In this quantitative review, the literature was systematically evaluated to determine the effect of air filter use on respiratory outcomes in patients with chronic respiratory diseases. The results demonstrated that air filter use significantly reduced indoor PM2.5, and that a higher baseline PM concentration was associated with greater reductions. PM reduction improved predicted FEV1, but showed no significant effect on respiratory symptoms. No study about other respiratory diseases such as interstitial lung disease or bronchiectasis was founded in our searching strategy.
Although two studies showed significant changes in respiratory symptoms with air filter use, (15, 24) the remaining studies showed no statistically significant improvement, and the pooled estimate failed to reach statistical significance in this review. Previous meta-analyses have shown that short-term exposure to PM2.5 could aggravate respiratory symptoms and reduce peak expiratory flow in patients with COPD (8) and asthma (8, 30). The odds ratio of the respiratory symptoms was 1.22‒1.57 with a 50 µg/m3 increase in PM2.5, depending on the model used (8). The baseline concentration of PM2.5 (6.2‒39.5 µg/m3) was relatively low in the studies evaluated here (14, 15, 24–26, 29), and the pooled change in PM2.5 was − 11.45 µg/m3. A minimal reduction in PM due to low baseline levels may have influenced the results and caused the lack of statistical significance. The measure of respiratory symptoms was differing between studies (14, 15, 24–26) such as from cough to emotional function. The previous systemic review (8) used cough and wheeze as measure of respiratory symptoms. However, for the estimates of respiratory symptoms as a whole asthma status in this review, there is no significant change by using air filters.
The previous meta-analysis was a panel study to evaluate the effect of short-term exposure to PM (a few days) on respiratory outcomes (8). Two studies (31, 32) evaluated the effect of long-term PM (decades of year) exposure on respiratory symptoms. The current study evaluates the effect of PM reduction (less than 6 weeks to 12 months) in various periods. In subgroup analysis, however, there was no tendency or significant difference in respiratory symptoms by the duration of air filter use (Supplementary Fig. 1).
Edginton et al. reviewed the effect of PM on FEV1 in healthy adults (33). Short-term exposure (over several days) resulted in FEV1 changes of -7.02 mL with an increase of 10 µg/m3 PM2.5. Gauderman et al. (34) evaluated the effect of PM on lung development in adolescents over an 8 year period and showed that high PM2.5 concentration could reduce lung growth, measured by FEV1 and FVC. In another study, long-term exposure to PM2.5 reduced FEV1 by 0.24% each year (35). Our data support the previous exposure studies and suggest that reducing PM can improve lung development in polluted areas. Subgroup analysis showed that longer duration of use (1 year) had a greater effect on FEV1 than shorter duration of use (less than 6 months), although these results were not statistically significant. A greater number of studies with an observation period of more than 6 months is required to confirm the significance of these data.
Other interventions can ameliorate the adverse effect of PM, such as facial mask, omega-3, and vitamin D. A facial mask is an easy way to avoid exposure to PM, and Langrish et al. demonstrated that the use of facial masks in healthy subjects exposed to PM results in lower systolic blood pressure than in those without facial masks (36). During the study period, the concentration of PM2.5 was 86–140 ug/m3 in Beijing. Other studies also support the cardiovascular effects of facial mask use to avoid PM, including decreased systolic blood pressure, increased heart rate variability, and reduced ST segment depression (27, 36, 37), even when considering the confounding effect of traffic noise (38). However, the observation period of the studies was just a few weeks and none of the studies showed the effect on respiratory outcomes such as lung function and symptom. In one review article, omega-3 oil was shown to have an anti-inflammation effect and reduced asthma symptoms (39). Brigham et al. (40) demonstrated a correlation between omega-3 intake and PM2.5, and showed a marginal effect in reducing respiratory symptoms in children with asthma. Vitamin D is also reported to help control asthma. (39) Bose et al. demonstrated that a decreased level of vitamin D in obese children with asthma is associated with a higher level of symptoms following a 10 unit increase in PM2.5. (41) However, these studies were observational study and no other studies of vitamin D or omega-3 oil as interventions are available. Therefore, we realized after brief literature search that a tool other than air filter could not be analyzed as an intervention for PM.
This study has some limitations. First, the studies included in this analysis were primarily conducted in the USA, where PM concentration is relatively low. It is possible that studies conducted in areas of high PM concentration may have shown a greater effect with air filter use. Second, some studies (23, 24) may have deviated from the intended intervention to allow comparison with a control group (Supplementary Fig. 3). Third, since evaluating respiratory symptoms are easy to be affected by a placebo effect, some of the subjects in studies (23, 24) might favor reporting their symptoms when they are using an air filter. However, there was no tendency to see high estimates in these groups so that the result of respiratory symptom is free from this bias.
Despite these limitations, our study has several strengths. To our knowledge, this is the first meta-analysis of the effect of intervention on respiratory outcomes associated with reduced indoor PM in patients with respiratory diseases. Heterogeneity in our analyses was addressed by stratifying baseline PM concentration, air filter use duration, and study design.
In conclusion, indoor air filter use successfully reduced indoor PM levels, and this reduction had a beneficial effect on lung function in patients with chronic respiratory disease including asthma and COPD. Further studies of varying duration are required in areas of high PM concentration to confirm these observations.