Impaired lung function in individuals chronically exposed to biomass combustion

Environmental Research 112 (2012) 111–117

Contents lists available at SciVerse ScienceDirect

Environmental Research journal homepage: www.elsevier.com/locate/envres

Impaired lung function in individuals chronically exposed to biomass combustion$, $$ Luiz Fernando Ferraz da Silva a,n, Silvia Regina Dias Me´dici Saldiva b, Paulo Hila´rio Nascimento Saldiva a, Marisa Dolhnikoff a, Bandeira Cientı´fica Projecta a b

Department of Pathology, University of Sao Paulo School of Medicine, Sao Paulo, Brazil Health Institute, State Health Secretariat, Sao Paulo, Brazil

a r t i c l e i n f o

abstract

Article history: Received 13 October 2010 Received in revised form 12 September 2011 Accepted 25 October 2011 Available online 30 November 2011

Background: The use of biomass for cooking and heating is considered an important factor associated with respiratory diseases. However, few studies evaluate the amount of particulate matter less than 2.5 mm in diameter (PM2.5), symptoms and lung function in the same population. Objectives: To evaluate the respiratory effects of biomass combustion and compare the results with those of individuals from the same community in Brazil using liquefied petroleum gas (Gas). Methods: 1402 individuals in 260 residences were divided into three groups according to exposure (Gas, Indoor-Biomass, Outside-Biomass). Respiratory symptoms were assessed using questionnaires. Reflectance of paper filters was used to assess particulate matter exposure. In 48 residences the amount of PM2.5 was also quantified. Pulmonary function tests were performed in 120 individuals. Results: Reflectance index correlated directly with PM2.5 (r ¼ 0.92) and was used to estimate exposure (ePM2.5). There was a significant increase in ePM2.5 in Indoor-Biomass and Outside-Biomass, compared to Gas. There was a significantly increased odds ratio (OR) for cough, wheezing and dyspnea in adults exposed to Indoor-Biomass (OR ¼ 2.93, 2.33, 2.59, respectively) and Outside-Biomass (OR ¼ 1.78, 1.78, 1.80, respectively) compared to Gas. Pulmonary function tests revealed both NonSmoker-Biomass and Smoker-Gas individuals to have decreased %predicted-forced expiratory volume in the first second (FEV1) and FEV1/forced vital capacity (FVC) as compared to Non-Smoker-Gas . Pulmonary function tests data was inversely correlated with duration and ePM2.5. The prevalence of airway obstruction was 20% in both Non-Smoker-Biomass and Smoker-Gas subjects. Conclusion: Chronic exposure to biomass combustion is associated with increased prevalence of respiratory symptoms, reduced lung function and development of chronic obstructive pulmonary disease. These effects are associated with the duration and magnitude of exposure and are exacerbated by tobacco smoke. & 2011 Elsevier Inc. All rights reserved.

Keywords: Biomass Lung function Chronic obstructive pulmonary disease Respiratory symptoms Particulate matter

1. Introduction More than 3 billion people around the world use biomass as their primary source of domestic energy for cooking and heating (WHO, 2006). Exposure to indoor air pollution is a significant cause of morbidity and mortality in developing countries (World Health Organization, 2002), responsible for approximately 2.6% of global morbidity and 1.5 million deaths annually (WHO, 2006).

$

Funding: This work were founded by FAPESP, LIM-HC-FMUSP, CNPq. This study was approved by the institutional review board for human studies – CAPPesq. Written consent was obtained from all subjects. n Correspondence to: Faculdade de Medicina da Universidade de Sa~ o Paulo, Av. Dr. Arnaldo, 455–11 Andar – Sala 1155 – Sa~ o Paulo – SP 01246-903 – Brasil. Fax: þ 55 11 3061 8521. E-mail addresses: [email protected], [email protected] (L.F. da Silva). $$

0013-9351/$ - see front matter & 2011 Elsevier Inc. All rights reserved. doi:10.1016/j.envres.2011.10.012

The indoor use of biomass fuels can release 50 times more pollutants during cooking compared to gas stoves (Smith, 2000) and leads to high concentrations of air pollutants including carbon monoxide, nitrogen oxide, sulfur oxide, aldehydes, polycyclic aromatic hydrocarbons and inhalable particulate matter of different sizes (Pierson et al., 1989). Small particles less than 2.5 mm in diameter (PM2.5) have been increasingly related to respiratory and cardiovascular morbidity and mortality (Dennekamp et al., 2010; Park et al., 2010). Several studies show that exposed individuals have a higher prevalence of cough, phlegm, coryza, wheezing, shortness of breath and itching eyes (Ellegard, 1996; Ellegard, 1997). Furthermore, exposure to biomass combustion has been implicated as a causal/contributory agent for several diseases in developing countries including acute respiratory infections (Bruce et al., 2000; Ezzati and Kammen, 2001; Mishra and Retherford, 1997;

112

L.F. da Silva et al. / Environmental Research 112 (2012) 111–117

Smith et al., 2000), pulmonary arterial hypertension (Sandoval et al., 1993), COPD exacerbations, lung cancer (for coal smoke), asthma, nasopharyngeal and laryngeal cancers (Sapkota et al., 2008) and diseases of the eye (Smith and Mehta, 2003). In children, exposure to biomass pollution is associated with a higher prevalence, duration and poorer outcome of acute respiratory infections (Ezzati and Kammen, 2001; Mishra, 2003; Smith et al., 2000). Few studies have investigated the impact of biomass combustion exposure on pulmonary function, showing conflicting results. Rinne et al. (2006) did not observe any differences in pulmonary function among Ecuadorian women using biomass or liquefied petroleum gas for cooking. Conversely, studies from Turkey and India show that women exposed to biomass present a higher odds ratio of chronic bronchitis and decreased forced expiratory volume in the first second (FEV1)/forced vital capacity (FVC) compared to women using liquefied petroleum gas (Dutt et al., 1996; Ekici et al., 2005; Kiraz et al., 2003). In children, the use of biomass fuel has also been associated with decreased FEV1 and FVC (Qian et al., 2007; Rinne et al., 2007). Although the aforementioned studies report a decreased FEV1 associated with biomass combustion exposure, airflow limitation (Pauwels et al., 2001) was observed in none of them. Most epidemiologic studies that have evaluated the health effects of biomass combustion (prevalence of symptoms and respiratory diseases or changes in pulmonary function) use qualitative assessments of exposure, such as fuel or housing type. Few of these studies also evaluated the exposure levels of particles less than 10 mm in diameter (PM10) (Regalado et al., 2006). No study has yet evaluated all these effects in the same population and taken into account the intensity of exposure to PM2.5. In the present study we report the respiratory effects of chronic exposure to biomass combustion in a Brazilian population that uses wood and cashew nut bark as the main cooking fuel and compared the results with those of individuals living in the same community that use liquefied petroleum gas instead. The study

was designed in order to simultaneously quantify the magnitude of exposure, the prevalence of respiratory symptoms and the effects of exposure on pulmonary function in this population.

2. Materials and methods 2.1. Study population A schematic representation of the study design is presented in Fig. 1. The study was conducted in Joao Camara, a city in the northeast countryside of Brazil. Joao Camara is a city with an area of 715 km2 that is 120 m above sea level and is situated in the Brazilian state of Rio Grande do Norte. The proportion of the population living in extreme poverty in this Brazilian state is around 35%, almost triple that of Brazil as a whole. Demographic and social data show that black people, houses headed by young people and houses with more than four children are the poorest in this state. Income inequality in Joao Camara is similar to the whole country, which is one of the highest in the world. Joao Camara has 30,989 inhabitants and a demographic density of 43.34 inhabitants/km2 according to the Brazilian Institute of Geography and Statistics. Thirty-five percent of Joao Camara’s inhabitants live in rural areas and the proportion of individuals less than 20 years of age is 35.63%, while individuals older than 60 years of age comprise only 8.9% of the population. In the ‘‘urban’’ area of the city, the main economic activities are commerce and public service. Instead of urban, Joao Camara should actually be considered as a community center, since basic sanitation is poor, the majority of the streets are not paved and electricity is not available in all houses. In the rural areas, the main activity is the production and processing of cashew nuts, which are abundant in this region. After the crop is harvested the nut is extracted from the cashew and then cooked for a period varying from 2 to 3 h in bonfires or stoves, usually located close to or inside the house, using biomass as a fuel. After cooking, the nut is stripped and the bark covered by the residual smut is used as biomass fuel for the next round of cooking. When considering the whole process, we expect that the exposure to the products of biomass combustion may have particularly pronounced effects in these individuals due to the increased biomass exposure. 2.1.1. Population sample For sample size calculations, we used information concerning the number of inhabitants and residences in urban and rural areas of the city obtained by the demographic census (2000) and projected to 2005 by the census. We

Fig. 1. Schematic representation of study design. y (years of age), BD (bronchodilator).

L.F. da Silva et al. / Environmental Research 112 (2012) 111–117

113

Fig. 2. Representative pictures of an indoor stove (A) and outside bonfire (B), both using biomass as fuel. calculated the sample size for assessing the effects of exposure on respiratory symptoms, considering a sample power of 80%, using the Epitable module of the Epiinfo& 2000 software (CDC, Atlanta, USA, 2000), and then we defined the sample fraction for the rural and urban population to obtain the weighted sample. For subject selection in the urban and rural areas, we elected to perform systematic sampling of residences and then evaluate and interview all individuals living in the selected residences. In this way, the probability of selecting a specific individual is the same as that of his/her residence being selected; thus, because initially all residences have the same probability of being selected, all individuals have the same probability of selection. The number of residences in each area was obtained using the total population and the mean number of individuals per residence from estimates for each stratum. An extra 10% of residences were included in the calculated number to consider possible losses. Sample size calculations suggested that 260 residences should be sampled. In each residence, a questionnaire was given to collect infrastructure and socioeconomic data as well as data concerning the main source of energy and the characteristics of the place of cooking. The subjects were stratified by age (adults Z20 years; younger individuals o 20 years) and then divided into three groups according to the fuel and the site used for cooking: liquefied petroleum gas (Gas), Indoor-Biomass or and Outside-Biomass (Figs. 1 and 2). Gas was defined as the control fuel because this is the cleanest available source of energy for cooking for this population.

2.2. Exposure assessment The magnitude of particulate matter exposure was indirectly assessed using an index of reflectance of paper filters, which was defined as the difference between the reflectance of non-exposed white filters and the reflectance of exposed filters. In this process, the emitted light reaches the filter, is reflected and is captured by a photovoltaic cell that generates a depolarization followed by a current that is measured, amplified and translated into a specific value (Reid et al., 1998). Filters with a diameter of 110 mm (cellulose filter no1, Grade 1: 11 mm, Whatman&) were divided into quarters with a 2375.8 mm2 area, attached to a paperboard and sequentially numbered for identification. These filters were individually packaged in plastic bags during storage and transport. In each residence, one quarter of the paper filter was placed near the cook stove during a four-day period. The reflectance of 15 white (non-exposed) reference filters was measured for quality control purposes, showing maximum variability of 1.3%. To validate this exposure assessment, the PM2.5 concentration was directly measured by real time particle mass estimation from light scattering for 1 h during the cooking period in 48 of the 260 houses. After correlation analysis between the reflectance index and direct PM2.5 measurements, the reflectance index was used to estimate PM2.5 concentrations in all residences (ePM2.5). We also used an index (ePM2.5-years) to take into account both the magnitude and the duration (in years) of exposure.

2.3. Symptom analysis An individual questionnaire was given to all residents in each of the 260 houses to collect personal and clinical data (smoking history, time of exposure in hours per week, respiratory/ocular symptoms, chronic diseases, medications and hospitalizations). The questions used for the evaluation of respiratory/ocular symptoms were obtained from the validated and adapted Portuguese versions of the ISAAC and British Medical Council questionnaires (Cotes, 1987; Sole et al., 1998). For children, the questions were answered by the responsible adult (Sole et al., 1998). The respiratory/ocular symptoms assessed were cough with sputum, wheezing, dyspnea, eye itching/tears and coryza.

2.4. Pulmonary function test (PFT) After symptom and exposure analysis, pulmonary function tests were performed in 80 adults from 80 different residences with the Koko& Spirometer (OHD, AL, USA) attached to a laptop computer. The system was calibrated 2 times a day following the manufacturer’s instructions. All tests were performed during a 6-day period from 9:00 AM to 4:00 PM with similar weather conditions (average daily temperature ¼ 28.37 2.1 1C and average humidity ¼82 7 2%). For each patient, three acceptable spirometric curves were obtained (pre-bronchodilator tests). Individuals received a dose of 200 mg of salbutamol inhaled through a 500mL spacer and three other acceptable curves were obtained 15–20 min after bronchodilator (post-bronchodilator tests). All tests were performed with the individuals in the sitting position using a nasal clip and dischargeable bocal. Pulmonary function tests parameters (FEV1, FVC, FEV1/FVC) were compared among four groups (Smoker-Gas, Non-Smoker-Gas, Smoker-Biomass and NonSmoker-Biomass; n¼ 20 in each group). The 20 individuals from the SmokerBiomass group were randomly selected. The individuals from the other 3 groups were selected to ensure that groups were matched by age, sex and, in case of smoker groups, also by pack-years. Only one individual was selected per residence, and individuals included in the non-smoker groups lived in smoking free residences to avoid the bias of passive smoking. Pulmonary function tests were also performed in 40 non-smokers (active or passive, living in smoking free residences) less than 20 years of age who were divided in two groups according to type of cooking fuel and were then matched by age and sex (Young-Gas, Young-Biomass, n¼ 20 in each group). When present in these individuals, chronic obstructive pulmonary disease (COPD) was defined as a ratio of less than 0.7 of post-bronchodilator forced expiratory volume in the first second over forced vital capacity. The predicted values were calculated based on normal reference values for the Brazilian population (Pereira et al., 1992; Polgar and Promadhat, 1971). 2.5. Statistical analysis Demographical data were compared using the chi-square test for categorical variables. The linear numeric variables were compared using analysis of variance or the Kruskal–Wallis test and Student’s T-test or Mann–Whitney test, depending on the distribution of data and the number of groups. The Spearman test was used to assess correlations between pulmonary function tests and particulate matter exposure parameters (years of exposure, ePM2.5 and ePM2.5-years). Results are expressed as mean 7 standard deviation or median (interquartile range). The analysis of symptoms was performed using logistic regression considering symptoms as dependent variables and fuel, place of cook stove, time of exposure per week (in hours), smoking status, age and sex as covariates. These covariates were selected from the complete set of epidemiological and socioeconomic variables included in the described questionnaire (online appendix), which were tested individually among groups with univariate analysis. Those that presented any difference and those known to be related to the development of chronic obstructive pulmonary disease or to biomass use were included in the regression analysis. These variables were age, gender, family income, educational status, exposure and smoking. The same analysis was performed using ePM2.5 (continuous variable) instead of fuel and place of the cook stove as a covariate. Results are expressed as odds ratios (95% confidence interval). All statistical analyses were performed using SPSS 15.0 (SPSS Inc&, Chicago, IL, USA, 2004). The level of significance was set at p o 0.05.

3. Results We collected data from a total of 1402 individuals. The demographic data in each group are shown in Table 1.

114

L.F. da Silva et al. / Environmental Research 112 (2012) 111–117

Table 1 Demographic data of adults and younger individuals according to exposure groups. A—Adults Z20 years Gender: M/F Smokers (%) Age (years) Pack-years Incomen Years of school

Table 2 Prevalence of symptoms (%) in adults and younger individuals according to exposure groups. A—Adults

Gas (466) 156:310 210 (45.1) 43.38 718.20 32.55 717.65 53.94 712.00 5.927 4.21

Outside-Biomass (218) Indoor-Biomass (60) 88:130 107 (49.1) 45.74 7 18.46 31.09 716.29 55.47 7 17.34 5.43 74.85

21:39 28 (46.7) 42.94 7 18.14 34.05 718.20 51.46 7 15.58 5.18 73.24

Z 20 years

Gas (466)

Coughþ sputum 132 (28.3) Dyspnea 49 (10.5) Wheezing 68 (14.6) Coryza 156 (33.5) Eye itching/tears 121 (26.0)

Outside-Biomass (218) 91 39 52 64 50

#

(41.7) (17.9)# (23.9)# (29.4) (22.9)

Indoor-Biomass (60) 32 14 17 30 24

(53.3)# (23.3)# (28.3)# (50.0)n,# (40.0)n,#

B—Young individuals

B—Young individuals o 20 years o20 years

Gas (360)

Outside-Biomass (229) Indoor-Biomass (69)

Gender: M:F Smokers/% Age (years) Incomea Years of School

171: 189 4 (1.1) 9.247 5.70 50.617 12.27 4.357 3.56

110: 119 5 (2.2) 9.87 75.66 49.16 7 15.41 4.21 73.92

31: 38 2 (2.9) 9.21 75.32 52.60 711.10 4.48 73.55

a

Outside-Biomass (229)

Coughþ sputum 114 (31.7) 99 (43.2)# Dyspnea 44 (12.2) 37 (16.2) Wheezing 51 (14.2) 52 (22.7)# Coryza 94 (26.1) 82 (35.8)# Eye itching/tears 70 (19.4) 36 (15.7) n

M, male; F, female. Results presented in mean 7 SD unless otherwise specified.

Gas (360)

#

Indoor-Biomass (69) 24 17 20 30 20

(34.8)n (24.6)n,# (29.0)# (43.5)# (29.0)n,#

Significant difference with Outside-Biomass. Significant difference with Gas.

Income per capita per month in Brazilian Reais (BRL).

Fig. 3. Estimated PM2.5 according to exposure group. *Significant difference p o0.001.

There were no significant differences in sex, age, smoking history, income and literacy between groups. The PM2.5 concentration during the cooking period in each group was 3.0 73.6 mg/m3 (LPG, n ¼16), 151.1 7114.8 mg/m3 (Outside-Biomass, n ¼16) and 230.37157.0 mg/m3 (Indoor-Biomass, n ¼16). The reflectance index of the exposed filters showed a significant positive correlation with PM2.5 concentration (r¼ 0.92) (Figure in online appendix) and was therefore used for the estimation of PM2.5 exposure (ePM2.5). Values of ePM2.5 for each group are shown in Fig. 3. A significantly higher ePM2.5 was observed in Indoor-Biomass compared to Gas and Outside-Biomass groups. The prevalence of symptoms in each group and the odds ratio (OR) for presenting symptoms in the Biomass groups compared to Gas are presented in Table 2. Table 3 shows that there was a significantly positive OR for cough, wheezing and dyspnea in adults exposed to Indoor-Biomass (OR ¼2.93, 2.33 and 2.59, respectively) and Outside-BM (OR¼1.78, 1.78 and 1.80 respectively) compared to Gas. A significant OR for eye itching/tears and

coryza was observed in adults exposed to Indoor-Biomass compared to Gas (OR ¼1.90 and 1.98, respectively) and to Outside Biomass (OR ¼2.27 and 2.47, respectively). The analysis using ePM2.5 instead of groups showed similar results; we observed a significantly positive OR for cough [2.32(1.58–3.12)], wheezing [1.85 (1.21–2.36)], dyspnea [2.12(1.64–2.90)] and coryza [1.82(1.12–2.45)] for increasing values of ePM2.5. Table 3 also demonstrates that younger individuals presented a significant OR for cough, wheezing, dyspnea and coryza when exposed to Indoor-Biomass as compared to Gas (1.64, 2.46, 2.29 and 2.20, respectively). A significant OR for wheezing and coryza was observed in younger individuals exposed to Outside-Biomass as compared to Gas (1.77 and 1.58, respectively). There was also a positive OR (2.17) for eye itching/ tears in younger individuals exposed to Indoor-Biomass compared to Outside-Biomass. The analysis using ePM2.5 instead of groups showed similar results; we observed a significantly positive OR for wheezing [1.86(1.15–2.57)], dyspnea [1.76 (1.22–2.30)] and coryza [1.68(1.19–2.17)] for increasing values of ePM2.5. The point prevalence of other health conditions in the entire study population as demonstrated by the questionnaires included asthma (120, 8.6%), systemic arterial hypertension (273, 19.5%) and diabetes (52, 3.7%). There were no significant differences among exposure groups. Parameters of post-bronchodilator pulmonary function tests are shown in Table 4 and Fig. 4. There were no significant differences in body mass index among groups. Among adults, the NonSmokers-Gas group presented a significantly higher predicted FEV1 and FEV1/FVC compared to all other groups (p¼0.002). The Smokers-Biomass group showed a significantly lower predicted FEV1 and FEV1/FVC compared to Smokers-Gas group (p¼0.007). Compared to the Non-Smokers-Gas group, Non-Smokers-Biomass group showed a similar decrease in pulmonary function tests parameters as did the Smokers-Gas group. There were significantly more individuals with airway obstruction according to Global Initiative for Chronic Obstructive Lung Disease (GOLD) criteria (FEV1/FVCo0.70) in the Smokers and/or Biomass groups compared to the Non-Smoker-Gas group. The point prevalence of airway obstruction was 20% (4/20) in both the Non-SmokerBiomass and Smoker-Gas groups and 35% (7/20) in the SmokerBiomass group (individual COPD patient data is available online).

L.F. da Silva et al. / Environmental Research 112 (2012) 111–117

115

Table 3 Odds ratio (95% CI) for symptoms among groups in adults and younger individuals adjusted for gender, age, smoking and socioeconomic parameters. A—Adults Z20 years

Indoor-Biomass:Gas

Cough þsputum Dyspnea Wheezing Coryza Eye itching/tears

2.93 2.59 2.33 1.98 1.90

Outside-Biomass:Gas

(1.68–5.10)n (1.32–5.09)n (1.25–4.38)n (1.15–3.42)n (1.09–3.32)n

1.78 1.80 1.78 0.83 0.85

Outside-Biomass:Indoor-Biomass

(1.27–2.50)n (1.14–2.86)n (1.18–2.69)n (0.59–1.17) (0.58–1.24)

1.60 1.40 1.26 2.47 2.27

(0.90–2.83) (0.70–2.79) (0.66–2.40) (1.37–4.46)n (1.23–4.18)n

B—Young individuals o20 years Cough þsputum Dyspnea Wheezing Coryza Eye itching/tears n

Indoor-Biomass:Gas 1.64 2.29 2.46 2.20 1.69

n

(1.16–2.31) (1.21–4.32)n (1.34–4.47)n (1.29–3.75)n (0.95–3.02)

Outside-Biomass:Gas

Outside-Biomass:Indoor-Biomass

1.64 (0.92–2.36) 1.38 (0.86–2.22) 1.77 (1.15–2.72)n 1.58 (1.10–2.27)n 0.77(0.50–1.12)

0.70 (0.40–1.22) 1.70(0.89–3.25) 1.39 (0.76–1.54) 1.38 (0.80–2.38) 2.17 (1.15–4.09)n

Significant difference.

Table 4 Pulmonary function test data in adults and younger individuals according to exposure and smoking [mean (95% CI)]. A—Adults Z20 years

Non-Smoker-Gas

Smoker-Gas n

0.85 (0.82–0.88) 94.65 (93.03–96.26)n

0.76 (0.73–0.80) 87.15 (83.90–90.40)#

o20 years

Young-Gas

Young-Biomass

FEV1/FVC % FEV1 Pred.

0.89 (0.87–0.91) 92.15 (91.37–92.92)

0.88 (0.86–0.90) 94.60 (92.51–94.23)

FEV1/FVC % FEV1 Pred.

Non-Smoker-Biomass #

0.79 (0.76–0.82) 88.50 (84.73–92.26)#

Smoker-Biomass 0.72 (0.68–0.77) 76.18 7 84.92

B—Young individuals

FEV1, forced expiratory volume in the first second; FVC, forced vital capacity. n

#

Significant difference with all other groups Significant difference with Smoker-BM.

Fig. 4. Comparison of FEV1/FVC ratios among exposure and smoking groups in individuals with 20 years of age or more (means and 95% CI). *Significant difference.

116

L.F. da Silva et al. / Environmental Research 112 (2012) 111–117

There was a significant negative correlation between the pulmonary function test parameters and the duration of exposure in years (FEV1: r ¼  0.46, p o0.001; FEV1/FVC: r ¼ 0.63), ePM2.5 (FEV1: r¼  0.48; FEV1/FVC: r ¼ 0.52), and ePM2.5years (FEV1: r ¼  0.56; FEV1/FVC: r ¼  0.63). Among younger individuals, no significant differences in pulmonary function test parameters were observed between the Biomass and Gas groups.

4. Discussion In the present study we performed for the first time a simultaneous evaluation of respiratory symptom prevalence, pulmonary function changes and magnitude of exposure in a population chronically exposed to biomass combustion. Our main findings were an increased prevalence of respiratory symptoms, impaired pulmonary function and increased prevalence of chronic obstructive pulmonary disease in individuals exposed to biomass. We also observed a synergistic effect between biomass exposure and smoking in terms of loss of pulmonary function. There is increasing evidence that the exposure to biomass combustion can affect respiratory health (Liu et al., 2008; Mishra, 2003; Rinne et al., 2007; Rumchev et al., 2007; Salvi and Barnes, 2009; Smith et al., 2000; Zhou et al., 2009). Similar to previous reports, we observed a higher prevalence of respiratory symptoms in the exposed population compared to individuals that used gas, with cough and coryza being the most prevalent symptoms in both adults and younger subjects. It is known that the indoor concentration of pollutants directly depends on the ratio between the rate of emission and room ventilation (Johnson et al., 2004). In order to evaluate the effect of particulate matter dispersion, residences using biomass as fuel were divided into two groups according to the site of combustion (indoor or outside). Indeed, the symptom odds ratios showed different patterns according to the site of exposure; symptoms associated with the upper airways (coryza/wheezing and eye itching/tears) presented significant ORs only in the Indoor-Biomass group while symptoms associated with the lower respiratory tract (cough, wheezing and dyspnea) presented significant ORs in both the Indoor-BM and Outside-Biomass groups. Previous studies investigating the impact of biomass combustion exposure on pulmonary function have shown either decreased FEV1/FVC without airflow limitation (Salvi and Barnes, 2009; Zhou et al., 2009) or no effects of BM on pulmonary function tests. Rinne et al. (2006) did not observe differences in pulmonary function among Ecuadorian women using biomass or gas for cooking. Ekici et al. (2005) showed in Turkish women that 23.1% of chronic bronchitis cases were attributable to biomass exposure. Decreased values of FEV1 were reported in rural Turkish women using biomass as compared to urban residents; however, both rural and urban women had pulmonary function tests within normal limits (Kiraz et al., 2003). Using predicting models, Dutt et al. (1996) observed that Indian women exposed to biofuels were more likely to have reduced pulmonary function as compared to those using kerosene and gas. In children, the use of biomass fuel has also been associated with decreased FEV1 and FVC (Qian et al., 2007; Rinne et al., 2006). Although all the above mentioned studies report a decrease in FEV1 associated with biomass combustion exposure, airflow limitation was observed in none of them. Recently, some studies have reported exposure to biomass to be a significant risk factor for COPD as evidenced by pulmonary function tests, using standardized criteria (Pauwels et al., 2001). Zhou et al. (2009) reported an overall 5.2% prevalence of COPD among Chinese non-smokers, a 6.0% prevalence among non-smokers who used biomass for cooking and an OR of 1.31 for COPD in the latter population. Another Chinese study

(Liu et al., 2008) evaluated the COPD prevalence among nonsmoker women living in rural (7.2%) and urban (2.5%) areas, demonstrating a significant OR of 1.72 for COPD in women using biomass as fuel. In a recent meta-analysis Kurmi et al. (2010) showed an OR of 2.96 for COPD in individuals exposed to indoor air pollution. Our results showed a significant loss of pulmonary function in Non-Smokers-Biomass compared to Non-Smokers-Gas. Interestingly, the prevalence of COPD among Non-Smokers-Biomass (20%) was similar to that observed among Smokers-Gas. These findings indicate that biomass combustion is an independent etiologic factor for COPD in non-smokers. We also observed that the effects of biomass combustion on pulmonary function are exacerbated by cigarette smoking. Taken together, the aforementioned studies indicate that biomass combustion is involved in COPD development among non-smokers. In the present study, we further demonstrated a direct association between biomass combustion exposure and COPD. Passive smoking has been considered an important risk factor for COPD (Chen, 2008; Menezes and Hallal, 2007). Since the nonsmoker groups consisted of individuals living in non-smoking residences, our study design allowed the evaluation of the effects of biomass combustion exposure alone on pulmonary function, avoiding passive smoking as a confounding variable. Younger individuals exposed to biomass combustion showed a lower prevalence of cough compared to that in exposed adults and preserved pulmonary function. These results, together with the significant correlations between the loss of pulmonary function and exposure parameters, reflect the cumulative effect of biomass combustion and evidence the relevance not only of the magnitude (PM2.5) but also of the duration of exposure. The mechanism by which biomass combustion impairs pulmonary health is not clear. It is suggested that biomass smoke can elicit damaging oxidation reactions in both the lung and the systemic circulation, which may account for many of the toxic responses in exposed individuals (Padhy and Padhi, 2009). A pulmonary and systemic inflammatory response has also been shown as a mechanism for the increased cardiopulmonary morbidity that epidemiological studies have associated with biomass smoke inhalation (Swiston et al., 2008) . One limitation of our study was the number of pulmonary function tests in each group. Epidemiological studies that evaluated COPD prevalence in different populations included from 100 to few thousands spirometries, according to study design (Brostrom et al., 2009; Perez-Padilla et al., 1996). However, since the groups for pulmonary function tests comparison were matched for several confounding variables, the calculated sample power in the present study to evaluate the effect of exposure in decreasing lung function parameters to COPD levels was 90%. The lack of mean values of daily PM2.5 could be considered another limitation. However, since the subjects living in the same residence stay at home during different periods, the mean daily PM2.5 concentration in the residence does not necessarily reflect the actual individual exposure. Thus, to avoid the possibility of sampling bias due to a single measurement in different periods of the day, measurements were taken for a one hour period during active biomass combustion considered the peak-PM2.5 exposure in each residence. Another important consideration is that, although the level of education and income was similar among groups, we cannot eliminate other socio-economic determinants related to the use of gas. Previous studies that evaluated the effects of chronic exposure to biomass combustion suggest that there is a relationship between exposure magnitude and health effects (Ezzati and Kammen, 2001). However, different methods of exposure assessment have been used, varying from qualitative analysis to direct

L.F. da Silva et al. / Environmental Research 112 (2012) 111–117

quantification of pollutants (Jin et al., 2006; Smith-Sivertsen et al., 2009; Viegi et al., 2004; WHO, 2006). Direct measurements of particulate matter, such as PM2.5 concentrations, usually imply the use of sophisticated and expensive equipments that require many hours or days of sampling, making this procedure less feasible in poor rural areas. In the present study we assessed PM2.5 exposure using an index of reflectance of exposed filters. The high correlation coefficient between the PM2.5 concentration and the filter reflectance index validated this cheap and feasible method for use in the estimation of the magnitude of PM2.5 exposure.

5. Conclusions We conclude that chronic exposure to biomass combustion is associated with increased prevalence of respiratory symptoms, reduced lung function and the development of chronic obstructive pulmonary disease. Although there is a synergistic effect, the loss in pulmonary function due to biomass combustion is not dependent on smoking and is associated with the duration and magnitude of biomass exposure.

Acknowledgments The authors would like to thank Carlos Sampaio for clinical support in pulmonary function tests, Regiani Oliveira for filter reflectance measurements, Klaus F. Rabe and Prof. Thais Mauad for their relevant suggestions, and the members of the ‘‘Bandeira Cientı´fica’’ Project organizing committee: Andreza Cadima Silva, Carla Romagnolli, De´bora Terribili, Fabı´ola Ole´a Albieri, Gabriella Paiva Bento, Lilian Zancheta Castelli, Livia Caroline Barbosa Mariano, Maı´ra Grizzo, Marilena Nakaguma, Nata´lia Sassaki Takahashi, Fernando Arturo Effio Sollis, Higor Alexandre Pavoni Gomes e Igor Tadeu da Costa.

Appendix A. Supplementary materials Supplementary data associated with this article can be found in the online version at doi:10.1016/j.envres.2011.10.012.

References Brostrom, E.B., et al., 2009. Obstructive lung disease in children with mild to severe BPD. Respir. Med.. Bruce, N., et al., 2000. Indoor air pollution in developing countries: a major environmental and public health challenge. Bull. World Health Organ. 78, 1078–1092. Chen, R., 2008. Passive smoking exposure and risk of COPD in China. Lancet 371, 201 (author reply 201-2). Cotes, J.E., 1987. Medical research council questionnaire on respiratory symptoms (1986). Lancet 2, 1028. Dennekamp, M., et al., 2010. Outdoor air pollution as a trigger for out-of-hospital cardiac arrests. Epidemiology 21, 494–500. Dutt, D., et al., 1996. Effect of indoor air pollution on the respiratory system of women using different fuels for cooking in an urban slum of Pondicherry. Natl. Med. J. India 9, 113–117. Ekici, A., et al., 2005. Obstructive airway diseases in women exposed to biomass smoke. Environ. Res. 99, 93–98. Ellegard, A., 1996. Cooking fuel smoke and respiratory symptoms among women in low-income areas in Maputo. Environ. Health Perspect. 104, 980–985. Ellegard, A., 1997. Tears while cooking: an indicator of indoor air pollution and related health effects in developing countries. Environ. Res. 75, 12–22. Ezzati, M., Kammen, D.M., 2001. Quantifying the effects of exposure to indoor air pollution from biomass combustion on acute respiratory infections in developing countries. Environ. Health Perspect. 109, 481–488. Jin, Y., et al., 2006. Exposure to indoor air pollution from household energy use in rural China: the interactions of technology, behavior, and knowledge in health risk management. Soc. Sci. Med. 62, 3161–3176.

117

Johnson, T., et al., 2004. A pilot study using scripted ventilation conditions to identify key factors affecting indoor pollutant concentration and air exchange rate in a residence. J. Expo. Anal. Environ. Epidemiol. 14, 1–22. Kiraz, K., et al., 2003. Chronic pulmonary disease in rural women exposed to biomass fumes. Clin. Invest. Med. 26, 243–248. Kurmi, O.P., et al., 2010. COPD and chronic bronchitis risk of indoor air pollution from solid fuel: a systematic review and meta-analysis. Thorax 65, 221–228. Liu, Y., et al., 2008. Outdoor and indoor air pollution and COPD-related diseases in high- and low-income countries. Int. J. Tuberc. Lung. Dis. 12, 115–127. Menezes, A.M., Hallal, P.C., 2007. Role of passive smoking on COPD risk in nonsmokers. Lancet 370, 716–717. Mishra, V., 2003. Indoor air pollution from biomass combustion and acute respiratory illness in preschool age children in Zimbabwe. Int. J. Epidemiol. 32, 847–853. Mishra, V., Retherford, R.D., 1997. Cooking smoke increases the risk of acute respiratory infection in children. Natl. Fam. Health Surv. Bull., 1–4. Padhy, P.K., Padhi, B.K., 2009. Effects of biomass combustion smoke on hematological and antioxidant profile among children (8–13 years) in India. Inhal. Toxicol. Park, S.K., et al., 2010. Particulate air pollution, metabolic syndrome, and heart rate variability: the multi-ethnic study of atherosclerosis (MESA). Environ. Health Perspect. 118, 1406–1411. Pauwels, R.A., et al., 2001. Global strategy for the diagnosis, management, and prevention of chronic obstructive pulmonary disease. NHLBI/WHO Global Initiative for Chronic Obstructive Lung Disease (GOLD) Workshop Summary. Am. J. Respir. Crit. Care Med. 163, 1256–1276. Pereira, C.A.C., et al., 1992. Valores de referˆencia para espirometria em uma amostra da populac- a~ o brasileira adulta. J. Pneumol. 18, 10–22. Perez-Padilla, R., et al., 1996. Exposure to biomass smoke and chronic airway disease in Mexican women. A case-control study. Am. J. Respir. Crit. Care Med. 154, 701–706. Pierson, W.E., et al., 1989. Potential adverse health effects of wood smoke. West J. Med. 151, 339–342. Polgar, G., Promadhat, V., 1971. Pulmonary Function Testing in Children: Techniques and Standards. Saunders, Philadelphia. Qian, Z., et al., 2007. Respiratory responses to diverse indoor combustion air pollution sources. Indoor Air 17, 135–142. Regalado, J., et al., 2006. The effect of biomass burning on respiratory symptoms and lung function in rural Mexican women. Am. J. Respir Crit. Care Med. 174, 901–905. Reid, J.S., et al., 1998. Comparison of techniques for measuring short-wave absorption and black carbon content of aerosol from biomass burning in Brazil. J. Geophys. Res. 103, 32031–32040. Rinne, S.T., et al., 2006. Relationship of pulmonary function among women and children to indoor air pollution from biomass use in rural Ecuador. Respir. Med. 100, 1208–1215. Rinne, S.T., et al., 2007. Use of biomass fuel is associated with infant mortality and child health in trend analysis. Am. J. Trop. Med. Hyg. 76, 585–591. Rumchev, K., et al., 2007. Indoor air pollution from biomass combustion and respiratory symptoms of women and children in a Zimbabwean village. Indoor Air 17, 468–474. Salvi, S.S., Barnes, P.J., 2009. Chronic obstructive pulmonary disease in nonsmokers. Lancet 374, 733–743. Sandoval, J., et al., 1993. Pulmonary arterial hypertension and cor pulmonale associated with chronic domestic woodsmoke inhalation. Chest 103, 12–20. Sapkota, A., et al., 2008. Indoor air pollution from solid fuels and risk of hypopharyngeal/laryngeal and lung cancers: a multicentric case-control study from India. Int. J. Epidemiol. 37, 321–328. Smith-Sivertsen, T., et al., 2009. Effect of reducing indoor air pollution on women’s respiratory symptoms and lung function: the RESPIRE Randomized Trial, Guatemala. Am. J. Epidemiol. 170, 211–220. Smith, K.R., 2000. Inaugural article: national burden of disease in India from indoor air pollution. Proc. Natl. Acad Sci. USA 97, 13286–13293. Smith, K.R., Mehta, S., 2003. The burden of disease from indoor air pollution in developing countries: comparison of estimates. Int. J. Hyg. Environ. Health 206, 279–289. Smith, K.R., et al., 2000. Indoor air pollution in developing countries and acute lower respiratory infections in children. Thorax 55, 518–532. Sole, D., et al., 1998. International Study of Asthma and Allergies in Childhood (ISAAC) written questionnaire: validation of the asthma component among Brazilian children. J. Investig. Allergol. Clin. Immunol. 8, 376–382. Swiston, J.R., et al., 2008. Wood smoke exposure induces a pulmonary and systemic inflammatory response in firefighters. Eur. Respir. J. 32, 129–138. Viegi, G., et al., 2004. Indoor air pollution and airway disease. Int. J. Tuberc. Lung. Dis. 8, 1401–1415. WHO, Fuel for life. In: W.H. Organization (Ed.), Household Energy and Health, New York, 2006. World Health Organization, U.N., Addressing the links between indoor air pollution, household energy and human health. In: W.H. Organization (Ed.), WHOUSAID Consultation on the Health Impact of Household Energy in Developing Countries, Geneva, 2002. Zhou, Y., et al., 2009. COPD in Chinese nonsmokers. Eur. Respir. J. 33, 509–518.