Page 233 - fbkCardioDiabetes_2017

Cardio Diabetes Medicine 2017 209

of disease and outcomes of tuberculosis may vary diabetes had delayed sputum conversion and trend-
between diabetics and non diabetics. Some stud- ed towards poorer outcomes including death at the
ies have shown that diabetics with TB tend to be end of treatment in the diabetic subset. Retrospective
of older age. Such co-existent disease occurs at cohort studies from Maryland, USA and in separate
a higher frequency among men. Others have also work published by Wang and colleagues, showed
demonstrated increased symptoms at presentation, that there was an increased risk of death among
such as higher levels of hemoptysis and weight loss diabetics with tuberculosis than patients with tuber-
being reported among diabetics compared to non culosis alone. These studies suggest that treatment
diabetic controls. Diabetes has been shown to be failure and death are more frequent in diabetics.
associated with lower lobe predominant, more cavita- However it is not clear whether aggressive glycemic
tory and multi-focal tubercular disease in some stud- control would improve treatment response. Further-
ies. However extra-pulmonary TB is considered to be more most of these studies do not report cause of
of lower occurrence in patients with diabetes, than death, thus we do not know if the excess mortality
those without. However more recent studies have not is explained by increased severity of tuberculosis
shown many of these differences to be significant. in diabetics or the presence of other vascular and
Stark variations in presentation of disease between non vascular complications of diabetes per se. What
diabetics and non diabetics is possibly overstated. it tells us is that, diabetes and tuberculosis present
This variation amongst studies is probably due to the a lethal combination that poses a double jeopardy
fact that levels of glycemic control were not adjusted to the patient. The clinician must have heightened
for and the screening test for the diagnosis of diabe- awareness that this combination has been associat-
tes has also varied amongst studies. (10)and with the ed with poorer outcomes, including death and thus
increasing prevalence of type 2 DM in less developed engage in efforts to an early diagnosis and effective
regions, many patients with TB will have concomitant therapy for both diseases.
DM. Presently, little is known about the effect of DM
on the clinical presentation and treatment outcome 8. Issues in the co-management of TB and
of TB.\nMETHODS: In an urban setting in Indonesia, diabetes:
737 patients with pulmonary TB were screened for
DM and were followed up prospectively during TB In the scenario of coexistence of TB and Diabetes in
treatment. Clinical characteristics and outcome were a patient, the physician is expected to manage both
compared between patients with TB who had DM diseases with a fine balance, not allowing the treat-
and patients with TB who did not have DM.\nRE- ment of one to adversely affect the other. There are
SULTS: DM was diagnosed in 14.8% of patients with many practical points to note in the clinical manage-
TB and was associated with older age and a great- ment of such a patient.
er body weight. On presentation, diabetic patients A. Influence of tuberculosis on diabetic medica-
with TB had more symptoms but had no evidence tions:
of more-severe TB. After 2 months, results of sputum
microscopic examination was more often positive in Infections like tuberculosis cause poorer glycemic
diabetic patients (18.1% vs. 10.0% (11) control and therefore need careful monitoring of
blood sugars during the acute phase of illness to
titrate anti diabetic medications to effective dos-
7. Microbiology of Tuberculosis in diabetes ages. Drug interaction is an important factor to
and overall outcomes: consider with Rifampicin being the usual suspect.
It was hypothesised and demonstrated in one study Rifampicin is a powerful enzyme inducer of cyto-
that rates of sputum positivity are lower in diabetics chrome P450. The iso-enzyme 2C9 metabolises
but other studies have shown no such significant Glibenclamide and Glipizide. The serum concen-
difference. Some small retrospective studies have trations of both these oral sulfonylureas have
shown that the initial bacterial burden among diabetic been shown to be reduced by 39% and 22 % re-
tuberculosis patients is higher. However this does not spectively in the case of rifampicin co-treatment.
seem to have any impact on the final sputum con- Rifampicin can also reduce the effective levels of
version beyond 3 months, though the median time to thiazolidenediones and meglitinides, causing hy-
culture negativity was significantly longer in diabet- perglycemia in diabetics who were earlier having
ic patients than in controls (42 vs 37 days; p=0∙03). blood sugars within acceptable limits. Moreover,
This difference of a few days probably does not as- Rifampicin has also been demonstrated to cause
sume clinical significance. In a study from India that early phase hyperglycaemic and hyper-insulin-
looked at 316 patients with tuberculosis, those with emia even in non diabetic patients. (10)and with

Cardio Diabetes Medicine