Epidemiological and clinical characteristics of multidrug-resistant tuberculosis in children and adolescents from 2016 to 2022

INTRODUCTION

Tuberculosis (TB) remains a public health problem worldwide, particularly in developing countries. Before the advent of the COVID-19 pandemic, TB was the leading cause of death due to a single infectious agent in the world, ahead of human immunodeficiency virus (HIV) infection. TB is caused by the mycobacteria of the TB Complex. Depending on the sensitivity of the bacilli to anti-TB drugs, the World Health Organization (WHO) distinguishes between drug-susceptible and drug-resistant TB.^[1] The most common form of drug-resistant TB is multidrug-resistant TB (MDR-TB). It is defined as TB caused by a strain of Mycobacterium tuberculosis that is resistant to at least the two most effective first-line drugs used in the treatment of TB: Rifampicin and isoniazid.^[2] The risk factors for MDR-TB are multiple and vary from region to region. These include inappropriate prescriptions for anti-TB drugs, irregularity of anti-TB treatment, and the existence of previous anti-TB treatment.^[3,4] MDR-TB affects both adults and children, in whom it is a serious form and a major health problem, especially in developing countries.^[5] TB in children poses many concerns, particularly those related to diagnosis, sometimes due to the difficulty of accessing samples to be analyzed, depending on the localization of the TB. This is the origin of an underestimation of cases of MDR-TB in children. Usually, TB is expressed by general signs of TB impregnation regardless of the localization. Functional signs depend on the localization. Thus, pulmonary localization, which is the most common form, is manifested by a chronic cough. MDRTB essentially poses two problems in developing countries: The first is related to treatment in relation to the absence of the usual therapeutic dose of anti-TB drugs for children^[1] which can cause adverse effects, and the second concerns the evolution of the disease, most often grafted with an unsatisfactory therapeutic success rate. Due to the under-reporting of cases of MDR-TB in children, most studies of MDR-TB are focused on adults. Very few data therefore exist on the MDR-TB of children in developing countries, and particularly in Côte d’Ivoire, where the latest data are at least 6 years old. To update these data, we carried out the present study, whose objective was to describe the epidemiological, clinical, of MDR-TB in children and adolescents from 2016 to 2022.

MATERIALS AND METHODS

RESULTS

Epidemiological characteristics of the study population

During our study period, we recorded 1741 cases of pediatric TB of all forms combined, including 46 cases of MDR-TB, representing a rate of 2.6% [Figure 1]. The mean age of patients was 7.8 ± 5.0 years with an interquartile range of 5 [8–13 quartile]. Patients whose ages ranged from 13 to 15 years old represented 60.9% of the study population. Female patients accounted for 65.2% of the population with a sex ratio of 0.5. The patients who were of school age (39/46) were all in school [Table 1].

Close

Figure 1:

Multidrug-resistant tuberculosis flow diagram. TB: Tuberculosis, MDR: Multidrug-resistant TB.

Export to PPT

Table 1: Sociodemographic characteristics of patients.

	Squad (n)	Percentage
Age groups
29 days–2 years	4	8.7
3 years–5 years	6	13.0
6 years–12 years	8	17.4
13 years–15 years	28	60.9
Sex
Male	16	34.8
Female	30	65.2
School status
No educated	7	15.2
Preschool	2	4.3
Primary	7	15.2
Secondary	30	65.2

DISCUSSION

TB affects people of all ages, but particularly adults and young people, in whom the notification of new cases is increased because of the positive diagnosis, which is most often easy. In our study, the rate of MDR-TB in children and adolescents was 2.7%. In Senegal, Ibrahim and col et al. found a similar rate of 1.8%.^[6] Indeed, according to the WHO, less than half of TB cases in children and adolescents are reported, which means that there are significant gaps in case detection.^[7] One of the reasons for these gaps is the difficulties in collecting samples and bacteriological confirmation of TB in young children, due to the paucibacillary nature of the disease in this age group. This prevalence is not negligible because these children, who are mostly of school age, could spread the disease at school to their peers and teachers, thus creating a real public health problem. These children are themselves most often contaminated by the family environment.

In our series, the mean age of patients was 7.8 ± 5.0 years. The age groups of 13–15 years (60.9%) and 6–12 years (17.4%) were the most represented. Bakayoko-Yéo and col et al.^[8] in Abidjan (Côte d’Ivoire), Diallo and col et al.^[9] in Guinea, and Drobac et al.^[10] in Peru found higher mean ages of 15, 13, and 11 years, respectively, probably due to the sampling fluctuation of each of these studies. However, the mean age is more than 5 years in all these studies. Indeed, before the age of 5, the child is mainly influenced by the family circle. From the age of 5, the child leaves the family environment and multiplies his or her contact with the outside environment (immediate neighborhood, nursery, or primary school), which increases the risk of exposure to TB.^[11]

In our study, the notion of TB in the family and friends was specified in only one patient out of 46, due to the non-mention of this variable in the patient file of the national TB control program in Côte d’Ivoire: A case of familial TB was reported and came from the family and friends. It has also been shown that the risk of developing TB after exposure to Koch’s bacillus is not the same for everyone. In fact, in the entourage of a subject spitting out Koch’s bacillus, 50% of children under 1 year of age will develop TB, and about 25% of children under 5 years of age. In adolescents, the risk can range from 15% to 25% due to wide variations in exposure.^[11] Regarding sex, our study population consisted mainly of female children (65.2%) with a sex ratio of 0.5. This clear female predominance is found in Ivorian and Peruvian studies with rates of 61% and 55%, respectively.^[8,10] These results differ from those of Satti and col et al.^[12] where the male sex predominated (53%) with a sex ratio of 1.1. The distribution of TB cases by sex is not of great value in pediatrics because of the risk of contamination, which is the same in the African family context.^[11]

TB usually manifests itself by general signs of TB impregnation associated with functional signs, depending on the location. In our study, the symptoms found in children and adolescents were identical to those observed in adults. These were fever (97.8%), weight loss (73.9%), asthenia (54.3%), and night sweats (43.5%). Chronic cough was found in 93.5% of patients, especially those with a pulmonary localization of TB. Ibrahim and col et al.^[6] found fever and deterioration in general condition in all patients, and chronic cough in 92%. Chan et al.^[13] found fever in 75% and dyspnea in 33% of patients. This variation in data could be explained by the lack of knowledge of the signs of TB by patients and their entourage, and sometimes by the difficulty in children under 5 years of age to be precise in the description of the clinical manifestations presented. The pulmonary form of MDR-TB predominated in our study. The same observation was made by Ibrahim and col et al. in Senegal.^[6] Indeed, despite the multiple localizations of TB, the lung remains the organ of choice because of the mode of contamination of the TB bacillus, which is mainly through the air. Among patients with a history of TB, 46% had been reported to have failed treatment after first-line anti-TB treatment. Diallo et al. had similar results in Guinea in children and adolescents.^[9] Indeed, certain factors such as inadequate treatment, poor adherence, history of anti-TB treatment, and treatment failure to a first anti-TB treatment are recognized as risk factors for developing MDR-TB.^[3,4]

In accordance with the therapeutic protocol for MDR-TB in force in Côte d’Ivoire, the children all benefited from the short protocol of 9 months with extensions to 12 and 15 months in two children with osteoarticular TB. This treatment was successful in 78% of patients, with a death rate of 10.9%. Bakayoko-Yéo and col et al. reported treatment success in 91.3% of patients with zero mortality in Abidjan.^[8] This zero mortality was also found by Ibrahim and col et al. in children with MDR-TB in Senegal.^[6] This difference in mortality could be explained by sampling variation. Respiratory rate >20 cycles/min has been recognized as a factor associated with mortality. In our study, children who had a respiratory rate higher than 20 cycles/min had a 39 times greater probability of death. Indeed, the appearance of dyspnea during TB reflects significant parenchymal involvement. In our context, this could be explained by the delay with which sick children are taken by parents to medical centers, while specifically MDR-TB patients benefit from free care. The delay could be explained by the fact that often in our communities, people consult traditional medicine before resorting to modern or conventional medicine because of the cultural conception of the disease. Second, in terms of the referral of MDR-TB patients who, although residing in urban areas such as Abidjan, the economic capital, may be faced with evacuation difficulties due to the insufficiency or unavailability of medical transport to the reference centers for care. In Cameroon, Bitchong and co et al. had made the same observation.^[14] However, our result differs from that of Ntumba et al. in the Democratic Republic of Congo^[15], where sex, weight loss, HIV seropositivity, negative direct microscopy, and leukocytosis ≥12000/mm³ were associated with a high risk of death in MDR-TB children. According to Hailu et al. in Ethiopia, the mortality factors were age under 5 years old and HIV infection.^[16] These mortality risk factors are therefore multiple and varied depending on the geographical area.

As limitations of this study, because of its retrospective nature, some information such as BCG vaccination and TB infection was only provided in very few medical records. This could influence our results. However, our selection criteria allowed us to have data to better characterize MDR-TB in children and adolescents in Abidjan.

CONCLUSION

The prevalence of MDR-TB in children and adolescents was low in our study. Clinical manifestations were similar to those in adults. Pulmonary localization is predominant. Two out of three patients had experienced adverse effects, dominated by digestive disorders. The treatment success rate was satisfactory, with a high mortality rate. Respiratory rate >20 cycles/min was associated with the patient death rate in our study.