed in the small intestine and excreted unchanged in the bile. The remaining 70% enters the colon intact, where the azo linkage is split by bacterial enzymes to sulfapyridine, which is absorbed and excreted in the urine, and 5-aminosalicylate, which reaches high concentrations in the feces. Approximately 90% of sulfapyridine is absorbed from the colon. Sulfapyridine is tightly protein bound and acetylated, hydroxylated, and conjugated with glucuronic acid in the liver. Sulfasalazine and sulfapyridine reach synovial fluid in concentrations comparable with those in serum. About one third of 5-aminosalicylic acid is absorbed, acetylated, and excreted in the urine. The rest is eliminated unchanged in the stool. The small amount of salicylate absorbed is insufficient to reach antiinflammatory levels in the plasma.

Dosing/efficacy.

Sulfasalazine is a recommended treatment for ERA following a trial of NSAIDs and/or intraarticular steroid injections. The suggested dosage in children is 30 to 50 mg/kg/day in two to three divided doses, usually taken with food or milk. Treatment is initiated at a lower dosage (10 to 15 mg/kg/day) and increased weekly over 4 weeks to achieve maintenance levels. A satisfactory clinical response may occur within 4 to 8 weeks ( Table 12-2 ).

Several studies have investigated sulfasalazine use in children and results have been mixed. A double-blind, randomized, placebo-controlled study of 69 Dutch children with oligoarticular or polyarticular juvenile chronic arthritis showed significant improvement in overall articular severity score, global, and laboratory parameters. A follow-up study showed sustained benefits in these same individuals 7 to 10 years later. Alternatively, in a small, 26-week, randomized, double-blind, placebo-controlled study, no significant differences in active joint count, tender entheses count, pain visual analogue scale, or spinal flexion were seen with sulfasalazine compared with placebo.

Safety.

Intolerance and toxic reactions occur in approximately 20% of sulfasalazine-treated adults with rheumatoid arthritis (range 5% to 55%). In a placebo-controlled study of 35 children with JRA, 29% developed adverse effects that led to discontinuation of the drug. Rashes occur in 1% to 5% of patients. A maculopapular rash occurring within 2 days after institution of therapy, especially on sun-exposed skin, is the most common dermatological complication.

Oral ulcers and Stevens–Johnson syndrome are uncommon but important complications. The drug should not be used in infants or in patients with known hypersensitivity to sulfa drugs or salicylates, impaired renal or hepatic function, or specific disease contraindications (e.g., porphyria, glucose-6-phosphate dehydrogenase deficiency). Other rare side effects have been reported and include cytopenias, drug-induced SLE, Raynaud phenomenon, interstitial pneumonitis, fibrosis, alveolitis, pulmonary syndromes, hepatitis, hypogammaglobulinemia, and IgA deficiency. Serious infections have not been reported, however. Most authors also consider sulfasalazine to be contraindicated in patients with systemic JIA because of an apparent increased risk of toxicity reported in patients with adult-onset Still disease.

A reversible decrease in sperm count has been observed, but there are no reports of teratogenicity, and it can be safely used in pregnancy. Sulfasalazine enters breast milk in negligible amounts and thus is not an absolute contraindication to breast-feeding. However, caution should be exercised because its metabolite, sulfapyridine, is present at significant levels in breast milk, and there has been a single case report of bloody diarrhea in an infant exposed to sulfasalazine through lactation. In addition, as sulfapyridine can displace bilirubin, this medication should be avoided in nursing mothers of premature infants, infants who are ill or suffering from hyperbilirubinemia, or glucose-6-phosphate dehydrogenase deficiency.

Mechanisms of action.

Leflunomide is an immunomodulatory agent that, through its active plasma metabolite, A77-1726, inhibits de novo pyrimidine synthesis by inhibiting the enzyme dihydroorotate dehydrogenase. As a result of the inhibition, p53 in the cytoplasm translocates to the nucleus and initiates cellular arrest in the G ₁ phase of the cell cycle. It also inhibits tyrosine kinase, inhibits leukocyte-endothelial adhesion, affects cytokine production, and decreases serum metalloproteinase activity similarly to MTX. In vitro, leflunomide inhibits the production of prostaglandin E ₂ , matrix metalloproteinase (MMP)-1, and IL-6, and modulates various tyrosine kinases and growth factor receptors.

Pharmacology.

Leflunomide is rapidly converted to A77-1726, which is highly protein bound and has a prolonged half-life of up to 18 days. As a result, loading doses have been recommended for the first 3 days of administration to achieve steady state rapidly in RA, although it may increase the frequency of GI toxicity ; however, a loading dose has not been adopted consistently in children with JIA.

Dose/efficacy.

Dosing guidelines for children are limited and based on studies by Silverman and colleagues ( Table 12-2 ) that compared the safety and efficacy of leflunomide to MTX in patients with polyarticular course JIA. Pharmacokinetic studies were performed. At week 16, 68% of patients receiving leflunomide showed an improvement according to the ACR Pediatric 30 versus 89% of patients treated with MTX; the improvements achieved were maintained at a similar rate in a 32-week extension study. The median time to ACR Pediatric 30 did not differ between the two groups. Body weight was a significant determinant of response, and patients weighing less than 20 kg showed the greatest discrepancy, with the clinically active metabolite notably lower in this group. The incidence of treatment-related adverse events was similar in both groups, although there was an increased frequency of liver transaminase elevations in the MTX group. Furthermore, in a long-term open label study of leflunomide in polyarticular JIA patients who were MTX intolerant or refractory, 52% of patients met ACR Pediatric 30 by 12 weeks, with 65% of those who entered the extension phase maintaining ACR Pediatric 30 status up to 2 years.

An additional retrospective review of 58 German JIA patients treated with leflunomide either alone (n = 48) or in conjunction with MTX (n = 10) were examined over a mean of 1.5 years and found that leflunomide was well tolerated and effective; approximately 30% of patients on leflunomide attained remission.

Safety.

Mild and dose- related side effects in adults include GI side effects (abdominal pain, dyspepsia, anorexia, diarrhea, gastritis), allergic rash, reversible alopecia, mild weight loss, and elevated hepatic transaminases. Side effects in children have included transiently elevated liver enzymes, abdominal pain and nausea, diarrhea, headaches, mouth ulcers, and alopecia; these symptoms tend to be dose related.

Leflunomide is teratogenic. Because of the very long half-life of this drug, it has been recommended that cholestyramine be administered, and that drug levels less than 0. 02 mg/L be verified on two separate tests at least 2 weeks apart in men and women before attempting to conceive. Breast-feeding is contraindicated.

Mechanism of action

Colchicine’s action is thought to depend on binding of two of its rings to cellular microtubules, inhibiting the movement of intracellular granules and preventing secretion of various components to the cell exterior. Interaction between endothelial cells and neutrophils is inhibited by reducing the expression of adhesion molecules on the neutrophil membrane. The drug is present in granulocytes to a much greater extent than in lymphocytes or monocytes.

Pharmacology.

Peak plasma levels are reached 1 to 3 hours after oral administration. Colchicine’s bioavailability is less than 50%, and its half-life after oral administration is 9 ± 4 hours. It is predominantly eliminated by biliary excretion through the stool. The multidrug transporter molecule ABCB1 (also known as P-glycoprotein and multidrug transporter 1) mediates the extrusion of colchicine into the GI tract, and polymorphisms in ABCB1 may explain some of the differences in treatment response in patients with FMF. Enteric and hepatic cytochrome P-450 3A4 (CYP3A4) is also important in colchicine metabolism. Drug interactions may also occur either at the level of the transporter or through the cytochrome P-450 system, and thus inhibition or competition for CYP3A4 may lead to colchicine accumulation and toxicity. Dose reduction algorithms have been proposed for adults.

The therapeutic dose of colchicine ranges from 0.5 to 2 mg/day as needed to prevent or reduce significantly the frequency of FMF attacks and is administered once or twice daily. Toxicity is extremely rare with oral administration and is generally limited to the GI tract (nausea, vomiting, abdominal pain, diarrhea); this can be helped by administering colchicine in two divided doses and reducing dietary lactose intake. In the case of serious overdose, treatment with colchicine-specific Fab could be considered. Severe toxicity can result in dehydration, multiorgan failure, and a disseminated intravascular coagulation–like syndrome. Colchicine is safe to take during pregnancy and while breast-feeding. Concerns about chromosomal and gonadal aberrations resulting from its effect upon microtubules have not been supported.

Pharmacology.

Glucocorticoid drugs are structural variants of the naturally occurring glucocorticoid, cortisol. Synthetic compounds, such as prednisone and cortisone, must be hydroxylated to form therapeutically active prednisolone and hydrocortisone. Topical glucocorticoids, such as dexamethasone, or those administered by intraarticular injection (e.g., triamcinolone hexacetonide) already have a hydroxyl group at C11 and are thus in active form. The different relative potencies and durations of biological action of the various synthetic analogues are outlined in Table 12-3 .

Orally administered glucocorticoids (prednisone, prednisolone) are rapidly absorbed. Prednisone is converted to prednisolone in the liver and reaches a peak plasma concentration within 2 hours. Hydrocortisone and prednisolone bind to the serum proteins transcortin (high affinity) and albumin (low affinity). Methylprednisolone and dexamethasone are bound primarily to albumin. Prednisolone has a large volume of distribution; about two thirds is taken up by muscle. After metabolism in the liver, excretion occurs principally via the bile.

Physiological and pharmacological effects.

Glucocorticoids are unique among pharmacological agents used to treat rheumatic diseases because they are synthetic analogues of endogenous molecules that are produced by the body that perform important physiological and pharmacological functions through glucocorticoid receptors (GRs) and genomic and nongenomic mechanisms.

Antiinflammatory and immunosuppressive actions.

Glucocorticoids have antiinflammatory and immunosuppressive effects. Steroids inhibit the early stages of inflammation (e.g., edema, fibrin deposition, capillary dilation, migration of lymphocytes into inflamed areas, phagocytic activity) and the later manifestations (e.g., proliferation of capillaries and fibroblasts, deposition of collagen). Many of these effects are mediated by inhibition of numerous chemokines and cytokines, including arachidonic acid and its metabolites, platelet-activating factor, TNF, IL-1, mitogen-activated protein kinase (MAPK) phosphatase 1, and NF-κB.

Glucocorticoid effects on the immune system are mediated principally through T lymphocytes. Acute administration of hydrocortisone produces a 70% decline in circulating lymphocytes. T lymphocytes are affected more than B lymphocytes, and T-helper cells are affected more than T-suppressor cells. Lymphopenia is probably a result of sequestration of cells in the bone marrow rather than cell lysis, although drug-induced apoptotic cell death may also be involved. Corticosteroids have been shown to result in a profound and transient lymphocytopenia, maximal at 4 hours after the dose and resolved by 24 hours, due to a redistribution of these cells to the bone marrow. There is also a 90% decline in circulating monocytes within the initial 6 hours. Proliferative T-cell responses to antigens (streptodornase-streptokinase), mitogens (concanavalin A), and cell surface antigens (as in the mixed leukocyte reaction) are reduced by glucocorticoids. IL-2 production by T cells in vitro is also reduced. Glucocorticoids cause an increase in the numbers of blood neutrophils by increasing the release of cells from the marginated neutrophil pool, prolonging their stay in the circulation, and reducing chemotaxis of neutrophils to sites of inflammation.

Intravenous glucocorticoid causes a decrease in circulating IgG but has little discernible effect on the serum titer of specific antibodies. The protein catabolic effects of long-term administration may have consequences on the humoral immune system. Endothelial secretion of C3 and factor B of the complement cascade are also inhibited.

Indications for systemic glucocorticoid therapy.

When considering glucocorticoid use in children with rheumatic diseases, the risk/benefit ratio must be carefully weighed because these agents are associated with substantial toxicity when used systemically in the long term ( eBox 12-1 ). The overall aim is to limit the dose and duration of steroid therapy as much as possible while achieving disease control.

Adverse effects.

Two broad categories of adverse effects are associated with the therapeutic use of systemic glucocorticoids: effects resulting from prolonged use of large doses and effects resulting from withdrawal of therapy. The mechanisms involved in the development of these adverse events are reviewed in detail elsewhere.

Cushing syndrome.

Cushing syndrome, a term used originally to identify the effects of idiopathic hypercorticism, may also be induced by prolonged glucocorticoid administration. It is characterized biochemically by high plasma glucocorticoid levels and suppression of the hypothalamic–pituitary–adrenal axis. It is characterized clinically by many features, including truncal obesity ( Fig. 12-6 ), osteoporosis, thinning of the subcutaneous tissues, and hypertension. The distribution of fat in Cushing syndrome is predominantly in the subcutaneous tissue of the abdomen and upper back (buffalo hump) and in the face (moon facies). Weight gain reflects fluid retention and increased caloric intake. Skin changes, in addition to the characteristic purple striae, include hirsutism and acne. Hypertension is usually mild but occasionally requires treatment or reduction of the glucocorticoid dose. With the exception of skin striae, all of these physical features are reversible after cessation of glucocorticoid therapy.

An 11-year-old girl with severe systemic-onset juvenile idiopathic arthritis requiring high-dose corticosteroid treatment. Cushingoid features shown include moon facies, truncal obesity, and cutaneous striae.

Effects on bone: osteoporosis and avascular necrosis.

Osteoporosis is one of the most troublesome consequences of long-term, high-dose glucocorticoid therapy, although there are multiple other contributing factors to consider, including inadequate dietary intake of calcium and vitamin D, underlying disease activity, reduced physical activity, and low body weight.

Glucocorticoids are associated with a reduction in bone formation and an increase in bone resorption; the reduction in bone formation seems to be more important and is caused by a direct inhibitory effect on, and apoptosis of, osteoblasts. Glucocorticoids also directly inhibit gut absorption of calcium and cause increased urinary calcium excretion, potentially resulting in secondary hyperparathyroidism and increased bone resorption. Additional mechanisms by which glucocorticoids result in bone loss are depicted and detailed in eFig. 12-7 . These mechanisms include effects on the production of local growth factors, reduction of matrix proteins, increase in the production of enzymes that break down matrix, increase in apoptosis of osteoblasts and osteocytes, increase in osteoclastogenesis secondary to decreased production of osteoprotegerin, and an increased production of receptor-associated NF-κB ligand (RANK ligand).

Approaches to the prevention and treatment of steroid-induced osteoporosis. FSH, Follicle-stimulating hormone; GI, gastrointestinal; LH, luteinizing hormone; LHRH, luteinizing hormone–releasing hormone; OPG, osteoprotegerin; PTH, parathyroid hormone; R receptor-associated NF-κB ligand.

The extent of bone loss seems to be related to the dose and duration of glucocorticoid therapy, although these factors do not have a consistent relationship with fracture risk. Significant bone loss occurs with dosages of 7.5 mg/day or greater in most adults. In adults, bone loss is predominantly trabecular (e.g., spine and ribs) rather than cortical, whereas in children the osteoporotic effects of glucocorticoids are more generalized. Bone loss seems to occur rapidly within the first 6 to 12 months of therapy and then reaches a plateau. Alternate-day glucocorticoid therapy may not be protective. Not all patients exposed to long-term glucocorticoid therapy develop bone loss ; there are, however, no reliable biochemical predictive markers. Bone densitometry may be used to screen children who are at high risk for osteoporosis, although there are challenges with misinterpretation due to adult norms. There are limited guidelines on what frequency to perform densitometry and also controversy on the utility of other types of imaging modalities.

High-dose glucocorticoids have also been associated with avascular necrosis of bone (AVN), although the exact mechanism is unknown. Intramedullary vascular compromise may result from increased osteocyte apoptosis induced by glucocorticoids. In the absence of the clearance of these apoptotic osteocytes, reduced blood flow and bony ischemia may result. Glucocorticoids induce adipocyte differentiation via the increased production of PPAR-γ ₂ , which may result in increased fat in the marrow. Finally, glucocorticoids also increase the expression of endothelin-I, which may also lead to reduced intramedullary blood flow. The most common and clinically significant location for AVN is the femoral head, and the underlying disease process (such as in SLE) can be a contributing factor as well as age.

Minimizing toxicity.

The deleterious effects of glucocorticoids can be minimized by choosing a drug with a short half-life (see Table 12-3 ). Prednisone is the drug most often given for oral therapy, as its prominent glucocorticoid and minimal mineralocorticoid actions give it the lowest risk/benefit ratio of any of the analogues in general use.

The antiinflammatory effect and the toxicity of glucocorticoids increase with larger doses and more frequent administration ( Table 12-4 ). Short-acting glucocorticoids given in the morning do not suppress the pituitary as much as glucocorticoids given later in the day (which suppress the normal surge of adrenocorticotropic hormone [ACTH] that occurs during sleep) so once daily administration should always be in the morning.

Reduction in glucocorticoid dose must be individualized for the child and the disease, and is often fraught with difficulty because of the adaptation of the patient’s metabolism to chronic steroid excess. At high dosages (e.g., 60 mg/day), reductions of 10 mg are usually well tolerated; at lower dosages (e.g., 10 mg/day), reductions of only 1 or 2 mg may be possible. An alternate-day regimen should be the goal to minimize toxicity, although some patients do not tolerate this regimen. In some children, steroid pseudorheumatism may result from a rapid dose decrease. These withdrawal effects gradually resolve over 1 or 2 weeks and are minimized if each decrement in daily prednisone is 1 mg or less per week (at the lower dose levels).

Many approaches for the prevention and treatment of corticosteroid-associated osteoporosis have been studied in adults, and several guidelines have addressed these issues. Vitamin D and its analogues, calcitonin, and various bisphosphonates have been used. Calcitriol (vitamin D ₃ ) or cholecalciferol (vitamin D), with or without calcitonin, were shown to prevent bone loss from the lumbar spine better than calcium alone in several randomized controlled clinical trials of adults who started long-term glucocorticoid therapy. Treatment with calcium and vitamin D in adults who receive glucocorticoids effectively slows lumbar and forearm bone loss, and treatment with calcium and vitamin D supplementation has become standard practice in most centers for children with rheumatic disease who receive glucocorticoids.

Bisphosphonates have also been studied as a potential treatment for glucocorticoid-induced osteoporosis. Etidronate, pamidronate, alendronate, and risedronate have been shown in randomized controlled trials to increase lumbar spine bone mineral density in adults receiving long-term glucocorticoids for various diseases These trials did not include children, however, and did not show any significant reduction in fracture incidence, which is the most clinically relevant outcome. Bisphosphonates have been studied in children with osteogenesis imperfecta and seem to be beneficial in reducing bone resorption, increasing bone density, and reducing the chronic bone pain associated with this condition. In addition, although there are concerns regarding their effects on growth and remodeling, bisphosphonates have been found to be useful and safe in open-label studies of children with idiopathic juvenile osteoporosis, or osteoporosis associated with connective tissue diseases or induced by glucorticoids. Binding to bone and prolonged renal excretion (mean 7 years) continues to concern clinicians for long-term safety of these agents in children. This concern necessitates larger, prospective trials to evaluate bisphophonates for the prevention and treatment of glucocorticoid-induced osteoporosis in children.

Preventing acute adrenal insufficiency (addisonian crisis).

The use of pharmacological doses of glucocorticoids for a 2-week period may result in transient suppression of endogenous cortisol production, and prolonged therapy may lead to suppression of pituitary–adrenal function that can be slow in returning to normal. This is potentially the most serious and life-threatening adverse effect associated with glucocorticoid therapy. The actual doses and duration of therapy that are associated with suppression and the length of the recovery period after cessation of therapy are not well defined. The recovery period may even be affected by the underlying inflammatory process.

If not recognized, suppression of the hypothalamic–pituitary–adrenal axis places a child at risk for vascular collapse, adrenal crisis, and death in situations that demand increased availability of cortisol. Under conditions of stress (e.g., serious infection, trauma, surgery), all children who may be at risk for hypothalamic–pituitary–adrenal axis suppression require additional glucocorticoids.

The “stress dose” regimen is based on the body’s requirements for hydrocortisone during stress. Hydrocortisone (6 to 9 mg/m ² /day divided three times daily) is needed for physiological maintenance. For febrile or severe illnesses, hydrocortisone requirements increase to 40 mg/m ² /day. With induction of anesthesia or in a resuscitation situation, 100 mg/m ² IV of hydrocortisone is required initially, and then 25 mg/m ² IV every 6 hours for the following 24 to 48 hours.

If the patient is currently receiving glucocorticoids such as prednisone or prednisolone at a dosage equivalent to or greater than 40 mg/m ² /day of hydrocortisone (see Table 12-3 for conversion), then the current glucocorticoid dose prescribed for disease management may be enough for febrile or severe illnesses. However, as steroids are weaned or discontinued after prolonged use, glucocorticoid replacement may be required to prevent adrenal crisis if the body’s corticosteroid needs exceed the dose prescribed for disease management.

High-dose intravenous glucocorticoid therapy.

Intravenous glucocorticoid “pulse” therapy is sometimes used to treat more severe, acute, systemic connective tissue diseases. The rationale of this approach is to achieve an immediate, profound antiinflammatory effect and to minimize toxicity related to long-term continuous therapy in moderate to high daily doses. Pulse methylprednisolone has been shown to inhibit cytokine generation and dissolve in cell membranes, altering membrane-associated proteins. Differences are seen in the alpha-interferon gene expression signature in pediatric lupus patients receiving pulse methylprednisolone therapy as compared with oral glucocorticoid therapy, which may explain an advantage this dosing regimen may have over oral doses. However, this observation will need further validation, and additional investigation is needed to optimize efficacy and minimize toxicity with this therapeutic modality.

Although oral pulse regimens have been reported, IV methylprednisolone has been the drug of choice, given in a dosage of 10 to 30 mg/kg/pulse up to a maximum dose of 1 g, administered according to various protocols ( eBox 12-2 ): a single administration as clinical circumstances warrant, a pulse each day for 3 to 5 days, or alternate-day pulses for three doses. Intravenous glucocorticoid pulse therapy may be associated with potentially serious complications ( eBox 12-2 ).

Intraarticular steroids.

Injection of long-acting glucocorticoids directly into inflamed joints has emerged as a major advance in the management of various types of arthritis. Although intraarticular steroid (IAS) therapy has not been studied in randomized controlled clinical trials, multiple reports have documented its efficacy and safety in children.

IAS therapy has been used most often in children with oligoarticular disease; indications for use have included lack of response to NSAIDs; significant NSAID toxicity; and the presence of joint deformity, growth disturbance, or muscle wasting, or as an alternative to NSAID therapy in children with oligoarticular disease. In polyarticular disease, multiple IAS injections at one time can be used as a temporizing measure while awaiting response to second-line systemic agents to take effect. IAS may also be useful as an alternative to increasing systemic therapy in children with polyarticular disease who have significant inflammation in only a few joints.

Virtually all patients experience rapid resolution of symptoms and signs of joint inflammation within a few days after injection, resulting in improved physical function. About two thirds achieve remission for at least 12 months after a single injection. A longer duration of response has been described in children with oligoarticular JIA, and in those who are younger, with shorter disease duration, and with higher mean erythrocyte sedimentation rates. Early use of IAS injections have been associated with less leg-length discrepancies in patients with asymmetrical pauciarticular JRA. More recent data suggest the effectiveness of using radiographic assistance to guide IAS injections into involved temporomandibular and subtalar joints.