Optimizing compliance with latent tuberculosis infection screening among kidney transplant recipients

Abstract

Introduction

The increased risk of tuberculosis (TB) reactivation in solid organ transplant recipients supports the recommendation of screening for latent tuberculosis infection (LTBI). Adherence to available screening tests has not been studied in the kidney transplant (KT) population. We aimed to assess screening compliance within the ATALANTA-DOS population study.

Methods

ATALANTA-DOS studied an intervention bundle aimed at preventing infection in KT recipients. We compared LTBI screening rates between the pre-intervention (February 2016 – September 2017) and intervention (February 2018 – September 2019) cohorts and evaluated adherence rates between the interferon-gamma release assay (IGRA) and the tuberculin skin test (TST).

Results

A total of 307 KT recipients were included (155 in the pre-intervention cohort; 148 in the intervention cohort). A systematic assessment of screening compliance by an infectious disease specialist on day +30 post-KT improved LTBI screening adherence (82.6% [114/138] vs 1.3% [2/155]; p-value <0.001). In the intervention cohort, compliance was higher with IGRA (83.3% [52/62]) than with TST (68.1% [49/72]). Two cases of LTBI were detected in the pre-intervention cohort and five in the intervention cohort (4.4% [5/114]). All patients completed LTBI treatment after ruling out active TB. No cases of active TB were identified during follow-up.

Conclusions

Systematic evaluation of LTBI screening compliance significantly increased screening completion rates among KT recipients. IGRA-based strategies increased screening compliance, supporting their implementation over TST for LTBI screening among KT recipients. Increased adherence would allow a more targeted and effective treatment of LTBI.

Introduction

An estimated 25% of the global population is infected with Mycobacterium tuberculosis with a lifetime reactivation risk of 5-10% [1–5]. Solid organ transplant (SOT) recipients have a significantly higher risk of developing active tuberculosis (TB), with reported incidence rates ranging from 20 to 74-fold higher than those of the general population [3,6–11]. Most cases occur within the first year post-transplantation due to latent tuberculosis infection (LTBI) reactivation [2,3,7].

In transplant recipients, TB incidence ranges from 0.35% to 15% [2,7,8] and is associated with high morbidity and mortality rates, with mortality reaching 25–40% and allograft loss risk as high as 33% in patients receiving antituberculosis therapy [2,8,11,12]. Consequently, the American Society of Transplantation guidelines recommend screening for LTBI in all SOT candidates [13] to facilitate targeted treatment for patients at the highest risk of developing active TB [14]. In Spain, the incidence of TB in the general population was 8.07 cases per 100,000 inhabitants per year in 2022, highlighting the importance of LTBI screening [15].

The World Health Organization (WHO) [1] recommends LTBI screening strategies, including interferon-gamma release assays (IGRAs) and the tuberculin skin test (TST) [6,9,16]. However, neither test is considered a gold standard [4,14]. Despite this, there is a growing preference for IGRA over TST, given its advantages, which include the absence of a booster effect, single-visit completion, standardized positivity thresholds, and lack of cross-reactivity with the Bacillus Calmette-Guérin (BCG) or non-tuberculous mycobacteria [9,13].

Some studies have yielded mixed results on the comparative efficacy of IGRA and TST in the transplant population. While some report [2,3,10] greater sensitivity for QuantiFERON TB (QFT) compared to TST, even suggesting that a TST-only strategy would miss 50—78% of LTBI [10]; others [17] indicate similar performance in kidney transplant (KT) recipients, with 14 (21.9%) positive TST results, identical to the 14 (21.9%) positive QFT results. Systematic reviews [3,18] among haemodialysis patients support the former findings, showing significantly higher pooled sensitivity with QTF-G or T-SPOT.TB, while specificity remain similar. Concordance between tests varies, with kappa statistics ranging from 0.16 to 0.61 in end-stage renal disease (ESRD) patients and from 0.09 to 0.57 in transplant candidates [19].

In ESRD patients, ELISA-based IGRA demonstrated the strongest association with a clinical history of TB infection (OR 6.01, 95% CI 2.66–13.56), radiologic evidence of TB (OR 2.97, 95% CI 1.30–6.82), and contact with active TB (OR 3.52, 95% CI 1.69–7.31), compared to ELISPOT-based IGRA or TST [14]; Maung Myint et al. [11] also reported a correlation between IGRA results and radiological findings.

Thus, IGRA is preferred in SOT candidates due to its higher sensitivity and negative predictive value (NPV) [12,20], with a higher prevalence of LTBI reported with IGRA compared to TST [19]; notably, only 20–25% of all cases of active TB after transplantation have a prior positive TST reaction [5]. Moreover, the IGRA test appears to be more beneficial for ESRD patients [2,19], where high anergy rates [13] (ranging from 22.6 to 81%) reduce TST sensitivity, as well as in immunosuppressed individuals. While positive responses to IGRA are also diminished in these populations, they remain significantly higher than those observed with TST [8,16].

As there are no previous studies focusing on LTBI screening adherence in the transplant population, we present a sub-analysis of the ATALANTA-DOS study [21] to explore which screening strategy –TST or IGRA–achieves higher compliance among KT recipients.

Patients and methods

This subanalysis is part of ATALANTA-DOS, a quasi-experimental, prospective study conducted at the University Hospital “12 de Octubre” (Madrid, Spain) [21]. It included 307 KT recipients: 155 in the pre-intervention cohort (a historical cohort of ESRD patients who underwent KT at our centre between February 2016 and September 2017) and 148 patients in the intervention cohort (all consecutive ESRD patients aged >18 years who underwent KT between February 2018 and September 2019).

The mean age (SD) of transplant recipients was 55.7 (15.8) and 56.1 (15.7) years in the intervention and pre-intervention cohorts (p-value 0.67), with a male predominance of 64.2% and 67.9% (p-value 0.62), respectively. The most common cause of ESRD was glomerulonephritis (23.0% and 19.5%, p-value 0.54) followed by diabetic nephropathy (22.3% and 15.1%, p-value 0.44) in the intervention and pre-intervention cohorts, respectively. Overall, 19.6% (28/148) and 18.1% (28/155) of the patients included in the intervention and pre-intervention cohorts came from countries with a high TB burden, specifically from Africa and Central and South America (p-value 0.65). No data were recorded on TB vaccination rates, as this intervention is not part of routine practice in Spain.

Patients in the pre-intervention cohort underwent screening with TST (with booster administration) or IGRA, at discretion of the physician in charge, if no previous documented exposure was available.

Patients in the intervention cohort received a personalized infection prevention strategy at day +30 after KT, which included LBTI screening, vaccination, antimicrobial prophylaxis, and immunological assessments. Compliance with these measures was evaluated at months +2 and +12 post-transplantation. Initially, TST was the primary LTBI screening method, but low compliance led to a switch to IGRA after an adherence evaluation at the end of 2018. During the transition phase, both tests were performed.

TST was administered via the intradermal injection of 0.1 mL of purified protein derivative, with results interpreted after 48-72 hours. The cut-off for a positive TST result was defined as an induration >5 mm, regardless of BCG vaccination status [22]. Per previous recommendations, in place in our centre TST was performed, and if a negative result was obtained, a booster was administered 4 weeks later to rule out false negatives. IGRA was performed using the QuantiFERON-TB Gold Plus (Qiagen, Hilden, Germany) following standard protocols in an accredited diagnostic laboratory. Results were interpreted and reported as positive, negative and indeterminate according to the manufacturer’s instructions [23].

In case of a positive result either for TST or IGRA, LTBI treatment with isoniazid 300 mg daily for 9 months was recommended, provided active TB had been ruled out. The latter was achieved through clinical evaluation, chest radiography and Ziehl-Neelsen staining of three sputum samples.

Ethics. Compliance was defined as completion of the prescribed screening strategy. The study adhered to ethical guidelines and was approved by the local Clinical Research Ethics Committee (ref.no.16/342). Further details on the methods and statistical analysis used are available in the main article [21].

Results

LTBI screening was part of the care bundle performed on day +30 after KT, despite recommendations for its application before transplantation to avoid sensitivity loss resulting from immunosuppressive therapies, which were not followed due to logistical constraints.

Screening rates for LTBI using TST and/or IGRA markedly improved in the intervention cohort (82.6% [114/138] vs 1.3% [2/155]; p-value <0.001). Test compliance was significantly higher with IGRA (83.3% [55/66]) compared to TST (68.1% [49/72]) in the intervention cohort. This may be attributed to the need for a TST booster, which resulted in up to four hospital visits to complete the screening process. In three patients (75% [3/4]) both tests were performed due to the transition period between the two strategies.

Five cases of LTBI were detected post-transplant in the intervention cohort (4.4% [5/114], 80% [4/5] detected via IGRA and 20% [1/5] with TST), and all of them received treatment (100% [5/5]) with the previously described regimen. In contrast, the only two patients in the pre-intervention cohort who underwent screening tested positive for LTBI and completed treatment (100% [2/2]). However, additional cases of latent infection in the pre-intervention cohort may have remained undetected due to low screening rate.

No significant conclusions can be drawn regarding the relative efficacy of IGRA and TST due to limited data. Additionally, although treatment adherence was assessed, treatment failures or adverse effects were not recorded. Active TB was not detected in either cohort during the 12-month follow-up. The complete results can be found in the main article [21]. The study flow of this article is represented in Figure 1.

Figure 1.

Study flow diagram.

TST, Tuberculin Skin Test; IGRA, Interferon Gamma Release Assay; TB: Tuberculosis.

Discussion

The use of immunosuppressive drugs in transplantation, coupled with the rising number of transplant procedures, has led to a growing population of individuals at high risk of severe infectious diseases. To mitigate these risks, preventive strategies have been developed, with a strong emphasis on TB prevention. Alsdurf et al. [24] proposed a cascade of care for TB and identified four steps where the most losses occurred: initial testing of screening candidates, completing medical evaluation after a positive result, recommendation of treatment and completing therapy.

Historically, treatment adherence has been the main focus. Length of therapy, need for close follow-up and risk of hepatotoxicity reduce completion rates. Consequently, significant efforts have been allocated to developing shorter courses of therapy for LTBI. However, improvements in earlier steps have been suggested to have a greater impact on treatment completion than these new regimes.

As such, Alsdurf et al. [24] reported that while shorter regimens were associated with a 20% greater treatment completion rate, a greater benefit could be obtained by implementing measures designed to improve the proportion of patients undergoing screening, medical evaluation and initiation of treatment.

Studies directly evaluating adherence to LTBI screening are scarce and mainly focus on high-risk groups. Among active drug users, a study [25] found that return for TST reading was of 90% in the group receiving a $10 incentive and 85% in the group receiving a $5 incentive, but only reached 33% among those not receiving any monetary incentive. This contrasts with other studies[26] that report an adherence to LTBI screening of 87.3%, in part due to the presence of a nurse dedicated to the periodic monitoring of compliance and recovery of subjects that failed to keep their appointments. When comparing TST to IGRA adherence, Adams JW et al. [27] found that patients of an HIV primary care clinic were 1.45 times more likely to adhere to guidelines under IGRA protocol.

In KT recipients, no studies have directly investigated adherence to LTBI screening strategies. The only available evidence comes from studies that reported it as accessory data to the main endpoint. Singh et al reported only 47% of KT candidates undergoing TST [8], another study reported 12.3% of patients failing to return within 72 hours for TST reading, leading the authors to suggest QFT as the recommended test in high-risk population [17], and Casas et al. [7] reported that out of 95 liver transplant candidates undergoing both tests, eight failed to return for TST reading, compared to only one IGRA that was incomplete due to missing the positive control.

Our study [21] also found that compliance with LTBI screening was significantly higher with IGRA than with TST, likely due to fewer required hospital visits and reduced logistical burdens. Failure to return for TST readings suggests that IGRA, with its single-visit requirement, is a more practical option in transplant populations.

An important failure of transplant candidates to comply with screening strategy recommendations in the pre-intervention cohort was also reported [21], significantly improving after systematic evaluation of compliance with LTBI screening by an infectious disease (ID) specialist at day +30 after KT and active request if not previously performed. This underscores the importance of continuous reassessment of protocol compliance. However, poor compliance in the pre-intervention cohort may have been due to prior TST or IGRA conducted at the patients’ hospital of origin during haemodialysis, though this information was unavailable at the time of the study.

Although some clinical guidelines continue to favour TST, possibly due to cost-effectiveness considerations [18], recent reviews [20] on the cost-effectiveness of different screening strategies for LTBI concluded that, provided the higher sensitivity and prognostic value of IGRA were confirmed, the IGRA-only screening strategy should prove the more cost-effective option in high-risk populations [20].

Limitations of this subanalysis include its single-centre design and reliance on retrospective comparisons. However, both cohorts were managed by the same clinical team, under similar clinical protocols (baring the applied bundle), strengthening the validity of our findings. Secondly, due to logistical needs, all bundle components were evaluated at a single visit on day +30 after KT, rather than at pre-transplant evaluation or at waiting list inclusion. Although this time point was more beneficial for evaluation of some interventions, we acknowledge that other components of the bundle, particularly LTBI screening, should ideally be implemented before transplantation.

In conclusion, LTBI screening is an essential component of transplant evaluation, making strategies to increase adherence critical. The ATALANTA-DOS study found that systematic review by an ID specialist at day +30 after KT increased screening adherence and that IGRA-based strategies had a higher compliance rate that TST-based methods, making the former a preferable option for KT recipients. A more solid screening strategy will allow a more targeted and effective treatment of LTBI.

Funding

None to declare.

Conflicts of interest

No author reports any conflict of interest with regards to this manuscript.