Implementation of multidrug-resistant tuberculosis (MDR-TB) treatment in Gabon: lessons learnt from the field

Multidrug-resistant tuberculosis (MDR-TB) and higher degrees of resistant tuberculosis are an emerging health concern including populations in sub-Saharan Africa, and a massive threat to TB programmes in many low- and middle-income countries [1]. However, in the majority of these countries, patients often only have very limited access to MDR-TB treatment and data are still scarce on incidence, prevalence, molecular characteristics, treatment, and outcome [1]. In Gabon, despite case reports of treatment failure to the first-line TB therapy [2], it was only in 2013 that a more comprehensive assessment of the epidemiology of MDR-TB was conducted for the first time [3]. This study, carried out in Lambaréné, a semi-urban town of around 30.000 inhabitants, indicated an MDR rate of 4/91 (4%) and 4/13 (31%) in new and retreatment TB cases, respectively [3]. This rate is quite high and is comparable to prevalences reported from highly MDR-TB endemic countries [4, 5]. At the time of the official report on the first MDR-TB cases in Lambaréné and their growing number, the Gabonese TB programme lacked access to the second-line anti-tuberculosis drugs as well as specialized and dedicated treatment facilities [6]. In response to this epidemic, MDR-TB was recognized as a public health emergency in Gabon; and in September 2015, the first MDR-TB treatment center was opened in Lambaréné, at the Georges Rawiri Hospital (GRH), with technical support provided by CERMEL’s clinical and mycobacteriology laboratory teams. In parallel, the Centre de Recherches Médicales de Lambaréné (CERMEL) obtained funding from the German Ministry of Health for the acquisition of the second-line anti-tuberculosis drugs for the evaluation of a 9–12 month short-course treatment regimen of MDR-TB. This short-course treatment was already evaluated by the International Union Against Tuberculosis and Lung Disease (IUALTD, ‘The Union’) in several countries in Africa and in Bangladesh prior to its endorsement by the WHO in May 2016 [7]. As a result of the collaboration between CERMEL and the Ministry of Health in Gabon, the first MDR-TB patients were initiated on a short-course treatment regimen in September 2015; in a study aimed at evaluating the implementation of the 9–12 month short regimen for the treatment of MDR-TB. We report here the experience gained during the treatment of those first MDR-TB patients in Gabon’ and comment on the appropriateness of the standard 9-month regimen in our setting.

MDR-TB represents an emerging public health concern in Gabon. Despite the increasing number of cases, few, if any, treatment options were available in the country until recently [2, 3, 10]. Here, we report on the outcome of a pilot treatment phase that aimed to introduce and implement the 9–12 month short-regimen scheme before its endorsement by the WHO in May 2016 [7].

The appropriateness of the abridged WHO MDR-TB regime has been disputed [11]. Lange et al. [11] point out that in view of the respective prevailing drug resistance patterns e.g., in the Western European region and in the Eastern European region, only 7.8% [12] and 4.2% [13] of patients with DR-TB would be eligible for the short-course regimen. In Singapore, Indonesia and other South East Asian countries, the figure would be 30% [14]; of the other regions examined, only Brazil would yield 50–55% eligible patients [15]. However, with until recently an almost complete absence of second- and third-line TB drugs [3], our setting renders patient cohorts such as the one presented here as an ideal candidate for the abridged WHO regimen, which is in situations where due to previous lack of access to second-line treatments, resistance to second-line drugs is by-and-large absent.

The medical history of the patients included in this case series clearly indicates a delay in the diagnosis of MDR-TB that led to unnecessary and ineffective treatment with the first-line drugs. Consequently, this led to profound health deterioration, extensive lung lesions, and irreversible sequelae such as chronic respiratory failure. Moreover, DST indicated that mycobacteria isolated from all patients were resistant to at least three of the first-line anti-tuberculosis drugs, which could be a consequence of repeated treatment [16] emphasising the importance of early diagnosis and treatment of MDR-TB.

Under the standardized treatment regimen, sputum culture conversion was obtained already after 3 months of treatment in nine patients, whereas sputum conversion by microscopy, possibly due to the presence of dead bacteria, was observed after 4 months of treatment in seven patients. This confirms the necessity of culture for the monitoring of treatment outcome. A similar delay in smear conversion over culture conversion has been reported elsewhere [17] and is probably linked to the presence of dead bacilli hidden in cavities and later on released into the airways. Given that infectiousness is associated with positive sputum culture results and is highest when the smear results are also positive [18], data from our small patient cohort indicate that MDR-TB patients’ infectiousness was already reduced by month 2 and reached null by month 3 of treatment (data not shown).

All patients experienced at least one AE during the treatment phase that was managed accordingly. As expected, the most common AEs were gastro-intestinal and consisted mainly of nausea and vomiting. One patient experienced a cutaneous hypersensitivity reaction that was not clearly attributable to an individual drug. This reaction resolved after 2.5 months of symptomatic management without discontinuation of therapy. Towards the end of the continuation phase, two patients complained about a significant hearing loss. Although this is surely caused by the prolonged administration of kanamycin, it is possible that these patients who already received streptomycin several times before already had established sub-clinical cochlear lesions. Unfortunately, due to logistical constraints, we were not able to quantify the hearing loss. Management of MDR-TB patients is multidisciplinary and usually requires expertise and medical equipment not available in remote hospitals. This could affect the outcome of their treatment particularly with regard to the monitoring, prevention, and treatment of drug adverse effects. This limitation should be addressed in the future by encouraging the policy maker to provide appropriate logistic tool for efficient monitoring.

One patient died in the first month of treatment. Clinical and radiological features on admission depicted extensive lung damage that could be linked to delay in receiving adequate MDR-TB treatment. Liver failure occurred in one patient towards the end of the intensive phase. This patient who worked in Libreville was referred to our team in Lambaréné for the initiation of her MDR-TB treatment. Hospitalisation in Lambaréné was originally planned to last for the whole duration of the intensive phase. However, due to family and professional issues, the patient was discharged from our MDR-TB treatment unit and referred to a treating physician in Libreville. Liver enzymes before discharge in Lambaréné were within normal ranges.

The cost of treatment in our patient cohort is lower compared to a study in South Africa which reported 17,164 USD per patient [19], but our findings equate to those reported from, e.g., the Philippines, Peru, and Nigeria (3613; 2423, 2095 USD, respectively) [20, 21]. The treatment cost is strongly correlated with the country’s socio-economic profile and the level of hospital care, as well as the sourcing/financing of the drugs. However, all of these studies agree that the largest cost of care is attributable to the prolonged hospitalisation. Despite evidence that hospitalisation is desirable during the infective period and to achieve the highest rate of adherence to treatment, ambulatory care approaches should be encouraged where possible and as early as possible to reduce the cost of treatment.

In summary, the implementation of the MDR-TB short-treatment regimen led to clinical improvement and sputum conversion in most of the patients. Early treatment could prevent extensive lung damage and death. All patients experienced at least one adverse event that included adverse cutaneous drug reaction and impaired hearing function, but did not lead to an interruption of the treatment. One patient experienced a severe and fatal adverse event. The problems of managing this patient highlight the importance of expanding the MDR-TB treatment capacity to other provinces to facilitate country-wide appropriate patient care. At the time of this study, Lambaréné was the only site of MDR-TB treatment. Patients could not be transferred safely to other province near to their residence to continue the treatment.