Double Trouble: Diabetes and Tuberculosis

There is evidence of an important association between diabetes mellitus (DM) and tuberculosis (TB). DM increases the risk of active tuberculosis by a factor of 2 to 3. Also people with TB, who have diabetes, have been found to have worse TB treatment outcomes. The WHO - The Union (The International Union Against Tuberculosis and Lung Disease) Framework for Collaborative Activities recommends bilateral screening for TB and Diabetes -- screening for active tuberculosis in patients with diabetes and for diabetes in patients with TB in order to make earlier diagnoses, which in turn may lead to better treatment outcomes and control of both diseases.

It was interesting to come across two such studies presented at the 44th Union World Conference on Lung Health organised by The Union and currently being held in Paris. Both were conducted in 2012 in South India—one in a tribal TB Unit in Vizianagram, and the other in a sub district unit of Kolar Karnataka. And yet both reported very different outcomes with some commonality. 

In terms of absolute numbers and given the size of the population, India is a high diabetes and TB burden country. According to the RNTCP Annual Report 2013 people with diabetes have a 2-3 times higher risk of TB compared to people without diabetes and about 10% of TB cases globally are linked to diabetes. Also, in a large proportion of people with diabetes as well as TB is not diagnosed, or is diagnosed too late. DM can lengthen the time to sputum culture conversion and theoretically this could lead to the development of drug resistance if a 4-drug regimen in the intensive phase of therapy is changed after 2 months to a 2-drug regimen in the presence of culture-positive TB. People with diabetes who are diagnosed with TB have a higher risk of death during TB treatment and of TB relapse after treatment. DM is complicated by the presence of infectious diseases, including TB, and good glycemic control in TB patients can improve treatment outcomes.

A pilot feasibility study, conducted by India Tuberculosis-Diabetes Study Group (ITDG) across 8 tertiary care hospitals and more than 60 peripheral health institutions in 8 tuberculosis units, screened nearly 98% of the TB patients for Diabetes. About13% were diagnosed to have DM based on fasting blood glucose, which included 8% of registered TB patients with a diagnosis of DM already known, and 5% having a new diagnosis of DM. There was a higher prevalence of DM in patients with TB diagnosed in tertiary care hospitals (16%) than in those diagnosed in tuberculosis units (9%) and amongst those from South India (20%) than from North India (10%). The results showed that it is important and feasible to screen patients with TB for DM in the routine setting, resulting in earlier identification of DM in some patients and opportunities for better management of comorbidity. A policy decision has since been made by the RNTCP to implement this intervention in 100 districts where National Programme for Prevention and Control of Cancer, Diabetes, Cardiovascular Diseases and Stroke (NPCDCS) activities are being implemented.

The Vizianagram study done by the RNTCP and The Union, done by WHO India, London School of Hygiene and Tropical Medicine and The Union, pertained to the screening of all the 381 TB patients diagnosed and initiated on treatment at 10 Peripheral Health Institutions of a tribal TB Unit.  Out of these 381 patients, 374 were assessed for DM and 19 (5.1%) were found to have DM. 32 cases of pre diabetes were also detected. Of the TB-DM cases 15 were referred to diabetes care. The prevalence of DM was found to be higher in those above 40 years of age but did not vary by sex, smoking status, HIV status and type of TB—new or previously treated. Thus in this study, DM prevalence among TB patients was low at 5% compared to 9% assessed by ITDG. But the screening process was found to be effective and implementable.

The Karnataka study assessed feasibility and results of screening 362 TB patients at peripheral level in all the 17 Peripheral Health Institutions of a sub district unit (0.5 million population) for DM. Out of these 358 (99%) were assessed for DM and 62 (17.1%) were found to have it—53 (14.6%) with previously known DM and 9 (2.9%) newly diagnosed with DM. All the newly diagnosed DM patients were enrolled into care. Nearly 44% of the TB patients were smokers and DM prevalence in smokers was found to be higher as compared to non- smokers. Higher DM was also found in patients who were above 40 years of age, and also in those with smear positive TB. Thus in this case overall DM prevalence among TB patients at 17% was higher as compared to 9% assessed by ITDG, again indicating a heterogeneous DM prevalence across India.

These two studies indicate a heterogeneous DM prevalence in TB patients across India—one showing the prevalence to be more than that in the pilot study and the other came up with just the opposite results. However in both studies DM prevalence was found to be higher in those over 40 years old. As one of the researchers indicated, perhaps more data is needed to understand if linking TB patients to pre-diabetes and diabetes care can lead to better treatment outcomes.

Shobha Shukla, Citizen News Service - CNS
November 2013
(The author is supported by Lilly MDR TB Partnership to provide conference coverage from 44th Union World Conference on Lung Health)