Prevalence of Pulmonary Tuberculosis and Associated Factors among Patients Admitted to the Amissa Bongo University Hospital Center in Franceville, Gabon ()
1. Introduction
Caused by Mycobacterium tuberculosis, tuberculosis (TB) is a serious bacterial disease that mainly affects the lungs (pulmonary TB), but can also affect other organs (extra-pulmonary TB) [1] . The disease is transmitted when a person with tuberculosis coughs, speedés, sings or talks. The germs of Mycobacterium tuberculosis are spread through the air and can be inhaled by other people. An estimated 10.6 million people fell ill with tuberculosis (TB) in 2021, an increase of 4.5% on 2020, and 1.6 million people died from TB (including 187,000 HIV-positive people), according to the World Health Organization’s Global TB Report 2022 [2] . However, tuberculosis is a very common opportunistic disease that kills infected people, but can be cured if people go to hospital early for screening and follow the advice of healthcare staff until treatment is complete [3] . The WHO recently reported that 12% of the 9.6 million new cases of tuberculosis were HIV-positive [4] . In HIV-positive patients, active tuberculosis is due to the reactivation of endogenous latent disease and reinfection with a new strain [5] . Epidemiological studies have shown that co-infection with HIV can increase the risk of reactivation of latent tuberculosis by a factor of 20, and is the strongest known risk factor for the progression of M. tuberculosis infection to active disease [6] . Furthermore, the WHO has indicated that tuberculosis, along with HIV, is the most deadly infectious disease in the world, with both diseases estimated to be responsible for approximately 1.5 million deaths in 2022 [4] . Because of the HIV pandemic, poverty, the movement of displaced people and the emergence of multi-drug resistant strains, tuberculosis remains an enormous public health problem in developing countries [7] . Moreover, a previous study showed that in most developing countries, HIV pandemics, diabetes, malnutrition, alcoholism, smoking, contact with active TB, extreme poverty and homelessness are common risk factors identified in relation to TB [8] . Tuberculosis has been identified as an important cause of morbidity and mortality in Gabon. [9] . Considered a developing country in Central Africa, economic imbalances are widespread, the literacy rate is borderline average and basic health services are not only under-equipped but sometimes scarce. In Gabon, according to the World Health Atlas, the incidence of tuberculosis was estimated at 521/100,000 in 2019 [10] . According to the National Tuberculosis Control Programme (PNLT), the WHO ranks Gabon 4th among African countries and territories for the incidence of tuberculosis, and with a high morbidity rate [11] . Another problem facing the country is the simultaneous prevalence of TB and HIV. In a study conducted in the south of the country in 2021, 73 patients (6.4%) were co-infected with HIV+/TB+ [12] . The mortality rate for people infected with both TB and HIV was 25% [13] . Faced with this situation, since 4 May 2021, at a workshop to validate its national strategic plan 2021-2025, the Gabonese government has made the fight against tuberculosis one of the priorities of the country’s health sector development programme [10] . However, a review of the literature indicates that there is a paucity of studies on the prevalence of TB in patients in south-eastern Gabon. For this reason, the aim of the present study was to determine the prevalence of pulmonary tuberculosis and associated factors among patients attending the Centre Hospitalier Universitaire Régional Amissa Bongo in Franceville.
2. Patients and Methods
2.1. Study Design, Period and Description of the Study Area
This is a prospective, cross-sectional study designed and conducted from 19 June 2021 to 22 January 2022 at the biomedical analysis laboratory of the Centre Hospitalier Universitaire Amissa Bongo in Franceville, the provincial capital of Haut Ogooué, Gabon.
2.2. Study Population
Determination of Sample Size and Sampling Technique
To estimate the prevalence and risk factors associated with pulmonary TB in the city of Franceville and surrounding departments, the study sample size was determined using the single proportion population formula as used elsewhere, positing the formula:
[14]
In this, n represents the sample size, Zα/2 is the standard normal deviation (1.96), corresponding to a 95% confidence interval (CI), P is the prevalence of pulmonary tuberculosis. In the absence of P values obtained elsewhere, or in a previous study in the town of Franceville and surrounding departments, this P value was taken to be 50%. d is considered to be the precision/marginal error (d = 0.05) or 5%. Initially, the sample size determined for the study was 281. As applied in two studies elsewhere, errors resulting from the probability of non-compliance or drop-out were minimised. As the sample size was increased by 10% [15] [16] , the sample size finally used in this study was 309 participants.
2.3. Data Collection, Sampling Procedure and Sample Processing
The data used for this study were collected from people who came for diagnosis to the biomedical analysis laboratory of the Centre Hospitalier Universitaire Régional Amissa Bongo in Franceville as part of the national tuberculosis control programme.
2.3.1. Study Inclusion and Exclusion Criteria
Only people who had consented to participate in the study by correctly answering the survey questionnaire, and whose results of the two diagnoses were usable, were included in the present study. People who did not wish to take part in the study or whose test results could not be used were excluded.
2.3.2. Sociodemographic Data
Data on patients’ socio-demographic characteristics were obtained by means of a face-to-face interview in the laboratory with the patient using a pre-tested semi-structured questionnaire on 10% of the subjects, who were not included in the study. The questionnaire took into account the patient’s date of diagnosis, age, sex, marital status, professional status, level of education, family size, place of residence and monthly income.
2.3.3. Laboratory Diagnoses
The diagnosis of tuberculosis was made on the basis of clinical evidence, i.e. sputum was collected in leak-proof sterile plastic containers, decontaminated with sodium hypochlorite and concentrated by centrifugation to release the bacteria ready for staining. Sputum samples from adults or gastric fluid from children were examined by the Xpert MTB/RIF test, using the standard protocol. This is a rapid diagnostic tool for multidrug-resistant tuberculosis recommended by the WHO, which stipulates that once the sputum has been collected, it is mixed in the cartridge supplied with the sample reagent buffer in a volume ratio of 1:2 (sample: sample reagent buffer). The cartridge is then hermetically sealed, vortexed for 15 seconds and left to stand at room temperature for 10 min. After this, the cartridge is vortexed again after 10 minutes and left to stand for 5 minutes. Using the Pasteur pipette supplied with the kit, just over 2 ml of the treated sample is placed in the cartridge, which is then loaded into the Xpert machine and the results collected via an Xpert computer after 2 hours. The results are then collected using an Xpert computer [17] . Secondly, by smears prepared, fixed and stained using Ziehl-Neelsen staining methods as described elsewhere [18] . To confirm the presence of acid-fast bacilli, the smears were then carefully inspected using a zoom stereomicroscope. When two or more smears showed acid-base staining, the sample was defined as positive. In addition, the questionnaire submitted to the patients also included the risk factors for tuberculosis to which they were exposed, i.e. smoking, patients who had already received anti-tuberculosis treatment, HIV infection, asthma, nutritional status, and alcohol consumption.
2.3.4. Clinical Symptoms of Tuberculosis
Clinical symptoms of tuberculosis such as chest pain, fever, cough, duration of cough, and night sweats were observed in some patients in the study.
2.4. Data Analysis
Entered in a Microsoft Excel 2016 format, the data were then analysed using R software version 3.6.1, including the measurement of rates to determine risk factors associated with TB prevalence. A 95% confidence interval was estimated and a p ≤ 0.05 value was considered statistically significant. Crude odds ratios (ORs), and adjusted odds ratios (AORs) to a 95% confidence interval (95% CI), were calculated for each risk factor variable. A p-value ≤ 0.05 was considered statistically significant.
2.5. Ethical Considerations
The study obtained authorisation from the South-East Regional Director of Health, by letter No. 0345/PHO/SG/DRSSE/SGP/D, and endorsement from the managers of the University Hospital Centre of Franceville (CHUF). The data collected did not include patient identities or personal information. Patients’ anonymity was guaranteed, to avoid any stigmatization.
3. Results
3.1. Sociodemographic Characteristics of Study Patients
A total of 309 patients were included in this study. Of these, 50.81% (n = 157) were men and 49.19% (n = 152) were women. It was found that 55.66% (n = 172) were aged between 15 and 49 years, followed by the ≥50 years age group with 26.53% (n = 82). The remainder (n = 55, i.e. 17.08%) were in the 1 to 4 years and 0 to 11 months age groups. The mean age of the respondents in this study was 43.31 years (mean standard deviation = 20.21). Regarding the marital status of the study population, 44.01% (n = 136) of respondents were single and 22.65% (n = 70) were divorced. The majority (n = 189; 61.17%) of respondents lived in urban areas (in Franceville) and 38.84% (n = 120) lived in rural areas in the surrounding departments. 27.51% (n = 85) were unemployed, followed by 23.95% (n = 74) civil servants, 18.77% (n = 58) pensioners and 18.12% (n = 56) housewives. Only 45.31% (n = 140) of the patients were illiterate, but 29.13% (n = 90) had completed secondary school and 8.41% university. 138 patients (44.66%) lived in large families of 6 or more people. Monthly income was high for only 13.26% of patients (Table 1).
3.2. Prevalence of Tuberculosis in Study Patients
Of 309 sputum samples from adults or gastric fluid samples from children in the
Table 1. Socio-demographic characteristics of study patients.
CFA = African Financial Community; 1 US Dollar = 607.74 CFA Franc.
study, 98 samples, giving an overall prevalence rate of 31.72% (95% CI: [0.26 - 0.26]), were positive for both diagnoses of pulmonary tuberculosis. There were 40 females or 40.82% (95% CI: [0.30 - 0.51]) and 58 males or 59.18% (95% CI: [0.48 - 0.69]). The most infected age group was between 15 and 49, with 38 patients or 19.74% (95% CI: [0.29 - 0.49]).
3.3. Prevalence of Tuberculosis According to Risk Factors and Clinical Signs Observed in Study Patients
A univariate analysis of tuberculosis prevalence according to risk factors and clinical signs observed in study patients indicated that, patients in the 5 - 14 age group (OR = 3. 24; 95% CI: [1.47, 7.15] p = 0.002*), 15 and 49 years (OR = 0.36; 95% CI: [0.22; 0.59] p < 0.001*), male (OR = 1.64; 95% CI: [1.01, 2.66] p = 0.045*), who smoked (OR = 1. 75; 95% CI: [1.03; 2.91] p = 0.03*), who consumed alcohol (OR = 3.62; 95% CI: [2.11; 6.22] p = 0.001*), with asthma history (OR = 5.71; 95% CI: [2.39, 13. 66] p < 0.001), having previously received anti-tuberculosis treatment (OR = 2.57; 95% CI: [[1.24, 5.33] p = 0.006*) and being HIV-infected (OR = 3.2; 95% CI: [1.87, 5.51] p < 0.001) had a very high probability of being tuberculosis patients. Similarly, patients with clinical symptoms such as cough (OR = 7.53; 95% CI: [4.19, 13.54]) p < 0.001, chest pain (OR = 62.81; 95% CI: [29.07: 135. 72] p < 0.001), fever (OR = 21.1; 95% CI: [10.62, 42.2] p < 0.001), and night sweats (OR = 13.24; 95% CI: [7.36: 23.82] p < 0.001), were at high risk of tuberculosis infection (Table 2).
3.4. Multivariate Analysis of the Prevalence of Tuberculosis, According to the Risk Factors and Clinics Observed in the Study Patients
An in-depth analysis of the prevalence of tuberculosis, according to the risk factors and clinics observed in the patients in the study, indicated that the age group 15 and 49 years (Adjusted OR = 47.77; 95% CI: [4. 4; 519.7] p = 0.001*), fever (Adjusted OR = 4.83; 95% CI: [1.16, 20.12] p = 0.031*), alcohol consumption (Adjusted OR = 12.2; 95% CI: 12.2 [2.5; 5.9] p = 0. 002*), smoking (Adjusted OR = 9.83; 95% CI: [1.78; 54.24] p = 0.037), HIV infection (Adjusted OR = 4.63; 95% CI: [1.25; 17.1] p = 0.022), cough (Adjusted OR = 4.31; 95% CI: [1.21, 15.4] p = 0. 025*), chest pain (Adjusted OR = 103.6; 95% CI: [19.4; 55.2] p = 0.000), night sweating cough (Adjusted OR = 10.84; 95% CI: [3.18; 36.98] p = 0.000*) were potential risk factors for tuberculosis in the present study (Table 3).
4. Discussion
Although the recent WHO report on tuberculosis indicated that the prevalence of this disease has declined worldwide, we are far from achieving a world without tuberculosis, even by 2050 [19] . The present study indicated a global prevalence of 31.72%. Confirming Gabon’s position in the zone of 300 or more cases of tuberculosis per 100,000 inhabitants worldwide [20] , this result is comparable to that of an earlier study carried out in Ethiopia, which found a prevalence of
Table 2. Univariate analysis of tuberculosis prevalence according to risk factors and clinical signs observed in study patients.
OR = odds ratio; CI = confidence interval; * = significant test.
Table 3. Multivariate logistic regression analysis of tuberculosis prevalence as a function of risk factors associated with tuberculosis and clinical symptoms observed in study patients.
OR = odds ratio; CI = confidence interval; * = significant test.
33% [21] . It is well above those found in Central African Republic 10.99%, East Amhara, Ethiopia: 2015-2019 which was 11%, South Africa 13% [22] [23] [24] .. However, it is well below those obtained in other studies, carried out in other African countries such as Congo, 79.1% [25] , Togo, 57% [26] . The variability of these results could be due not only to the different diagnostic methods used in each study, but also to the quality of the participants in the different studies, the study period, the sample size, the geographical location and the differences in the methods and practices put in place for tuberculosis control in each region or country. The high prevalence of tuberculosis in the present study could be due to the multiple screening awareness campaigns promoted by the PNLT. As a result, the majority of study participants (33.33%) were newly infected patients (who had never taken anti-tuberculosis treatment). Contrary to the results obtained in studies which indicated that the age groups most affected by pulmonary tuberculosis were those aged 21 - 30 (28.24%) and 31 - 40 (26.55%) [27] , a univariate and then multivariate analysis indicated that the prevalence of pulmonary tuberculosis was statistically associated with patients in the 15 - 49 age group. In line with studies carried out in many countries that have shown a predominance of tuberculosis in young adults [28] , this result can be justified by the fact that, with a mean age of 43.31 ± 20.21 years, the patients in the present study were relatively young adults. Socially and economically active, they represent a period of life that is unfortunately often conducive to numerous infections, including that caused by Koch’s bacillus, the causative agent of tuberculosis [29] . Just as fever was reported in 60% - 85% of patients with pulmonary tuberculosis in a study elsewhere [30] , the present study indicated that fever was statistically associated with the disease. This corroborates well with the fact that generally, every case of active pulmonary tuberculosis presents some degree of pyrexia, which is one of the important signs of tuberculosis activity [31] . In contrast to studies such as Arnould, which denied the etiological influence of alcoholism on tuberculosis [32] , alcohol consumption was shown to be significantly associated with tuberculosis in the present study. This is in line with a study carried out in France, in which it was stated that most tuberculosis sufferers were alcoholics, and that tuberculosis and alcoholism constituted a serious morbid couple [33] . In addition, in vivo and in vitro studies have shown that alcohol consumption considerably disrupts the immune response, increasing susceptibility to respiratory diseases such as tuberculosis [34] . Despite the opinion of some systematic reviews and meta-analyses of observational studies that have shown an unfavourable association between global epidemics of tuberculosis and smoking [35] , smoking was associated with pulmonary tuberculosis in the present study. This finding corroborates the fact that both passive and active exposure to cigarette smoke are associated with an increased risk of developing active tuberculosis and with mortality from this disease [9] . In line with the literature according to which HIV infection is accompanied by an immune deficiency, which favors the occurrence of numerous opportunistic infections including tuberculosis, it was indicated in the present study that HIV infection was significantly associated with tuberculosis and presented an increased risk of active tuberculosis [36] . This may also be justified by the fact that in Gabon, Koula-Moutou is a town with a high rate of HIV cases [37] . Cough and chest pain were significantly associated with the prevalence of pulmonary tuberculosis in this study. This result is similar to that found in a study in the Central African Republic, where the most frequent functional signs were chronic cough (71.81%) and chest pain in 127 cases (57.73%) [38] . This could be explained by the fact that our patients came to the clinic very late. As in a study which reported that TB patients present with symptoms, in particular weight loss, fever, dyspnoea and night sweats [38] , night sweats were significantly associated with pulmonary TB in the present study. This may be justified by the fact that TB symptoms can vary from patient to patient, but a few key signs are generally associated with the disease. Sweating profusely at night is one of them, and often indicates that the body’s infection levels are potentially very high.
5. Study Limitations
Stemming solely from a hospital-based analysis of patients infected with pulmonary tuberculosis, the results of the present study may not be generalizable. Having used only Ziehl-Neelsen smear microscopy and the Xpert MTB/RIF test, the use of a chest X-ray, as a complementary diagnostic tool for all HIV-infected individuals, would have avoided missing certain cases of tuberculosis. In addition, the culture technique and CD4 assay were not performed due to limited laboratory facilities.
6. Conclusion
The present study has established that the prevalence of tuberculosis is predominant among males in the 15 to 49 age group in Franceville and surrounding departments. Many risk factors and clinical signs were significantly associated with tuberculosis in the present study. This suggests possible gaps in symptom-based case-finding that may need to be addressed in similar settings. Therefore, TB screening of HIV-positive patients, public awareness and community mobilization should be encouraged. In addition, large-scale studies on trends in TB/HIV co-infection and associated factors should also be implemented throughout the country.
Acknowledgements
The authors are very grateful to the Department of Biology, the Laboratory of Molecular and Cellular Biology, and the Biochemistry Research Laboratory of the University of Science and Technology of Masuku Franceville, who made their laboratory facilities available during the research.
Authors’ Contributions
All authors contributed equally to this work. They have read and appreciated the final version of this manuscript.
Informed Consent
Informed verbal consent was obtained from each study participant.