Background: Anti-tumour necrosis factor (Anti-TNF) therapy is effective for both inducing and maintaining remission in IBD patients. Recent meta-analyses have demonstrated their safety in the general IBD population. However, little is reported about the safety of these drugs in elderly patients.
Therefore, we aimed to investigate the safety of management of IBD in elderly patients using anti-TNF treatment by systematic review and meta-analysis of available data.
Methods: A literature search was conducted for papers and conference proceedings through November 2016 regarding elderly IBD patients and anti-TNF therapy. All studies were appraised using the adapted Newcastle-Ottawa Scale (NOS), which contains 9 criteria for cohort studies and is adapted to 6 criteria for case series and case reports. Three reviewers independently extracted data on anti TNF-exposed older and younger patients, with number of serious infections, dead during follow-up and cessation of therapy due to adverse events as outcomes of interest. Poisson regression was used to compare the occurrence of outcomes of interest per patient year follow up between older and younger IBD patients.
Results: From 454 found titles, four papers and 5 conference abstracts were included, totalling 1276 patients: 470 older and 806 younger patients. Data on combined steroid use was provided in 2 papers, data on IM combo-therapy in 5 papers. Papers used either 60 years or 65 years as cut off.
The rate ratio for serious infections was similar between older and younger IBD patients (2.1, p=0.084) and was not influenced by IM use (rate ratio 2.8, p=0.054). However, when steroid use was added to the model, the rate ratio for serious infections was 2.35 (p<0.001, 95% CI 2.3–2.4 and use of steroid increased the risk (p<0.001). Risk of death did not significantly differ between older and younger patients during follow (risk ratio 10.4, p=0.19) and was not affected by use of IM (p=0.5). When use of steroids was added to the model, there was a trend towards an increased risk of death in older IBD patients during anti-TNF exposure (p=0.06, ratio 3.3, 95% CI 0.94–11.8)
Older patients were 3.1 x more likely to stop anti-TNF therapy due to AE (p=0.008, 95% CI 1.34–7.4) compared to younger patients, and this was not influenced by use of IM. Too few data were available on steroid use.
Conclusions: Although anti-TNF therapy is more often discontinued in older IBD patients due to adverse events, the use of anti-TNF was only associated with an increased risk for serious infections when combined with steroids. More data on safety of anti-TNF in elderly are needed, with special attention to indices of physical, frailty and mental and social impairment. This would enable a more personalised assessment of risk.