0
Communications to the Editor |

Does Diabetes Predispose to the Development of Multidrug-Resistant Tuberculosis? FREE TO VIEW

Rupak Singla, MD, DNB; Nazeer Khan, PhD
Author and Funding Information

Affiliations: Department of Tuberculosis and Chest Diseases Sahary Hospital,  College of Dentistry King Saud University Riyadh, Saudi Arabia,  NYU School of Medicine New York, NY,  Program in Urban Public Health Hunter College of the City University of New York New York, NY

Correspondence to: Rupak Singla, MD, DNB, L.R.S. Institute of Tuberculosis and Allied Diseases, Sri Aurobindo Marg, New Delhi-110030, India; e-mail: drneetasingla@rediffmail.com.



Chest. 2003;123(1):308-309. doi:10.1378/chest.123.1.308-a
Text Size: A A A
Published online

To the Editor:

We read with interest the article by Bashar et al (November 2001).1 The authors have used “odds ratio” interchangeably with “relative risk” while discussing multidrug-resistant tuberculosis (MDR-TB) among diabetes patients, which is not correct.

Among diabetes patients, 13 of the 18 MDR-TB patients were enrolled in directly observed therapy (DOT), compared to only 4 of 10 patients in the control group. Therefore, the enrollment in DOT itself becomes an important confounding factor for studying the incidence of MDR-TB among diabetic patients receiving DOT. Without controlling for this confounding factor, it is not correct to report that patients with tuberculosis (TB), who are both diabetic and receiving DOT, will have 49-fold greater chance of having MDR-TB. The positive aspect of DOT intervention, that all 13 MDR-TB patients with diabetes receiving DOT were successfully treated, needs to be highlighted instead.

Reasons for higher rate of MDR-TB among diabetic patients (36%) compared to 10% among control group are difficult to explain. The authors have not mentioned any plausible scientific explanation with published reference for this observation. Possible malabsorption of anti-TB drugs among diabetic patients, as mentioned by the authors, has not been documented so far. During the study period, from 1987 to 1997, New York City was going through rapid changes in the incidence of HIV infection. Also, the implementation of DOT as a standard therapy in 1992 led to a drastic fall in the incidence of MDR-TB in New York City.2 These factors have not been controlled properly and might have influenced the results of the study. It is surprising that the number of MDR-TB cases in the study remained the same during the 11-year period from 1987 to 1997. This study was also limited by its small number of subjects and by retrospective data collection.

At Sahary Chest Hospital, Riyadh, Saudi Arabia, a retrospective review (unpublished data) revealed that among sputum smear-positive pulmonary TB patients, no MDR-TB case was detected among 126 diabetic patients compared to 10 MDR-TB cases among 389 nondiabetic control patients. We feel that we should wait for more studies before we believe that diabetes is a significantly high risk factor for the development of MDR-TB.

Bashar, M, Alcabes, P, Rom, WN, et al (2001) Increased incidence of multidrug-resistant tuberculosis in diabetic patients on the Bellevue Chest Service, 1987 to 1997.Chest120,1514-1519
 
Frieden, TR, Fujiwara, PI, Washko, RM, et al Tuberculosis in New York City: turning the tide.N Engl J Med1995;333,229-233
 
To the Editor:

We agree with Drs. Singla and Khan about the importance of directly observed therapy (DOT) when assessing determinants of multiple drug-resistant tuberculosis (MDR-TB) cases. We recognized DOT as a potential confounder in our assessment of the link between diabetes and MDR-TB because, as pointed out, DOT was increasingly common in New York during our study period.

More frequent contact with the health-care system would make diabetic patients more likely to receive DOT if they acquired tuberculosis. Accordingly, we stratified our analyses by receipt of DOT, as well as other possible confounders (homelessness and HIV positivity). Therefore, our estimate that diabetic patients with tuberculosis were dramatically more likely to have multidrug-resistant disease was not a consequence of the association of DOT with diabetes, nor was it an artifact of changing prevalence of HIV infection.

We agree with the assertion that the odds ratio is not interchangeable with the relative risk. Obviously, relative risk could not be calculated directly, as our study was of a case-control design. The odds ratio estimates the relative risk, with a slight systematic bias (arising from the difference between odds and risk). We therefore chose to use the term relative risk to denote its (odds ratio) estimator. To be perfectly correct, we could have used the term odds ratio only. We regret any confusion that our use of terms might have caused.

With regard to the observation of Drs. Singla and Khan of no MDR-TB among a sample of patients in Riyadh, we are not surprised by the disparate findings in different samples. Further and more conclusive investigations should be done to examine the possible link between diabetes and drug-resistant tuberculosis in different locales. As widespread as tuberculosis is, its occurrence remains an intensely local phenomenon.


Figures

Tables

References

Bashar, M, Alcabes, P, Rom, WN, et al (2001) Increased incidence of multidrug-resistant tuberculosis in diabetic patients on the Bellevue Chest Service, 1987 to 1997.Chest120,1514-1519
 
Frieden, TR, Fujiwara, PI, Washko, RM, et al Tuberculosis in New York City: turning the tide.N Engl J Med1995;333,229-233
 
NOTE:
Citing articles are presented as examples only. In non-demo SCM6 implementation, integration with CrossRef’s "Cited By" API will populate this tab (http://www.crossref.org/citedby.html).

Some tools below are only available to our subscribers or users with an online account.

Related Content

Customize your page view by dragging & repositioning the boxes below.

  • CHEST Journal
    Print ISSN: 0012-3692
    Online ISSN: 1931-3543