BACKGROUND: Vitamin D deficiency is associated with progression of latent tuberculosis (TB) infection to active disease. The impact of preventive therapy on this association is unknown.METHOD: Serum 25-hydroxyvitamin D (25(OH)D) levels were retrospectively linked to adults diagnosed with latent TB between April 2010 and January 2019 in a hospital in London, UK. Individuals in the cohort who progressed to active TB were identified by matching to a national notification register. A logistic regression model was used to examine baseline vitamin D deficiency and use of preventive therapy with subsequent incidence of TB disease.RESULTS: Of 1509 latently infected individuals with 3902 patient-years of follow-up, 687 (45.5%) were identified as vitamin D deficient and 691 (45.8%) individuals had a LTBI regimen prescribed. There were 29 (1.9%) instances of TB reactivation. On multivariate analysis, profound (<25 nmol/L) vitamin D deficiency (aHR 5.68, 95%CI 2.18-14.82; P = 0.0003) and the absence of preventive therapy (aHR 3.84, 95%CI 1.46-10.08; P = 0.006) were associated with progression to active TB disease. There was no evidence that preventive therapy modified the association between vitamin D status and TB reactivation.CONCLUSION: Our results show an independent association between vitamin D deficiency and progression from latent TB infection to active disease.

Introduction Age-related immunosenescence influences the presentation of tuberculosis (TB) in older patients. Here, we explore the clinical and radiological presentation of TB in the elderly and the factors associated with time to treatment for TB. Methods This is a retrospective cohort study comparing the clinical, radiological and demographic characteristics of TB patients aged ≥65 years with TB patients aged 18–64 years in a large cohort of TB patients in the UK. Factors associated with the time to presentation and time to treatment were identified using a multivariable analysis model. Results 1023 patients were included in the analyses: 679 patients aged 18–64 years and 344 patients aged ≥65 years. “Classical” symptoms of TB (cough, haemoptysis, fever, nights sweats and weight loss) were less common among older patients with pulmonary TB (PTB) (p<0.05), but dyspnoea was more common among older patients (p=0.001). Time from presenting in secondary care to starting treatment was shorter in younger compared with older patients: 3 versus 15 days (p=0.001). When adjusted for age, factors associated with shorter time to treatment from symptom onset include sex (male versus female) (hazard ratio (HR) 1.23 (95% CI 1.05–1.46)), UK born (HR 1.23 (95% CI 1.05–1.46)) and HIV (HR 2.07 (95% CI 1.30–3.29)). Only age remained an independent predictor of time to treatment in a multivariable model (HR 0.98 (95% CI 0.98–0.99)). For those with PTB, chest radiography findings showed that cavitation and lymphadenopathy were more common among younger patients (p=0.001). Conclusions Older patients aged ≥65 years with TB had fewer “classical” clinical and radiological presentations of TB, which may explain longer times to starting treatment from symptom onset compared with younger patients aged <65 years.

2016 TB National Institute for Health and Care Excellence (NICE) guidelines imply that contacts of extrapulmonary TB do not require screening for latent TB infection. At our high TB prevalence site, we identified 189 active cases of TB for whom there were 698 close contacts. 29.1% of the contacts of pulmonary TB and 10.7% of the contacts of extrapulmonary TB had active or latent TB infection. This supports screening contacts of extrapulmonary TB at our site and presents a way to access high-risk individuals. We propose to continue to screen the contacts of our patients with extrapulmonary TB and recommend other TB units audit their local results.

Background: Anti-TNF treatments have improved the outcomes for patients with inflammatory bowel disease (IBD). However, they are associated with an increase in risk of Tuberculosis (TB). The aim of this study is to determine the proportion of latent TB infections (LTBI) in our IBD cohort treated with anti-TNFs and the management of these patients. We also examined the effect of different immunosuppressive drugs on the indeterminate rate of the interferon-gamma release assays (IGRA) in LTBI screening Methods: We conducted a retrospective review of all patients treated with biologics between March 2007 and November 2015. Patient notes and electronic records were reviewed. LTBI screening was assessed using a risk assessment form, chest x-ray, and tuberculin skin test (TST) or IGRA for the screening for LTBI and the nature of the immunosuppressive treatment was documented. Results: 732 patients with IBD were screened for LTBI before starting a biological treatment during the study period. 31 of 732 (4%) IBD patients were diagnosed and treated for LTBI with no significant side effects. 596 of 732 received their biologic treatment with a median delay of 86 days in initiating biologics. 6 of 596 patients who received biologic treatment developed active TB; 5 of whom were Caucasian. 247 patients were screened with an IGRA test. 6 were positive, 162 were negative and 79 patients were indeterminate. 45% with indeterminate IGRA had a repeat; half were negative and half remained indeterminate. 73% of the patients were receiving immunosuppressive (IMM) medication(s). There was a higher indeterminate rate of the IGRA in the IMM group compared to the no-IMM group (32% compared to 9%). The combination of steroids and thiopurines gave the highest rate of an indeterminate IGRA. High and low doses of steroids were equally likely to result in an indeterminate IGRA result. Conclusions: 4% of patients screened had LTBI and 1% of patients treated with biologics developed active TB. Delays in initiating biologics due to LTBI screening may have detrimental effects on the IBD management. Almost one third of patients on immunosuppressants had an indeterminate IGRA, which supports the need for a guideline incorporating a risk stratification strategy based on factors such as country of birth and local TB prevalence. There is a need to remain vigilant for TB regardless of baseline LTBI results or epidemiological factors.

Untreated, miliary tuberculosis (TB) has a mortality approaching 100%. As it is uncommon there is little specific data to guide its management. We report detailed data from a UK cohort of patients with miliary tuberculosis and the associations and predictive ability of admission blood tests with clinical outcomes.

Background: We describe drug-induced liver injury (DILI) secondary to antituberculous treatment (ATT) in a large tuberculosis (TB) centre in London; we identify the proportion who had risk factors for DILI and the timing and outcome of DILI.Methods: We identified consecutive patients who developed DILI whilst on treatment for active TB; patients with active TB without DILI were selected as controls. Comprehensive demographic and clinical data, management and outcome were recorded.Results: There were 105 (6.9%) cases of ATT-associated DILI amongst 1529 patients diagnosed with active TB between April 2010 and May 2014. Risk factors for DILI were: low patient weight, HIV-1 co-infection, higher baseline ALP, and alcohol intake. Only 25.7% of patients had British or American Thoracic Society defined criteria for liver test (LT) monitoring. Half (53%) of the cases occurred within 2 weeks of starting ATT and 87.6% occurred within 8 weeks. Five (4.8%) of seven deaths were attributable to DILI.Conclusions: Only a quarter of patients who developed DILI had British or American Thoracic Society defined criteria for pre-emptive LT monitoring, suggesting that all patients on ATT should be considered for universal liver monitoring particularly during the first 8 weeks of treatment.

Introduction Anti-TNF therapies for inflammatory bowel disease (IBD) can lead to a five fold increase of tuberculosis (TB) reactivation in patients with latent TB infection (LTBI). Current European Crohn’s and Colits Organisation recommends LTBI screening with Interferon Gamma Release Assay (IGRA) according to local prevalence, national recommendations and preferred in BCG immunised individuals. In this study, we determine the proportion of LTBI in our cohort of IBD patients treated with biologics and any complications of treatment, particularly drug-induced liver injury (DILI). Methods All patients with IBD who were treated with biologics between March 2007 and November 2015 were identified from the high cost funding database held at the Pharmacy of St Mark’s Hospital. Case notes and electronic records were retrospectively reviewed to identify records of LTBI screening. The following were excluded from the analysis: those who were treated for active TB and those who did not receive treatment either for latent or active TB. Of note, at this site, in July 2013, an IGRA testing replaced the Mantoux on our screening algorithm. DILI was defined as deranged liver function tests with no other identifiable cause. Results Seven hundred and thirty-two IBD patients were screened for TB prior to starting a biologic. 31 patients (4.2%) were identified as having LTBI and had prophylactic treatment for TB (19 with single agent isoniazid; 12 dual agent isoniazid and rifampicin). Of these 31 patients, 22 went on to have anti-TNF treatment with a median delay of 86 days (range = 7–336 days). Of the 31 that went on to have treatment for latent TB, 3% (n = 1) had side effects, dizziness which was not severe to be stopped the LTBI treatment and non had DILI. Conclusion Rate of LTBI in this population receiving biologics for IBD was 4%; all identified patients were treated and none experienced significant side effects. The median time to starting biologics after screening was 86 days, which may represent a significant delay in starting biologics for the IBD patients; this could have detrimental effects to their IBD related morbidity. There remains the risk of false negative IGRA results in an immunosupressed patient and, as such, place of birth and likely previous exposure should be taken into account. Reference 1 Rahier J, Moutschen M, Van Gompel A, Van Ranst M, Louis E, Segaert S, et al. Vaccinations in patients with immune-mediated inflammatory diseases. Rheumatology (Oxford) 2010;49:1815–1827. Disclosure of Interest None Declared

Background Non-attendance at TB contact screening clinics has been highlighted as a common phenomenon across a number of sites during recruitment to the PREDICT TB Study. This has obvious implications for the safety of patients, their communities and for NHS resources. The objective of this study was to explore why adults who have been in contact with TB do, and do not, attend their screening appointment, thereby allowing identification of interventions to reduce non-attendance.Methods A multi-method approach was taken using 15 questionnaires with adults who attended for screening, 15 telephone questionnaires with adults who did not attend and in-depth interviews with 8 TB nurses. Interviews were coded to trace emerging descriptive themes, then refined through an iterative process of interpretation and recoding.Results Findings from the questionnaires and interviews were categorized into three principle themes following analysis: awareness, hospital factors and leadership. These themes deconstruct the complex phenomena of patients' lack of attendance at this TB contact screening service.Conclusion Recommendations related to issues of leadership, outreach services, flexibility of clinic timing and awareness amongst both the local community and GPs were made.

Introduction Biologic treatment has improved outcomes of patients with complex inflammatory bowel disease (IBD). Anti-TNF therapy is associated with a fivefold increased risk of reactivation of tuberculosis (TB). When TB occurs, it is commonly extra-pulmonary and disseminated with an atypical presentation and can be diagnostically challenging. An indeterminate interferon gamma release assay (IGRA) result occurs in <2% of the healthy population, however this is higher in immunosuppressed patients. Given the increased risk of TB, screening for latent TB infection (LTBI) remains vital before commencing biologics. Methods All IBD patients who had an IGRA test between July 2013 and November 2015 before commencing biologic therapy were identified from the high cost funding database held in the Pharmacy Department at St Mark’s Hospital. Clinical and electronic case records were reviewed. Results 247 patients were screened for TB with an IGRA test during the study period. The mean age was 36.9 years (range 8–85), 54% were male and 70% had Crohn’s disease. 78 patients (32%) had an indeterminate IGRA result and 35/78 (45%) patients had a repeat test; 17/35 (49%) had a second indeterminate results and the remaining 18 /35 (51%) had a negative test. 8/78 with an indeterminate test had a Mantoux as per our algorithm; 3 were positive, 4 were anergic and 1 was negative. Of the 247 patients, 210 patients received biologic treatment. Thirty-two patients (15%) delayed receiving their biologic treatment for various reasons, including receiving LTBI treatment. Of 210 patients, 158 patients (75%) had Infliximab,38 patients (18%) had Adalimumab, 13 had Vedolizumab and 1 had Golimumab. 30 patients were referred for an infectious diseases opinion; 7 with a positive IGRA and 7 with a negative IGRA but abnormalities on their chest X-ray and a high epidemiological risk for TB. One patient who had an indeterminate IGRA result developed active TB during the study period. Conclusion Almost one third of IBD patients had an indeterminate IGRA with half of those having a second indeterminate result. Sampling and processing factors also influence the prevalence of indeterminate results in this population. This supports risk stratification with more weight added to the risk than IGRA alone. Reactivation of LTBI may lead to delays in starting treatment hence it is vital to take into account the patient’s pre-test probability of LTBI before using an IGRA to avoid such delays. Reference 1 Wong H, Ip M, Tang W, et al. Performance of the interferon-gamma release assay for Tuberculosis screening in inflammatory Bowel disease patients. IBD 2014;20(11):2067–2072. Disclosure of Interest None Declared

•A third of tuberculous spondylodiscitis patients had more than one region affected. •Age and ethnicity differs between tuberculous and pyogenic spondylodiscitis. •Thoracic then lumbar involvement were commonest in tuberculous spondylodiscitis. •Lumbar then thoracic involvement were commonest in pyogenic spondylodiscitis.

Introduction and objectives It has been suggested that TB has a different phenotype in older patients with age-related changes to the cell-mediated immune response and co-existent organ dysfunction. Older patients with tuberculosis (TB) may have different radiographic features than younger patients; this may lead to less immediate suspicion of TB resulting in delays to diagnosis and starting treatment. We wanted to identify if there are differences in the most common radiological differences in older and younger patients with pulmonary TB (PTB). Methods Patients with PTB > 65 were noted from the London TB register between 2002 and 2015. A random selection of younger patients aged 18–40 with PTB were also identified. All available chest x-ray (CXR) reports were obtained from online radiology systems. CXR features were classified according to reported features with particular note of cavitation, nodules and miliary changes, consolidation, lymphadenopathy and effusions. Results The CXR reports of 239 patients with PTB < 65 and 99 patients with PTB > 65 were collated. Demographic details as well as CXR changes are detailed in Table 1. Cavitation, lymphadenopathy and effusions were more common in younger patients whereas consolidation was more evident in older patients. Upper zone involvement was similar in both groups. Conclusions Studies by other groups have suggested a higher proportion of cavitation and upper zone changes in younger patients with TB with less specific changes in older patients. This may lead to less suspicion of TB and potentially a longer infective period; this is important given that 23% and 19% of younger and older patients have smear positive PTB. In our study, the proportion with upper zone changes are similar though cavitation is more frequent in younger patients. Of note, is the much higher presence of lymphadenopathy and effusions seen in younger patients. This may potentially be related to differences in the immune function of both groups or primary infection versus reactivation. These findings re-enforce the need for clinical suspicion for PTB in both older and younger patients with both specific and non-specific radiographic changes.

Introduction Anti-tumour necrosis factor agents (anti–TNF) are major advances in the management of inflammatory bowel disease (IBD) however they are associated with a 5 fold increased risk of Mycobacterium tuberculosis (TB) infection. All patients should be screened for latent TB infection (LTBI) prior to anti-TNF therapy. Here, we review active TB cases after anti-TNF therapy to identify lessons to be learned. Methods All patients with IBD treated with anti-TNF between March 2007-November 2015 were identified from pharmacy database. Those who developed active TB were identified from the London TB register. Clinical and electronic notes were reviewed. LTBI screening is with history, Chest X-Ray (CXR) and tuberculin skin test (TST) before July 2013 or Quantiferon Gold Assay (QFT-G) after July 2013. Results Of 596 patients treated with anti-TNF, 6 had active TB. 5 had Crohn’s and 1 had Ulcerative Colitis. 3 were male, none had HIV. Age range was 24–49 years. 5 were Caucasian, 1 was UK born but of Pakistani origin. 3 were on Adalimumab (ADA) at time of TB diagnosis and 2 on Infliximab (IFX), 3 were on at least 1 other immunosuppressive. 1 was not on anti-TNFs at TB diagnosis but received IFX 3 months earlier. Time from initiation of anti-TNF to TB diagnosis ranged from 3–41 months (median:13, IQR = 32). 3 had culture confirmed TB, 1 was MTB complex PCR positive but culture negative and 2 had presumed TB. All isolated cultures were fully sensitive. 2 had miliary TB, 2 abdominal TB, 1 pleuro-pulmonary TB and 1 both pulmonary and pericardial TB. Treatment duration was 6–12 months, 5 patients completed treatment and 1 remains on treatment. 3 had prior vaccination for TB, 1 did not and the vaccination status of 2 was unknown. 4 patients had negative TSTs pre-anti-TNF (3 while immunosuppressed), 1 had an indeterminate QFT-G test (while immunosuppressed) and 1 had neither. All patients had normal CXR prior to anti-TNF. No patients received LTBI treatment. Conclusion All cases were considered low epidemiological risk for LTBI . None had a positive TST or QFT-G, however the risk of false negatives is high in immunosuppression. As some patients screened may receive prolonged and recurrent courses of anti-TNF, we recommend discussion around when patients should be rescreened. It is unclear if these cases represent de novo infection or reactivation of latent disease but re-screening may have identified them at the latent stage. As it is not possible to prevent all cases of active TB, there must be continued focus on prompt diagnosis and treatment, alongside comprehensive screening by working with local TB services. Disclosure of Interest None Declared

Introduction and objectives In 2013, Public Health England (PHE) reported that 39% of pulmonary tuberculosis (TB) patients >65 years old started treatment >4 months after symptom onset compared to 25% of patients aged 15–44 years. A longer symptomatic period, particular if smear positive, may result in increased transmission and poorer treatment outcomes. We investigated age-related differences in symptom duration and time to starting TB treatment in a large UK cohort of TB patients. Methods The cohort comprised patients on the London TB register (LTBR) who were treated at Northwick Park Hospital between 2002–2015. Patients aged >65 years were compared with a random sample of patients aged 18–40 years, with respect to symptoms, symptom duration at presenting to secondary care and time to starting treatment after presenting to secondary care. Results 357 patients over 65 and 517 younger patients were identified. Demographics for the total number of patients are included in Table 1. Data were available for 489 of the younger group and 73 of the older group. 26 of the >65s (35.6%) and 9 (1.8%) of the 65s were diabetic. Conclusions This study has identified that a potential cause for a delay in diagnosis in the elderly may be related to the decreased frequency of ‘classical’ TB symptoms in those >65 years of age. Clinicians need to be vigilant despite the lack of these symptoms. Median duration of symptoms was twice that in older patients versus younger patients (90 days versus 45) with a longer time to starting treatment (7 days versus 2). The total duration (in months) from symptom onset was also higher at 3 months versus 2 months. A higher proportion of older patients (32%) started treatment greater than 4 months from symptom onset compared to 13.1% in younger patients. Biases in this work including recording bias due to the retrospective nature of symptom recording, the effect of selecting younger patients from more recent years (to improve data availability) and the low proportion of patients over 65 which had symptom and treatment data available.

Introduction Indeterminate interferon-gamma release assay (IGRA) results when screening for latent tuberculosis infection (LTBI)prior to biologic use in inflammatory bowel disease (IBD) may delay biologic treatment initiation. Concomitant steroids affect IGRA result, however the impact of other immunosuppressive medications is less clear. We determine the effect of immunosuppressives on results of IGRAs. Methods All patients treated with biologics from July 2013-November 2015 were identified from a pharmacy database. Electronic and clinical records were reviewed for IGRA result and concomitant immunosuppressive use. X2 was used to compare categorical data and univariate logistic regression using SPSS was used to determine likelihood of indeterminate tests with different immunomodulator medications (IMM). Results 247 patients with IBD had an IGRA. Mean age: 36.9 years (range 8–85) and 54% were male. 78/247 (32%) had an indeterminate result and 35/78 (45%) patients had a repeat test; 17/35 (49%) had a second indeterminate result. 181/247 (73%) received at least one IMM prior to screening and 66 patients (27%) were not on any IMM (IMM-free). In the IMM group: 121/181 (67%) patients had thiopurines, 18/181 (10%) had corticosteroids, 25/181 (14%) had both, 9/181 (5%) had methotrexate and 7/181 (4%) had other IMMs at screening. 72/181 (40%) had a first indeterminate IGRA 59/181 (33%) had a second indeterminate result. In the IMM-free group: 6/66 (9.0%) had an indeterminate IGRA. Patients in IMM group were more likely to have an indeterminate results than patients in IMM-free group (33% v 9.0%, p = 0.00001). Each separate IMM group was more likely to be associated with an indeterminate IGRA result compared with those in IMM-free group: thiopurines (23%) p = 0.020, steroids (42%) p = 0.001, thiopurines and steroids (64%) p = 0.001, other (43%) p = 0.01 and methotrexate (44%) p = 0.008). The combination of steroids and thiopurines together was the strongest factor associated with an indeterminate result. High dose (Prednisolone >20 mg or intravenous Hydrocortisone) and low dose steroids (Prednisolone <20 mg or Budesonide) were equally likely to cause an indeterminate result (66.6% v 50.0%, p = 0.68). Conclusion A combination of thiopurines and steroids gave the highest likelihood of an indeterminate IGRA result, although significant results occurred with all IMMs. This has implications for LTBI screening in IBD patients. Guidelines to address indeterminate IGRA results, perhaps with more focus on epidemiological risk may be helpful. There is an unmet need to have improved assessment tools for TB in patients on IMMs. Disclosure of Interest None Declared

Background: Clinical, radiological and microbiological criteria inform diagnosis of pulmonary Non-Tuberculous Mycobacteria (NTM) disease and treatment decisions. This multicentre, review aims to characterise NTM disease meeting ATS/IDSA criteria and define factors associated with initiation of treatment.Methods: Sputum samples growing NTM from 5 London hospitals between 2010 and 2014 were identified. Data for HIV-negative individuals meeting ATS/IDSA guidelines for pulmonary NTM disease were extracted. Associations between clinical variables and treatment decision were investigated using Chi-squared, Fishers-exact or Mann Whitney tests. Factors associated with treatment in univariate analysis (p < 0.150) were included in a multivariate logistic regression model.Results: NTM were identified from 817 individuals' sputum samples. 108 met ATS/IDSA criteria. 42/108 (39%) were initiated on treatment. Median age was 68 (56-78) in the cohort. On multivariate analysis, factors significantly associated with treatment of pulmonary NTM infection were: Cavitation on HRCT (OR: 6.49; 95% CI: 2.36-17.81), presenting with night sweats (OR 4.18; 95% CI: 1.08-16.13), and presenting with weight loss (OR 3.02; 95% CI: 1.15-7.93). Of those treated, 18(43%) have completed treatment, 9(21%) remain on treatment, 10(24%) stopped due to side effects, 5(12%) died during treatment. Mortality was 31% (n = 13) in treated versus 21% (n = 14) in the non-treated cohort. Subgroup analysis of individual NTM species did not observe any differences in treatment initiation or outcomes between groups.Discussion: Decision to treat pulmonary NTM infection requires clinical judgement when interpreting clinical guidelines. Factors independently associated with decision to treat in this HIV-negative cohort include cavitation on HRCT and presenting with night sweats or weight loss. (C) 2016 Elsevier Ltd. All rights reserved.

Introduction Non-Tuberculous Mycobacteria (NTM) are ubiquitous in the environment meaning clinical, radiological and microbiological criteria are important in diagnosing NTM lung disease. A multicentre, retrospective review was performed to characterise NTM disease within our region and describe the outcomes of current management. Methods All NTM positive sputum samples received by the National Mycobacterium Reference Laboratory (NMRL) from Imperial College NHS Healthcare and North West London Hospitals NHS Trusts between 2010–2014 were extracted. HIV-negative individuals with ≥2 positive sputum samples or ≥1 positive bronchoalveolar lavage were included. Demographic, clinical, radiological, microbiological, management and outcome data was obtained from electronic records. Results 1190 NTM sputum samples were identified from 822 individuals. 152 individual patients met inclusion criteria for analysis. Table 1 describes cohort demographics. Within the cohort 48/152 (32%) were treated for NTM disease. All treated subjects and 74/104 (71%) non-treated subjects met international guidelines for diagnosis of NTM infection, which included positive clinical, radiological (cavities or bronchiectasis +/- nodules or infiltrates) and microbiological criteria. Mycobacterium avium complex (MAC) was the most commonly isolated (68/152; 45%) and treated organism (21/48; 44%) followed by Mycobacterium kansasii (11/48; 23%). 19/48 (40%) completed treatment (median duration: 17 months [IQR: 12–24]). 10/48 (21%) remain on treatment (median duration: 18 months [IQR: 11–36]), 11/48 (23%) stopped treatment due to side effects and 13/48 (27%) were either lost to follow up or treated for Mycobacterium tuberculosis . Of those treated, 29/48 (60%) culture converted; 23/29 (79%) remain negative at 12 months post culture conversion. Of 19/48 who completed treatment, 5/19 (26%) had symptomatic or radiological disease progression compared to 11/28 (39%) who did not complete treatment. 11/48 (23%) patients died within the treatment group. Within the untreated subjects who met international guidelines for NTM infection (74/104), mortality was 19/74 (26%) (p = 0.83). Discussion NTM is a challenging disease with only 39% of eligible subjects receiving treatment and a high associated mortality. Furthermore, only 40% starting treatment completed it and the 21% who remain on treatment have been treated for a median duration of 18 months to date. Unlike similar HIV-negative UK cohorts, MAC pulmonary disease is the most prevalent.

We read with interest the article by Alvaro-Meca and colleagues who measured rates of Mycobacterium tuberculosis (TB) recurrence in Spain 1 Alvaro-Meca Alejandro Rodriguez-Gijon Lorena Asuncion Diaz Ángel Gil Salvador Resino Incidence and mortality of tuberculosis disease in Spain between 1997 and 2010: impact of human immunodeficiency virus (HIV) status. J Infect. Apr 2014; 68: 355-362 Abstract Full Text Full Text PDF PubMed Scopus (10) Google Scholar where overall TB incidence is similar to UK levels. Recurrence rates of TB can help evaluate the effectiveness of a TB program and identify vulnerable patients, however little recent data describes this in the UK. We performed a retrospective, longitudinal descriptive study of all TB notifications at Northwick Park Hospital over 12 years and identified potential risk factors. 3515 patients were identified and the rate of recurrence was 0.65–1.16 per 1000 patient years follow up, depending on case definition. Results from this study will inform local practice and encourage comparison with other centres both nationally and internationally.