Patients at Risk
Patients Taking Glucocorticoids
The increased risk for infection with glucocorticoid use is significant and highly dose dependent. One study identified minimal thresholds for a daily dose and total exposure dose at which prednisone apparently increases the risk for infection. In the 71 clinical trials pooled together in this meta-analysis, when the total cumulative dose was <700 mg (21 of 71) or with daily doses <10 mg (5 of 71), there was no statistically significant increased risk for infection detected with prednisone compared to the control group. Sixteen out of 47 trials with prednisone doses exceeding these thresholds demonstrated an increased risk of infection in the prednisone group compared to the control group. The increased risk of infection was particularly high in patients with neurologic diseases (relative risk [RR] 2.8; 95% CI, 1.9–4.3; P <.001). When looking at the daily dose only, a glucocorticoid dose ≥20 mg/day (prednisone equivalents) is a universally recognized point at which the risk for infections increases.
Likewise, a daily prednisone dose ≥20 mg is commonly considered the cutoff at which vaccine efficacy is substantially lost. Short durations of glucocorticoid therapy (<2 weeks) and lower doses (<10–20 mg/day of prednisone equivalents) do not usually contraindicate vaccination, even with live-virus vaccines. On the other hand, live vaccines may be ineffective (or even dangerous) and are contraindicated in patients on higher doses and longer durations of glucocorticoids. When a patient has been on high-dose prednisone at doses >10 to 20 mg per day for longer durations, there should be a waiting period of up to 3 months after discontinuation of higher-dose therapy before administering a live-virus vaccine. If a vaccine is administered during high-dose glucocorticoid use, patients should be considered unvaccinated and be revaccinated if and when they achieve immunocompetence.
Patients Taking Disease-modifying Antirheumatic Drugs (DMARDs)
Many patients with immune-mediated diseases, like rheumatoid arthritis, already have a baseline risk for infections independent of drug therapy. Most of the scientific literature aimed at determining the risk of infections with the antirheumatic drugs compares the various drugs against other drugs in this group, and not against placebo. Some of the more useful comparisons available are when drugs that may be immunosuppressive (e.g., methotrexate, leflunomide, tumor necrosis factor [TNF]-alpha inhibitors, and other biologics) are compared against agents unlikely to have immunosuppressive effects (e.g., anti-inflammatory drugs like sulfasalazine.)
Methotrexate (MTX). This drug may increase infection risk and impair response to vaccines, especially at chemotherapy doses. There is also some concern in the medical community about the risk for varicella zoster (VZ) and herpes zoster (HZ) infections in patients on lower doses as used in immunologic conditions like rheumatoid arthritis. However, a recent analysis detailed three studies in particular that failed to discover a link between MTX use (vs. other DMARD therapy) and these infections.
TNF-alpha Inhibitors. The TNF-alpha inhibitors (infliximab [Remicade], etanercept [Enbrel], adalimumab [Humira], golimumab [Simponi], certolizumab pegol [Cimzia]) contain numerous warnings in their prescribing information, medication guides, Web sites, and various advertisements regarding the risk of infections. A black box warning assigned to the labeling of all of these products contains multiple warnings about infection risk. Infliximab may be associated with the greatest risk for infection, as it has been associated with a statistically significant increase in serious infections (especially pneumonia and soft-tissue infections) compared to etanercept and adalimumab.
It is less clear how TNF-alpha inhibitors affect immunization considerations. There is incomplete or conflicting information as to whether TNF-alpha therapy (or other DMARDs for that matter) poses an increased risk for infection with vaccine-preventable diseases and/or vaccine failure. When TNF-alpha antagonists, MTX, glucocorticoids, and other DMARDs are compared, only glucocorticoid use (baseline and initiation) stands out for association with a dose-dependent increase in serious infections and herpes zoster.