Rifampicin levels strongly reduced in patients with TB and diabetes

Levels of the anti-TB drug rifampicin are significantly lower in people with tuberculosis and type II diabetes, Dutch and Indonesian researchers report in the October 1^st edition of Clinical Infectious Diseases. They believe their finding may explain why people with type II diabetes have a less favourable response to TB treatment, and warn that people with diabetes, especially those with higher body weights, may need higher doses of rifampicin.

Type II diabetes is known to be an important risk factor for the development of active TB, and it has been estimated that by 2025, 75% of people with diabetes will live in countries that harbour the majority of TB cases, such as India and South Africa.

Researchers from Indonesia and Radboud University Nijmegen Medical Centre in the Netherlands selected 17 patients with diabetes and TB who were in the continuation phase of TB treatment and matched them by age and sex with 17 TB patients who did not have diabetes. Diabetes was confirmed at the time of study dosing. All participants were receiving treatment with rifampicin 450mg and isoniazid 600mg three times a week. No patients were using diabetes medication known to have an effect on rifampicin levels. HIV-positive participants were excluded from the study. Male diabetic patients (9/17) were significantly heavier than non-diabetic patients (55kg vs 50kg, p=0.03).

Rifampicin levels were measured after an average of two months of continuation phase treatment on a single day at the time of dosing and two, four and six hours after dosing. Patients fasted from the preceding evening until four hours after dosing on the study day. Plasma samples were frozen and shipped to the Netherlands for analysis.

Patients with diabetes had significantly lower rifampicin levels than non-diabetic patients, whether measured by total exposure over the six hour period of sampling (AUC) (12.3mg X h/L vs 25.9 mg x h/L, p=0.03) or peak levels (Cmax) (3.49 mg/L vs 6.74mg/L, p=0.004).

Exposure to rifampicin was 53% lower in diabetic patients, and the authors comment: “A 50% lower exposure to a key TB drug such as rifampicin may cause clinical failures and may favour the emergence of drug resistance. In fact, the findings of this study suggest that the unfavourable response of patients with [diabetes] to anti-TB drugs may at least partially be explained by differences in pharmacokinetics.”

Multivariate analysis showed that 57% of the difference could be attributed to the difference in weight between the diabetic and non-diabetic groups, but fasting blood glucose levels were also shown to be independently associated with rifampicin AUC.

“The mean difference of approximately 4 mmol/L in fasting plasma glucose between patients with TB only and and patients with TB and [diabetes] corresponds with 61% of the difference in exposure to rifampicin,” the authors note.

They say that if these findings are confirmed, higher fixed doses of rifampicin may be advisable for patients with diabetes and higher body weight, and they sugest that if available, pharmacokinetic monitoring of rifampicin levels at multiple time points may be warranted in order to individualise dosing.

Regarding the mechanism by which diabetes might affect rifampicin levels, the authors point out that diabetes can reduce secretion of gastric hydrochloric acid, so reducing stomach acidity. This in turn might reduce rifampicin absorption, which is dependent on a high pH in the stomach.