Pulmonary function assessment of a community exposed to air pollution in an industrial area in Ouagadougou, Burkina Faso

Abstract

Industrial air pollution has been linked to respiratory diseases due to pulmonary function impairment, which are poorly documented in low-and middle-income countries. This study aimed to assess the pulmonary function of a community exposed to air pollution in an Industrial Area (IA) in Ouagadougou, Burkina Faso. Thus, a cross-sectional study was conducted among 146 residents of the IA and 146 residents of Nioko 2, constituting respectively, the target and control populations, from November to December 2023. A medical team collected data on respiratory symptoms experienced by the study population through a questionnaire followed by a clinical examination. Pulmonary function tests (PFTs) were then performed by spirometry according to ATS/ERS guidelines. The results showed that the study population suffered from both upper airway (87% IA, 70.50% Nioko 2) and lower airway (89% IA, 64.40% Nioko 2) symptoms whose prevalence was higher among the residents of the IA. The PFTs revealed a decrease in the predicted average percentage of FEV1% (p = 0.002) and FET % (p = 0.002) before bronchodilation among the residents of the IA compared to Nioko 2. These spirometric parameters still remained decreased among residents of the IA compared with Nioko 2 even after bronchodilation, but with only a significant difference for FET % (p = 0.011). Furthermore, the pulmonary impairment diagnosed among the residents of the IA before bronchodilation was restrictive defects (p = 0.035), with a strong representation of females (37.50% IA, 11.11% Nioko 2). Univariate logistic regression revealed that residence in the IA was significantly associated with restrictive ventilatory defects (OR, 1.697; p = 0.035). The restrictive defects diagnosed were the main pulmonary function impairment observed in the residents of the IA notably in women. The findings in this study highlight the impact of air pollution on the pulmonary function of the residents of the IA specifically on the lower respiratory tract.

Graphical Abstract

1. Introduction

Air pollution is a major contributor to global morbidity and mortality, with nearly 6.7 million deaths worldwide in 2019 [1]. Globally, it is the first environmental risk to health according to the World Health Organization (WHO) and is considered a "silent killer", causing more deaths than other "big killers" such as HIV/AIDS, tuberculosis and malaria [2,3]. The WHO continues to declare it as a public health emergency and a leading cause of premature death for up to 7 million people annually from stroke, heart disease, lung cancer, and respiratory infections [4]. Thus, air pollution remains an environmental concern worldwide with adverse toxicological effects on human health increasingly well-documented [[5], [6], [7], [8], [9], [10]]. The burden of disease caused by air pollution is increasing, and Particulate Matter (PM) has been estimated to be responsible for 9 million premature deaths, representing one in six deaths per year [11].

Among the pathologies contributing to these deaths are obstructive pulmonary diseases, in particular Chronic Obstructive Pulmonary Disease (COPD) and asthma, which are a significant source of morbidity and mortality [12,13]. Indeed, links have been established between exposure to PM and respiratory health problems such as asthma, COPD, and declining lung function [14,15]. Studies have reported short-term (cough, wheezing, phlegm, and shortness of breath) and long-term (reduced lung function) respiratory effects from PM exposure [[16], [17], [18], [19]]. According to the WHO, PM was responsible for 4.2 million premature deaths worldwide in 2019, of which 68% were attributable to ischemic heart disease and stroke, 14% to chronic obstructive pulmonary disease, 14% to acute lower respiratory tract infections, and 4% of deaths due to lung cancer [20].

Due to their relatively small size, PM, especially PM_2.5, can penetrate deep into the lungs and significantly affect human respiratory health [21,22]. Therefore, pulmonary function tests are often used as indicators to assess the effects of air pollution on the lungs. Previous studies have used spirometry to measure forced lung function parameters, mainly Forced Expiratory Volume in the first second (FEV1) and Forced Vital Capacity (FVC), to obtain diagnostic information on airway conduction function and the severity of the obstruction [[23], [24], [25]]. A meta-analysis of the European Study of Cohorts for Air Pollution Effects (ESCAPE) study in adults showed that increased PM₁₀ levels were associated with decreased FEV1 and FVC [26]. In a community that is located near cement plants, reductions in FEV1 and FVC were observed after assessing [27]. A study by Kim et al. (2015) found reductions in these parameters and diagnosed obstructive and restrictive ventilatory disorders in two populations exposed to emissions from a cement plant. Living near steel industries was also associated with reduced lung function as ultrafine particle levels increased [28]. These investigations show that atmospheric pollutants emitted by industries, are likely to affect the respiratory health of surrounding populations by inducing obstructive pulmonary pathologies, particularly chronic obstructive pulmonary disease and asthma [29], one of the most important causes of morbidity and mortality in industrialized countries [1].

In Africa, the desire to establish industry as a key economic sector has led Burkina Faso to revitalize the industrial sector by creating industrial areas, including the Kossodo industrial area of Ouagadougou [30]. This industrial area (IA) concentrates various potentially polluting industrial units (Cement plants, foundries, thermal power plants, metallurgical plants, etc.) that are located alongside residential areas, posing a high risk to the health of local communities [31]. Studies conducted in industrial areas have, however, reported air pollution levels above standards both within factories and in nearby communities [19,32]. This indicates that populations that are located near industrial areas are strongly exposed to atmospheric pollutants from industries and encourages further investigations in order to detect the pulmonary disorders that this particulate air pollution could lead. This is especially relevant because previous studies have found reduced lung function and pulmonary ventilation problems in communities living near industrial areas [23,27,[33], [34], [35]]. Although some studies have examined the health effects of industrial pollution on nearby communities [36], they remain insufficient hindering the adoption of strategies for reducing industrial air emissions and preventing health risks. In Burkina Faso, in particular, very few studies have documented the impact of air pollution on the respiratory health of populations living near an IA. The aim of this study is therefore to explore the pulmonary function by spirometry of the residents of the Kossodo industrial area in Ouagadougou, Burkina Faso in order to contribute on addressing a gap in respiratory disorders due to industrial air pollution exposure in Low-and Middle-Income Countries (LMICs).

2. Material and methods

2.1. Study frame

This study was conducted in the Kossodo industrial area, one of the country's major industrial hubs. (Fig. 1). Long before the establishment of industries, the IA was mainly composed of non-formal habitats of Kossodo and Polesgo [37]. At the start of industrial activities, the industries were surrounded by the informal settlements of Kossodo and Polesgo, but at a reasonable distance. With the city's demographic and spatial growth, the Kossodo industrial area, with its multiple industries of varied profiles, now sits at the center of residential areas. In fact, it is limited to the north by the informal houses of Polesgo; to the north-east, the university city of Kossodo; to the south-west, the informal houses of Kossodo and Somgandé and to the south-east, national road no.3 [38]. For comparison, a control area whose population had the same characteristics as those of the IA was selected. This is the Nioko 2 area, located to the southeast at three km from the IA and separated from it by National Road No. 3. To rule out the influence of pollution linked to road traffic, the parts located more than 500 m from the National Road No. 3 were considered in each area to conduct the study.

Fig. 1.

Mapping of study areas.

2.2. Study design and period

This study is part of an overall study aiming to explore the health effects of air pollution in an urban area in Ouagadougou, Burkina Faso. This study, particularly, assesses the pulmonary function of a community living in the Kossodo industrial area. It was a cross-sectional study conducted from November to December 2023, in which populations from the Kossodo industrial area and Nioko 2 took part.

2.3. Study population and participant sampling

The populations of Kossodo and Polesgo in the IA constituted our target population. These increasingly cornered populations, given the uncontrolled expansion of industries towards homes and vice versa, find themselves trapped in clouds of dust and black smoke during the operation of certain industries. With the extension of the IA in its north-eastern part, some of the inhabitants and their homes are sandwiched between the old and the new IA, increasing the exposure. A control population was also retained for data comparison. This is the population of Nioko 2. The sampling included both sexes of residents in the two study areas. Individuals aged over 18 years, residing in one of the two study areas, who were able to answer the study questions and undergo the respiratory functional exploration were included in the study. Workers of a hydrocarbon station, welders, carpenters and persons presenting a contraindication to spirometry including a recent or current pneumothorax, recent abdominal or ocular disease, recent abdominal or ocular surgery, recent heart attack, unstable cardiovascular state (aneurysm, recent heart attack, etc.) were excluded from the study.

2.4. Sample size determination

The population size for the respiratory function test by spirometry was determined based on the mean ± standard deviation of FEV1. Considering the FEV1 of 91.56 ± 22.45 and 96.54 ± 14.53 obtained respectively in exposed and control populations in a study by Nkhama and colleagues [39], the population size was determined as follows by the formula of comparing 2 means [40]:

$$N=\frac{r+1}{r}\frac{S{D}^2{\left(Z_{\beta }+Z_{\frac{\alpha }{2}}\right)}^2}{d^2}$$

SD: Difference in standard deviations of FEV1 between exposed and control populations

d: Difference in FEV1 means of the exposed and control populations

r = Ratio between the number of cases and the number of controls (r = 1)

Zβ = Standard normal value for a power of 90% or 1.28

Zα/2 = Critical value on the reduced centered normal distribution for a risk of error α (α= 5%, Zα/2 = 1.96)

For an adjusted 20% non-response rate, a minimum of 66 participants were retained for the spirometry test. The same calculated sample size was considered for each area.

Households were randomly selected in each study area. In each household, one individual was randomly selected for respiratory functional exploration. In case of refusal, the next household was selected to continue the study.

2.5. Data collection

2.5.1. Questionnaire applying

Respiratory health effects were explored using the adapted, standardized Occupational Safety and Health Administration (OSHA) Respirator Medical Evaluation Questionnaire [41] according to the method of Abdel-Rasoul and collaborators [42]. The questionnaire was validated with the local population before started the survey. It was applied face-to-face to each participant by the medical team in French and local languages according to the participant's convenience at their respective homes. In each of the two study areas, the first household surveyed was randomly selected for the interview. In each household, the direct participant in the study was also randomly selected for the interview. At the end of the interview or in case of refusal, the next household was immediately selected to continue the survey.

2.5.2. Clinical constants measurement

Clinical constants such as systolic and diastolic arterial pressure, heart rate, and pulse oximetry were recorded preferably in the morning. After ensuring that the participant had not smoked, consumed tobacco, energy drinks (coffee, tea, etc.), or engaged in physical activity, the medical team measured arterial pressure. Systolic and diastolic arterial pressures and heart rate were recorded using an electronic device (OMRON Hem8402). These clinical parameters were measured in the right arm, supported at heart height, in a seated position. The recording was made after the subject had observed a minimum rest period of 10 min. Hypertension was defined as systolic blood pressure ≥ 140 mmHg and/or diastolic pressure ≥ 90 mmHg, a history of physician-diagnosed hypertension, or use of antihypertensive medication at the time of the survey [43]. Tachycardia was defined as a heart rate ≥ 100 beats/min. Oximetry was performed using an oximeter placed on the fingertip for one minute. Hypoxemia was considered for an oxygen saturation level < 95% and desaturation for an oxygen saturation level < 90%. A height gauge and a scale were also used to measure height and weight, respectively. These constants were used to determine the Body Mass Index (BMI). BMIs were then classified according to WHO recommendations [44]. Following the constants, the clinical examination, including auscultation, allowed the exploration of respiratory and cardiac function. The respiratory rate was also determined at the end of the auscultation.

2.5.3. Pulmonary function tests

Pulmonary function tests (PFTs) were carried out in accordance with the recommendations of the "American Thoracic Society (ATS)" and the "European Thoracic Society (ERS)" [45] using a Spirobank II-BLE spirometer [46]. After ensuring that the participant did not have any contraindications to spirometry, sociodemographic characteristics including age and gender as well as history of smoking, recent illness and medication were recorded for the purposes of the test. To carry out the test, the participant was asked, in a seated position, with their back straight and their neck aligned with their back, to hold the spirometer with the mouthpiece firmly in their mouth, behind the dental arch, with their lips well closed. A nose clip was used to prevent any nasal leakage. During the test maneuver, each participant was given clear instructions and encouraged to breathe properly following the instructions in each of the following three phases: In phase 1 of maximum inspiration, the participant is encouraged to take a maximum and deepest possible inspiration. In phase 2 of the breathing phase, the participant is asked to perform a forced exhalation as quickly, as hard, and as long as possible. In phase 3 of continuation, the participant is asked to hold the exhalation for at least 6 s [47]. The parameters assessed were mainly FVC, FEV1, FEV1/FVC ratio, Peak Expiratory Flow (PEF) and Forced Expiratory Time (FET). These parameters were determined before and after bronchodilation. Bronchodilation consisted of pulmonary administration of three inhaled puffs of salbutamol at a dose of 100 µg to each participant. The post bronchodiladation phase was performed 15 min after pre bronchodilation phase. Three minimum acceptable flow-volume measurement maneuvers were performed taking into account the acceptability and repeatability criteria. The PEF was considered acceptable when the peak was well-defined in the flow-volume curve and the exhalation time was higher than 4 s in the volume-time curve. The three test curves satisfying acceptability and repeatability criteria, were used for analysis. Predicted value’s percentage for each participant was then obtain from measured value and theoretical or predicted value as follows:

$$\%\mathit{Theor}=\frac{\mathit{PRE}}{\mathit{Pred}}\times 100$$

%Theor: Predicted value’s percentage;

PRE: Measured value;

Pred: Theoretical or predicted value.

Participants with a FEV1/FVC ratio below the lower predicted limit after bronchodilation were classified as having an obstructive impairment according to the ATS/ERS guidelines. For these cases, ATS/ERS defines them. Therefore, spirometric results were interpreted according to ATS/ERS guidelines as follows:

- Normal spirometry when FEV1 ≥ 80%, FVC ≥ 80%, and FEV1/FVC > 70% of predicted values

- Obstructive defect for FEV1/FVC ≤ 70% and FVC > 80% of predicted values

- Restrictive defect for FEV1/FVC > 70% and FVC < 80% of predicted values

- Mixed defect for FEV1/FVC < 70% and FVC < 80% of predicted values.

The degree of severity was classified according to FEV1, as follows:

- Mild when FEV1 ≥ 70%

- Moderate when 70% > FEV1 ≥ 50%

- Severe when FEV1 < 50% [45].

2.6. Statistical analysis

For statistical analysis of data, SPSS software version 20 (SPSS Inc., Chicago, IL, USA) was used. Qualitative variables were presented as frequency and proportion while continuous variables were presented as means ± standard deviation. The independent variables including socio-demographic characteristics, lifestyle, smoking, area of residence, etc. of the residents of the IA were compared with those of Nioko 2. Thus, the proportions of qualitative variables like gender, smoking, body mass index, etc. were compared using the chi-square/fischer exact test. Student's t-test was used to compare continuous variables, including age and the mean predicted values of spirometric parameters, etc. The dependent variables including pulmonary disorders of the two study areas, were also compared using the chi-square/fischer exact test. The association between respiratory symptoms and restrictive ventilatory disorders was assessed using the 2 × 2 ×N Chi-square test. Next, the factors associated with restrictive ventilatory disorders and the degree of association were determined by performing univariate and multivariate binary logistic regression by assessing the Odds ratio and the p-value of the association. Sociodemographic characteristics, area of residence, smoking, BMI, were the independent variables while pulmonary disorders were the dependent variables. The difference was considered statistically significant for a p-value less than 5% for a confidence interval maintained at 95%.

2.7. Ethics

This study was conducted in accordance with the Declaration of Helsinki guidelines and was approved by the Ethics Committee for Health Research of Burkina Faso by deliberation n˚ 2023–10–255 of November 6th, 2023. Free and informed consent was obtained from each study participant before enrollment in the study. For reasons of confidentiality, the identity of the individuals surveyed, as well as the results of the spirometric tests were kept anonymous. The medical examination data was also kept confidential.

3. Results

3.1. Characteristics of the study population

Regarding the characteristics of the two study populations, it emerged that diastolic arterial pressure (p = 0.096), heart rate (p = 0.785), pulse oxygen saturation (p = 0.249) and respiratory rate (p = 0.798) did not differ significantly according to the area of residence. The same applies to the average age of the study populations (p = 0.444) and the average number of years of residence (p = 0.1233) in the two study areas, which showed no statistically significant difference. In contrast, BMI (p = 0.002) and systolic blood pressure (p = 0.032) differed significantly depending on the area of residence. The results also showed that gender (p = 0.404) and smoking (p = 0.259) did not differ significantly by area of residence. In contrast, BMI distribution differed significantly by area of residence (p = 0.026) (Table 1).

Table 1.

Characteristics of the two study populations.

Variables	Industrial area (n = 146)		Nioko 2 (n = 146)		Statistical analysis
	Mean ± Dev Std		Mean ± Dev Std
Body mass index (kg/m2)	22.45 ± 3.40		24.06 ± 4.95		0.002a**
Systolic blood pressure (mmHg)	125 ± 12.97		128.60 ± 15.41		0.032a*
Diastolic blood pressure (mmHg)	78.62 ± 9.68		80.55 ± 10.15		0.096a
Heart rate (bat/mn)	82 ± 12.16		81.60 ± 12.64		0.785a
Pulse oxygen saturation (%)	98.01 ± 0.95		98.13 ± 0.87		0.249a
Respiratory rate	14.93 ± 5.40		15.08 ± 4.09		0.798a
Age (Year)	31.15 ± 10.15		32.15 ± 12.04		0.444a
Year of residence	9.10 ± 10.56		10.77 ± 7.57		0.123a
	n	(%)	n	(%)
Gender					0.404b
Male	90	61.60	83	56.80
Female	56	38.40	63	43.20
Smoking					0.259c
No	109	74.70	114	78.10
Currently smoker	27	18.50	28	19.20
Former smoker	10	6.80	4	2.70
Body Mass Index					0.026b*
Underweight	12	8.20	12	8.20
Normal	103	70.50	81	55.50
Overweight	23	15.80	36	24.70
Obese	8	5.50	17	11.60

n: Frequency, ^aStudent test, ^bChi-square test, ^cFisher exact, *p < 0.05, **p < 0.01

3.2. Self-report respiratory symptoms and auscultation results

Several symptoms have been reported by residents of the IA that impact their respiratory systems. These symptoms can be grouped into upper and lower airway symptoms, and their prevalence was significantly higher in the IA than in Nioko 2. For health effects related to the upper airways which differed significantly according to the area of residence (p = 0.001), the most felt symptoms were rhinitis with respective proportions of 80.10% in the IA versus 53.40% in Nioko 2. For health effects related to the lower airways for which a statistically significant difference was observed according to the area of residence (p = 0.0001), the symptoms that predominated were dry cough which proportions of 63% and 19.2% reported respectively in the IA and in Nioko 2. It was seconded by wet cough with 58.20% of complaints in the IA versus 28.80% in Nioko 2 (Table 2).

Table 2.

Respiratory health effects experienced by the study populations (n = 292).

Reported health effects	Industrial area (n = 146)		Nioko 2 (n = 146)		Statistical analysis
	n	(%)	n	(%)	(Chi-square/Fisher exact)
Respiratory system
Upper airway	127	87	103	70.50	0.001**
Throat irritation	62	42.50	30	20.50	0.0001
Nasal cavities irritation	77	52.70	39	26.70	0.0001
Nasal cavities obstruction	99	67.80	61	41.80	0.0001
Rhinitis	117	80.1	78	53.40	0.0001
Lower airway	130	89	94	64.40	< 0.001***
Hacking cough	30	20.50	11	7.50	0.001
Wet cough	85	58.20	42	28.80	0.0001
Dry cough	92	63	28	19.2	0.0001
Dyspnea	38	26	14	9.60	0.0001
Shortness of breath on exertion	80	54.80	29	19.90	0.0001
Wheezing	14	9.60	6	4.10	0.064
Morning expectoration	70	47.90	23	15.80	0.0001
Hemoptysis	7	4.80	2	1.40	0.173
Chest pain	69	47.30	40	27.40	0.0001

n: Frequency, **p < 0.01, ***p < 0.001

4. Pulmonary function tests

4.1. Average percentage of predicted values before and after bronchodilation

Table 3 shows the mean predicted values of spirometric parameters in percent in the two study populations before and after bronchodilator use. A significant decrease in FEV1% was observed among residents of the IA (84.34 ± 16.93) compared to residents of Nioko 2 (93.60 ± 32.26) (p = 0.002). Also, FET% was significantly reduced in the IA (53.15 ± 30.18) compared to Nioko 2 (70.36 ± 28.08) (p = 0.002). In contrast, FVC% (p = 0.153), FEV1/FVC (p = 0.166) and PEF% (p = 0.726) did not differ significantly depending on the area of residence.

Table 3.

Average percentage of predicted values of study population before and after bronchodilation.

Spirometric parameters	Industrial area (n = 146)	Nioko 2 (n = 146)	t-test
	Mean ± SD	Mean ± SD	p-value
Before bronchodilation
FVC %	88.07 ± 26.45	91.89 ± 19.46	0.153
FEV1%	84.34 ± 16.93	93.60 ± 32.26	0.002**
FEV1/FVC	99.24 ± 15.02	97.02 ± 12.51	0.166
PEF %	77.59 ± 36.34	76.10 ± 29.95	0.726
FET%	53.15 ± 30.18	70.36 ± 28.08	0.002**
After bronchodilation
FVC %	86,84 ± 17,43	89,93 ± 21,39	0,177
FEV1%	87,14 ± 17,99	89,36 ± 29	0,437
FEV1/FVC	101,27 ± 10,13	98,85 ± 10	0,041*
PEF %	80,73 ± 29,42	76,78 ± 30,07	0,257
FET%	62,08 ± 34,03	71,38 ± 27,31	0,011*

**p < 0.01, *p < 0.05

Despite the use of bronchodilator, the proportions of FVC% (86.84 ± 17.43), FEV1% (87.14 ± 17.99) of residents of the IA remained low compared to Nioko 2, where the proportions were 89.93 ± 21.39 and 89.36 ± 29, respectively, but without any significant difference (p˃0.05). The FET % also remained lower in the IA (62.08 ± 34.03) compared to Nioko 2 (71.38 ± 27.31) but differed significantly according to the residential area (p = 0.011). The bronchodilator led to an increase in FEV1/FVC in the IA (101.27 ± 10.13) while this parameter did not experience any notable variation in Nioko 2 (98.85 ± 10). However, a statistically significant difference was observed (p = 0.041).

4.2. Interquartile range of FVC, FEV1, PEF, FET, FEV1/FVC before and after bronchodilation

After administration of the bronchodilator to residents of the IA, a variation in the spirometric parameters evaluated was observed but without any statistically significant difference (p˃0.05). At the 75th quartile, FVC was 93.97 before bronchodilation and 96.04 after bronchodilation (p = 0.489). FEV1 was 96.12 before bronchodilation and 98.08 after bronchodilation (p = 0.522). PEF was 88.4 and 98.14 before bronchodilation and after bronchodilation, respectively (p = 0.472). Still at the 75th quartile, the FET was 76.29 and 82.71 before bronchodilation and after bronchodilation, respectively (p = 0.512). On the other hand, the 75th quartiles of the FEV1/FVC ratio were 109.03 before bronchodilation and 107.69 after bronchodilation respectively (p = 0.506) (Fig. 2a).

Fig. 2.

Box plot of comparison of FVC%, FEV1%, PEF%, FET and the FEV1/FVC% ratio interquartile before and after bronchodilation among: a) The residents of the IA; b) The residents of Nioko 2. The boxes show the interquartile range (IQR, 25th − 75th percentile) and the horizontal line inside the box represents the median. The circles show the outlier values.

Among Nioko 2 residents, no increase in FVC, FEV1, PEF, and FET was observed after administration of the bronchodilator at the 75th quartile (p > 0.05). However, the FEV1/FVC ratio increased from 103.74 before bronchodilation to 105.24 after bronchodilation, but without any significant difference (p > 0.05) (Fig. 2b).

4.3. Pulmonary function disorders before and after bronchodilation

The results in Table 4 show the pulmonary disorders diagnosed among the residents of the IA and Nioko 2 before and after bronchodilation according to ATS/ERS guidelines. It emerged that a low proportion of men (34.44%) and women (42.86) in the IA had normal spirometry compared to Nioko 2, where the proportions were 54.22% and 68%, respectively. This difference was not statistically significant (p = 0.078). Abnormal spirometries were mainly due to restrictive defects which differed significantly according to the area of residence (p = 0.035). Among those who had a restrictive defect, the mild form was found mainly in women from the IA (37.50%) compared to women from Nioko 2 (11.11%) (p = 0.280). On the other hand, in its moderate form, it was men who were mainly affected (16.67%) in the IA compared to 9.64% in Nioko 2 (p = 0.086) (Table 5).

Table 4.

Pulmonary function disorders before bronchodilation.

ATS/ERS Interpretation		Gender	Industrial area	Nioko 2	X2 /Fisher exact test
Spirometric disorders			n (%)	n (%)	p-value
Normal spirometry		Male	49 (34.44)	45 (54.22)	0.078
		Female	24 (42.86)	43 (68.25)
Restrictive defect Level of severity		Male	30 (33.33)	26 (31.32)	0.035*
		Female	27 (48.21)	13 (20.63)
	Sever	Male	0 (0.00)	0 (0.00)	0.045*
		Female	1 (1.78)	0 (0.00)
	Moderate	Male	15 (16.67)	8 (9.64)	0.086
		Female	5 (8.93)	6 (9.52)
	Mild	Male	15 (16.67)	18 (21.69)	0.280
		Female	21 (37.50)	7 (11.11)
Obstructive defect Level of severity		Male	2 (2.22)	1 (1.20)	-
		Female	1 (1.78)	0 (0.00)
	Sever	Male	0 (0.00)	0 (0.00)	-
		Female	0 (0.00)	0 (0.00)
	Moderate	Male	1 (1.11)	0 (0.00)	-
		Female	0 (0.00)	0 (0.00)
	Mild	Male	1 (1.11)	1 (1.20)	-
		Female	1 (1.78)	0 (0.00)
Mixed defect Level of severity		Male	2 (2.22)	0 (0.00)	-
		Female	0 (0.00)	0 (0.00)
	Sever	Male	1 (1.11)	0 (0.00)	-
		Female	0 (0.00)	0 (0.00)
	Moderate	Male	0 (0.00)	0 (0.00)	-
		Female	0 (0.00)	0 (0.00)
	Mild	Male	1 (1.11)	0 (0.00)	-
		Female	0 (0.00)	0 (0.00)

Industrial area: 146 (Male: 90 and Female: 56), Nioko 2: 146 (Male: 83 and Female: 63), *p < 0.05

Table 5.

Pulmonary function disorders after bronchodilation.

ATS/ERS Interpretation		Gender	Industrial area	Nioko 2	X2 /Fisher exact test
Spirometric disorders			n (%)	n (%)	p-value
Normal spirometry		Male	57 (64.77)	42 (52.50)	0.287
		Female	37 (63.79)	37 (56.06)
Restrictive defect Level of severity		Male	21 (23.86)	27 (33.75)	0.304
		Female	20 (34.48)	20 (30.30)
	Sever	Male	2 (2.27)	1 (1.25)	0.498
		Female	2 (3.45)	1 (1.51)
	Moderate	Male	10 (11.36)	10 (12.5)	0.159
		Female	4 (6.90)	9 (13.63)
	Mild	Male	9 (10.23)	16 (20)	-
		Female	14 (24.14)	10 (15.15)
Obstructive defect Level of severity		Male	0 (0.00)	0 (0.00)	-
		Female	0 (0.00)	0 (0.00)
	Sever	Male	0 (0.00)	0 (0.00)	-
		Female	0 (0.00)	0 (0.00)
	Moderate	Male	0 (0.00)	0 (0.00)	-
		Female	0 (0.00)	0 (0.00)
	Mild	Male	0 (0.00)	0 (0.00)	-
		Female	0 (0.00)	0 (0.00)
Mixed defect Level of severity		Male	1 (1.14)	0 (0.00)	-
		Female	0 (0.00)	0 (0.00)
	Sever	Male	1 (1.14)	0 (0.00)	-
		Female	0 (0.00)	0 (0.00)
	Moderate	Male	0 (0.00)	0 (0.00)	-
		Female	0 (0.00)	0 (0.00)
	Mild	Male	0 (0.00)	0 (0.00)	-
		Female	0 (0.00)	0 (0.00)

After bronchodilation (Table 5), we observed reversibility of certain ventilatory defects in both study populations. Indeed, after the administration of the bronchorelaxant, normality of spirometry was observed in a greater number of participants of both genders in the IA compared to Nioko 2 (p = 0.287). Normal spirometry was observed in 64.77% of men versus 63.79% of women in the IA compared to 52.50% of men versus 56.06% of women in Nioko 2. Like before bronchodilation, respondents who presented abnormal spirometry suffered mainly from a restrictive defect but with relatively reduced proportions in both sexes in the IA. Except for women, who had a consistently high proportion in the IA (34.48%) compared to those in Nioko 2 (30.30%), a large majority of men in Nioko 2 (33.75%) presented a restrictive defect compared to those in the IA (23.86%) (p = 0.304). The proportions were also lower for those with moderate ventilatory defect in men (11.36%) and women (6.90%) in the IA compared to men (12.50%) and women (13.63%) in Nioko 2 (p = 0.159). Also, in the mild form of restrictive defect, a decrease in proportions was observed in the IA in both men (10.23%) and women (24.14%) compared to Nioko 2 where the proportions were 20% and 15.15%, respectively.

Industrial area: 146 (Male: 88 and Female: 58), Nioko 2: 146 (Male: 80 and Female: 66)

4.4. Relationship between respiratory symptoms and restrictive ventilatory defects

The results in Table 6 show the relationship between respiratory symptoms and restrictive ventilatory defects. It emerges that symptoms reported by the population of the IA such as wet cough (p = 0.009) and shortness of breath on exertion (p = 0.018) were significantly associated with the restrictive ventilatory defects.

Table 6.

Relationship between respiratory symptoms and restrictive ventilatory defects of the study population.

Symptoms	Frequency	Restrictive ventilatory defects		Statistical analysisa
		Before bronchodilation	After bronchodilation
Industrial area
Respiratory system	n	n	n	p-value
Throat irritation	62	22	20	0.849
Nasal cavities irritation	77	27	25	0.864
Nasal cavities obstruction	99	36	34	0.881
Rhinitis	117	47	35	0.131
Hacking cough	30	13	10	0.595
Wet cough	85	37	20	0.009**
Dry cough	92	35	26	0.206
Dyspnea	38	13	10	0.224
Shortness of breath on exertion	80	34	19	0.018*
Wheezing	14	6	5	0.698
Morning expectoration	70	31	21	0.11
Hemoptysis	7	5	2	0.285
Chest pain	69	23	18	0.456
Nioko 2
Throat irritation	30	3	1	0.604
Nasal cavities irritation	39	10	9	0.791
Nasal cavities obstruction	61	23	21	0.850
Rhinitis	78	23	22	0.859
Hacking cough	11	3	2	0.695
Wet cough	42	11	9	0.797
Dry cough	28	8	7	0.762
Dyspnea	14	4	3	0.662
Shortness of breath on exertion	29	7	5	0.745
Wheezing	6	2	1	0.504
Morning expectoration	23	9	7	0.756
Hemoptysis	2	0	0	0.512
Chest pain	40	9	8	0.784

^a2 × 2 ×N Chi-square, n: Frequency, *p < 0.05, **p < 0.001

4.5. Factors associated with pulmonary function impairment

Logistic regression analysis was performed to assess the factors associated with restrictive defects. In univariate analysis, it emerged that only residence in the IA was significantly associated with restrictive defects (p = 0.035).

In the multivariate analysis, variables that showed a significance of less than 20% in univariate analysis were included in the model. Sociodemographic characteristics including age and gender were systematically included. The results indicate that no variable was associated with restrictive defect (p < 0.05) (Table 7).

Table 7.

Factors associated with restrictive ventilatory defects in univariate and multivariate logistic regression analysis.

Variables	Restrictive ventilatory defects
	Univariate binary logistic regression		Multivariate binary logistic regression
	p-value	OR (CI 95%)	p-value	OR* (CI 95%)
Area of residence
Nioko 2	1		1
Industrial area	0.035*	1.697 (1.037 – 2.778)	0.063	0.615 (0.368 – 1.026)
Age	0.064	0.978 (0.965 – 1.001)	0.085	0.976 (9.949 – 1.003)
Year of residence	0.284	0.985 (0.957 – 1.013)	0.777	0.996 (0.966 – 1.027)
Gender
Female	1		1
Male	0.906	0.970 (0.591 – 1.592)	0.812	1.076 (0.591 – 1.959)
Smoking
No	1		1
Current smoker	0.785	0.919 (0.486 – 1.737)	0.814	0.916 (0.439 – 1.910)
Former smoker	0.193	2.055 (0.695 – 6.077)	0.120	2.564 (0.781 – 8.415)
Body Mass Index
Normal	1		1
Underweight	0.113	2.000 (0.849 – 4.713)	0.206	0.443 (0.125 – 1.564)
Overweight	0.581	0.837 (0.446 – 1.572)	0.900	1.071 (0.368 – 3.114)
Obese	0.414	0.667 (0.252 – 1.765)	-	-

OR: Odds ratio, OR*: Adjusted Odds Ratio, CI: Confidence Interval, *p < 0.05

5. Discussion

This study aimed to assess the health impact of atmospheric pollution in the Kossodo industrial area by assessing the pulmonary function of a community living within the perimeter of the IA.

The study showed that in terms of socio-demographic characteristics, the populations included in this study were essentially young. Since the integration of the neighborhood housing the IA within the perimeter of the capital, Ouagadougou, it has become a center of attraction for young people looking for employment opportunities in factories, to trade but also for housing opportunities. A similar study conducted by Nkhama et al. (2017) also included predominantly young populations. A massive participation of men was observed in this study to the detriment of women in both study areas. In a large number of households visited, women refrained from answering the study questions, unlike men who were enthusiastic about participating in the study given the topic covered.

Pulmonary function exploration via the questionnaire undertaken by the medical team and clinical examination revealed health effects impacting the upper and lower airways of the respiratory system with higher prevalence in the IA compared to Nioko 2 (p < 0.05). The potentially polluting industries, particularly cement plants, which predominate in the IA, and metallurgy plants [31], are known to emit a mixture of pollutants including PM [48,49], gases (nitrogen oxides, sulfur oxides, carbon monoxide) [50], heavy metals [[51], [52], [53], [54], [55]] and persistent organic pollutants (polychlorinated dibenzo-p-dioxins, dibenzofurans, polychlorinated biphenyls) [56,57]. A recent study measuring the level of particulate air pollution in this same IA revealed average daily concentrations of PM_2.5, PM₁₀ and TSP that exceeded national and WHO daily standards [31].The proximity of populations to industries continuously exposes them to these pollutants, which are likely to cause respiratory health effects [[58], [59], [60]]. The predominance of rhinitis in the upper airways shows the induction of an inflammatory mechanism due to dusty cement particles by stimulation of neutrophils and lymphocytes which release chemical mediators [[61], [62], [63]]. Particle-clearance mechanisms such as coughing and expectoration reported in this study confirm particle exposure among IA residents, where prevalences were higher [64]. These same respiratory health effects have also been reported by Eom et al. but with low prevalence than our findings for cough (6.6), Dyspnea (7.9), Wheezing (3.1), Rhinitis (20.5), sputum (8.2). These differences in prevalence between the two studies could be explained by the difference in sample size, with more than 26,000 participants included in the study by [65].

A decrease in the mean predicted values of FEV1% and FET% was observed in the Kossodo industrial area with a statistically significant difference in bivariate analysis. The respiratory system is the preferred target for particulate pollutants with the respiratory tract as the major route of entry [66]. The extent of health damage depends on the area of deposition, duration of exposure, particle size and composition, individual susceptibility, and biological responses initiated by particles in various regions of the respiratory tract [[67], [68], [69]]. PM₁ and PM_2.5 in particular, can cross upper airway barriers and deposit in the bronchial and alveolar regions [[70], [71], [72]]. Extrathoracic and intrathoracic particle levels above 4.5 mg/m³ and 2.2 mg/m³ respectively have been shown to lead to reduced lung function [73]. Prolonged exposure to heavy metals such as mercury, arsenic, zinc, copper, nickel, cadmium and chromium has been linked to impaired lung function and reduced lung function [74]. Accumulation of chromium and silica in the alveolar region can induce irritation and inflammation of the lungs with hypersecretion of mucus [14,75]. Due to inflammation, bronchial smooth muscle and vascular endothelium are affected due to the release of histamine and serotonin from degranulated amines [76]. The reduced size of the dusty cement particles (<10 µm) capable of adsorbing chromium and silicas allows them to be deposited in the tracheobronchial tree and in the alveoli to induce these same dysfunctions and result in a reduction in pulmonary function [77]. In a study by [78], reduced lung function was linked to dysplasia, a squamous metaplasia of the lung epithelium. In studies by [39] and [34], FEV1 was also decreased in populations residing near an industrial complex.

Restrictive defects were the pulmonary disorders most experienced, particulary in pre-bronchodilation, by residents of the IA who had higher prevalences than those of Nioko 2 (p = 0.035). Indeed, 33.33% of men and 48.21% of women had a predicted value of FVC < 80% and a predicted value of the FEV1/FVC ratio ˃ 70%. Among those who presented a restrictive defect, 16.67% of men and 37.50% of women had predicted values of FEV1 ≥ 70%. These proportions were the highest among those of other forms of restrictive defects indicating that the mild form of restrictive defects predominated over the other forms. The proportions of women were higher than those of men for restrictive defects in general and in its mild form in the IA. This could be explained by the constant assiduity of women in households and whose trade turns out to be the main activity carried out on site in the IA. They are therefore continually exposed to industrial emissions unlike men who are more constrained to mobility. A proportion of 4% of residents of the IA had a predicted value of FVC ˃ 80% and a predicted value of the FEV1/FVC ratio ≤ 70%. This indicates that obstructive defects were diagnosed in a small proportion of 4% in the IA compared to 1% in Nioko 2. The results interestingly showed a significant association between restrictive ventilatory defects and symptoms such as dry cough and shortness of breath on exertion among residents of the IA. Cough, sputum production, wheezing, and dyspnea are known to be characteristic symptoms of COPD, including both obstructive and restrictive ventilatory defects [79]. The association, particularly between dry cough and restrictive ventilatory defects, therefore underscores the presence of these defects among residents of the IA.

After administration of the bronchodilator to the residents of the IA, FEV1%, PEF%, and FET% increased to the 75th interquartile range after bronchodilation. Although no statistical difference was observed, these results indicate that the bronchodilator allowed a reversibility of the ventilatory defects that have been observed well before its use. This can be seen in the results of Table 5 & 6 where the proportions of normal spirometries which were 34.44% and 42.86% in men and women respectively before bronchodilation increased to 64.77% and 63.79% in men and women respectively after bronchodilation. Also, the proportions of restrictive defects decreased after taking the bronchorelaxant. From 33.33% and 48.21% obtained in men and women respectively before bronchodilation, these proportions decreased to 23.86% and 34.48% in men and women respectively in after bronchodilation.

The study of associated factors in univariate logistic regression analysis showed that residence in the IA was significantly associated with restrictive defects (p = 0.035). This suppose that the pollution induced in the IA has a real impact on the lung health of the population living there. Former smokers were twice as likely to experience restrictive ventilatory defects but with no statistically significant difference (p˃0.05). In a study conducted among workers exposed to cement dust particles, it was found that reduced lung and chest wall compliance was responsible for the restrictive defect diagnosed. Indeed, an alteration in the performance of the intercostal muscles was confirmed by surface electromyography [80]. Also, exposure of macrophages to dust has been shown to cause generation of Reactive Oxygen Species (ROS), induction of apoptotic cells, and decreased glutathione and secretion of interleukin-1β and TNF-α. In other words, pulmonary toxicity is induced by endocytosis, oxidative stress-induced apoptosis and induction of pro-inflammatory cytokines when alveolar macrophages are exposed to dust [81]. In a study quantifying ROS and Nitric Oxide (NO) levels in the serum of residents exposed to cement particulate emissions, an increase in ROS was observed at the expense of NO [19]. The proportions of restrictive defects in this study are higher than those found by [33] who reported 21.6% and 12.4% of restrictive defects respectively in highly exposed and low-exposed populations. In contrast to our findings, restrictive defects were diagnosed predominantly in men in their study. The proportion of obstructive defects found in their study in highly exposed populations (9.7%) did not differ significantly from that of low-exposed populations (8.5%) [33]. However, these proportions remain higher than those obtained in this study which were 4% in the IA versus 1% in Nioko 2.

6. Conclusion

This study provides additional data on the impact of industrial air pollution on health, particularly on populations residing in the perimeter of an IA. The findings reveal health effects related not only to lower airways but also to upper airways, as highlighted by self-reported health effects. The pulmonary function assessment revealed both restrictive and obstructive defects among the residents of the IA. Restrictive ventilatory defects, in particular, were the main pulmonary disorders experienced by residents of the IA. These restrictive ventilatory defects were diagnosed predominantly in women. Both restrictive and obstructive defects undoubtedly explain the high prevalence of self-reported respiratory symptoms in the IA. Given the pulmonary disorders observed, it would be useful to continue this study by performing a pulmonary X-ray to better assess the extent of damage to the pulmonary compartments.

Abbreviations

ATS	:	American Thoracic Society
AIDS	:	Acquired Immune Deficiency Syndrome
BMI	:	Body Mass Index
CI	:	Confidence Interval
COPD	:	Chronic Obstructive Pulmonary Disease
ERS	:	European Thoracic Society
ESCAPE	:	Study of Cohorts for Air Pollution Effects
FET	:	Forced Expiratory Time
FEV1	:	Forced Expiratory Volume in the first second
FVC	:	Forced Vital Capacity
HIV	:	Human Immunodeficiency Virus
IQR	:	Interquartile Range
LMIC	:	Low-and Middle-Income Countries
NO	:	Nitric Oxide
OR	:	Odds Ratios
OSHA	:	Occupational Safety and Health Administration
PEF	:	Peak Expiratory Flow
PFT	:	Pulmonary Function Test
PM	:	Particulate Matter
PM10	:	Particulate Matter with an aerodynamic diameter equal to or less than 10 μm
PM2.5	:	Particulate Matter with an aerodynamic diameter equal to or less than 2.5 μm
ROS	:	Reactive Oxygen Species
SD	:	Standard Deviation
TSP	:	Total Suspended Particle
WHO	:	World Health Organization

CRediT authorship contribution statement

Marin Moïse Zouma: Writing – original draft, Methodology, Investigation, Formal analysis, Data curation, Conceptualization. Geoffroy Gueswindé Ouédraogo: Writing – review & editing, Validation, Supervision, Resources, Methodology, Investigation, Conceptualization. Arnaud Jean Florent Tiendrébéogo: Supervision, Resources, Methodology, Conceptualization. Jean Claude Romaric Pingdwindé Ouédraogo: Validation, Supervision, Methodology, Formal analysis, Conceptualization. Jean Noel Dado Koussé: Visualization, Formal analysis. Aude Nikiema: Resources, Methodology, Conceptualization. Rasmané Semdé: Resources. Sylvain Ilboudo: Writing – review & editing, Visualization, Validation, Resources, Methodology, Funding acquisition, Conceptualization.

Declaration of Competing Interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Data Availability

Data will be made available on request.