Effect of exposure to ambient particulate matter on risk of developing pulmonary tuberculosis: A systematic review and meta-analysis

Abstract

Tuberculosis (TB) is a major global health challenge, particularly in polluted areas. The relationship between ambient particulate matter and TB risk remains unclear, making this systematic review and meta-analysis (SRMA) vital for assessing this link. This SRMA aimed to estimate the association between exposure to ambient particulate matter (PM10 and PM2.5) and the risk of pulmonary tuberculosis (PTB) infection. A literature search was conducted in PubMed, Web of Science, and Cochrane (English-language studies) on January 29, 2024. The review followed PRISMA Guidelines (2020) for comprehensive literature searches, data extraction, and quality assessment of included studies. A random-effects model was used for meta-analysis to estimate pooled effect sizes and assess heterogeneity. Study quality and publication bias were also evaluated. Of the 507 articles identified, 25 met the inclusion criteria. Long-term PM2.5 exposure was linked to a 26% increase in PTB risk (RR =1.26, 95% CI: 1.07–1.48), while short-term exposure raised the risk by 10% (RR =1.10, 95% CI: 0.98–1.25). Long-term PM10 exposure increased PTB risk by 7% (RR =1.07, 95% CI: 1.02–1.12), with short-term exposure showing a similar increase (RR =1.07, 95% CI: 0.95–1.17). Subgroup analysis revealed PTB risk increased by 15% in males and 29% in females for PM2.5, and by 10% in males and 7% in females for PM10. A 10 µg/m³ increase in Particulate matter is associated with a higher risk of pulmonary tuberculosis, highlighting the need for targeted public health measures to reduce particulate exposure, especially in high-risk urban and industrial areas.

INTRODUCTION

Tuberculosis (TB) remains a major global health issue, affecting millions. According to the World Health Organization (WHO), TB is the leading cause of death from infectious diseases, with about 10.6 million new cases and 1.3 million deaths in 2020.[1] TB, caused by Mycobacterium tuberculosis, primarily impacts the lungs (pulmonary tuberculosis, PTB) and spreads through air droplets when an infected person coughs, sneezes, or talks.[2] Focusing on pulmonary tuberculosis (PTB), this form of TB specifically targets lung tissue, leading to severe respiratory symptoms.[3] In 2015, WHO launched the End TB Strategy to reduce TB deaths, incidence, and costs by 2035.[4,5,6] Despite this, TB remains a global challenge. Research on PTB has traditionally focused on the immune system’s role.[7,8,9,10] Recently, however, air pollution has been recognized as a factor influencing PTB.[11,12,13] Ambient particulate matter (PM) in the air can enter the respiratory tract, causing inflammation, oxidative stress, and immune dysregulation.[14,15]

PM exposure has increased a lot in recent decades. According to the World Health Organization (WHO), about 4.2 million premature deaths worldwide were linked to fine particulate matter exposure in 2019.[16,17,18] Despite WHO guidelines for air quality, 99% of the global population lives in areas that do not meet these standards.[19] Studies show a strong connection between PM exposure and a higher risk of pulmonary tuberculosis (PTB). Research suggests that the risk of TB may depend on the size of the particulate matter, seasonal changes, and how long someone is exposed.[20,21,22,23] However, results are mixed, and there is no clear agreement on how PM exposure affects TB risk over time.[24,25,26]

Collective evidence from studies is limited regarding the increased risk of PTB due to PM exposure. A previous meta-analysis found that particulate matter ≤2.5 µg/m³ (PM2.5) was not significantly linked to a higher risk of tuberculosis.[27] In the same study, long-term exposure to particulate matter ≤10 µg/m³ (PM10) was associated with an increased risk of TB. Another meta-analysis indicated a significant relationship between PM2.5 and PM10 and the incidence of PTB.[28] However, this study did not consider the long- and short-term associations of PM with pulmonary TB. Establishing a clear link between ambient particulate matter and pulmonary tuberculosis is crucial for developing effective TB prevention measures. This systematic review and meta-analysis (SRMA) aims to estimate the association between ambient particulate matter exposure and risk of pulmonary tuberculosis infection. This study attempts to address the gap in knowledge by synthesizing all accessible evidence.

METHODOLOGY

The review followed the PRISMA (Preferred Reporting Items for Systematic Review and Meta-Analysis) guidelines,[29] as outlined in Table S1. Additionally, it was registered in PROSPERO (International Prospective Register of Systematic Reviews)[30] with the registration ID: CRD42023385393. As our analysis did not include patient-specific or individual data, the acquisition of informed consent and ethical approval was not applicable.

Table S1.

PRISMA Checklist (2020)

Section and Topic		Item #		Checklist item		Location where item is reported
Title
Title		1		Identify the report as a systematic review.		Pg 1
Abstract
Abstract		2		Made as per the Journal guidelines		Pg 1, 2
Introduction
Rationale		3		Describe the rationale for the review in the context of existing knowledge.		Pg 2,3
Objectives		4		Provide an explicit statement of the objective(s) or question(s) the review addresses.		Pg 3 & Table S2
Methods
Eligibility criteria		5		Specify the inclusion and exclusion criteria for the review and how studies were grouped for the syntheses.		Supplementary Table S2
Information sources		6		Specify all databases, registers, websites, organisations, reference lists and other sources searched or consulted to identify studies. Specify the date when each source was last searched or consulted.		Pg 4 & Supplementary Table S3
Search strategy		7		Present the full search strategies for all databases, registers and websites, including any filters and limits used.		Supplementary Table S3
Selection process		8		Specify the methods used to decide whether a study met the inclusion criteria of the review, including how many reviewers screened each record and each report retrieved, whether they worked independently, and if applicable, details of automation tools used in the process.		Pg 4,5
Data collection process		9		Specify the methods used to collect data from reports, including how many reviewers collected data from each report, whether they worked independently, any processes for obtaining or confirming data from study investigators, and if applicable, details of automation tools used in the process.		Pg 4, 5
Data items		10a		List and define all outcomes for which data were sought. Specify whether all results that were compatible with each outcome domain in each study were sought (e.g. for all measures, time points, analyses), and if not, the methods used to decide which results to collect.		Pg 5
		10b		List and define all other variables for which data were sought (e.g. participant and intervention characteristics, funding sources). Describe any assumptions made about any missing or unclear information.		Pg 4, 5
Study risk of bias assessment		11		Specify the methods used to assess risk of bias in the included studies, including details of the tool(s) used, how many reviewers assessed each study and whether they worked independently, and if applicable, details of automation tools used in the process.		Pg 6
Effect measures		12		Specify for each outcome the effect measure(s) (e.g. risk ratio, mean difference) used in the synthesis or presentation of results.		Pg 5
Synthesis methods		13a		Describe the processes used to decide which studies were eligible for each synthesis (e.g. tabulating the study intervention characteristics and comparing against the planned groups for each synthesis (item #5)).		Pg 5
		13b		Describe any methods required to prepare the data for presentation or synthesis, such as handling of missing summary statistics, or data conversions.		Pg 5
		13c		Describe any methods used to tabulate or visually display results of individual studies and syntheses.		Pg 5
		13d		Describe any methods used to synthesize results and provide a rationale for the choice(s). If meta-analysis was performed, describe the model(s), method(s) to identify the presence and extent of statistical heterogeneity, and software package(s) used.		Pg 4,5
		13e		Describe any methods used to explore possible causes of heterogeneity among study results (e.g. subgroup analysis, meta-regression).		Pg 4,5
		13f		Describe any sensitivity analyses conducted to assess robustness of the synthesized results.		Pg 4, 5,6
Reporting bias assessment		14		Describe any methods used to assess risk of bias due to missing results in a synthesis (arising from reporting biases).		Pg 7
Certainty assessment		15		Describe any methods used to assess certainty (or confidence) in the body of evidence for an outcome.		Pg 4
Results
Study selection		16a		Describe the results of the search and selection process, from the number of records identified in the search to the number of studies included in the review, ideally using a flow diagram.		Figure 1
		16b		Cite studies that might appear to meet the inclusion criteria, but which were excluded, and explain why they were excluded.		Pg 6 & Figure 1
Study characteristics		17		Cite each included study and present its characteristics.		Pg 6 & table 1
Risk of bias in studies		18		Present assessments of risk of bias for each included study.		Pg, 6 & Table S4
Results of individual studies		19		For all outcomes, present, for each study: (a) summary statistics for each group (where appropriate) and (b) an effect estimate and its precision (e.g. confidence/credible interval), ideally using structured tables or plots.		Pg 7, 8,9 & Figure 2
Results of syntheses		20a		For each synthesis, briefly summarise the characteristics and risk of bias among contributing studies.		Table I & Table S4
		20b		Present results of all statistical syntheses conducted. If meta-analysis was done, present for each the summary estimate and its precision (e.g. confidence/credible interval) and measures of statistical heterogeneity. If comparing groups, describe the direction of the effect.		Pg. 7,8,9
		20c		Present results of all investigations of possible causes of heterogeneity among study results.		NA
		20d		Present results of all sensitivity analyses conducted to assess the robustness of the synthesized results.		NA
Reporting biases		21		Present assessments of risk of bias due to missing results (arising from reporting biases) for each synthesis assessed.		Pg 7, 8, 9
Certainty of evidence		22		Present assessments of certainty (or confidence) in the body of evidence for each outcome assessed.		Pg 7, 8, 9
Discussion
Discussion		23a		Provide a general interpretation of the results in the context of other evidence.		Pg 9,10, 11
		23b		Discuss any limitations of the evidence included in the review.		Pg 11
		23c		Discuss any limitations of the review processes used.		Pg 11
		23d		Discuss implications of the results for practice, policy, and future research.		Pg 11, 12
Other information
Registration and protocol		24a		Provide registration information for the review, including register name and registration number, or state that the review was not registered.		Pg 3
		24b		Indicate where the review protocol can be accessed, or state that a protocol was not prepared.		Pg 3
		24c		Describe and explain any amendments to information provided at registration or in the protocol.		NA
Support		25		Describe sources of financial or non-financial support for the review, and the role of the funders or sponsors in the review.		Pg 12
Competing interests		26		Declare any competing interests of review authors.		Pg 12
Availability of data, code and other materials		27		Report which of the following are publicly available and where they can be found: template data collection forms; data extracted from included studies; data used for all analyses; analytic code; any other materials used in the review.		Pg 12

Data source

The review assessed the influence of particulate matter (PM10 and PM2.5) on pulmonary tuberculosis occurrence across all age groups, as detailed in Table S2. The selection criteria focused on the question, “What is the risk of developing tuberculosis associated with exposure to ambient particulate matter?” Authors searched PubMed, Cochrane, and Web of Science [Table S3] on January 29, 2024, using terms like “tuberculosis,” “TB,” “air pollution,” and “particulate matter.” Mendeley 1.19.50 software was used to manage citations, remove duplicates, and organize the review process, with results exported to a Microsoft Excel spreadsheet after filtering.

Table S2.

Inclusion and exclusion criteria

	Inclusion			Exclusion
Participants		Diagnosed with Pulmonary tuberculosis		Diagnosed with • Extra pulmonary tuberculosis • Multi drug resistanttuberculosis
Exposure		Air pollutants:- • Ultrafine Particulate matter ≤0.1 µm (UFPM0.1) • Particulate matter ≤2.5 µm (PM2.5) • Particulate matter ≤10 µm (PM10)		• Indoor Air pollutants • Ambient air pollutants including Sulphur dioxide, Nitrogen dioxide, carbon dioxide, carbon monoxide and ozone
Outcome		1. Incidence 2. Hospital admission 3. Mortality
Study Designs		Cross-sectional study, Time-series study, Observational study, case crossover study		Expert opinions, case reports, reviews, case-control and cohort studies, randomize control trials.
		Geography: Global Date of Search: 29-01-2024 Language: English Human studies		Unpublished data
		Published data from 01-01-2014 till 29-01-2024		Published before 01-01-2014

Table S3.

The adjusted search terms used to extract literature from electronic databases

Database		No		Search Query		Results
Date: 29.01.2024
PubMed		#1		((((“Tuberculosis”[Mesh] OR “Tuberculosis, Pulmonary”[Mesh]) OR (Pulmonary TB[Title/Abstract])) OR (Tuberculosis[Title/Abstract])) OR (Mycobacterium tuberculosis[Title/Abstract])) OR (pulmonary tuberculosis[Title/Abstract])		279,778
		#2		((((((((((“Air Pollution”[Mesh]) OR (particulate matter[MeSH Terms])) OR (Ambient air pollution[Title/Abstract])) OR (airborne particulate matter[Title/Abstract])) OR (ambient particulate matter[Title/Abstract])) OR (particulate air pollutants[Title/Abstract])) OR (air pollutants[Title/Abstract])) OR (PM2.5[Title/Abstract])) OR (PM10[Title/Abstract])) OR (ultrafine particles[Title/Abstract])) OR (PM0.1[Title/Abstract])		131,254
		#3		Search: (#1) AND (#2)		740
		#4		(#1) AND (#2) Filters used: 1. Language: English 2. Publications period: 1/1/2014-29/01/2024 3. Species: Humans		259
Cochrane		#1		MeSH descriptor: [Tuberculosis, Pulmonary] explode all trees		1268
		#2		(“tuberculosis”):ti, ab, kw OR (“pulmonary TB”):ti, ab, kw OR (“pulmonary tuberculosis”):ti, ab, kw OR (“Mycobacterium tuberculosis”):ti, ab, kw OR (“TB”):ti, ab, kw		9411
		#3		MeSH descriptor: [Particulate Matter] explode all trees		1286
		#4		(“particulate matter”):ti, ab, kw OR (“ultrafine particle”):ti, ab, kw OR (“air pollutants”):ti, ab, kw OR (“air pollution”):ti, ab, kw OR (“ PM”):ti, ab, kw (Word variations have been searched)		8873
		#5		#1 OR #2		9412
		#6		#3 OR #4		9677
		#7		#5 AND #6		44
		#8		#5 AND #6Filters used: 1. Language: English 2. Publications period: 1/1/2014-29/01/2024 (29 trials 1 editorial)		30
WOS Advanced		#1		TI=((tuberculosis) OR (pulmonary tuberculosis) OR (mycobacterium tuberculosis)) OR AB=((tuberculosis) OR (pulmonary tuberculosis) OR (mycobacterium tuberculosis))		143942
		#2		TI=((air pollution) OR (air pollutant) OR (particulate matter) OR (PM 2.5) OR (PM10) OR (ultrafine particles)) OR AB=((air pollution) OR (air pollutant) OR (particulate matter) OR (PM 2.5) OR (PM10) OR (ultrafine particles))		16102
		#3		#1 AND #2		276
		#4		#1 AND #2 Filters used: 1. Language: English 2. Publications period: 1/1/2014-29/01/2024		218

Data extraction

Two authors (NK and NKE) looked at the titles and abstracts of articles found in the search to see if they were suitable. If they disagreed about including a study in the full-text review, they discussed it to reach an agreement. If they could not agree, they asked the third author (DK) to make the final decision. The two authors (NK and NKE) also examined the full-text publications and extracted relevant information on their own. After extracting the data, they held a meeting to resolve any conflicts among the authors, with the third author (SBS) helping to settle any remaining issues.

Data management

To help with further analysis, a data extraction table was created using Microsoft Excel. The following details were collected from each eligible article: author’s first name, publication year, study design, study duration, particulate size, duration of exposure (long- or short-term), mean PM concentration, data on PTB cases (including incidence, hospital admissions, outpatient visits, and mortality), and effect estimates (relative risk, odds ratio, and percentage change) with a 95% confidence interval. PM concentration was measured in µg/m³. Eligible studies were entered into an Excel spreadsheet after removing ineligible ones [see Table 1]. The analysis used the PRISMA flowchart (2020) and checklist, shown in Figure 1 and Table S1, respectively.

Table 1.

Basic characteristics of the studies included for systematic review and meta-analysis (n=25)

First Author (Year)		Year of publication		Country		Study design		Study Duration		Tb Incidence (n)		Mortality (n)		Hospital Admission (n)
Smith[31]		2014		Carolina		Ecological		1993−2007		3028		NA		NA
Álvaro-Meca[32]		2016		Spain		Case Crossover		1992−2012		45,427		NA		NA
Liu Y[33]		2018		China		Time Series		2011−2015		9344		NA		NA
Zhu S[34]		2018		China		Time Series		2010−2015		36,108		NA		NA
Li Z[22]		2019		China		Time Series		2014−2017		7,282		NA		NA
Tian L[35]		2019		China		Cross-Sectional		2010		NA		NA		2,940
Wang H[36]		2019		China		Time Series		2013−2017		NA		NA		NA
Carrasco-Escobar[11]		2020		Peru		Ecological		2015−2017		28 381		NA		NA
Huang S[23]		2020		China		Time Series		2015−2016		12,648		NA		NA
Kim H[37]		2020		Korea		Time Series		2010−2016		120,280		NA		NA
Liu F[38]		2020		China		Ecological		2006−2016		9183		NA		NA
Yang J[39]		2020		China		Time Series		2013−2017		10238		NA		NA
Wang W[40]		2021		China		Time Series		2014−2018		21,205		NA		NA
Liu Y[41]		2021		China		Time Series		2013−2017		83,555		997		NA
Feng Y[42]		2022		China		Cross-Sectional		2005–2017		653,373		NA		NA
Reyna MA[43]		2022		Mexico		Time Series		2003−2007		1748		NA		NA
Xiong Y[44]		2022		China		Time Series		2014−2019		1746		NA		NA
Zhu S[45]		2022		China		Time Series		2015−2019		NA		NA		NA
Xu M[46]		2022		China		Time Series		1999−2018		NA		NA		NA
Deng X[47]		2023		China		Cross-Sectional		2014−2020		170,934		NA		NA
Popovic I[48]		2023		China		Time Series		2005−2017		38,942		NA		NA
Wang J[17]		2023		China		Time Series		2011−2020		849,676		NA		NA
Wu GH[49]		2023		China		Cross-Sectional		2004−2017		NA		39 216		NA
Zhu PP[50]		2023		China		Case-Crossover		2016−2019		NA		NA		4562
Li JX[51]		2023		China		Cross-Sectional		2014−2016		9759		NA		NA

n=numbers, NA=not available

Figure 1.

PRISMA flow diagram for Systematic Reviews, 2020

Data synthesis and representation

Studies presented various estimates for exposure duration, categorized as long-term (over 7 days) and short-term (7 days or less). We obtained suitable effect sizes using specific criteria: if a multiple-stage trial provided only one effect measure, it was included. For different lags, we used the most frequently cited lag. For short-term exposure, we chose the effect measure linked to the shortest lag, while for long-term exposure, we selected the measure associated with the longest lag. If both single and multiple-pollutant model outcomes were available, we prioritized the single-pollutant model data. We accepted findings from the fully adjusted model when studies reported both crude and adjusted outcomes. Percentage change in excess risk (ER) was converted into relative risk (RR) using this formula[52]:

ER = [RR - 1]×100

The incidence of PTB following exposure to the selected PM size was assessed across multiple studies by converting risk ratios and their confidence intervals into natural log values. Standard errors were derived from these log-transformed values. Combined summary estimates for the log-transformed association measures were calculated and displayed using forest plots. The pooled logarithmic value was then transformed back into a normal value for easier interpretation. Heterogeneity among studies was evaluated using the I² test statistic to understand variability.[53] To examine publication bias, a funnel plot was used for statistical assessments. Additional analyses included the Equivalent test, leave-one-out plot, and Q plot. Subgroup analysis was performed based on data availability. All statistical tests and graphical representations were done using Jamovi version 2.4.14 statistical software.

Quality assessment

Two authors (SBS and DK) appraised each study using JBI’s critical assessment techniques, based on its design.[54] After assessment, publications were categorized as good, fair, or poor quality. A sensitivity analysis was conducted based on the assigned rating [refer to Table S4].

Table S4.

Quality assessment of the included studies by JBI tool (n=25)

Author’s Name (Year)		Year of publication		Q1		Q2		Q3		Q4		Q5		Q6		Q7		Q8		Quality
Smith (2014)		2014		Y		Y		Y		Y		Y		Y		Y		Y		Good
Álvaro-Meca (2016)		2016		Y		Y		Y		Y		Y		Y		Y		Y		Good
Liu (2018)		2018		Y		Y		Y		Y		Y		Y		Y		Y		Good
Zhu (2018)		2018		Y		Y		Y		Y		Y		Y		Y		Y		Good
Li (2019)		2019		Y		Y		Y		UC		Y		Y		Y		Y		Good
Tian (2019)		2019		Y		Y		Y		UC		N		N		Y		Y		Fair
Wang (2019)		2019		Y		Y		Y		Y		Y		Y		Y		Y		Good
Carrasco (2020)		2020		Y		Y		Y		Y		N		N		Y		Y		Fair
Huang (2020)		2020		Y		Y		Y		Y		Y		Y		Y		Y		Good
Kim (2020)		2020		Y		Y		Y		UC		Y		Y		Y		Y		Good
Liu (2020)		2020		Y		Y		Y		Y		Y		Y		Y		Y		Good
Wang (2020)		2021		Y		Y		Y		UC		Y		Y		Y		Y		Good
Yang (2020)		2020		Y		Y		Y		UC		Y		Y		Y		Y		Good
Liu (2021)		2021		Y		Y		Y		Y		Y		Y		Y		Y		Good
Feng (2022)		2022		Y		Y		Y		Y		UC		UC		Y		Y		Fair
Reyna (2022)		2022		Y		Y		Y		Y		N		N		Y		Y		Fair
Xiong (2022)		2022		Y		Y		Y		Y		N		N		Y		Y		Fair
Zhu (2022)		2022		Y		Y		Y		Y		Y		Y		Y		Y		Good
Deng (2023)		2023		Y		Y		Y		UC		Y		Y		Y		Y		Good
Popovic (2023)		2023		Y		Y		Y		Y		Y		Y		Y		Y		Good
Wang (2023)		2023		Y		Y		Y		Y		UC		UC		Y		Y		Fair
Wu (2023)		2023		Y		Y		Y		Y		N		N		Y		Y		Fair
Zhu (2023)		2023		Y		Y		Y		Y		Y		Y		Y		Y		Good
Xu (2022)		2022		Y		Y		Y		Y		Y		N		Y		Y		Good
Li (2023)		2023		Y		Y		Y		Y		Y		Y		Y		Y		Good

Y: Yes, N: No, NA: Not Applicable, UC: Unclear. Q1:Were the criteria for inclusion in the sample clearly defined? Q2:Were the study subjects and the setting described in detail? Q3:Was the exposure measured in a valid and reliable way? Q4: Were objective, standard criteria used for measurement of the condition? Q5: Were confounding factors identified? Q6: Were strategies to deal with confounding factors stated? Q7: Were the outcomes measured in a valid and reliable way? Q8: Was appropriate statistical analysis used?

RESULT

Search insights

Following a thorough systematic search, 507 articles were initially identified out of 127 duplicates were found. Figure S1 ^{(2.8MB, tif)} illustrates the distribution of studies from each database and highlights overlaps between databases. Following the screening of titles and abstracts, and subsequent review of full texts, the systematic review included 25 papers that met the inclusion criteria. The article review and selection processes were depicted in the PRISMA flowchart [Figure 1]. The methodological quality assessment revealed that 18 articles were deemed to be of good quality, while 7 were rated as fair. The comprehensive scores of the suitable articles have been presented in Table S4.

Study characteristics

In the 25 selected studies, 15 were time-series studies,[17,22,23,33,34,36,37,39,40,41,43,44,45,47,48] five were cross-sectional studies,[35,42,46,49,51] three ecological study,[11,31,38] and two were case-crossover studies.[32,50] The included studies collectively reported a total of 2,084,476 cases of PTB incidence, 997 deaths attributed to PTB, and 7,502 hospital admissions due to PTB. Of the studies, 20 provided data on PTB incidence, while three focused on hospital admissions. Additionally, one study reported mortality rates related to PTB, and another one study presented both incidence and mortality data. Four articles examined PTB incidence with exposure to PM10 only, seven studied exposure to PM2.5 only, and 10 explored exposure to both PM10 and PM2.5 separately. Details of the study and participant characteristics are summarized in Table 1.

Analysing PTB incidence

The association between PM2.5 and PTB incidence

In total, 17 articles reported incidence of PTB in relation to PM2.5 exposure for systematic review.[11,17,22,23,31,33,36,39,41,42,43,44,45,46,47,48,51] For the meta-analysis, 12 studies were included, reporting RR (95% CI) for an increase of 10 µg/m³ of PM2.5. Among them, eight studies indicated a long-term association[11,22,31,36,43,45,47,48] while one study suggested a short-term association,[23] and three studies reported both long- and short-term associations.[33,42,46] Another five studies were there that had different effect estimate having different value of increment of PM2.5 [Table 2].[17,39,41,44,51]

Table 2.

Effect estimate of pulmonary tuberculosis incidence in correlation with increment in air pollutant concentrations

First Author (Year)		Particulate Matter Size		Increment in air pollutant concentration (µmg/m3)		Effect estimate (95% CI)
Kim H[37] (2020)		PM10		5.63		RR: 1.17 (1.13−1.20)
Yang J[39] (2020)		PM2.5		50		RR: 0.9993 (0.9977−1.0010)
		PM10		50		RR: 0.9996 (0.9987−1.0004)
Liu Y[41] (2021)		PM2.5		1		Percentage change: 3.04 (2.98, 3.11)
Xiong Y[44] (2022)		PM2.5		50		RR: 3.101 (1.096–8.777)
Wang J[17] (2023)		PM2.5		-		RR: 1.17 (1.10−1.24)
Li JX[51] (2023)		PM2.5		7.78		OR: 1.11 (1.09−1.13)
		PM10		17.3		OR: 1.03 (1.01−1.05)

PM2.5=particulate matter <2.5 µm, PM10: particulate matter <10 µm, µm=micrometre, µm/m³=microgram per metre cube, CI=confidence interval, RR=relative risk, OR=odds ratio

Long-term exposure: The pooled effect estimate for PTB incidence with a 10 µg/m³ increase in long-term PM2.5 exposure was RR =1.26 (95% CI: 1.07–1.48, P = 0.005), using a random-effects model. Significant heterogeneity was observed [I² =99.92%, Figure 2]. Leave-one-out analysis and Q-Q plot showed no major changes in pooled effect or heterogeneity. Subgroup analysis by gender showed a higher risk in males (RR = 1.15, 95% CI: 1.00–1.30, P < 0.001) and in females (RR = 1.29, 95% CI: 1.29–1.74, P = 0.038; Figure S1 ^{(2.8MB, tif)} ) but did not reduce overall heterogeneity.

Figure 2.

Forest plot of the analysis of the long-term effect of ambient particulate matter 2.5 on pulmonary tuberculosis incidence

Short-term exposure: For a 10 µg/m³ rise in PM2.5, the combined effect estimate for PTB incidence was RR = 1.10 (95% CI: 0.98–1.26, P < 0.001), using a random-effects model. Significant heterogeneity was observed [I² = 87.86, Figure 3].

Figure 3.

Forest plot of the analysis of the short-term effect of ambient particulate matter 2.5 on pulmonary tuberculosis incidence

The association between PM10 and PTB incidence

Fourteen articles examined the incidence of PTB in relation to PM10 and were eligible for systematic review.[22,23,31,38,43,45,47,34,36,37,39,40,46,51] Eleven studies were included in meta-analysis, reporting RR (95% CI) for an increase of 10 µg/m³. Among these eight reported long-term effects[22,31,34,36,38,43,45,47] one reported short-term effects,[23] and two reported both long- and short-term effects.[40,46] Three additional studies[37,39,51] reported on PM exposure and the incidence of PTB, presenting diverse effect estimates at varying increment units, as detailed in Table 2.

Long-term exposure: The pooled effect estimate for PTB incidence with a 10 µg/m³ increase in long-term PM10 exposure was RR =1.07 (95% CI: 1.02–1.12, P = 0.002), using a random-effects model. Substantial heterogeneity was observed [I² = 98.48%, Figure 4]. Leave-one-out analysis and the Q-Q plot showed no major changes in pooled effect or heterogeneity. Gender-based subgroup analysis did not reduce heterogeneity but indicated higher PTB risk for males (RR =1.10, 95% CI: 1.00–1.23) and females [RR =1.07, 95% CI: 1.02–1.15; Figure S2 ^{(3.1MB, tif)} ].

Figure 4.

Forest plot of the analysis of the long-term effect of ambient particulate matter 10 on pulmonary tuberculosis incidence

Short-term exposure: For a 10 µg/m³ increase in PM10, the combined effect estimate for PTB incidence was RR =1.07 (95% CI: 0.95–1.17, P = 0.002), using a random-effects model. Significant heterogeneity was noted [I² = 79.25%, Figure 5].

Figure 5.

Forest plot of the analysis of the short-term effect of ambient particulate matter 10 on pulmonary tuberculosis incidence

Publication Bias: Publication bias was observed in studies examining the long-term exposure to PM2.5 and PM10 in relation to PTB incidence. Evidence of publication bias was indicated by the asymmetrical funnel plots for PM2.5 and PM10 [Figure 6]. Furthermore, equivalence testing revealed significant findings for both PM2.5 [Figure S3 ^{(1.6MB, tif)} ] and PM10 [Figure S4 ^{(1.6MB, tif)} ]. However, publication bias for the short-term association between particulate matter (PM2.5 and PM10) and PTB incidence could not be assessed due to the limited number of studies available.

Figure 6.

Funnel plots of long term effect of ambient particulate matter (PM2.5 and PM10) on Pulmonary tuberculosis incidence

Analysing PTB hospital admission and mortality

Three studies examined the impact of PM exposure on hospital admissions for PTB cases[32,35,50] while two studies focused on PTB mortality.[41,49] Two of these studies found a significant correlation between PM exposure and hospital admissions. One of these study found that one inter quartile range (IQR) increase in PM2.5 and PM10 concentrations was associated with a 15.5% and 14.2% increase in hospital admission risks for PTB, respectively.[50] Another study revealed a favourable relationship between PM10 levels and tuberculosis hospitalizations, with a RR of 1.010.[35] Conversely, another study reported an insignificant association with an odds ratio (OR) of 0.97.[32] Regarding mortality due to PTB associated with PM exposure, one study found a significant association with an incidence rate ratio (IR%) of 0.742,[49] while another study found no association with an RR of 0.00.[41]

DISCUSSION

This meta-analysis included 25 studies that reported a total of 2,084,476 cases of PTB incidence, 997 deaths from PTB, and 7,502 hospital admissions due to PTB. The findings suggest that prolonged exposure to both PM2.5 and PM10 increases the likelihood of PTB. Specifically, for each 10 µg/m³ rise in PM2.5, the risk of PTB increased by 26%. For every 10 µg/m³ rise in PM10, the risk of PTB increased by 7%. This indicates that exposure to ambient particulate matter significantly raises the risk of PTB, with PM2.5 presenting a higher risk than PM10.

In cases of short-term exposure, the analysis showed that both PM2.5 and PM10 also raised the risk of PTB. For each 10 µg/m³ increase in PM2.5, the risk of PTB rose by 10%. Similarly, for PM10, a 10 µg/m³ increase also led to a 7% rise in PTB risk. A recent time series study from 2023, which looked at 849,676 PTB cases in 22 cities in China, found that people exposed to high levels of PM2.5 had a 26% higher risk of developing TB.[17] Another time series study from 2020 found that a 10 µg/m³ increase in PM2.5 and PM10 raised the risk of PTB by 17.03% and 11.08%, respectively.[23]

However, in both long and short term exposure, the results highlight considerable heterogeneity and potential publication bias. A 2016 nested case-control study of 6,913 participants found no consistent association between PM10/PM2.5 levels and pulmonary tuberculosis incidence.[55] Conversely, a 2016 prospective cohort study of 106,678 participants found a significant association between PM2.5 and TB incidence (adjusted hazard ratio: 1.39 per 10 μg/m³),[56] highlighting the need for further rigorous research on PM and active pulmonary TB. Meta-analysis of these high-evidence studies can yield clearer insights than time series or cross-sectional studies, aiding in more effective public health guidance.

Subgroup analyses by gender revealed varying effect estimates and heterogeneity levels, indicating gender-specific differences in response to particulate matter. Males had a higher PTB risk with PM10 exposure, while females showed a greater risk with PM2.5 exposure. Similar findings in recent studies indicated that increased PM2.5 exposure raised incidence risk for men (RR =1.013), with no effect in women; males also faced greater risk with PM10 exposure compared to females.[47]

Beyond ambient particulate matter, PTB can arise from various environmental factors, with air pollutants like traffic emissions linked to respiratory, cardiovascular, and neurological diseases across all ages.[57,58,59] These pollutants weaken the body’s defenses by disrupting respiratory barriers, inhibiting mucociliary clearance, impairing macrophages, and promoting chronic inflammation, increasing vulnerability to infections like tuberculosis. Particulate matter, in particular, hinders immune control of mycobacterial infections, heightening tuberculosis susceptibility and severity.[60]

Other factors such as diabetes, smoking, alcohol use, and drug use also raise TB risk.[61,62,63] One study linked each additional year of solid fuel exposure to a 3% increase in active TB history, while another found TB odds nearly five times higher in urban slums.[64,65] Reducing air pollution is essential to lowering respiratory and infectious disease risks, highlighting the importance of environmental policies and public health interventions for global health.

Limitations

The findings of the meta-analysis should be interpreted carefully because of publication bias. This bias is shown by asymmetrical funnel plots, which indicate a lack of studies with null or negative results. The subgroup analyses were limited and did not explore all possible effect modifiers. Including time series and cross-sectional studies, which provide lower-quality evidence than cohort studies and RCTs, may also introduce biases. The SRMA index used only three databases (PubMed, Cochrane, and Embase) due to resource limits. These limitations highlight the need for careful interpretation and more research. Future meta-analyses should include unpublished data and studies with null or negative findings to reduce publication bias.

CONCLUSION

The study reveals that on increasing 10 µg/m³ of both PM10 and PM2.5 incidence increases the risk of pulmonary tuberculosis. These findings underscore the urgent need for comprehensive air quality management strategies to mitigate the adverse health effects posed by particulate matter pollution and to safeguard public health against tuberculosis and related respiratory ailments. Moreover, fostering further research and encouraging high-evidence studies are imperative steps in addressing this urgent public health concern. Concerted efforts towards pollution reduction are essential to mitigate these risks and safeguard the well-being of communities.

Conflicts of interest

There are no conflicts of interest.

SUPPLEMENTARY FILE

Research Question: What is the risk of developing pulmonary tuberculosis on exposure to ambient particulate matter?

Figure S1

Forest Plot of subgroup analysis for long term effect of ambient particulate matter (PM2.5) on Pulmonary Tuberculosis incidence

Figure S2

Forest Plot of subgroup analysis for long term effect of ambient particulate matter (PM10) on Pulmonary Tuberculosis incidence

Figure S3

Equivalence Test Plot for long term effect of ambient particulate matter (PM2.5) on Pulmonary Tuberculosis incidence

Figure S4

Equivalence Test Plot for long term effect of ambient particulate matter (PM10) on Pulmonary Tuberculosis incidence

Funding Statement

Nil.