Inflammatory bowel disease (IBD) currently represents a substantial economic burden1–4 affecting nearly 4 million people worldwide.5 However, since the emergence of anti-tumor necrosis factor alpha (anti-TNF-α) therapy, patient outcomes have shown improvement,6, 7 facilitating their establishment in IBD management.8–11 While these treatments have been shown to improve prognosis, increased utilization and earlier administration of these therapies12, 13 has led to a growing concern about the risk of opportunistic infections.14 Of particular concern is the potential reactivation of latent tuberculosis infection (LTBI), which in this clinical scenario often presents in an aggressive and disseminated fashion.15, 16 Therefore, current guidelines mandate screening of LTBI prior to commencing anti-TNF therapy,17, 18 as early diagnosis can facilitate effective preventative therapy.19
The predominant tool for the diagnosis of LTBI has been the tuberculin skin test (TST), which utilizes purified protein derivative (PPD) to elicit a delayed-type T-cell-mediated hypersensitivity reaction.20 However, the TST has a number of limitations including false positivity due to the nonspecific nature of PPD leading to crossreactivity with nontuberculosis mycobacterium (NTM) and individuals vaccinated with Bacille–Calmette–Guerin (BCG); false positivity due to a boosting phenomenon; subjectivity of reaction size; and the requirement of two healthcare visits to obtain results.21–24 Of particular concern to the IBD population is an increased likelihood for false-negative results with impaired cellular immunity, such as in those under immunosuppressive therapy (IST).25
Recently made commercially available, interferon gamma release assays (IGRAs) have provided an alternative method to diagnose LTBI. IGRAs measure the release of interferon-gamma from whole blood (QuantiFERON-TB Gold [QFT-2G] and QuantiFERON-TB Gold In-Tube [QFT-3G]; Cellestis, Carnegie, Australia) or peripheral blood lymphocytes (T-SPOT.TB; Oxford Immunotec, Abingdon, UK) after stimulation by tuberculosis antigens. These tests have garnered support because their results are not prone to a boosting effect, and their antigens are more specific for Mycobacterium tuberculosis.26 However, IGRAs are not without their drawbacks, such as their higher cost and the need for appropriate infrastrutucture. Furthermore, they carry the potential for indeterminate results which can impair decision-making, specifically in the immunosuppressed population, where indeterminate results appear to be more frequent.27
A number of studies have recently assessed the performance of IGRAs in IBD. However, uncertainty remains regarding the superiority of IGRAs over the TST for diagnosing LTBI in IBD as well as the appropriate timing for screening. While guidelines universally support screening prior to starting anti-TNF therapy, many patients are already receiving IST, which may impact the performance of both tests. Due to these uncertainties, we conducted a systematic review and meta-analysis to assess the utility of IGRAs in the diagnosis of LTBI in patients with IBD.
- Top of page
- MATERIALS AND METHODS
This is the first systematic review and meta-analysis to assess the performance of IGRAs in IBD. It highlights the inability to determine the superiority of the IGRAs or the TST. This may be due to the lack of a gold standard in the diagnosis of LTBI and a lack of important outcomes required to assess test performance.30 While individuals with IBD are more likely to develop active TB after initiating anti-TNF therapy,14–16 most individuals do not.44 Therefore, the ability of LTBI diagnostic tests to predict which cases are more inclined to progress to active TB dictates their utility.30 Within our included studies, there was only one case of active TB, therefore limiting our ability to assess the capacity of IGRAs to predict active TB. The lack of these cases also prevented us from measuring sensitivity as current methodology for quantifying sensitivity uses active TB as a substitute marker in the absence of a gold standard. Furthermore, there are currently no randomized trials showcasing the benefit of prophylactic therapy based on IGRA results (+ vs. −).45
Our analysis provides a number of important outcomes. It shows a modest to strong agreement between IGRAs and the TST. However, when agreement was stratified by IST, differing results were seen. This may be partly explained by the discrepancy in BCG vaccination status between the two studies. In a study with a high rate of vaccination,36 concordance could be dramatically affected in the IST− population. With the introduction of IST, this could suppress both tests, leading to a higher likelihood of agreement. This is in contrast to a population with a low rate of BCG vaccination,39 where the differing effect of IST between the TST and the IGRAs may increase discordance in the IST+ population. Higher rates of indeterminate results were found in our analysis when compared with a number of studies assessing the performance of IGRAs in more generalized populations.46–49 This is possibly due to the higher frequency of IST in our included studies. Most important, our study demonstrates that both the IGRAs and the TST appear to be impaired by IST, with the impact being greater on TST results. These findings are in agreement with a recent meta-analysis assessing the performance of IGRAs in HIV.30
The lack of evidence to clearly support either the IGRAs or the TST helps to explain the discrepancies among current recommendations.17, 18, 50–55 Recommendations vary from suggesting either IGRAs or TST as first-line to utilizing both tests to increase detection. The European Crohn's and Colitis Organization (ECCO) endorses the utilization of TST alongside patient evaluation (including assessing TB risk factors, physical examination, and chest x-ray) for assessment of risk of LTBI in patients with IBD. IGRAs were suggested in BCG-vaccinated individuals. In individuals at high risk of LTBI, ECCO suggests screening prior to other forms of IST, not just anti-TNF therapy. Based on our findings, we feel that either the IGRAs or the TST can be used to screen for LTBI in IBD but one should consider performing these tests when patients are not using any IST. For patients who are BCG-vaccinated, IGRAs may be a more favorable choice, as BCG increases false TST positivity.56 This is further supported by a recent analysis that showed that among multiple screening strategies, utilizing QFT-2G in BCG-vaccinated individuals is the most cost-effective.57
Our study supports the suggestion that individuals, particularly those who are at high risk, who are starting other forms of IST, should be considered for screening prior to initiation of therapy. Both the IGRAs and the TST were shown to be negatively affected by IST. However, because our subanalysis had a limited number of studies, and many associations were nonsignificant, it is difficult at this time to determine before which specific forms of IST should LTBI screening be undertaken.
Our study has a number of limitations. Due to the limited number of full-text studies, we included relevant abstracts. This hindered our ability to thoroughly assess study methodology. The overall limited number of included studies hindered our ability to rigorously assess the affect of specific forms of IST on both IGRAs and the TST. The results of our pooled estimates suffered from heterogeneity. Reasons for heterogeneity between studies could be variation in BCG vaccination status, the percentage of individuals with risk factors for LTBI, possible variation among TST/IGRA test cutoff parameters, and the difference in IST definitions among studies. The difference among studies with regard to IST definitions further impaired our ability to pool studies. Ultimately, this limited the strength of conclusions that could be made. Lastly, this is not a robust review to evaluate the role of the TST in IBD since our focus was on IGRAs.
While studies continue to emerge, we are currently unable to universally recommend either the IGRAs or the TST for diagnosis of LTBI in patients with IBD. The strengths and weaknesses of both the TST (relatively inexpensive, the nonspecific nature of PPD, hypersensitivity reactions, subjectivity of reaction size, two healthcare visits required for interpretation) and the IGRAs (not prone to a boosting effect, specific nature of TB antigens utilized, increased cost, and increased laboratory capabilities required) should be weighed to facilitate the selection of LTBI diagnostics tests for regional/national screening programs. Our study shows that both the IGRAs and the TST are negatively affected by IST. Therefore, this should provide further evidence for committees and policymakers when suggesting screening protocols in IBD. Moreover, this study emphasizes that for individuals on IST, clinicians should not rely solely on either the TST or IGRAs for diagnosing LTBI. Further research in this area is needed to devise an effective screening strategy encompassing key factors such as TB risks, IST, and BCG vaccination. Moreover, studies focusing on the sensitivity and the predictive capacity of IGRAs for active TB will aid in the decision regarding using them as first-line diagnostic tests for LTBI in IBD.
In conclusion, although we are not able to identify a superior test among current LTBI diagnostic methods, both IGRAS and the TST were impaired by IST. Therefore, concerning the effective management of IBD, screening for LTBI prior to commencing patients on IST should be a consideration. Ultimately, regardless of what method is chosen to screen for LTBI, it is imperative that screening be performed in all patients prior to commencing anti-TNF therapy, which currently is not being universally achieved.