Tuberculosis (TB) is a chronic bacterial infection of worldwide distribution, caused by the Mycobacterium tuberculosis complex ¹. 90% of infected people are asymptomatic ^2,3, with pulmonary localization being the most frequent clinical form ⁴. Although the disease affects especially men of productive age in low-income countries, it is increasingly common in children. According to the World Health Organization (WHO), about one million cases of childhood TB are reported annually, equivalent to 11%, with an estimated mortality of 136,000 cases, of which 40% are HIV positive ⁵.

According to the Pan American Health Organization (PAHO), TB is the second cause of death by an infectious agent, after HIV- AIDS ⁵. In Colombia in recent years, the incidence of TB has ranged between 24 and 26 cases per each 100,000 inhabitants ⁶. Buenaventura reported an incidence of 90 cases per 100,000 inhabitants in 2011, the highest incidence in the country ⁷. For the year 2010, there were reported 719 cases of childhood TB (6.3%), for an incidence of 5.47 cases per 100,000 children aged under 15 years ⁸; and in 2016, there were reported 584 cases of childhood TB (18%).

The diagnosis of pulmonary TB in children is complex, due to the diversity of non-specific symptoms and the difficulty in obtaining sputum samples in a spontaneous manner that allow the observation and isolation of the microorganism, so the diagnosis is based on the joint evaluation of epidemiological, clinical, radiological, immunological and microbiological criteria, which raises costs and delays diagnosis ⁹.

The exposure, infection and development of the disease in children is determined by close contact with baciliferous adults (parents or caregivers) ¹⁰; for this reason, accuracy and timeliness in diagnosis and treatment, as well as an effective contact study, are essential to cut the chain of transmission and achieve the objectives of TB prevention and control programs. However, the adverse social, political and security conditions of depressed areas of Buenaventura are the main obstacle that hinder the actions of diagnosis, treatment, follow-up and notification for the TB control program of the municipality ⁽¹¹.

The objective of this research was to describe demographic and clinical characteristics and treatment outcomes of the infant population diagnosed with pulmonary tuberculosis enrolled between 2009 and 2013 in the municipal TB control program of Buenaventura - Colombia.

Between January 2009 and December 2013, there were registered 238 children aged 1 to 15 years with a diagnosis of Tuberculosis Pulmonary in the municipal Tuberculosis control program in Buenaventura. After evaluating the selection criteria, the information of 163 patient records was analyzed (Figure 1).

Figure 1:

According to the demographic characteristics, there were no differences by sex; however, when stratifying for age, it was found that the age group of 6-10 years had an increase in the number of cases in males in a ratio of 1.6:1; situation that is reversed in the age group of 11-15 years, in which females have a higher percentage of TB in a 1.5:1 ratio. Taking into account the previous classification, the age group of 1 to 5 years contributed 51% of cases of childhood TB, and the remaining 49% was evenly distributed between the two other groups (Table 1). When comparing the variable sex with age group, these were independent, according to the chi-square test: 3.83 (p-value: 0.15).

Table 1

Characteristic	%
Sex (total of population)
Male	50.0
Female	50.0
Age (years)
1-5	51.0
6-10	24.5
11-15	24.5
Health system
Subsidized	55.0
None	31.0
Contributive	12.0
Special	2.0
Commune
12	23.0
Rural	18.0
3	10.0
11	10.0
Other Communes	39.0

Regarding the health system, only 14% of the population belonged to the contributory or special regime; 55% of the population belonged to the subsidized regime; while 31% did not have any type of insurance. Analyzing the geographical distribution, the largest numbers of cases were concentrated in communes 3, 11 and 12; and the rural area; corresponding to 61% of the location of cases.

Tuberculosis in children is difficult to diagnose; its management and follow-up are challenging for TB control programs. This is the first study that describes the demographic, diagnostic and programmatic outcomes of pulmonary tuberculosis in a cohort of children enrolled in the Buenaventura TB control program, between 2009 and 2013.

The characteristics of the population according to the sex distribution did not show significant differences, reflecting the population distribution of Buenaventura, where there were no changes in the gender registered by the 2005 census with a projection to the year 2011 ¹⁴; also resembling a study conducted in Cuba from 2009-2011 ¹⁵.

According to age, a greater percentage of cases occurred between 1-5 years, which probably does not imply a higher incidence, but rather corresponds to a better screening strategy and epidemiological surveillance, based on the strengthening of the IMCI strategy ¹⁶ and the support of the NGO Medicos sin Frontera (Médecins Sans Frontières, France) in the generation and capture of patients and home visits, identifying all children under 5 years of age living in the home of an adult with pulmonary tuberculosis, and then referring them to a health center for their clinical evaluation ¹¹. Similar results have been reported for 2012 throughout Colombia ¹⁷; and in a study conducted in Cuba, 2001-2010 ¹⁸. Buenaventura has limitations in sanitary conditions and difficulties in territorial organization, which makes it difficult to classify socio-economic levels, so this study explored the relationship of the health system as a social indicator. This is correlated with the finding of 55% in the subsidized health system, followed by 31% of uninsured persons, situation that evidences the nature of the disease. This is similar to a study carried out in Cali in 2014 ¹⁹, where 86.4% of children with TB lived in neighborhoods of low socioeconomic status.

Cases from rural areas represented 18% of the reported cases, and the urban sector presented the highest figures in communes 12, 3 and 11, since these accounted for 43% of the cases. Such information would be useful to generate prevention strategies in high burden communities, as it has been expressed in some studies ^18,20 that worked in TB neighborhoods based on the preparation of the communities and health personnel, in order to detect new cases, so diminishing the incidence based on new Public Health strategies ²¹; this has great importance in the active search of contact sources, since the diagnosis of a child with TB represents a sentinel event that reflects recent transmission within a limited circle of social contact ². It is defined that the exact quantification of the burden of pediatric TB disease provides a key parameter of control of the epidemic ¹⁸.

The clinical diagnosis represents the highest percentage of cases, in addition to a high interrelation with the epidemiological diagnosis, which evidences the gaps mentioned by the Panel of Experts in 2012, since the tests used conventionally do not achieve microbiological confirmation of the disease ²². However, the diagnosis by laboratory showed an increase during the evaluated period, which could be explained by the introduction of microbiological tests to the program. This form of diagnosis is associated in a dependent manner at age >7 years (p-value: 0.0002), since there’s a greater possibility of detecting bacilli per field in a smear microscopy in this population, which could also explain the slight increase in notifications of cases for the age of 12 years.

A large percentage of new cases was evidenced thanks to a sieve format, which allowed pulmonary tuberculosis to be detected in respiratory, symptomatic patients. In Colombia, during the last years, about 6% of the new detected cases of TB occurred in the population aged under 15 years ⁶. It was evidenced re- entry of 7%, something that however does not correlate with the 32% of abandonment, possibly as a consequence of lack of education for parents and environment. According to Medicos sin Frontera’s (Médecins Sans Frontières), 58% of abandonment in this population depends on social, family and cultural factors ¹¹; however, new studies are required to expand the possible causes. The abandonment and living in urban or rural areas show dependence (p-value: 0.0007), an association that moves away the compliance of the national goal, according to PAHO ⁵.

The Pulmonary and Extra-pulmonary Tuberculosis Care Guide - 2007 ²³ given by the Ministry of Social Protection, recommends initiation of post-diagnosis treatment in a timely manner, in addition to a treatment based on a 180-day period, divided into two phases of supervision, a situation that varies in Buenaventura. Such behavior possibly occurred due to ignorance of the norm, lack of role of the support networks, and little education to the caregivers (regarding treatment). It is observed that the relationship between the start of treatment and the treatment time do not depend on the Assurance Regime or the Territorial Organization, despite the limitations described for Buenaventura.

The percentage of compliance for phase I in this study was 23%. The lower doses may be part of the Abandonment cases, but those that exceeded the doses would express the little support and control of the follow-up. Alarmingly, for phase II, the study showed no compliance in any of the cases, reflecting seriously the poor application of the regulations.

The exact quantification of pediatric TB provides a key parameter of control of the epidemic; according to the findings of the study, the incidence of Pediatric Pulmonary TB in Buenaventura was 2-4 x 100,000 inhabitants, (a figure) lower than that published in the Epidemiological Report of Colombia of the year 2012 ⁸, which shows an incidence of 5.5 x 100,000 inhabitants for those aged under 15 years. However, it is a high rate when compared with other countries, such as Cuba, with better prevention and control strategies, where a rate of 2.2 and 2.1 was registered in children from 0 to 14 years in the period 2001 to 2010 ¹⁸; or the Netherlands, corresponding to 3.6 for the year 1993 to 1.9 for the year 2012 ²⁴, evidencing that there is much to improve in the prevention of this disease.

At the beginning of the study, there was a 51% correlation between the Tuberculosis Program data and the National Public Health Surveillance System, but despite the shortcomings, a tendency towards improvement was observed, a situation that was unified in 2013, with a 100% correlation.

Due to the lack of data, a fact the researchers are not accountable for, it was possible to analyze (only) 68% (163/238) of the records of pediatric patients with pulmonary TB reported by the TB program, estimating the epidemiological situation of the TB children in Buenaventura.

One of the strengths of this study was the active search for information, managing to recover 93% of the records of patients not notified, which ensured the quality of the information.

The diagnosis, treatment and follow-up of the children population with pulmonary tuberculosis is a challenge for clinicians and for tuberculosis control programs. It is worth highlighting the role of Medicos sin Frontera (Médecins Sans Frontières, France), which contributed to the reduction of the incidence of pulmonary tuberculosis in children, in Buenaventura.

In conclusion, the social and political difficulties in depressed areas of Buenaventura have a negative impact on the control strategies of pulmonary tuberculosis in children, which are based on diagnosis and timely treatment, effective case monitoring, contact study, and efficient, timely and truthful notification.