A description of the pharmacokinetics of first line anti-tuberculosis drugs and adjunctive corticosteroid therapy in the pericardial compartment and their effect on cytokine profiles in the pericardial compartment of patients with tuberculous pericarditis

Shenje, Justin Tapiwa

A description of the pharmacokinetics of first line anti-tuberculosis drugs and adjunctive corticosteroid therapy in the pericardial compartment and their effect on cytokine profiles in the pericardial compartment of patients with tuberculous pericarditis

Thesis / Dissertation

2025

Publisher

University of Cape Town

Department

Department of Medicine

Faculty

Faculty of Health Sciences

Abstract

Tuberculous pericarditis is caused by Mycobacterium tuberculosis infection of the pericardium, which is associated with a higher mortality, compared with pulmonary tuberculosis. Constrictive pericarditis, a significant complication of tuberculous pericarditis, contributes to this elevated mortality. Adjunctive corticosteroid therapy with prednisolone, has been shown to reduce the incidence of constrictive pericarditis by more than 50%. However, this is associated with significant side effects, especially in patients with HIV co-infection. Constrictive pericarditis is only partially responsible for this increased mortality, whereas other factors including delayed diagnosis of tuberculosis, cardiac complications for example, pericardial tamponade and poor pericardial exposure of anti-TB drugs also likely contribute. The pharmacokinetics of anti-tuberculous (TB) drugs within the pericardial space remain poorly understood and until now have relied on studies of pulmonary tuberculosis. While adjunctive corticosteroids have been shown to be beneficial, there are no available data on the pharmacokinetics of prednisolone in the pericardium, and uncertainty remains as to whether an empiric dose of 120 mg prednisolone is the optimal dose for reducing the incidence of constrictive pericarditis or whether it is detectable at the site of infection, to exert its pharmacodynamic effect. Combining anti-TB treatment and high doses of prednisolone are likely to have a profound effect on the immune response, therapeutic effect and cortisol milieu in the pericardium. I was therefore interested in investigating the impact of this therapeutic combination, through direct sampling of the pericardial fluid in patients with tuberculous pericarditis. Methods: A total of 16 patients with newly diagnosed tuberculous pericarditis and a pericardial effusion requiring pericardial aspiration, were enrolled into the study. The patients were randomly assigned to 120 mg of prednisolone or matching placebo and were simultaneously started on first line World Health Organization (WHO) anti-tuberculosis treatment, namely, rifampicin, isoniazid, ethambutol and pyrazinamide. Intensive pharmacokinetic samples were collected at the following timepoints, just before administration of the anti-tuberculosis treatment and prednisolone or placebo and at 0.5 hour, 1 hour, 2 hour, 3 hour, 5 hour, 8 hour and 24 hour intervals thereafter for the first 24 hours. Intensive pharmacokinetic samples were collected from plasma, pericardial fluid and saliva. The pharmacokinetics of rifampicin, isoniazid, ethambutol and pyrazinamide in the pericardium were described so as to assess the adequacy of conventional doses of the WHO first line anti-TB treatment regimen. The study also described the pharmacokinetics of 120 mg of prednisolone and evaluated its effect on endogenous corticosteroids and cytokines. Furthermore, the study compared the immunological response to tuberculous pericarditis in plasma and pericardium. Results: While pericardial drug exposures to isoniazid and pyrazinamide were comparable to those of plasma, there was a reduction in rifampicin and ethambutol exposure in the pericardium, which were 20%; (p<0.0010) and 55%; (p<0.0010) of their respective plasma drug exposures. The median pericardial rifampicin concentration was 0.125 mg/L, which was lower than the reference minimum inhibitory concentration (MIC) for Mycobacterium tuberculosis isolates, which was 0.208 mg/L (p=0.0010). The pharmacokinetic profile of prednisolone in the pericardium was similar to that of the plasma, but there was a 1.60 hours delay (p=0.0320) in time to maximum concentration (Tmax) in the pericardial space. Adjunctive corticosteroid therapy administered as a single dose of 120 mg of prednisolone, was associated with 83% reduction in median plasma interleukin-6 (p=0.0360), a 50% reduction in median plasma interleukin-8 (p=0.0300) and a 50% reduction in median pericardial interleukin-8 (p=0.0400). Prior to administration of glucocorticoids, the pericardial cortisol/ cortisone ratio was twice that of plasma (p=0.0050). A similar elevation of the cortisol/ cortisone ratio at the site of infection has previously been described in the bronchial lavage fluid of patients with pulmonary tuberculosis. Among those participants who did not receive glucocorticoids, pericardial cytokine interferon-gama, tumour necrosis-alpha, interleukin-6, interleukin-8, interleukin-10, and interferon gamma-induced protein-10 concentration increases were multiples fold higher than their corresponding plasma concentrations, (p=0.0011), (p=0.0043), (p=0.0009), (p=0.0001), (p=0.0054) and (p=0.0031), respectively. Conclusion: By directly sampling the pericardial fluid among patients with tuberculous pericarditis, I was able to gain novel insights, which have never previously been described. The reduced pericardial rifampicin and ethambutol exposures, below the required minimum inhibitory concentrations in the pericardial space, are of great concern, particularly, considering the relative importance of rifampicin in the treatment of drug sensitive tuberculosis. This warrants a revision of the dosing strategy, or a review of therapeutic regimen used in tuberculous pericarditis, to enhance efficacy. Prednisolone administered at a dose of 120 mg daily orally is detectable with high exposure in the pericardium and may explain the beneficial effect observed in tuberculous pericarditis among HIV negative individuals, through its alteration in the cytokine milieu, and lower propensity to developing constrictive pericarditis. Additionally, the raised cortisol/ cortisone ratio at the site of infection, the pericardium, compared with other matrices especially plasma may indicate an immunomodulatory role either to dampen an excessive or evoke an immunological response.