Latent tuberculosis screening for inflammatory bowel disease in tuberculosis endemic region remains porous and suboptimal: A multicentre study

Abstract

BACKGROUND

Screening for latent tuberculosis (LTB) before initiating advanced therapy for inflammatory bowel disease (IBD) helps reduce the risk of tuberculosis (TB) development. However, there is limited data on screening practices from TB-endemic regions.

AIM

To study the practices of screening for LTB and study the incidence of TB in patients with IBD on biological and small molecule inhibitors.

METHODS

This retrospective multicentre study analyzed LTB screening practices in IBD patients starting advanced therapies between 2018 and 2022. We included patients who were initiated on biologics (infliximab, adalimumab, vedolizumab) or small molecule inhibitors (tofacitinib). We assessed compliance with LTB screening methods, including the tuberculin skin test, interferon-gamma release assay (IGRA), chest X-ray, and computed tomography chest, both at initiation and annually. We also evaluated the incidence of active TB and its predictors.

RESULTS

Of 378 patients (mean age: 36.9 ± 14.9 years, males: 56.9%), 158 (41.8%) and 216 (57.1%) had ulcerative colitis and Crohn’s disease, respectively. Advanced therapy used were anti-tumor necrosis factor in 309 (81.74%), tofacitinib in 41 (10.84%) and vedolizumab in 28 (7.40%). Standard screening and diligent screening strategy was employed in 59% and 33% of patients, respectively. Compliance with tuberculin skin test and IGRA was noted in 261 (69.04%) and 298 (78.83%) patients, respectively. Chest X-Ray and computed tomography chest were performed in 300 (79.36%) and 242 (64.02%), respectively. Annual screening in those on advanced therapy for > 1 year was performed in 27.2% (50/184). Active TB developed in 17 (4.49%); 15 (88.23%) were on anti-tumor necrosis factor. LTB was detected in 40 (10.72%), with most diagnosed on the basis of IGRA (21/40, 52.50%). Among 17 patients who developed active TB, LTB screen was negative in 12 (70.58%).

CONCLUSION

Standard screening practices for LTB, prior to starting advanced therapy, remain suboptimal (< 60%) in India despite high TB endemicity.

Key Words: Tuberculosis; Inflammatory bowel disease; Advanced therapy; Ulcerative colitis; Crohn’s disease

Core Tip: This multicenter study about inflammatory bowel disease patients on advanced therapy suggests that compliance with standard latent tuberculosis (LTB) screening, as recommended by LTB guidelines, is low at around 60%. Annual screening for LTB was found to be grossly neglected at 27%. Incidence of LTB was 10.7% and active tuberculosis (TB) developed in 4.5% of the patients on advanced therapy. Other interesting finding was occurrence of active TB with any advanced therapy, suggesting that there is no ‘safe’ therapy in TB endemic regions.

INTRODUCTION

Ulcerative colitis (UC) and Crohn’s disease (CD) are chronic immune-mediated inflammatory bowel disease (IBD) that may require immunosuppression with biologics and/or small molecule inhibitors for sustained remission[1]. Shifting targets in IBD, expanding armamentarium and increasing availability of these drugs have increased the usage of such advanced therapy[2,3]. Risk of developing active tuberculosis (TB) on anti-tumor necrosis factors (anti-TNFs) is reported to vary depending on the type of advanced therapy, TB endemicity, stringency of LTB screening and prophylactic treatment[4-9]. Latent TB (LTB) screening is one of the proposed actions to reduce the risk of TB reactivation before advanced therapy[4]. Screening for LTB is recognised to be porous, with many cases slipping through a rather imperfect screening tool.

Lack of adequate LTB screening is a strong predictor for the development of active TB[8,9]. Most guidelines recommend systematic screening for the presence of LTB or active TB before starting biologics and small-molecule inhibitors[10-12]. Systematic screening involves utilising the tuberculin skin test (TST) or interferon gamma release assay (IGRA) or both. Additionally, a chest X-ray (CXR) is advised to rule out active TB. These methods or combinations thereof for screening are believed to be inadequate in detecting subclinical TB[4]. Previous studies have shown that most patients who develop TB are negative for LTB screening[13-15]. LTB screening with CT chest in TB endemic regions was initially suggested for patients on anti-TNF for rheumatological diseases[16]. However, a gap remains between guidelines and real-world experience in LTB screening. We aimed to study the practices of screening for LTB and study the incidence, as well as the pattern of TB in patients with IBD on biologicals and small molecule inhibitors in a TB endemic region. Previous studies from this region have been single center and included only anti-TNF[9,13,15,17]. This study included a multicentric data, with large sample size and inclusion of both biologics and small molecules.

MATERIALS AND METHODS

Study design

This was a retrospective analysis of prospectively maintained databases at 17 centers across India. The database was maintained at participating centers through either a file-paper system or online records, wherein data was entered by a team of physicians running the IBD/gastroenterology clinic. All sites obtained institutional ethics committee clearances from respective committees and the need for informed consent was waived in view of the retrospective nature of the study. We followed the ethical guidelines of the Indian Council of Medical Research and Helsinki Declaration in the conduct of this study.

Setting and participants

Medical records of all patients with IBD (both UC and CD) who visited the hospital over the last 5 years were screened (2018-2022). Those who received advanced IBD therapy were included in the present study. All patients with IBD who received even a single dose of any biologic (infliximab, adalimumab, vedolizumab or biosimilar) or small molecule inhibitor (tofacitinib) were included for the study. We excluded patients if complete clinical details relevant to this study were not available.

Data extraction

The following parameters were retrieved from patient records through an electronic case report form: Demographic profile, IBD characteristics including location, extent, severity, and behaviour. Information on LTB infection screening methodology/tests utilised before initiating advanced therapy were recorded, including past history of TB, past antitubercular therapy (ATT), CXR, TST, IGRA, and computed tomography (CT) chest. Annual screening practices on patients with biologic/small molecule therapy for more than 1 year were also recorded. Results of the LTB screen, development of active TB on follow-up, and the interval between advanced therapy and development of TB were noted. A history of recent contact with active TB (someone in the family or close contact diagnosed and treated as TB) was actively sought for. If a patient had LTB, details of any therapy received, delay in onset of therapy and modification of biologic/small molecule therapy were recorded. Relevant information on the type of advanced therapy (anti-TNF agents, anti-integrin, Janus kinase inhibitor) and agents used in addition to the advanced therapy (steroids, thiopurines or methotrexate) was also noted. The principal investigators (Jha DK and Sharma V) regularly reviewed the data for completeness, and any missing information was retrieved telephonically from the centers and updated in the data records. Any discrepancies in data were rechecked and confirmed before analysis. Only records with complete information were included.

Outcomes assessed

We checked for compliance with LTB screening before starting biological therapy or small molecule inhibitors in patients with IBD. Incidence of active TB after advanced therapy in patients with IBD was noted, along with the impact of screening practices on the incidence of active TB. Outcome variables studied included: The frequency of TB screening at initiation for each advanced therapy, frequency of annual screening (using IGRA and CXR) when the patient was on advanced therapy for more than 1 year. We checked for the strategy used for LTB screening, as mentioned in the definitions below and the impact of using (or not using) a particular strategy. The positivity of LTB, as well as therapies administered (or not) for LTBI prior to starting advanced therapy, was noted. We also studied the frequency, as well as the pattern of TB, while the patient was on advanced therapy. We scrutinised from available data the relationship of development of TB while on advanced therapy with respect to completeness of LTB screening and treatment.

Definitions used

Diagnosis of UC and CD was made based on European Crohn’s and Colitis Organization guidelines[18]. Various definitions utilised for study outcomes are as under.

LTB: Patients were diagnosed as having LTB if any of the following was positive: (1) Medical history of previous TB or recent contact with active TB at home; (2) Positive TST: A TST reaction of ≥ 10 mm of induration at 72 hours in Bacillus Calmette-Guérin (BCG)-vaccinated individuals and a TST reaction of ≥ 5 mm of induration at 72 hours in BCG non-vaccinated; (3) Positive IGRA (QuantiFERON Gold or T-SPOT Test); and (4) CXR and/or CT chest findings consistent with post-infective sequelae. Although in the definitions used for the purpose of this study, we did not use past history of TB, as a qualifying condition, clinicians continue to use it as a marker of LTB. It is unclear if adequately treated past TB should be considered as LTB equivalent or not.

Strategy 1: Standard TB screening group (patients who had undergone all three standard LTB screening tests i.e., TST, IGRA, CXR) and incomplete TB screening group [patient who missed any single LTB screening test among the three standard tests (TST, IGRA, CXR)]. For the purpose of this analysis, we excluded any patients who had undergone CT of the chest.

Strategy 2: Diligent TB screening group (patients who had undergone all LTB screening tests, including TST, IGRA, and CT chest) and limited TB screening group (patients who missed any single LTB screening test among TST, IGRA, and CT chest). In this analysis, all patients were included and those who missed even one test of the three (TST, IGRA and CT chest) were considered to have undergone limited screening, irrespective of the CXR.

Strategy 3: Immunodiagnostic predominant screening - patients were categorised as those undergoing both TST and IGRA, either of these two tests or neither of these tests, irrespective of the status of CXR or CT chest.

Strategy 4: CT-based screening - patients were analyzed for risk of developing active TB, comparing patients who underwent CT chest for screening or not.

Active TB: Pulmonary TB was diagnosed in the presence of clinical symptoms (fever/cough/anorexia/weight loss) and evidence of fresh lesions suggestive of TB on the CXR/contrast-enhanced CT chest with demonstration of acid-fast bacilli (AFB) or positivity on culture or polymerase chain reaction on the sputum smear examination. In the absence of positive microbiological evidence, a consistent clinical and radiological picture with clinical and radiological response to ATT was considered as TB and labelled as clinically diagnosed TB. Extrapulmonary TB was diagnosed based on clinical features, suggestive radiologic findings, and demonstration of positive AFB/culture/polymerase chain reaction for M. TB or caseating or noncaseating granulomas on biopsy specimens. Diagnosis of pleural TB/peritoneal TB was based on biochemical evaluation of pleural/peritoneal fluid showing a high protein content along with a high adenosine deaminase content (≥ 40 IU/mL for pleural and ≥ 39 IU for peritoneal TB) and lymphocytic predominance or a positive microbiological test (AFB stain/culture/GeneXpert MT). Patients with evidence of TB at more than one site were diagnosed with disseminated TB.

Advanced therapy: IBD patient receiving any biologic or small molecule inhibitor (anti-TNF, vedolizumab and tofacitinib).

Annual screening: Screening for LTB annually (using IGRA and CXR) when the patient was on advanced therapy for more than 1 year.

Combined immunosuppression: Included patients on advanced therapy along with either azathioprine or methotrexate.

Concomitant steroids: Patients on advanced therapy receiving steroids irrespective of the dose used. Prednisone > 20 mg/day was considered high-dose steroid use.

Sample size

Being a retrospective study, we included all eligible patients.

Statistical analysis

Categorical variables were expressed as percentages, and continuous variables were expressed as mean ± SD or median (interquartile range) as appropriate. χ² test was used to compare categorical variables and Independent Sample T test/Mann-Whitney U test was used to compare continuous variables as appropriate. P < 0.05 was considered statistically significant. Statistical software SPSS was used for statistical analysis.

RESULTS

Patient population

We collected 395 electronic responses from 17 centers through the case report form. Seventeen incomplete records were excluded before analysis. This multicentric study enrolled 378 IBD patients on advanced therapy with a mean age of 36.9 ± 14.9 years, 215 (56.9%) being males. The number of patients with UC, CD and IBD-unclassified were 158 (41.8%), 216 (57.1%) and 4 (1.1%), respectively. The first advanced therapy used was anti-TNF in 309 (81.7%) [infliximab: 207 (54.8%), adalimumab: 102 (26.9%)], tofacitinib in 41 (10.8%), and vedolizumab in 28 (7.40%). Patients exposed to sequential multiple advanced therapy were 36 (9.52%). Combined immunosuppression was prescribed in 230 (60.8%) of patients and most of them were on azathioprine 222 (96.5%). Concomitant steroids were being utilised in 176 (46.6%) patients, with the majority of them being on low-dose steroids 102 (57.9%). In this study, no patient received a combination of advanced therapy. Past history of TB was present in 44 (11.6%) patients, with the majority 30/44 (68.18%) having TB > 1 year prior to starting advanced therapy. The history of BCG vaccination was present in 247 (65.34%) patients. Baseline characteristics of the patients are highlighted in Table 1.

Table 1 Baseline characteristics of inflammatory bowel disease patients on advanced therapy, n (%)/mean ± SD.

Number of patients screened	n = 378
Age at diagnosis, years	31.2 ± 14.4
Gender
Male	215 (56.9)
Female	163 (43.1)
IBD type
CD	216 (57.1)
UC	158 (41.8)
Unclassified	4 (1.1)
Age at screening, years	36.9 ± 14.9
IBD phenotype
CD (n)	216
Ileal (L1)	28 (12.96)
Colonic (L2)	24 (11.11)
Ileocolonic (L3)	118 (54.62)
Isolated upper GI (L4)	6 (2.77)
Unknown	41 (18.98)
UC (n)	158
Proctitis (E1)	4 (2.53)
Left sided (E2)	51 (32.27)
Extensive (E3)	97 (61.39)
Unknown	5 (3.16)
Unclassified	4 (1.1)
Initial advanced therapy used
Anti TNF	309 (81.74)
Infliximab	207 (66.99)
Adalimumab	102 (33.00)
Tofacitinib	41 (10.84)
Vedolizumab	28 (7.40)
Patients exposed to sequential multiple advanced therapy	36 (9.52)
Combined immunosuppression (AZA or methotrexate)	230 (60.84)
AZA	222 (96.52)
Methotrexate	8 (3.47)
Concomitant steroids	176 (46.56)
Low dose steroids	102 (57.95)
High dose steroids	74 (42.04)
Past tuberculosis
None	334 (88.35)
Yes, < 1 year	14 (3.70)
Yes, > 1 year	30 (7.93)
History of BCG vaccination
Yes	247 (65.34)
No	22 (5.82)
Unsure	109 (28.83)

IBD: Inflammatory bowel disease; CD: Crohn’s disease; UC: Ulcerative colitis; GI: Gastrointestinal; TNF: Tumor necrosis factor; AZA: Azathioprine; BCG: Bacillus Calmette-Guérin.

Compliance of screening practices for LTB

Audit of the LTB screening practices revealed compliance with TST in 261/378 (69%), IGRA in 298/378 (78.8%), CXR in 300/378 (79.1%) and CT chest in 242/378 (64.1%), as summarised in Table 2. Five patients did not undergo any type of screening before starting advanced therapy. In patients not undergoing a CT scan, standard screening was performed in 77 of 131 (58.77%) and incomplete screening in 54 patients (54/131, 41.22%). Among all patients, both TST and IGRA were performed in 208/378 (55.02%), 353/378 (93.38%) underwent either of two and 25/378 (6.61%) neither of the two tests. Patients on advanced therapy for more than 1 year were 184/378 (48.7%). Compliance with annual LTB screening with either TST or IGRA was 50/184 (27.17%) among patients with biologics for more than 1 year.

Table 2 Audit of screening before advanced therapy, n (%).

Latent TB screening1	n = 378
TST	261 (69.04)
IGRA	298 (78.83)
QuantiFERON Gold	295
TSPOT2	7
CXR	300 (79. 36)
CT chest (HRCT/CECT chest)	242 (64.02)
Standard screening3	77/131 (58.77)
Incomplete screening3	54/131 (41.22)
Immunodiagnostic predominant
Both	208 (55.02)
Either	353 (93.38)
Neither	25 (6.61)
Diligent screening	125 (33.06)
Limited screening	253 (66.93)
Patents on advanced therapy > 1 year	184 (48.67)
Annual screening of patients with advanced therapy > 1 year	50/184 (27.17)

¹Five patients no screening done.

²Four patients underwent both tests.

³Excluding all patients undergoing computed tomography chest.

TB: Tuberculosis; TST: Tuberculin skin test; IGRA: Interferon-gamma release assay; CXR: Chest X-ray; CT: Computed tomography; HRCT: High-resolution computed tomography; CECT: Contrast-enhanced computed tomography.

LTB

LTB was detected in 40 (10.72%, 40/373), with most being diagnosed on the basis of IGRA (21/40, 52.50%). Six (15%) patients were detected with LTB based only on CT chest, with other tests being negative. Only 25 of the 40 (62.5%) LTB screen-positive patients received prophylactic therapy. Progression to active TB was noted among 2 of 15 (13.3%) untreated LTB patients and 3 of 25 (12.0%) who received prophylaxis for LTB. Seventeen patients (4.49%) developed active TB on advanced therapy and most of them had pulmonary TB (9/17, 52.94%). Among active TB patients most were either on combined immunosuppression or steroids (16/317, 5.0% vs 1/61, 1.6%, P = 0.33). Among active TB patients, most were on anti-TNF biologicals 15/17 (88.23%). Median time to develop TB after advanced therapy initiation was 7 months (4-8 months). Data on LTB and active TB are highlighted in Tables 3 and 4.

Table 3 Characteristics of latent and active tuberculosis, n (%).

Variables
No of patients with LTB1	40/373 (10.72)
CD	23/216 (10.6)
UC	16/158 (10.1)
IBD-U	1/4 (25)
Basis of LTB2
IGRA	21/40 (52.50)
TST	9/40 (22.50)
CT chest	11/40 (27.50)
Past history of incomplete TB treatment	2/40 (5)
History of contact with pulmonary TB	1/40 (2.5)
LTB positive patients receiving treatment3	25/40 (62.50)
No of patients with active TB	17/378 (4.49)
Pulmonary	9 (52.94)
Extrapulmonary	8 (47.05)
Advanced therapy used among active TB
Anti-TNFi	15 (88.2)
Infliximab	12
Adalimumab	3
Tofacitinib	1 (5.88)
Vedolizumab	1 (5.88)
Median time to develop TB after advanced therapy initiation	7 (4-8) months

¹Five patients no screening done.

²Combination of test positive (n = 6).

³Total 28 received prophylactic treatment for latent tuberculosis, 3 patients received prophylactic therapy exclusively based on past history of tuberculosis.

LTB: Latent tuberculosis; CD: Crohn’s disease; UC: Ulcerative colitis; IBD: Inflammatory bowel disease; IGRA: Interferon-gamma release assay; TST: Tuberculin skin test; CT: Computed tomography; TB: Tuberculosis; TNFi: Tumor necrosis factor inhibitor.

Table 4 Predictors of active tuberculosis among inflammatory bowel disease patients on advanced therapy, n (%)/mean ± SD.

	Active tuberculosis (n = 17)	No tuberculosis (n = 361)	P value
Age (years)	40.93 ± 15.60	36.76 ± 14.96	0.263
Gender, male	13 (76.5)	203 (56.2)	0.095
IBD type			0.761
CD	11 (64.7)	205 (56.8)
UC	6 (35.3)	152 (42.1)
Unclassified	0	4 (1.1)
Duration of disease (months)	60 (12-72)	48 (24-72)	0.379
TNFi	15 (88.2)	294 (81.4)	0.748
Infliximab	12	195
Adalimumab	3	99
Standard screening	4 (23.52)	73 (20.22)	1.000
Diligent	7 (41.17)	118 (32.68)	0.444
Immunodiagnostic			0.905
TST and IGRA	11 (64.70)	197 (54.57)
TST or IGRA	16 (94.11)	337 (93.35)
Neither	1 (5.88)	24 (6.64)
CT done	10 (58.82)	232 (64.26)	0.617
Combination immunosuppression	16 (94.11)	301 (83.37)	0.329
Steroids			0.673
Any dose	5 (29.41)	171 (47.36)
High dose	1 (5.88)	73 (20.22)
Presence of LTB	5 (29.41)	35 (9.69)	0.033

IBD: Inflammatory bowel disease; CD: Crohn’s disease; UC: Ulcerative colitis; TNFi: Tumor necrosis factor inhibitor; TST: Tuberculin skin test; IGRA: Interferon-gamma release assay; CT: Computed tomography; LTB: Latent tuberculosis.

Comparing screening strategies for LTB

LTB was present in 1 of 77 (1.29%) and 11 of 54 (20.37%) patients among standard and incomplete screening groups, respectively. Similarly, LTB was present in 18 of 125 (14.40%) among the diligent screening group and 22 of 253 (8.69%) among the limited screening group. In the immunodiagnostic group, LTB was present in 24 of 208 (11.53%) in both test groups, 40 of 353 (11.33%) in either test group. Similarly, among 28 (11.57%) LTB patients of the CT scan group, 11 (39.28%) were identified using the CT scan strategy. This data has been summarised in Supplementary Table 1.

Comparison of screening strategies for development of active TB

We analysed various screening strategies for predicting active TB. Active TB developed in 4/77 (5.19%) in standard screening and 3/54 (5.55%) in incomplete screening (P = 1.00). Similarly, on including patients undergoing CT scan, comparing risk of active TB between diligent screening (7/125, 5.60%) and limited screening (10/253, 3.95%) was found to be non-significant (P = 0.444). In Immunodiagnostic predominant screening strategy comparing both tests (11/208, 5.28%) to either test (16/353, 4.53%) or neither of the tests (1/25, 4.0%) was found to be non-significant (P = 0.905). Comparing patients undergoing CT-based screening vs non-CT based screening also revealed no significant difference (10/242, 4.1% vs 7/136, 5.1%, P = 0.617). All screening strategies utilised to predict active TB are summarised in Table 4.

Predictors of active TB

Patients detected to have LTB were found to be at higher risk of developing active TB (12.50%, 5/40) as compared to patients with negative LTB screen (3.60%, 12/333) with a P value of 0.033. There was no significant relation between the use of anti-TNF, tofacitinib and vedolizumab with the development of active TB (P = 0.759). Patients on combination therapy had a higher risk of active TB (16/317, 5.0% vs 1/61, 1.6%) as compared to monotherapy with 5 aminosalicylic acid agents, but was found to be statistically insignificant. Completeness of screening, either immunodiagnostic predominant (TST/IGRA) or imaging predominant (CT chest done or not), showed no significance with P value of 0.905 and 0.617, respectively. The relationship of active TB with advanced therapy and LTB has been highlighted in Table 4.

Effect of active TB on advanced therapy for IBD

Among 17 patients who developed active TB, advanced therapy was stopped in 15 patients and continued in 02 patients. Among 15 patients who stopped advanced therapy, 10 patients were not restarted and 5 patients were restarted on advanced therapy. In 3 patients advanced therapy was restarted after completion of ATT and 2 patients restarted after 12 weeks of ATT. Among 2 patients who continued advanced therapy, one on tofacitinib underwent de-escalation of dose and one on vedolizumab continued the therapy.

DISCUSSION

This multicenter retrospective study evaluating the LTB screening practices before advanced therapy use in a TB endemic region suggests suboptimal compliance with the LTB screening guidelines. Standard screening, CT-based screening and annual screening for LTB was found grossly neglected, compliance being only 59%, 64% and 27%, respectively. Incidence of LTB was 10.7% and active TB developed in 4.5% of the patients on advanced therapy. Our study had a lower reactivation of TB compared to previous reports, which may be related to decreasing TB incidence[19]. Also, patients affording such treatment may not represent patients with typical risk factors for TB. Active TB developed primarily in patients on anti-TNF (4.85%, 15/309), with a case each with anti-integrin (3.57%, 1/28) and Janus kinase inhibitor (2.43%, 1/41). The presence of LTB showed significance with the development of active TB (P = 0.033). Type of advanced therapy and type of IBD showed no significant association with the development of active TB, although active TB was numerically higher in the anti-TNF group. The study has some interesting findings apart from the low compliance with TB screening guidelines. It shows a lower incidence of active TB on advanced therapy in India compared to the past and active TB can occur with any advanced therapy, suggesting that there is no ‘safe’ therapy in TB endemic regions.

TB is a dynamic infection with risk of reactivation on immunosuppression or de-novo infection due to re-exposure. Screening before advanced therapy is intended to identify those at risk of reactivation i.e., having LTB[4]. Limited data shows up to 20% of the patients on anti-TNF with negative LTB screening at baseline have conversion to positive test on annual screening and European Crohn’s and Colitis Organization suggests annual screening in TB endemic regions[12,20,21]. This study showed development of active TB was numerically higher in patients on anti-TNF compared to tofacitinib or vedolizumab but was not found significant (P = 0.759). Active TB was noted to be significantly higher among patients with LTB (P = 0.033). This data emphasises the role of adherence to LTB screening guidelines irrespective of the type of advanced therapy. However, our study noted suboptimal adherence to the LTB screening recommendation. Only a quarter (27.17%) of patients on advanced therapy for more than 1 year (48.67%) underwent annual screening. LTB screening compliance was higher with IGRA (79% vs 69%) as compared to TST, which may be explained by the ease of performing and avoidance of additional visit[4]. Screening with CXR should be done in all, but was done in only 79% of patients. Diagnostic limitation of CXR have led to increasing utilisation of CT chest in LTB screening and has been suggested to reduce reactivation[9,16]. Our study showed compliance with CT in only 64% of patients. Radiation exposure, practicality and cost-effectiveness related to CT screening has been poorly studied[22]. Low dose CT has been advocated to reduce radiation exposure but lacks standardisation, limiting applicability[22]. Although, stringent screening (inclusive of CT chest) has been recommended on the basis of a recent study showing a reduction in reactivation of TB from 17% to 1.7%[9]. A close reading of the above study shows that probably use of CT chest was not a determinant in reducing the TB rates- the CT was performed in 83% of the non-stringent group also. On the other hand, it is the increase in treatment of LTB (after detecting it) which might be responsible for the difference, from 45% to 88%. Even in the present study, a significant number of patients did not receive LTB prophylaxis. Although concerning, it is unclear why clinicians might not treat LTB after detecting it. Importantly 17 patients who developed active TB, 3 patients had received LTB prophylaxis before initiation of advanced therapy. This may be explained by either drug resistance, reinfection, or drug noncompliance.

A retrospective Spanish study analysed 50 patients who developed active TB on anti-TNF and noted 40% of them had failed compliance with national TB screening guidelines[8]. An early survey in France (2009) among gastroenterologists revealed only a quarter of them ordered TB screening prior to anti-TNF[23]. Data from the Netherlands among 611 CD patients between 2000-2010 receiving anti-TNF, revealed variable compliance with LTB screening (TST: 57%, IGRA: 17% and CXR: 63%, while only 2% had all three tests)[24]. Similarly, ten years of veteran data that included 3357 IBD patients till 2011, noted compliance with TST or IGRA in around 70% but CXR in only 38% (only 1% of them underwent testing with both TST and IGRA)[25]. Our data shows a low compliance in a TB endemic region that is almost similar to the Western data. Compliance to screening may be varied due to academic affiliation of hospitals, rural urban divide, frequency of hospital visits and number of comorbidities[25]. However, if we were to measure LTB screening practices as per time frame, it should have improved with increasing availability, larger evidence and clearer guidelines.

To the best of our knowledge this study is the first multicentre audit of LTB screening practices among IBD patients in high-endemic regions, increasing its generalizability. Limitations of the study include retrospective design, potential recall bias, missing data, and lack of microbiological confirmation in some TB cases. Because this was not a prospectively planned study, it best collects real-world evidence providing information about the current practices in India. However, although audits are typically retrospective, this comes at the cost of many details that can be better evaluated in a prospective design. A complete battery of tests for LTB may not be utilised by all clinicians as one of the initial immunological tests or easily available tests for LTB may have turned positive. Impact of LTB prophylactic treatment on active TB may not be reflective of actual benefit due to small numbers. More importantly, the patients receiving advanced therapy by virtue of better socio-economic status may not truly represent the patients with heightened TB risk e.g., those from lower socioeconomic status and living in overcrowded conditions or suffering from undernutrition.

CONCLUSION

The current study provides concerning data regarding the lack of optimal screening prior to initiation of advanced therapy. We also provide data to support LTB screening irrespective of the advanced therapy used in TB endemic regions. Future studies may focus on compliance with LTB prophylactic therapy among IBD patients with LTB and its effect on active TB outcome. LTB screening adherence may be improved by collaborative effort of institutional and pharmacological industry.

ACKNOWLEDGEMENTS

Dr Manoj Kumar Gupta, MD, Community medicine specialist for biostatistics review; Dr Sourabh Aggarwal and Ms Nikita Sharma for English editing of the manuscript.