Screening for tuberculosis and hepatitis B prior to the initiation of anti-tumor necrosis therapy

Authors

  • Byron P. Vaughn MD,

    Corresponding author
    1. Division of General Medicine and Primary Care, St. Vincent's University Hospital/University College Dublin, Ireland
    2. Beth Israel Deaconess Medical Center, St. Vincent's University Hospital/University College Dublin, Ireland
    • Beth Israel Deaconess Medical Center, 330 Brookline Ave., Boston MA 02215
    Search for more papers by this author
  • Glen A. Doherty MD,

    1. Beth Israel Deaconess Medical Center, St. Vincent's University Hospital/University College Dublin, Ireland
    2. Gastroenterology, Harvard Medical School, Boston, Massachusetts, St. Vincent's University Hospital/University College Dublin, Ireland
    3. Division of Gastroenterology, Center for Colorectal Disease, St. Vincent's University Hospital/University College Dublin, Ireland
    Search for more papers by this author
  • Shiva Gautam PhD,

    1. Division of General Medicine and Primary Care, St. Vincent's University Hospital/University College Dublin, Ireland
    2. Beth Israel Deaconess Medical Center, St. Vincent's University Hospital/University College Dublin, Ireland
    Search for more papers by this author
  • Alan C. Moss MD,

    1. Beth Israel Deaconess Medical Center, St. Vincent's University Hospital/University College Dublin, Ireland
    2. Gastroenterology, Harvard Medical School, Boston, Massachusetts, St. Vincent's University Hospital/University College Dublin, Ireland
    Search for more papers by this author
  • Adam S. Cheifetz MD

    1. Beth Israel Deaconess Medical Center, St. Vincent's University Hospital/University College Dublin, Ireland
    2. Gastroenterology, Harvard Medical School, Boston, Massachusetts, St. Vincent's University Hospital/University College Dublin, Ireland
    Search for more papers by this author

Abstract

Background:

Since the introduction of infliximab, anti-tumor necrosis factor alpha (anti-TNF-α) agents have been used with increasing frequency for the treatment of inflammatory bowel disease (IBD). Reactivation of latent Mycobacterium tuberculosis (TB) soon became recognized as a complication of therapy. More recently, reactivation of hepatitis B while on anti-TNF therapy has been documented. The aim of this study was to assess the adherence to screening for latent TB and hepatitis B by gastroenterologists prior to initiation of an anti-TNF.

Methods:

This is a retrospective analysis of all patients with IBD treated with an anti-TNF at a large urban academic hospital. In our population, 65% of patients were screened for latent TB prior to the initiation of anti-TNF therapy, while 25% of patients were screened for hepatitis B.

Results:

Failure to screen for latent TB was strongly correlated with prior exposure to an anti-TNF (odds ratio [OR]: 5.3; P < 0.0001) and initiation of treatment prior to 2006 (OR: 5.8; P < 0.0001). Failure to screen for hepatitis B was associated with lack of an abnormal alanine aminotransferase (OR: 2.6; P = 0.005) and treatment prior to 2010 (OR: 3.3; P = 0.02). Providers who had been in practice longer were less likely screen for TB or hepatitis B.

Conclusions:

The rate of screening for both latent TB and hepatitis B in this study was inadequate. While the rate of screening is increasing, further systems improvements and physician education is needed. (Inflamm Bowel Dis 2012;)

Anti-tumor necrosis factor (anti-TNF) agents have become an increasingly prominent option in the treatment of several immune-mediated inflammatory disorders. Infliximab, the first of these agents to receive Food and Drug Administration (FDA) approval, has largely revolutionized the management of inflammatory bowel disease (IBD) through its targeted inhibition of underlying inflammatory mediators. Increasingly recognized as first-line therapy for moderate-to-severe Crohn's disease (CD), infliximab has also demonstrated utility in moderate-to-severe ulcerative colitis (UC),1 which has led to its growing use in all aspects of IBD. More recent additions to the anti-TNF armamentarium include adalimumab and certolizumab pegol. All three anti-TNFs are now FDA-approved for the treatment of CD, although only infliximab is FDA-approved for UC.

While the TNF-α immune pathway can be inappropriately activated in IBD, it can also play a critical role in the control of certain infections. Shortly after the introduction of infliximab, numerous clinical reports documented the development of serious infections in patients receiving the drug.2–5 The most worrisome of these involved the reactivation of latent Mycobacterium tuberculosis (TB).6 TNF-α is central to the process of granuloma formation and its inhibition can result in reactivation of granulomatous disease.7, 8 While less well understood, but increasingly recognized, hepatitis B also has the potential to reactivate with anti-TNF treatment.9, 10 Reactivation of both TB and hepatitis B has resulted in substantial morbidity and mortality in patients receiving these agents.6, 11

In response to the potential infectious complications, a range of clinical guidelines for infection screening prior to the initiation of anti-TNF treatment have been developed by both national and international groups. The American College of Gastroenterology (ACG) recommends screening for latent TB with a tuberculin skin test (TST) and a chest x-ray.1, 12 The American Gastroenterological Association (AGA) recommends a TST prior to anti-TNF therapy.13 Both the ACG and the AGA emphasize the importance of a thorough risk factor assessment to TB given the high risk of anergy in this patient population.14 The American College of Rheumatology (ACR) recommends routine screening with a TST regardless of Bacille Calmette-Guérin (BCG) status combined with a thorough history and chest x-ray, as indicated.15 The European Crohn's and Colitis Organization (ECCO) guidelines differ in that there is less emphasis on TST given the high rate of BCG vaccination and more emphasis on obtaining a screening chest x-ray prior to anti-TNF initiation.16

In the most recent recommendation for UC management, the ACG recommends screening for hepatitis B prior to treatment with infliximab.12 While the ACR does not make a specific recommendation for screening for hepatitis B, they do recommend vaccination if risk factors are present.15 In 2009, the ECCO recommended baseline hepatitis B serologies and vaccination if not previously done.17 Testing for hepatitis B prior to initiation of chemotherapy in patients with underlying malignancies is advised and if the surface antigen is positive, prophylactic treatment with lamivudine is recommended. For patients who demonstrate hepatitis B core antibody, close monitoring with liver function tests and hepatitis B DNA titers are suggested.18

Data from British rheumatologists have indicated variable adherence to the recommendations for latent TB infection screening prior to anti-TNF therapy.19 It is unknown to what extent infection screening is practiced in the U.S. gastroenterology community. The aim of this study was to assess the level of adherence by gastroenterologists to recommended guidelines for screening for TB as well as hepatitis B prior to initiating anti-TNF therapy.

PATIENTS AND METHODS

Patient Population

The cohort was composed of IBD patients at a large urban academic medical center in Boston, Massachusetts treated with an anti-TNF agent between 1998 and 2010. The electronic medical records (EMR) were searched using the ICD-9 codes for IBD (555.x and 556.x) and cross-referenced with an anti-TNF logged as a pharmacy code in the electronic medications list from 1998 through 2010. In all, 464 patients were identified. The EMR at this center is extensive and has been in use for over 25 years. In the mid 2000s there was a transition requiring all billable encounters in the medical system to be accompanied by a note in the EMR system. Each EMR chart was reviewed for demographic information, disease-specific information, data regarding the anti-TNF, and screening for TB and hepatitis B. Patients were excluded if they were started on an anti-TNF prior to having care at the medical center, prior to when the patients chart was stored electronically, or if the gastroenterologist was not prescribing the anti-TNF. After exclusion, there were 287 patients for whom data were collected. The Institutional Review Board approved this study.

The academic practice is composed of five gastroenterologists who make up the Center for IBD and 16 general and subspecialty gastroenterologists. Demographic data regarding the gastroenterologists (gender, year graduated form medical school, and U.S. medical school and residency) were collected from the hospital's public Web page. To attempt to capture if provider experience was significant, physicians were categorized into those who had greater than 10 years of experience in 1998 versus those who did not. Anti-TNF prescribing volume was categorized into physicians who treated fewer than five patients with an anti-TNF and those who treated more.

Data Collection

Screening for latent TB was considered valid if 1) a TST was done within a year before starting anti-TNF therapy and 2) the result was documented by a healthcare provider in the electronic medical records. For patients with a known positive TST in the past, a chest x-ray was required for proper screening. Patient reporting of a verbal negative TST within the previous year was counted as “not screened.” Interferon gamma release assays (IGRA) were considered an acceptable means of screening for latent TB if done within the appropriate time frame. As it is difficult to assess physician awareness of the risk of latent TB reactivation over time, we used published guidelines as a surrogate for awareness. Thus, rates of screening before and after the AGA recommendation in 2006 were compared.

Hepatitis B screening was considered complete if a hepatitis B surface antigen was sent within 1 year prior to starting an anti-TNF. For hepatitis B surface antibody a minimum of 10 mIU/mL was considered positive. Hepatitis B core antibody was considered positive if there was any detectable level. Any alanine aminotransferase (ALT) above the laboratory normal of 40 IU/L was noted prior to and during therapy with anti-TNF. Since the most recent ACG guidelines for UC recommend screening for hepatitis B, we chose to specifically compare screening rates prior to 2010 with rates of screening after 2010.

Statistical Analysis

Differences in the populations of patients screened for tuberculosis or hepatitis B were compared using the chi-square test or the 2-tailed Fisher exact test. A P-value less than 0.05 was considered significant. Significant variables were then reanalyzed in a multivariate logistic regression. Tests of trend over time were done with linear regression and piecewise regression. All statistical analyses were performed using JMP software v. 8 and SAS v. 9.2 (Cary, NC).

RESULTS

Patient and Provider Characteristics

Characteristics of the patients and providers are displayed in Table 1. A total of 287 IBD patients who received an anti-TNF between 1998 and 2010 were included in the final analysis. CD was the most prevalent diagnosis (78%) and most patients received infliximab (72%). Three patients were started on an anti-TNF agent with a diagnosis of IBD-unclassified and were excluded from subsequent multivariate analysis. Most of the providers were male (86%) and trained in the U.S. (71%). Approximately half of the physicians had more than 10 years of experience in 1998.

Table 1. Characteristics of Patients and Providers
CharacteristicValue
  • TNF, tumor necrosis factor; IBD, inflammatory bowel disease; US, United States.

  • a

    Diagnosis at the time on initiation of anti-TNF.

  • b

    Based on 2006 AGA guidelines.

  • c

    Based on 2010 ACG guidelines.

Median age (range)41 (19–89)
Age group, no. (%) 
 <3057 (20)
 30–3962 (22)
 40–4979 (27)
 50–5957 (20)
 >6032 (11)
Sex-no. (%) 
 Female148 (52)
 Male139 (48)
Diagnosis,a no. (%) 
 Crohn's disease223 (78)
 Ulcerative colitis61 (21)
 Indeterminate colitis3 (1)
Anti-TNF initiated, no. (%) 
 Infliximab206 (72)
 Adalimumab72 (25)
 Certolizumab9 (3)
Treatment started prior to 2006,b no. (%)52 (18)
Treatment started prior to 2010,c no. (%)254 (89)
Treated at IBD Center, no. (%)187 (65)
Provider characteristics, no. (%) 
 Male18 (86)
 Trained in US medical school15 (71)
 Trained in US residency19 (90)
 Part of IBD center5 (25)
 Treated more than 5 patients with anti-TNF12 (57)
 More than 10 years of experience as attending in 199811 (55)

Tuberculosis Screening

Data regarding screening for latent TB is displayed in Table 2. In our population 65% (187/287) of patients were appropriately screened for latent TB. In only 48 cases (17%) was there documentation that risk factors for TB were discussed. Fifty-five percent of patients were on prednisone or an immunomodulator at the time of TST placement.

Table 2. Characteristics of Screening for Tuberculosis and Hepatitis B
CharacteristicValue, no. (%)
  • BCG, Bacille Calmette-Guérin; TB, tuberculosis; TST, tuberculin skin test; Ab, antibody; ALT, alanine aminotransferase.

  • a

    Screened indicates that a negative TST was documented or TST not indicated and appropriate screening done via other modality.

  • b

    Negative screen if hepatitis B surface antigen negative.

Tuberculosis 
 Screened 1 year prior to treatmenta187 (65)
 History of BCG5 (2)
 TB risks factors documented48 (17)
 Had chest imaging within 1 year122 (43)
 Had TST and chest imaging within 1 year89 (31)
 On prednisone when TST done66 (36)
 On immunomodulator when TST done62 (24)
Hepatitis B 
 Screened 1 year prior to treatmentb72 (25)
 Screened ever prior to treatmentb107 (37)
 Hepatitis B surface Ab positive prior to treatment33 (12)
 Hepatitis B core Ab positive prior to treatment3 (1)
 Hepatitis B core Ab ever positive7 (3)
 Hepatitis B core Ab ever sent123 (43)
 Abnormal ALT prior to treatment96 (33)

We analyzed those factors associated with a failure to screen for TB (Table 3). Prior exposure to any anti-TNF greater than 1 year prior to treatment was strongly associated with failure to screen for TB (odds ratio [OR]: 5.3; P < 0.0001), as was anti-TNF treatment that was initiated prior to 2006 (OR: 5.8; P < 0.001). Providers who had been attending physicians for more than 10 years in 1998 were significantly more likely not to screen for latent TB (OR: 2.5; P = 0.002). Patients not being treated by a physician at the Center for IBD were also more likely to be not screened (OR 1.9; P = 0.04).

Table 3. Factors Associated with Not Screening for Infection Prior to Initiating Treatment
CharacteristicUnivariate Odds Ratio (CI)P-valueMultivariate Odds Ratio (CI)P-value
  1. CI, confidence interval; IBD, inflammatory bowel disease, TNF, tumor necrosis factor; ALT, alanine aminotransferase, US, United States.

Tuberculosis    
 Not seen at IBD center2 (1.2–3.4)0.0071.9 (1 – 3.4)0.04
 Prior exposure to anti-TNF3.7 (2.1–6.7)<0.00015.3 (2.8–10.3)<0.0001
 Treatment started prior to 20065.3 (2.8–10.3)<0.00015.8 (2.8–11.9)<0.0001
 Provider with >10 years as attending in 19983.5 (2.2–6)<0.00012.5 (1.4–4.5)0.002
 Age >501.6 (0.4–1)0.06
 Male gender1.1 (0.6–1.4)0.7
 Diagnosis of Crohn's disease1.3 (0.2–2.1)0.5
 Initiating treatment with infliximab1.6 (0.3–1.1)0.08
 Male provider0.8 (0.6–2.8)0.4
 Provider trained at non-US medical school0.6 (0.4–1.1)0.08
 Provider trained at non-US residency0.6 (0.3–1)0.06
 Provider treats <5 patients with an anti-TNF1.3 (0.5–3.2)0.6
Hepatitis B    
 Age >502 (1.1–3.8)0.042.7 (0.3–1.1)0.1
 Not seen at IBD center2.8 (1.5–5.2)0.0022.3 (1.1–5)0.03
 Prior exposure to anti-TNF2.1 (1–4.3)0.052.4 (1.1–5.8)0.04
 Treatment started prior to 20102.9 (1.4–6.1)0.0093.3 (1.4–7.9)0.02
 No abnormal ALT prior to treatment2.5 (1.5–4.5)0.00072.6 (1.4–4.9)0.005
 Provider trained at non-US medical school2 (1.1–3.6)0.023.1 (1.6–6)0.0004
 Provider with >10 years as attending in 19984.2 (2.3–7.8)<0.00013.5 (1.8–7.2)0.0005
 Initiating treatment with infliximab1.8 (1.01–3.1)0.052.3 (1.1–4.7)0.03
 Provider treats <5 patients with an anti-TNF2 (0.6–8.6)0.3
 Diagnosis of Crohn's disease1 (0.3–4)1
 Provider trained at non-US residency1.3 (0.7–2.5)0.5
 Male provider0.6 (0.7–4.6)0.3
 Male gender0.6 (0.9–2.7)0.09

For the 100 patients (35%) who were not screened, data were collected on the reasons for failure to screen and presented in Figure 1. In 21% of these patients the TST was placed, but no result was documented. In another 21% the patients were instructed to get a TST or the physician accepted a verbal report of a negative TST. However, in most of the cases we could not determine the reason why patients were not appropriately screened.

Figure 1.

Assessment of why a patient did not receive a tuberculin skin test (TST) within 1 year of initiating an anti-TNF therapy.

A total of nine patients were treated for latent TB. Four patients were treated based on appropriate TST screening prior to treatment. One patient had a negative TST before a prior treatment with an anti-TNF and was found to have a positive TST on screening for reinitiating an anti-TNF. Another patient had a positive TST noted by the infusion nurse a week after it was placed. Although this patient was ultimately treated appropriately for latent TB, this was counted as a failure to screen given the lack of adequate follow-up on the TST. Of the five total positive TSTs done for screening, two were placed while the patients were on concomitant prednisone and a thiopurine. One patient had a negative TST before treatment was initiated and subsequently had a positive TST during a hospitalization for a pulmonary infection. The anti-TNF was held and the patient underwent treatment for what was ultimately felt to be latent TB. Of the remaining three patients treated for latent TB, two had a prior history of positive TST but had not previously been treated, while one had a chest imaging concerning for latent TB.

Hepatitis B Screening

Data regarding screening for hepatitis B is displayed in Table 2 and data regarding factors associated with failure to screen for hepatitis B is presented in Table 3. Twenty-five percent (72/287) of patients in this population were screened for hepatitis B in the year prior to an anti-TNF being initiated. Overall, 37% of patients had hepatitis B antigen checked at any time prior to treatment with an anti-TNF.

In the multivariate analysis, not having an abnormal ALT (including no ALT sent) was a significant predictor for not screening for hepatitis B (OR: 2.6; P = 0.005), as was prior exposure to an anti-TNF (OR: 2.4; P = 0.04), and treatment started prior to 2010 (OR: 3.3; P = 0.02). Initiating treatment with infliximab compared to another anti-TNF was associated with not screening (OR: 2.3; P = 0.03). Providers who trained at a non-U.S. medical school (OR: 3.1; P = 0.0004), had more than 10 years of experience as an attending physician in 1998 (OR: 3.5; P = 0.0005), and not being a treated at the Center for IBD were all associated with a lack of screening.

No patients were documented to be hepatitis B surface antigen-positive prior to or during treatment with an anti-TNF. However, there were a total of seven patients who were noted to be core antibody-positive. Three patients were found to be hepatitis B core antibody prior to treatment with an anti-TNF. Each of these patients still went on to receive anti-TNF treatment, but surveillance labs, including hepatitis B viral load and/or hepatitis B surface antigen, were monitored routinely. The other patients were found to be hepatitis B core antibody-positive after anti-TNF treatment was initiated (Table 2).

Screening Patterns Over Time

When screening rates were compared on an annual basis from 2000 to 2010, an increase in the proportion of patients screened for TB was noted (20% in 2002 vs. 80% in 2010, P = 0.002) (Fig. 2). Similarly, screening rates for hepatitis B were higher in 2010 than in 2002 (45% vs. 10%, P = 0.05). We evaluated trend by performing a linear regression analysis. The results showed that rate of TB screening increased by 8.6% per year (95% CI: 6.6%–10.6%, R-square = 91%, P < 0.001). Similarly hepatitis B increased by 4.3% per year (95% CI: 2.5%–6.1%, R-square = 77%, P < 0.001). Piecewise regression done before and after guidelines were published in 2006 for TB did not demonstrate a significant difference in the rates of screening. As we had information for only 1 year after hepatitis B guidelines we were unable to perform piecewise regression for hepatitis B screening.

Figure 2.

Percent adherence to screening for latent tuberculosis (TB) and hepatitis B over time.

DISCUSSION

It is well accepted that all patients should have appropriate screening for TB prior to initiating anti-TNF therapy. There is emerging evidence that patients should also be screened for hepatitis B prior to initiating anti-TNF therapy.9 To our knowledge, this is the first study to examine the adherence to latent infection screening in patients with IBD prior to starting anti-TNF therapy. In our study, 65% of patients were appropriately screened for latent TB, while only 25% were screened for hepatitis B infection. While the rates of screening have increased over time, assessing for potentially life-threatening complications prior to initiation of anti-TNF therapy should approach 100% compliance.

As there is a range of clinical guidelines for infection screening prior to initiation of an anti-TNF,1, 12, 13, 16 we chose to assess “obtaining a TST” as a simple surrogate for screening for latent TB. Even with this least stringent criterion, less than 80% of patients in our practice are currently being appropriately screened (Fig. 2). Both the ACG and the AGA emphasize the importance of a thorough assessment of risk factors for TB, including place of birth, travel history, known TB exposures, incarceration, and homelessness. Unfortunately, in our study we were only able to demonstrate that risk factors for TB were adequately assessed in 17% of patients. We agree that in addition to a TST or IGRA it is essential that a thorough history of TB risk factors should be ascertained for appropriate risk stratification and potential treatment of latent TB prior to the initiation of an anti-TNF. In our series, IGRAs were not used as a sole means for screening for latent TB. Interestingly, there is recent evidence that in patients on prednisone, IGRA may be superior to TST.20 However, both tests are still immune assays and subject to false-negative results when on immunosuppressive therapy. Even if not on immunosuppressive therapy, a substantial number of IBD patients are anergic prior to initiation of an anti-TNF.14 Because of this it is essential that a thorough history of TB risk factors should be ascertained for appropriate risk stratification and potential treatment of latent TB prior to the initiation of an anti-TNF.

We demonstrated that previous exposure to an anti-TNF was a strong predictor of failure to screen adequately for TB. This is potentially due to providers assuming the patient had previously been screened or that the patient's lack of reactivation of TB during their prior anti-TNF course infers an absence of latent TB. Regardless, patients should still be assessed for latent TB prior to reinitiating anti-TNF therapy. In our series, one patient was documented to have latent TB in this situation.

There was no documented clinical reason for a failure to screen for latent TB in the majority of cases (Fig. 1). However, in 21% of cases it was noted that a TST was placed, but no documented result. In one instance a positive TST was noted by an infusion nurse prior to initiation of infliximab. A number of patients were instructed to get a TST but the results were never documented, or the physician accepted a verbal report by a patient of a negative TST. As care for IBD patients can be complex and multiple healthcare specialists are often involved, there may be confusion with regard to responsibility for screening. We feel that it is the responsibility of the provider initiating treatment to ensure that appropriate screening has occurred. Methods to ensure patient and physician follow-through need to be examined.

Only recently has the ACG and ECCO incorporated screening for hepatitis B into their practice guidelines. In our population, screening for hepatitis B was low. Although the rates of screening for hepatitis B are increasing, even in 2010 they are below 50% (Fig. 2). Prior exposure to an anti-TNF was associated with not screening likely for similar reasons as TB.

Currently, screening for hepatitis B generally only includes a surface antigen. However, patients on chemotherapy who are antigen-negative but core antibody-positive have been known to reactivate.18 While it is unclear at this time what implications hepatitis B core antibody positivity has for IBD patients, we found that hepatitis B surface antigen and viral load were routinely followed. To date, none of our patients have reactivated, but long-term follow-up is necessary. Negative hepatitis B antibody titers can be useful and should prompt hepatitis B vaccination. For these reasons, we think it is reasonable to screen for hepatitis B prior to initiating treatment with a surface antigen, surface antibody, and core antibody.

As with all guidelines, dissemination of recommendations and education of physicians is paramount. The AGA mentioned screening for latent TB as early as 2003,21 although significantly elaborated on those guidelines in 2006.13 Recently, in 2010 the ACG recommended screening for hepatitis B prior to treatment of UC with infliximab.12 Our data show that screening has been increasing over time (Fig. 2). While screening increased over time, independent of the guidelines, we hypothesize that further support from societies and dissemination of these guidelines may help gastroenterologists create practice standards and increase adherence to accepted screening. However, further study is needed to assess this effect.

Our study has several limitations. First, this is a retrospective analysis of medical records. As a result, it may underestimate the true prevalence of screening. This is most likely for the TB risk factor assessment, which was not formalized, and less likely with hepatitis B screening, which was an objective lab value. As no paper charts were reviewed, it is possible that nonelectronic communications were missed. However, while the true prevalence of screening may be higher, the importance of documentation should not be understated, as it is vital to the multidisciplinary care of these complicated patients. Second, our outcome was a documented result in the electronic medical record. At the start of the study period there was a gradual transition to electronic medical records, which may account for “missing” results. In the middle of our study period electronic medical records were required for billable encounters, which certainly caused some of the increase in documentation over time. Third, this study only examined the records of one medical center. While many patients had all of their care at this center, others had only a GI provider at the center. As the responsibility of appropriate screening relies on the provider prescribing the medication, this should not have affected the primary outcome of adherence to screening. A final limitation is generalizability. In our population the rate of latent TB was low and there were no cases of reactivation of latent TB or reactivation hepatitis B. In a population where latent TB and hepatitis B are more common, screening habits may be different based on provider experience.

Adherence to screening for TB and hepatitis B, although improving over time, was not adequate in this study. Prior anti-TNF exposure is frequently associated with a failure to rescreen for latent TB and hepatitis B. In addition to screening with a TST or an IGRA, more emphasis needs to be placed on a TB risk factor assessment of patients prior to treatment with an anti-TNF. Although screening rates have improved over time, all patients starting anti-TNF therapy should be screened for evidence of latent infections. Further study is warranted to assess the optimal mechanism to improve screening.

Acknowledgements

Disclosures: Dr. Adam Cheifetz has served as an advisory board participant for UCB and Abbott. Dr Alan Moss has been an advisor and received honoraria from Abbott and UCB. Dr. Glen Doherty has served as an advisory board participant for Abbott and MSD. There are no other potential conflicts of interest to disclose. Contributions: Byron Vaughn: Study concept and design, acquisition and interpretation of data, statistical analysis, drafting the article, and critical revision of the article. Glen Doherty: Study concept and design, interpretation of data, and critical revision of article. Shiva Gautam: Statistical analysis, interpretation of results and critical revision of article. Alan Moss: Study concept and design, interpretation of data, and critical revision of article. Adam Cheifetz: Study concept and design, interpretation of data, and critical revision of article.

Ancillary