Global Epidemiology of Nontuberculous Mycobacterial Pulmonary Disease: A Review

Introduction

In this chapter we review approaches to studying the epidemiology of NTM pulmonary disease (NTM-PD), and update advances in the field since the last review published in 2015 ¹.

The importance of epidemiology: surveillance and research

Defining the NTM-PD disease burden is critical for justifying the need for resource allocation, both for clinical resources such as health care, as well as for the development of new therapeutics. A fuller understanding of the risk factors which contribute to changes in disease frequency is important for guiding and evaluating interventions, at either the individual-level through modification of behaviors, or through structural interventions beyond the individual. Epidemiologic surveillance is key for estimating the burden of disease (prevalence) as well as the frequency of new cases (incidence) in a given population ², which in turn can lead to developing hypotheses regarding individual or general environmental risk factors for infection and disease ^{3, 4}.

We here present a new approach for understanding the external and internal exposures which may lead to increased risk of NTM disease ⁵ (Figure 1). This paradigm is adapted from the fields of environmental epidemiology ⁵ and cancer epidemiology ⁶, whereby cumulative exposures to a variety of types of exposures will increase the risk of infection and disease. With respect to NTM-PD, specific environmental factors include those in an individual’s environment, such as those exposures from high-risk activities such as gardening, with aerosolization of soil. General environmental exposures include those beyond the control of a single individual which impact an entire population. Both factors interact with the human host, and host susceptibility, including biologic response, will influence the risk of developing NTM-PD ⁵ (Figure 1).

Figure 1.

NTM exosome framework. Adapted from The exposome: a new paradigm to study the impact of environment on health, Vrijheid M., 69, 876–8, 2014 with permission from BMJ Publishing Group Ltd.

Methodologic challenges

A key methodologic challenge is the lack of a global surveillance system with standard case definitions that would allow comparison within and across countries and regions. The NTM-PD case definition which includes microbiologic, radiographic, and clinical criteria, was developed by ATS/IDSA in 2007 for diagnostic and treatment purposes ⁷, and endorsed in the more recent ATS/ERS/ESCMID/IDSA guidelines ⁸. In this review we will use the simplified term “ATS-criteria” acknowledging the equal contribution of the other societies. However, different case definitions may be used for epidemiologic purposes of monitoring infection rates and risk factor identification, as the decision about whether and when to treat is different from the surveillance goal of identifying patterns of infection and disease.

Ascertaining the radiographic and clinical criteria for NTM-PD is generally time consuming and costly, and therefore not feasible at the scale needed to study national and regional patterns, particularly in areas with a higher disease burden. Therefore, microbiology data from centralized public health laboratory systems, as well as administrative claims data with International Classification of Disease (ICD) codes have been increasingly used to better understand the epidemiology of NTM-PD. Each approach has strengths and limitations. Microbiologic data provide an indicator of exposure from the environment, but will overestimate the true disease prevalence, as not all isolates from respiratory secretions are clinically significant. In this review we use the term “isolation” or “isolates” to refer to culture positivity of respiratory specimens for NTM without necessarily attributing these isolates to a disease status in a specific patient. We therefore also avoid the term “colonization” which suggests “no disease”, a status which has not been well studied. Although the term “NTM infection” has often been used to refer to cases which satisfy the ATS microbiologic criteria ^9–14 we generally avoid use of this term as the definition of this term has not been standardized.

The microbiologic component of the ATS criteria, as well as ICD codes (9 or 10) have been evaluated and found to have a high predictive value for NTM-PD (meeting the full ATS diagnostic criteria). In a large, population-based cohort from a centralized provincial laboratory in Ontario, Canada, 46% of those with an isolate of NTM met the microbiologic component of the ATS disease diagnostic criteria ¹⁴. Within the subset of patients being followed in a clinic setting, 73% of the patients who fulfilled the ATS microbiologic criteria met the full ATS disease criteria, indicating a high predictive value ¹⁴. Similarly, in a large national study using NTM isolates from the national referral laboratory in Denmark, among a small subset with isolates meeting ATS microbiologic criteria (termed “possible NTM disease”), 90% met the full ATS disease criteria ¹³. In a study in North Carolina, USA, based on mycobacterial isolation and full chart review, 55.7% of those with an isolate met the ATS microbiologic criteria, and 60.9% of these met the full criteria including microbiologic, clinical, and radiographic criteria ¹⁵. Thus, although analysis of isolates is important for understanding patterns and trends, case definitions based on 1 or 2 isolates will usually overestimate the burden of ATS defined NTM-PD.

ICD codes provide a more useful indicator of true disease, although they will tend to underestimate true disease prevalence, as administrative codes tend to lack sensitivity for rare disease. The sensitivity of the ICD codes relative to the ATS microbiologic criteria has been found to range from 27% ¹⁶ in a general population to 50% among persons with rheumatoid arthritis ¹⁷ and up to 69% among persons with bronchiectasis ¹⁸; the positive predictive value across various studies has ranged from 77% to 100% ^17–19. Although these codes may underestimate true disease rates, the direction of this bias is unlikely to change substantially across time and populations, such that these codes are useful tools for epidemiologic purposes.

Two noteworthy examples of implementation of NTM-PD surveillance are from Queensland, Australia, and the state of Oregon, USA. In Queensland, laboratory-based notification of NTM isolates has been mandated since the inception of the Tuberculosis control program in the 1950’s, and all cases of NTM isolation are notifiable under the Public Health Act ³. Data from this surveillance system has since facilitated the characterization of NTM-PD epidemiology, particularly with respect to geographic distribution, environmental risk factors, and increasing burden ³. In the state of Oregon, USA, a pilot surveillance program was implemented from 2005–2006 and provided important insights regarding NTM epidemiology ^{4, 19}. Currently, in the United States, only 4 states have mandated notification of NTM-PD, as part of a Centers for Disease Control and Prevention (CDC) through an Emerging Infections Network (EIN) pilot program ^{20, 21}.

Despite the known limitations of death certificate data, in the United States, recent data have shown an increase in non-HIV associated NTM-related deaths from 1999–2014, during a period in which both HIV-associated NTM-related mortality decreased, and TB-associated mortality also decreased significantly ²². In Japan, in the absence of nationally representative data, mortality data were useful for providing insight into patterns by age, sex, and region, and for estimating prevalence ²³. Several studies have identified the increased risk of mortality associated with NTM-PD ^{24, 25} In one study, persons with NTM-PD had a fourfold increased risk of death after adjusting for all other factors (HR=3.64) ²⁶.

Risk factors: Environment

General and specific environmental risk factors identified to date are summarized in Table 1. Specific exposures refer to those estimated from studies which associate individual behaviors or household factors with human NTM pulmonary isolation or disease. General exposures refer to those not unique to a single individual but rather which affect the broader population.

Table 1:

Host and Environmental Risk factors for NTM infection and disease

Risk Factor	Relative risk, Odds ratio, Relative Prevalence, or other measure	Case Definition
Environmental: specific individual exposures
NTM isolation from shower aerosols, Oregon, USA	4.0 [OR]27	MAC-PD, case-control study
Use of spray bottle to spray house plants, Oregon, USA	2.7 [OR]29	MAC-PD, case-control study
Use of Public Baths ≥ 1/week, S Korea	4.0 [OR]30	MAC-PD, case-control study
Indoor swimming pool use (in the past four months), USA	5.9 [OR]31	Incident pulmonary isolation persons with CF (pwCF)
Soil exposure, Japan (bronchiectasis with MAC vs bronchiectasis with no MAC)	5.9 [OR]33	MAC-PD, case-control study
Environment- general exposures, measured at population level: climate, soil, and water- related
Relative abundance of NTM in showerhead biofilm and population prevalence	Significant correlation, p<0.000135	NTM-PD and NTM pulmonary isolation (Medicare with ICD codes; CF ≥ 1 pulmonary isolate)
Higher average annual precipitation, FL, USA	1.34130	Incident pulmonary isolation (≥ 1 isolate) among persons with CF
Increased soil sodium, FL, USA	1.92130	Incident pulmonary isolation (≥ 1 isolate) among persons with CF
Increased soil manganese	0.59130	Incident pulmonary isolation (≥ 1 isolate) among persons with CF
Increased molybdenum concentrations in source water	1.4112	M. abscessus incident pulmonary isolation (≥2 isolates), non-CF
	1.7911	M. abscessus pulmonary isolation in persons with CF (≥1 isolate)
Increased vanadium concentrations in source water	1.4912	Incident MAC isolation (≥2 isolates), non-CF
	1.2236	Incident MAC pulmonary isolation, non-CF (≥2 isolates)
Percent hydric soil in census blocks, NC, USA	26.838	NTM pulmonary isolation (≥ 1 isolate), non-CF
Proportion of area as surface water	4.639	NTM-PD (ICD codes)
Mean daily potential evapotranspiration	4.039	NTM-PD (ICD codes)
Copper soil levels, per 1 ppm increase	1.239	NTM-PD (ICD codes)
Sodium soil levels, per 0.1 ppm increase	1.939	NTM-PD (ICD codes)
Manganese soil levels, per 100 ppm increase	0.739	NTM-PD (ICD codes)
Increased average topsoil depth	0.87 (M. intracellulare)42	NTM isolation (≥ 1 isolate), non-CF
Soil bulk density	1.8 (M. kansasii)42	NTM isolation (≥ 1 isolate), non-CF

^*Estimated from data in paper

^#Hazard ratio, fully adjusted for age, sex, income, rurality, and comorbidities for NTM (HIV, COPD, asthma, and GERD)

⁺Comparison of patients with NTM/TB with uninfected anti-TNF users

^**risk decreased with increasing duration since use; see full paper for specific estimates

Case-control studies in the United States, South Korea, and Japan have identified individual factors related to water and soil exposure. A recent study in Oregon, United States, found that the NTM isolation in the household shower aerosols was significantly associated with NTM-PD ²⁷, but that other home water and soil exposure were not. This finding is supported by a study in Queensland, Australia which found genetic matches between patient isolates and species identified from shower aerosols and household water supplies ²⁸. In the initial Oregon study, the only household-level factors significantly associated with NTM-PD risk was spraying plants with a spray bottle [OR=2.7] ²⁹. In South Korea, use of public baths at least weekly was associated with a fourfold increased risk of NTM-PD ³⁰. In the United States, a national study of NTM-PD among persons with CF found that indoor swimming pool use at least monthly was significantly associated with incident NTM isolation ³¹ (Table 1), and that other behavioral factors, including showering frequency, were not. A study among children with CF in Florida found that those who lived in households which were within 500 meters of a body of water had a significant 9.4-fold increased odds of having NTM pulmonary isolation ³². In Japan, exposure to soil more than twice weekly was significantly associated with the NTM-PD ³³. The high risk associated with frequent soil exposure is consistent with a population-based study of agriculture workers in Florida, which found that cumulative occupational exposure was significantly associated with infection, as defined by M. avium skin test sensitivity ³⁴. The risk associated with any given behavior will depend on the intensity of the exposure and the NTM abundance in environmental exposure source–generally soil or water. The risk associated with household NTM isolation from shower aerosols in the NTM-PD study in Oregon is supported by a national study of shower biofilms, which found a significant correlation between the relative abundance of NTM in showerhead biofilms and state-level NTM-PD prevalence in both the CF population and the US Medicare beneficiary population aged ≥65 years ³⁵.

Ecologic epidemiologic studies have found water, soil, and climate factors associated with an increased risk of NTM-PD; these studies have been conducted primarily in the United States and Australia (Table 1). In several studies conducted in the United States from 2020–2022, a consistently significant and strong association has been found between concentrations of molybdenum and M. abscessus infection, and vanadium and MAC infections in natural water sources (ground or surface) supplying municipal water systems: for every log increase in molybdenum concentrations the risk of M. abscessus isolation increased by 41% ¹² to 79% ¹¹; for every log increase in the concentration of vanadium, the risk of MAC isolation (≥ 2 isolates) increased 22% ³⁶ to 49% ¹² (Table 1). Interestingly, a case-control study in South Korea found NTM-PD patients had significantly higher median blood serum concentrations of molybdenum than controls ³⁷.

Several other studies have identified risks related to soil, land use, and other climate factors. In North Carolina, USA, the percent hydric soil in census blocks was significantly associated with NTM isolation: census blocks with >20% hydric soils had a significant 26.8% increased adjusted mean patient count relative to those with ≤20% hydric soils ³⁸. In Florida, in the CF population, increased soil sodium in the zip code of residence was associated with an increased risk of incident NTM isolation, whereas increased soil manganese was protective ³⁹. In the same study, persons living in counties with above average annual levels of precipitation were also at increased risk of incident infection ³⁹. In two national studies in the United States, vapor pressure (a measure of the water in the atmosphere at a given temperature) was predictive of disease prevalence among CF patients ³¹ as well as Medicare beneficiaries aged ≥65 years ^{31, 40}. In addition, in the national Medicare study, evapotranspiration (the potential of the atmosphere to absorb water), and the proportion of the area as surface water were predictive of high-risk areas ³⁹. In the Medicare study, higher manganese soil levels in the county were associated with lower incidence of NTM-PD, whereas higher copper and sodium levels in soil were associated with increased risk ³⁹. In Queensland, Australia, an effect of temperature and rainfall was observed, but this varied by region: cyclic incidence patterns were associated with temperature and rainfall ⁴¹. In a prior study in Australia, increased average topsoil depth was protective against M. intracellulare, while increased soil bulk density was positively associated with M. kansasii disease ⁴². In Japan, NTM-PD mortality, a surrogate measure for disease prevalence and distribution, was higher in the warmer and more humid coastal areas and in an area with a large amount of surface water (lake) ²³. A recent study found dust bioaerosols as a source of NTM, with higher relative abundance in East Asian inland cities in Japan and China than in desert areas ⁴³.

Epidemiology of NTM Isolation and Disease by Global Region

In our review of epidemiological data of NTM-PD, we strive to include the highest quality data published since 2014. Ideal studies comprise population-based investigations including both clinical and microbiologic data and encompassing an adequate temporal period to best capture the burden of infection in the population. However, because these types of data are often not available, other data sources are used as described previously.

Systematic Review and Meta-analysis – Global

A systematic review and meta-analysis was recently conducted for global culture-based microbiologic data, indicating an overall increase worldwide in the frequency of isolation, including that which satisfies the ATS microbiologic criteria⁹. This review included only those studies with culture-based data for at least three years, and with at least 200 samples, representing findings from 47 publications in more than 18 countries. Overall, 82% of studies reported increasing isolation, and 66.7% reported increasing disease, using either the adapted ATS microbiologic criteria or the full radiographic and clinical criteria. The overall rate of increase was 4% (3.2–4.8) per year for isolation and 4.1% (3.2–5) per year for disease, which was most often defined using the ATS microbiologic criteria. Most of these studies focused on MAC, the predominant species, and to a lesser degree the M. abscessus group⁹ (Figure 2).

Figure 2.

a. Forest plot of annual change of NTM infection per 100,000 persons/year. From Dahl VN, Mølhave M, Fløe A, et al. Global trends of pulmonary infections with nontuberculous mycobacteria: a systematic review. Int J Infect Dis. Oct 13 2022;125:120–131. doi:10.1016/j.ijid.2022.10.013; with permission. (Figure 3 in original) b. Forest plot of annual change of NTM disease per 100,000 persons/year (Dahl et al., 2022); with permission. (Figure 6 in original)

A recent Delphi survey including a physician survey as well as chart review allowed comparison of prevalence rates across 4 European countries (France, Germany, Spain, UK) and Japan, and found a remarkably similar prevalence of NTM-PD of 6.1/100,000 to 6.6/100 000 across European countries, but with a fourfold increased prevalence in Japan, suggesting true differences in prevalence ⁴⁴.

North America

In North America, national and subnational studies indicate increasing prevalence and incidence, and continue to demonstrate geographic heterogeneity in NTM isolation and disease. This section is based on five national studies from the United States, one provincial level study from Ontario, and nine state or territorial level studies in the United States. The national studies include three based on ICD codes to define disease, and two based on microbiologic data. The heterogeneity of methodologic approaches makes comparisons difficult, but overall, these studies indicate a picture of increasing prevalence dominated by MAC infections.

In the United States, a national study to estimate prevalence and incidence of NTM-PD defined this as two claims with NTM-PD ICD codes separated by 30 days. This study estimated a prevalence of 6.8/100,000 in 2008, increasing to 11.7/100,000 in 2015, with an estimated annual incidence from 3.1/100,000 to 4.7/100,000 during the same time period ⁴⁵ (Figure 3). A separate study which estimated prevalence using a single ICD code for NTM to define a case estimated a prevalence of 27.9/100,000 in 2010, with a projected estimate of 181,037 cases in 2014, assuming a continued 8% annual prevalence increase ⁴⁶. The discrepancy in estimates between the two studies for a similar year (2014/2015) is due to the more specific case definition in the former study. The geographic heterogeneity across states was consistent with prior reports and historic patterns, with the highest prevalence in the warm, humid areas in the Southeast and Southwest ⁴⁶. A study in a high risk population, among veterans with COPD, also used two ICD codes for NTM-PD to define disease and found an increasing incidence and prevalence NTM-PD from 2001 to 2015, with incidence increasing from 34.2/100,000 to 70.3/100,000 and prevalence from 93.1/100,000 to 277.6/100,000 patients ⁴⁷.

Figure 3.

Annual incidence of NTM-PD in national U.S. health insurance plan (Optum EHR database) 2008–2015. Reprinted with permission of the American Thoracic Society. Copyright © 2023 American Thoracic Society. All rights reserved. Cite: Winthrop KL, Marras TK, Adjemian J, Zhang H, Wang P, Zhang Q./2020/Incidence and Prevalence of Nontuberculous Mycobacterial Lung Disease in a Large U.S. Managed Care Health Plan, 2008–2015/Ann Am Thorac Soc./17/178–185. Annals of the American Thoracic Society is an official journal of the American Thoracic Society. (Figure 2 in original)

Two national microbiologic studies based on electronic health records demonstrated geographic variation in NTM species as well as increasing trends in AFB testing and NTM isolation prevalence in high-risk groups ^{48, 49}. Among 5 million unique patients of whom 7,812 had at least one isolate, MAC was the most common species, ranging from 61 to 91% of isolates, and was most frequent in the South and Northeast regions; M. abscessus/M. chelonae ranged from 2% to 18% of isolates and were most frequent in the West. A separate national study using microbiology data from a different large EHR system to study frequency of AFB testing and pulmonary NTM isolation during 2009–2015 found an overall AFB testing rate of 45/10,000 population with an increasing annual percent change (APC) of 3.2% ⁴⁹. The isolation rate for pathogenic NTM also increased with an APC of 4.5% and was highest among persons with CF and those with bronchiectasis ⁴⁹.

One subnational study used population-based surveillance data from 2014 in four states (Mississippi, Missouri, Ohio, and Wisconsin) to describe NTM isolation. Overall prevalence was 13.3/100,000 population (after excluding M. gordonae), with a predominance of MAC, at a prevalence of 8.5/100,000 ⁵⁰. An earlier study which estimated trends from 2008–2013 in the same four states plus Maryland found an increasing trend, with an APC of 9.9% ⁵¹.

Hawaii and Florida have been identified as a ‘hotspots’ for NTM in the United States. In Hawaii, microbiologic data from a large, linked health care system were used to associate microbiologic data with demographic and clinical factors for the period 2005 and 2013. Isolation was based on at least one pulmonary isolate, and the ATS microbiologic criteria were used to define NTM-PD ⁵². Isolation prevalence increased significantly during the study period, from 20/100,000 in 2005 to 44/100,00 in 2013, with an APC of 6% and a period prevalence of 122/100,000; MAC was the most commonly isolated species, comprising 64% of isolates. NTM-PD (ATS microbiologic criteria) increased from 9/100,000 in 2005 to 19/100,000 in 2013 (Table 2). NTM isolation prevalence varied by ethnic group, with NTM isolation rates approximately twofold higher among persons who identified as Chinese, Korean, Japanese, and Vietnamese relative to those who identified as white, and the prevalence was twofold lower among Native Hawaiian and other Pacific Islander (NHOPI) relative to whites ⁵² (Table 3). Another study used KPH data to estimate NTM infection incidence by ethnic group during the period 2005–2019 (Table 2) ⁵³. Cases were defined as ≥ 1 pulmonary isolate. Average annual isolation incidence was 44.8 cases/100,000 beneficiaries, with a cumulative incidence of 247 cases/100,000 over the study period. Beneficiaries who self-identified with only Asian ethnic groups had the highest NTM pulmonary isolation incidence (46 cases/100,000 person-years) and had a 30% increased risk after controlling for all other clinical and demographic factors ⁵³. A separate study from the US Affiliated Pacific Islands (USAPI), based on samples submitted to the Diagnostic Laboratory Services in Hawaii, typically for evaluation of suspected TB, found a significantly increasing rate of NTM isolation, increasing from 0.5% of isolates screened in 2007 to 11.3% in 2011, corresponding to an NTM isolation prevalence increase of 2/100,000 to 48/100,000. MTB isolation remained stable during the same period ⁵⁴.

Table 2:

Studies of Rates of Pulmonary NTM Isolation and Disease by Region

North America
Location (dates)	Case Definition	Data Source/Cohort	Annual Prevalence (per 100,000 population)		Period Prevalence		Incidence (per 100,000 population)
			Isolation (dates)	Disease (dates)	Period Duration (years)	Prevalence (per 100,000 population)
Canada
Ontario, Canada (1998–2010)131	Isolation: ≥ 1 isolate Disease: ATS microbiologic criteria	Public Health Ontario Laboratory Database	11.4–22.2 (1998–2010) (APC: 6.3%); Annual Change (MAC: 0.291, M. xenopi: 0.059, M. abscessus: 0.019)	4.65 – 9.08 (1998–2010) (APC: 8.0%)+**	-	-	-
United States
U.S. (2008–2015)45	ICD-9/ICD-10	Optum Clinformatics Data Mart Care Claims database	-	6.78–11.70 (APC: 7.5%) (2008–2015)	8	-	3.13–4.73 (2008–2015) (APC: 5.2%) (D)
Florida (2012–2018)55	ICD-9-CM/ICD-10	OneFlorida Clinical Research Consortium database	-	14.3–22.6 (2012–2018)	7	-	6.5–5.4 (2012–2018) (D)
Hawaii (2005–2013)52	Isolation: ≥ 1 isolate Disease: ATS microbiologic criteria	Kaiser Permanente Hawaii databases	20–44 (2005–2013) (APC: 6%)**	9–19 (2005–2013)+**	9	Range by Ethnic group: 50–300 (I)**	-
Hawaii (2005–2019)53	≥ 1 isolate	Kaiser Permanente Hawaii EHR databases	-	-	15	-	Overall Annualized: 44.8 (I)**, Range by Ethnic group: 17–63 (I)**, Cumulative: 247 (I)**
Maryland, Mississippi, Missouri, Ohio, Wisconsin (2008–2013)51	≥ 1 isolate	Surveillance data from Maryland, Mississippi, Missouri, Ohio, Wisconsin	Overall: 8.7–13.9 (2008–2013)§# (APC: 9.9%)§#, Maryland: 9.5–9.29 (2008–2013)§# Mississippi:10.2–13.69 (2008–2013)§#, Missouri: 5.5–13.39 (2008–2013)§#, Ohio: 5.8–11.39 (2008–2013)§#, Wisconsin: 15.4–24.69 (2008–2013)§#	-	-	-	-
North Carolina (2006–2010)58	≥ 1 isolate	All labs that test for NTM in 3 counties (hospital, commercial, public health)	Overall: 9.4**	-	-	-	-
North Carolina (2006–2010)38	≥ 1 isolate	All labs that test for NTM in 3 counties (hospital, commercial, publica health	-	-	5	-	Cumulative: 8.8 (I)
Oregon (2007–2012)10	ATS microbiologic criteria	All laboratories that test for NTM in Oregon	-	5.9 (2011–2012)+	6	-	5.0 (APC: 2.2%) (D)+; 3.8 (D)§+
U.S.–Affiliated Pacific Island Jurisdictions (2007–2011)54	≥ 1 isolate	Diagnostic Laboratory Services data	2–48 (2007–2011) (adjusted rate ratio 1.65)	-	4	Overall: 106 (I); American Samoa (22) (I), Federated States of Micronesia (164) (I)	-
Central and South America
Location (dates)	Case Definition	Data Source/Cohort	Annual Prevalence (per 100,000 population)		Period Prevalence		Incidence (per 100,000 population)
			Isolation (dates)	Disease (dates)	Period Duration (years)	Prevalence (per 100,000 population)
French Guiana (2008–2018)61	Isolation: ≥ 1 isolate Disease: Full ATS criteria	Three general hospitals EHR data	-	-	11	-	6.17 (I); 1.07 (D)
Mexico City, Mexico (2001–2017)62	Full ATS criteria	Tertiary care referral medical center EHR data	-	-	17	-	SGM: 0.6/1000 admissions (2001–2011)–1.9 (2012–2017) (D)#; RGM: 0.3/1000 (2001–2011)−1.1/1000 (2012–2017) (D)#
Uruguay (2006–2018)67	≥ 1 isolate	National tuberculosis reference laboratory EHR data	0.33 – 1.57 (2006–2018) (4.79-fold increase)#	-	-	-	-
Europe
Location (dates)	Case Definition	Data Source/Cohort	Annual Prevalence (per 100,000 population)		Period Prevalence		Incidence (per 100,000 population)
			Isolation (dates)	Disease (dates)	Period Duration (years)	Prevalence (per 100,000 population)
Czech Republic (2012–2018)81	Full ATS criteria	Public Health Institute Ostrava database for Moravian and Silesian Regions	-	-	7	-	1.1 (D)#
Denmark (1991–2015)68	Isolation: ≥ 1 isolate Disease: Modified ATS microbiologic criteria	International Reference Laboratory data	-	-	25	-	2.14 (I); Definite and Possible NTM-PD: 1.10 (D)+#
England, Wales and Northern Ireland (January 2007–July 2012)77	≥ 1 isolate	Public Health England database	-	-	6	-	3.4–5.0 (2007–2012) (I)**
France (2010–2017)69	ICD-10 or suggestive medication	National Health System database	-	-	8	5.92 (D)	1.025–1.096 (2010–2017) (D)
Germany (2009–2014)72	ICD-10	Health Risk Institute health services research database	-	2.3–3.3 (2009–2014)§	-	-	-
Germany (2010–2011)26	ICD-10	German statutory health insurances claims database	-	-	2	-	1.12–1.48 (2010–2011) (D); Cumulative: 2.6 (D)
Germany (2011–2016)71	ICD-10 (German modification)	InGef research database of statutory health insurances claims	-	ICD-10 coded: 3.79§; Predicted: 19.05§	5	-	ICD-10 coded: 1.56 (D)§; Predicted: 15.33 (D)§
Greece (January 2007 - May 2013)84	Isolation: ≥ 1 isolate Disease: Full (12 pts) ATS criteria and ATS microbiologic criteria (56 pts)	Sismanoglio-A. Fleming General Hospital of Attiki EHR data	-	-	7	-	Cumulative:18.9 (I); Cumulative: 8.8 (D)++
Portugal (2002–2012)73	Treatment for NTM-PD	National Tuberculosis Surveillance System database	-	-	11	-	0.54 (D)#
Scotland (2000–2010)86	ATS microbiologic criteria	Scottish Mycobacteria Reference Laboratory data	-	-	11	-	2.43 (D)+**#
Serbia (2010–2015)74	Isolation: 1 isolate Disease: ATS microbiologic criteria	The National Reference Laboratory of Serbia database	-	0.31 (2011–2012)–0.47 (2014–2015)+**	6	-	1.3 (I); 0.29 (D)+**
Spain (1994 –2014)75	Isolation: ≥ 1 isolate Disease: ≥2 positive cultures or antimicrobial chemotherapy	Referral Hospital (Bellvitge University Hospital) Laboratory data	-	-	21	113.2 (I)**; 42.8 (D)+++ **	-
Spain (1994–2015)76	Isolation: ≥ 1 isolate Disease: Full ATS criteria (only RGM)	Referral Hospital (Bellvitge University Hospital) Laboratory data	-	-	22	-	0.34–1.73 (2003–2015) (APC: 8.3%) (I)#
Spain (1997–2016)78	Full ATS criteria	Microbiology Laboratory of Cruces University Hospital data	-	-	20	-	10.6–1.8 (2016) (APC: 3.3% (MAC), −6.5% (M. kansasii)) (D)
The United Kingdom (2006–2016)79	‘Strict Cohort’, highly likely to have NTM-PD; ‘Expanded Cohort’ possible NTM-PD++++	Clinical Practice Research Datalink	-	Strict Cohort: 7.68–4.70 (2006–2016)#	10	Strict Cohort: 6.38 (D)#	Strict Cohort: 3.85–1.28 (2006–2016) (D)#; Expanded Cohort: 22.9–40.9 (2006–2016) (D)#
The Netherlands (2012–2019)70	Drug combination in drug dispensing database, ICD-10	IQVIA’s Real-World Data Longitudinal Prescription database, Outpatient Pharmacy Database of the PHARMO Database Network, IQVIA’s health insurance claims database, Hospitalization Database of the PHARMO Database Network	-	2.3–5.9 (databases); 6.2–9.9 (pulmonologist survey)	-	-	-
East Asia
Location (dates)	Case Definition	Data Source/Cohort	Annual Prevalence (per 100,000 population)		Period Prevalence		Incidence (per 100,000 population)
			Isolation (dates)	Disease (dates)	Period Duration (years)	Prevalence (per 100,000 population)
Japan (2001–2009)132	Full ATS criteria	Microbiological database of 11 hospitals in Nagasaki prefecture	-	-	9	-	4.6–10.1 (2001–2009) (D)
Japan (2004–2006)23	ICD-8–10, estimated from Death Statistics	Vital Statistics of Japan database	-	33–65 (2005)§	-	-	-
Japan (2009–2014) 107	ICD-10	The Japanese National Database of Health Insurance Claims	-	29 (2011)	6	-	8.6 (2011) (D)
Japan (2012–2013)105	ATS microbiologic criteria	3 major laboratories (SRL, Inc., LSI Medience Corporation, BML, Inc.)	-	12 (2012–2013)+**	2	24 (D)+**	-
South Korea (2001–2015)104	Full ATS criteria	NTM Registry of Tertiary Referral Hospital (Samsung Medical Center)	-	-	15	-	7.0–55.6 (2001–2015) (D)
South Korea (2003–2016)98	ICD-10	National Health Insurance Service database	-	1.2–33.3§	14	-	1.0–17.9 (2003–2019) (D)§
South Korea (2006–2016)100	Full ATS criteria	Tertiary Referral Hospital (Severance Hospital) EHR data	-	-	11	-	4.6–19.6 (2006–2016) (I); 1.2–4.8 (2006–2016) (APC: 14%) (D)
South Korea (2007–2018)99	ICD-10	Health Insurance Review and Assessment Service database	-	5.3–41.7	12	-	Diagnostic-based: 3.5–18.0 (2008–2018) (APC: 16.7%) (D); Clinically-refined: 2.9–12.3 (2008–2018) (APC: 13.2%) (D)
South Korea (2009–2015)133	ICD-10	Health Insurance Review and Assessment service database	-	-	7	-	6.6–26.6 (2009–2015) (D)
South Korea (2009–2015)103	Full ATS criteria	Tertiary care Hospitals (Pusan National University, Pusan National University Yangsan) EHR data	-	-	7	-	6.8–12.9 (2009–2015) (D)#
Taiwan (2000–2012)115	Full ATS criteria	Tertiary Medical Center (National Taiwan University) data	-	-	-	-	3.4–13 (D)
Taiwan (2003–2018)113	ICD-9-CM, treatment	National Health Insurance Research database	-	0.68–7.17# (Treatment Case)	-	-	0.54–3.35 (2003–2018) (D) (Treatment Case)#
Taiwan (2005–2013)112	ICD-9-CM, ≥ 2 cultures, treatment	National Health Insurance Research database	-	-	-	-	5.3–14.8 (2005–2013) (D)§#
Taiwan (2007–2010)114	≥ 1 isolate	National TB Registry of Taiwan CDC	-	-	-	-	8.6 (I)
Taiwan (2010–2014)116	Full ATS criteria	6-Hospital EHR database	-	-	-	-	46 (D)
West and Central Asia
Location (dates)	Case Definition	Data Source/Cohort	Annual Prevalence (per 100,000 population)		Period Prevalence		Incidence (per 100,000 population)
			Isolation (dates)	Disease (dates)	Period Duration (years)	Prevalence (per 100,000 population)
Iran (1990–2014)134	-	-	-	-	25	9.6–10.6 (D)#	-
Oceania
Location (dates)	Case Definition	Data Source/Cohort	Annual Prevalence (per 100,000 population)		Period Prevalence		Incidence (per 100,000 population)
			Isolation (dates)	Disease (dates)	Period Duration (years)	Prevalence (per 100,000 population)
Australia (2001–2016)41	≥ 1 isolate	Queensland Notifiable Conditions database	-	-	16	-	11.1–25.88 (I)#
Australia (2012–2015)3	≥ 1 isolate	Queensland Notifiable Conditions database	-	-	4	-	25.9 (I)#

^*Rates expressed as annual rates per 100,000 population, averaged over study period unless otherwise specified

^**Studies which excluded M. gordonae

ICD-9 – International diagnostic classification of diseases – ninth revision

ICD-10 – International diagnostic classification of diseases – tenth revision

SGM – slow-growing mycobacteria

RGM – rapid-growing mycobacteria

EHR – electronic health records

APC – annual percent change

⁺NTM-PD is defined using ATS microbiologic criteria

⁺⁺NTM-PD is defined using ATS microbiologic criteria and full ATS criteria

⁺⁺⁺NTM-PD is defined ≥2 positive cultures or antimicrobial chemotherapy

⁺⁺⁺⁺‘Strict Cohort’ is defined using evidence of treatment and/or monitoring of NTM-PD and ‘expanded cohort’ is defined using NTM disease clinical terminology codes

^#Included patients with NTM isolated from extrapulmonary sites/specimens from extrapulmonary sites

^§Adjusted

Table 3:

NTM isolations from respiratory specimen (I) and NTM pulmonary disease (D) by species and region

North America
Location (dates)	N	Most Common Species (%)
Canada
Ontario, Canada (1998–2010)131	I: 2631	MAC (50.5)	M. xenopi (21.6)	M. gordonae (12.8)	M. fortuitum (5.0)	M. abscessus (2.2) M. chelonae (1.5)
Toronto, Ontario, Canada (2003–2019)60	D: 252	M. avium (87.3)	M. xenopi (12.7)	-	-	-
United States
Florida (2011–2017)56	I: 396	M. abscessus subsp. abscessus (31.8), M. abscessus subsp. bolletii (<1) M. abscessus subsp. massiliense (4.5) M. chelonae (1.8)	M. gordonae (18.7)	M. avium (1.6) M. chimaera (4.7) M. intracellulare (1.8)	M. kansasii (4.7)	M. szulgai (3.9)
Hawaii (2005–2013)52	455 (I: 201, D: 254+)	MAC (63.7)	M. fortuitum (24.0)	M. abscessus (19.1)	-	-
Hawaii (2005–2019)53	I: 739	MAC (69)	M. fortuitum group (24)	M. abscessus (21)	M. kansasii (2)	-
Illinois, U.S. (2000–2012)135	I: 448	M. avium (54) M. chimera (28) M. intracellulare (18)	-	-	-	-
Iowa (1996–2017)136	D: 185	MAC (68.6)	M. kansasii (8.1)	M. abscessus (7.6) M. chelonae (5.4)	M. fortuitum (2.2); M. xenopi (2.2)	-
North Carolina (2006–2010)58	I: 750	MAC (50.9)	M. gordonae (20.4)	M. abscessus complex (13.6)	M. fortuitum (5)	M. mucogenicum (3.9)
Oregon (2007–2012)10	I: 806	M. avium/intracellulare complex (82.8)	M. abscessus/chelonae complex (4.1) M. chelonae (<1)	M. fortuitum complex (2.9)	M. lentiflavum (1.0)	M. kansasii (<1)
	D: 1146+	M. avium/intracellulare complex (85.7)	M. abscessus/chelonae complex (5.9) M. chelonae (1.2)	M. kansasii (1.2)	M. lentiflavum (1.0)	M. fortuitum complex (<1)
U.S. (2009–2013)48	I: 487	MAC (77)	M. abscessus/M. chelonae (9)	M. kansasii (6)	M. fortuitum (5)	-
US-Affiliated Pacific Island Jurisdictions (2007–2011)54	I: 35‡	MAC (31.4)	M. fortuitum (20.0)	M. gordonae (14.3)	M. abscessus/ chelonae (5.7); M. parascrofulace um/M. fortuitum (5.7)	M. florentinum (2.9); M. kansasii (2.9); M. mucogenicum (2.9); M. paraffinicum (2.9); M. simiae (2.9); M. terrae (2.9)
Central and South America
Location (dates)	N	Most Common Species (%)
Ceará, Brazil (2005–2016)65	I: 42	M. abscessus (4.8); M. avium (4.8); M. fortuitum (4.8)	M. kansasii (2.4); M. szulgai (2.4)	-	-	-
French Guiana (2008–2018)61	I: 147	M. fortuitum (23) M. avium (52)	M. avium (12) M. intracellulare (17)	M. gordonae (5); M. scrofulaceum (5)	M. abscessus (3); M. smegmatis (3)	M. interjectum (2); M. kansasii (2)
	D: 31	M. intracellulare (29)	M. abscessus (16)	M. genavense (3)	-	-
Mexico City, Mexico (2001–2017)62	D: 66	MAC (47)	M. abscessus (27.3); M. fortuitum (27.3)	M. kansasii (7.6); M. scrofulaceum (7.6)	-	-
Panama (2012–2014)66	I: 7	M. avium (57.1)	M. haemophilum (28.6)	M. tusciae (14.3)	-	-
Rio Grande do Sul, Brazil (2003–2013)63	D: 100	M. avium (26) M. intracellulare (9)	M. kansasii (17)	M. abscessus (12)	M. fortuitum (4)	M. gordonae (3)
São Paulo, Brazil (2011–2014)64	I: 2843	M. avium (18.3) M. intracellulare (14.9)	M. kansasii (15.9)	M. gordonae (13.2)	M. fortuitum (10.9)	M. abscessus (7.7) M. chelonae (3.4)
	D: 448+	M. kansasii (29.2)	M. avium (20.8) M. intracellulare (19.6)	M. abscessus (17.2) M. chelonae (<1)	M. gordonae (4.5)	M. fortuitum (3.6)
Uruguay (2006–2018)67	I: 255†	M. avium (23.9) M. intracellulare (33.7)	M. kansasii (8.2)	M. gordonae (5.9)	M. peregrinum (4.7)	M. abscessus (1.2) M. chelonae (3.1); M. fortuitum (3.1)
Europe
Location (dates)	N	Most Common Species (%)
Belgium (2015–2018)137	I: 264	M. avium (12.1) M. chimaera/M. intracellulare (20.8)	M. abscessus (12.9) M. chelonae (6.1)	M. gordonae (3.4)	M. fortuitum (1.9); M. xenopi (1.9)	M. paragordonae (1.5)
Belgium (2010–2017)138	I: 384	M. avium (25) M. chimaera (4.4) M. intracellulare (16.7) M. marseillense (<1)	M. gordonae (14.6)	M. xenopi (7.6)	M. fortuitum (3.9)	M. abscessus (3.1) M. chelonae (2.9)
	D: 165	M. avium (31.5) M. chimaera (4.2) M. intracellulare (24.8) M. marseillense (<1)	M. xenopi (8.5)	M. abscessus (6.7) M. chelonae (<1)	M. kansasii (4.2)	M. malmoense (3.6)
	I: 1926	M. gordonae (42.6)	M. xenopi (15.3)	M. fortuitum (11.4)	M. terrae (6.6)	M. abscessus (2.1)
Croatia (2006–2015)139	D: 137	M. xenopi (39.4)	M. avium (22.6) M. chimaera (1.5) M. intracellulare (14.6)	M. kansasii (5.8)	M. abscessus (4.4)	M. chelonae (5.0) M. fortuitum (1.5); M. gordonae (1.5)
Czech Republic (2012–2018)81	I: 2176†	M. xenopi (36.4)	M. avium-intracellulare complex (17.7)	M. gordonae (15.7)	M. fortuitum (9.7)	M. kansasii (6.2)
	D: 303†	M. avium-intracellulare complex (47.2)	M. kansasii (23.8)	M. xenopi (18.2)	M. abscessus (1.3) M. chelonae (2.3)	M. fortuitum (2); M. malmoense (2)
England, Wales and Northern Ireland (January 2007-July 2012)77	I: 16294	MAC (35.6)	M. gordonae (16.7)	M. abscessus (5.0) M. chelonae (9.7)	M. fortuitum (8.2)	M. kansasii (5.9); M. xenopi (5.9)
France (2002–2013)140	D: 92	M. avium (29.3) M. intracellulare (29.3)	M. kansasii (17.4)	M. xenopi (16.3)	M. abscessus (2.2)	M. fortuitum (1.1)
France (2009–2014)141	D: 477	M. avium (31.2) M. intracellulare (28.1)	M. xenopi (19.7)	M. kansasii (5.7)	M. abscessus complex (3.8) M. chelonae (<1)	M. fortuitum (2.7)
Germany (2006–2016) 142	I: 216	MAC (33.3)	M. gordonae (23.6)	M. xenopi (15.2)	M. abscessus complex (9.3)	-
Greece (January 2007-May 2013)84	I: 122	M. gordonae (13.9)	M. avium (13.1) M. intracellulare (9.8)	M. fortuitum (12.2)	M. lentiflavum (4.9); M. peregrinum (4.9)	M. abscessus (1.6) M. chelonae (2.4); M. xenopi (2.4)
	D: 12	M. avium (25) M. intracellulare (25)	M. abscessus (8.3); M. fortuitum (8.3); M. gordonae (8.3); M. xenopi (8.3)	-	-	-
Ireland (January 2007-July 2012)143	I: 37	M. avium (59.5) M. intracellulare (10.8)	M. gordonae (10.8)	M. abscessus (5.4) M. chelonae (5.4); M. szulgai (5.4)	M. malmoense (2.7)	-
The Netherlands (2008 – 2013)144	D: 63	M. avium (36.5) M. chimera (4.8) M. intracellulare (12.7)	M. malmoense (11.1)	M. kansasii (9.5)	M. abscessus (6.3)	M. simiae (4.8)
Poland (2010–2015)145	I: 73	M. avium (22) M. intracellulare (8); M. kansasii (22)	M. gordonae (19)	M. xenopi (11)	M. fortuitum (6)	M. abscessus (3)
	D: 36	M. kansasii (36)	M. avium (28) M. intracellulare (11)	M. xenopi (8)	M. abscessus (6)	M. gordonae (3)
Poland (2013–2017)146	I: 2799†	M. kansasii (27.2)	M. avium (22) M. intracellulare (6.2)	M. xenopi (16)	M. gordonae (14.5)	M. fortuitum (5.9)
Portugal (2005–2014)147	I: 365	MAC (48.2)	M. gordonae (13.7)	M. peregrinum (11)	M. abscessus (1.6) M. chelonae (7.9)	M. kansasii (5.8)
Scotland (2000–2010)86	I: 933	MAC (44.8)	M. malmoense (21.7)	M. abscessus (13.7) M. chelonae (2.6)	M. xenopi (4.5)	M. kansasii (3.9)
Serbia (2009–2016)148	I: 296†	M. gordonae (22.3)	M. fortuitum (21.3)	M. xenopi (13.5)	M. peregrinum (12.2)	MAC (9.8))
	D: 83	MAC (30.1)	M. xenopi (24.1)	M. kansasii (18.1)	-	-
Serbia (2010–2015)74	I: 565	M. xenopi (17.3)	M. gordonae (12.9)	M. fortuitum (11.3)	M. abscessus (4.6) M. chelonae (4.1)	M. kansasii (4.2)
	D: 126+	M. xenopi (28.6)	M. abscessus (15.1) M. chelonae (2.4)	M. kansasii (11.1)	M. fortuitum (10.3)	M. avium (7.9) M. intracellulare (6.3)
Spain (1994–2014)75	I: 680	M. kansasii (28.5)	MAC (20.4)	M. xenopi (14.1)	M. abscessus (2.5)	-
	D: 257**	M. kansasii (59.9)	MAC (26.1)	M. xenopi (6.2)	M. abscessus (4.3)	-
Spain (1994–2015)76 (only RGM)	I:116	M. fortuitum (45.7)	M. abscessus (5.2) M. chelonae (33.6)	M. mucogenicum (8.6)	M. mageritense (1.7); M. peregrinum (1.7)	-
	D: 15	M. abscessus (40) M. chelonae (26.7)	M. fortuitum (26.7)	M. mucogenicum (6.7)	-	-
Spain (1997–2016)78	D: 327	M. kansasii (83.8)	MAC (13.1)	-	-	-
Spain (2007–2013)149	I: 156 D: 34+	MAC (64.1) MAC (79.4)	M. gordonae (12.2) M. genavense (8.8)	M. abscessus (1.9) M. chelonae (8.3) M. abscessus (2.9) M. chelonae (5.9)	M. fortuitum (5.8) M. malmoense (2.9); M. xenopi (2.9)	M. lentiflavum (4.5); M. scrofulaceum (4.5) -
Spain (2013–2017)150	I: 314	M. avium (48.4) M. intracellulare (16.6)	M. fortuitum (12.1)	M. gordonae (8.0)	M. lentiflavum (4.1)	M. abscessus (<1) M. chelonae (3.2)
Switzerland (2015–2020)151	I: 236†	M. gordonae (24.6)	M. avium (23.3) M. chimera (3.4) M. intracellulare (6.8)	M. xenopi (11.9)	M. abscessus subsp. abscessus (6.4) M. chelonae (4.2)	M. fortuitum (4.2)
The United Kingdom (2007–2014)152	I: 853	M. avium (21.2) M. intracellulare (31.3)	M. gordonae (15.2)	M. abscessus (8.4) M. chelonae (8.4)	M. xenopi (7.9)	M. malmoense (4.7)
Africa
Location (dates)	N	Most Common Species (%)
Botswana (Aug 2012–Nov 2014) (HIV–infected)89	I: 228	M. avium (2.2) M. intracellulare (47.8)	M. gordonae (7)	M. malmoense (3.9)	M. simiae (3.5)	M. asiaticum (2.6); M. fortuitum (2.6); M. scrofulaceum (2.6)
Ethiopia (2017)93	I: 35	M. simiae (42.9)	M. abscessus complex (14.3)	M. fortuitum (11.4)	M. avium complex (5.7) M. intracellulare (8.6); M. gordonae (8.6)	M. kansasii (2.9); M. scrofulaceum (2.9); M. szulgai (2.9)
Gabon (Jan 2018–Dec 2020)97	I: 137	M. avium (5.1) M. intracellulare (54)	M. fortuitum (21.9)	M. abscessus (6.6) M. chelonae (2.2)	M. kansasii (4.4)	M. mucogenicum (1.5)
Ghana (2012–2014)90	I: 43	M. avium subsp. paratuberculosis (30.2); M. colombiense (7); M. intracellulare (41.9)	M. abscessus (11.3)	M. mucogenicum (7)	M. simiae (2.3)	-
Ghana (Jan 2013– Mar 2014)91	I: 38	MAC (23.7)	M. chelonae complex (7.9); M. simiae (7.9)	M. fortuitum complex (5.3)	M. arupense (2.6); M. flavescens (2.6); M. kansasii (2.6); M. terrae (2.6)	-
Kenya (2020)94	I: 146	MAC (31)	M. fortuitum complex (20)	M. abscessus complex (14)	-	-
Mali (2006–2013)92	I:41	M. avium (24.4)	M. specie (7.3)	M. simiae (4.9)	-	-
Nigeria, Cameroon, Ghana (Jan-Dec 2017)153	I: 14	M. fortuitum (35.7)	M. engbaekii (14.3); M. avium (7.1) M. colombiense (7.1) M. intracellulare (14.3)	M. gordonae (7.1); M. paraense (7.1); M. peregrinum (7.1)	-	-
Tanzania (Nov 2012–Jan 2013)95	I: 36	M. gordonae (16.7); M. interjectum (16.7)	M. avium (5.5) M. colombiense (2.8) M. intracellulare (11.1)	M. scrofulaceum (8.3)	M. fortuitum (5.5); M. kumamotonense (5.5)	-
Tanzania (Nov 2019 - Aug 2020)154	I: 24	M. avium (8.3) M. intracellulare (16.7)	M. abscessus subsp. abscessus (12.5) M. abscessus subsp. bolletii (4.2)	M. fortuitum group (8.3)	M. kansasii (4.2); M. simiae (4.2); M. szulgai (4.2)	-
Tunisia (2002–2016)88	I: 30	M. kansasii (23.3)	M. fortuitum (16.7); M. novocastrense (16.7)	M. chelonae (10)	M. gadium (6. 6); M. gordonae (6. 6)	M. flavescens (3.3); M. peregrinum (3.3); M. porcinum (3.3)
East Asia
Location (dates)	N	Most Common Species (%)
China (2008–2012)155	I: 616	M. kansasii (45)	M. avium (3.6) M. intracellulare (20.8)	M. chelonae-abscessus complex (14.9)	M. fortuitum (4.5)	-
China (2009–2019)156	I: 1102	M. avium (13.2) M. intracellulare (54.8)	M. chelonae-abscessus complex (16.5)	M. kansasii (8.2)	M. gordonae (3.3)	M. fortuitum (<1)
China (2010–2015)120	I: 232	M. avium (4.7) M. intracellulare (40.5)	M. abscessus (28.4)	M. kansasii (9.9)	M. fortuitum (8.6)	M. gordonae (4.3)
	D: 72*	M. avium (2.8) M. intracellulare (52.8)	M. abscessus (33.3)	M. kansasii (9.7)	M. szulgai (1.4)	-
China (2013–2016)157	D: 607	M. avium (16.3) M. intracellulare (28.2)	M. abscessus subsp. abscessus (23.9) M. abscessus subsp. massiliense (16.6)	M. kansasii (10.0)	M. fortuitum (2.8)	M. gordonae (1.0)
China (2014–2021)119	I: 1755	M. avium (7.8) M. intracellulare (51.6)	M. abscessus (22.2)	M. kansasii (8.3)	M. fortuitum (2.1); M. gordonae (2.1)	M. paragordonae (1.3)
China (2015–2020)118	I: 789	M. abscessus (35.1)	M. avium (6.8) M. intracellulare (12.8)	M. fortuitum (10.1)	M. kansasii (10.0)	M. gordonae (9.4)
China (2017–2018)158	D: 87	M. avium (11.5) M. intracellulare (70.1)	M. chelonae-abscessus complex (11.5)	M. kansasii (7.5)	M. gordonae (1.1)	-
China (2018)159	I: 24	M. avium (4.2) M. intracellulare (66.7)	M. abscessus (12.5)	M. kansasii (8.3)	M. fortuitum (4.2); M. szulgai (4.2)	-
China (2019–2020)117	D: 458	M. avium (8.5) M. intracellulare (52.6)	M. abscessus complex (23.1)	M. kansasii (8.1)	M. szulgai (2.6)	-
Japan (2000–2013)108	D: 592+	M. avium (70.6) M. intracellulare (22.3)	-	-	-	-
Japan (2001–2009)132	D: 975	M. avium (42.6) M. intracellulare (44.3)	M. abscessus (3.1) M. chelonae (<1)	M. gordonae (2.1); M. kansasii (2.1)	-	-
Japan (2006–2016)109	I: 3620	MAC (82.2)	M. abscessus complex (4.5) M. chelonae (<1)	M. kansasii (3.7)	M. fortuitum (2.1)	M. peregrinum (<1)
	D: 2155+	MAC (87.2)	M. abscessus complex (5.5)	M. kansasii (3.9)	M. fortuitum (1.3)	-
Japan (2009–2015)160	I: 416	M. abscessus (31) M. chelonae (11)	M. avium (5) M. intracellulare (20)	M. gordonae (18)	M. fortuitum (11)	-
	D: 114	M. abscessus (36) M. chelonae (10)	M. avium (6) M. intracellulare (27)	M. fortuitum (10)	M. gordonae (6)	-
Japan (2012–2013)105	I: 26059	M. avium (61.8) M. intracellulare (31.1)	M. kansasii (2.1)	M. abscessus (2.0) M. chelonae (<1)	M. fortuitum (1.1)	M. terrae (<1)
	D: 7167+	M. avium (65.1) M. intracellulare (32.4)	M. kansasii (3.4)	M. abscessus (2.7)	-	-
South Korea (2001–2015)104	D: 2329	MAC (75)	M. abscessus (22)	M. kansasii (3)	-	-
South Korea (2006–2016)100	D: 1017	MAC (63.6)	M. abscessus complex (10.8)	M. fortuitum (2.1); M. kansasii (2.1)	-	-
South Korea (2007–2019)102	I: 2807	M. avium (19.2) M. intracellulare (50.6)	M. fortuitum complex (4.7)	M. abscessus (4.5) M. chelonae (1.0)	M. gordonae (3.5)	M. kansasii (1.1)
South Korea (2009–2015)103	I: 5558	M. avium (23.1) M. intracellulare (38.9)	M. abscessus (8.4)	M. kansasii (7.7)	-	-
South Korea (2014–2019)101	I: 4962	M. avium (17.5) M. intracellulare (42.3)	M. abscessus (7.0) M. chelonae (3.3)	M. fortuitum (5.5)	M. kansasii (3.6)	M. mucogenicum (2.7)
Taiwan (2000–2012)115	D: 3317	MAC (41.5)	M. abscessus (21.4) M. chelonae (9.4)	M. fortuitum (13.4)	M. kansasii (7.1)	M. gordonae (5.0)
Taiwan (2007–2010)114	I: 894	MAC (32.4)	M. abscessus complex (9.2) M. chelonae complex (17.6)	M. fortuitum complex (17)	M. kansasii (9.8)	M. gordonae (7.0)
Taiwan (2010–2014)116	D: 1674	MAC (34.4)	M. abscessus (24.3)	M. kansasii (11)	M. fortuitum (9.8)	M. gordonae (4.7)
South & Southeast Asia
Location (dates)	N	Most Common Species (%)
India (1981–2020)161	I: 2071	MAC (18.9)	M. abscessus (8.8) M. chelonae (10.3)	M. fortuitum (9.8)	M. gordonae (6.9)	M. terrae (6.3)
India (2013–2015)126	I: 209	M. abscessus (31.1) M. chelonae (8.1)	M. fortuitum (20.6)	M. avium (8.1) M. intracellulare (13.4)	M. interjectum (4.3); M. simiae (4.3)	M. gordonae (3.3)
India (2014–2015)127	I: 164	M. chelonae (26.8)	M. fortuitum (12.8)	M. gordonae (9.1)	M. kansasii (6.1)	M. simiae (4.9)
India (2015–2020)125	I: 43	M. abscessus complex (37.2) M. chelonae (2.3)	M. fortuitum (23.3)	M. chimaera (2.3) M. colombiense (2.3) M. intracellulare (7.0) M. marseillense (2.3)	M. parascrofulace um (4.7); M. simiae (4.7)	-
Pakistan (2016–2019)122	I: 169	MAC (61)	M. abscessus (24)	M. fortuitum (5.5); M. kansasii (5.5)	M. gordonae (1.8); M. szulgai (1.8)	-
Singapore (2011–2012)162	I: 511	M. abscessus (39.5) M. chelonae (1.8)	M. fortuitum (16.6)	M. kansasii (15.5)	M. avium (15.1)	M. gordonae (7.0)
Singapore (2012–2016)163	I: 2026†	M. chelonae-abscessus complex (49.9)	M. fortuitum (17)	MAC (15.4)	M. kansasii (11.5)	M. haemophilum (2.0)
	D: 352	M. chelonae-abscessus complex (56)	MAC (28.1)	M. fortuitum group (25.9)	M. kansasii (20.2)	M. scrofulaceum (2.3)
West and Central Asia
Location (dates)	N	Most Common Species (%)
Iran (2011–2017)124	I: 236	M. abscessus (19.1)	-	-	-	-
Iran (2016–2018)123	I: 95	M. fortuitum (48.4)	M. simiae (16.8)	M. kansasii (15.7)	M. abscessus (7.3)	M. thermoresistibile (4.2)
Turkey (2015–2019)121	I: 45†	M. fortuitum (24.4)	M. abscessus (17.7) M. chelonae (8.9)	M. lentiflavum (11.1); M. simiae (11.1)	M. avium (4.4) M. intracellulare (8.9)	M. gordonae (6.7)
Saudi Arabia (2006–2012)164	I: 142	MAC (35)	M. fortuitum (24)	M. chelonae-abscessus complex (17)	M. gordonae (6)	M. kansasii (4)
	D: 40	MAC (47.5)	M. abscessus (25)	M. kansasii (10)	M. fortuitum (7.5)	M. szulgai (5)
Oceania
Location (dates)	N	Most Common Species (%)
Australia (2001–2016)41	I: 12219†	M. avium (9.8) M. intracellulare (39.1)	M. abscessus (8.5) M. chelonae (3.3)	M. fortuitum (8.3)	M. kansasii (2.4)	-
French Polynesia (2008–2013)128	I: 87†	M. fortuitum (11.5) M. porcinum (18.4) M. senegalense (13.8)	M. abscessus (29.9) M. bolletii (1.1) M. massiliense (1.1)	M. mucogenicum complex (9.2)	M. avium (1.1) M. chimaera (3.4) M.intracellular e (1.1)	-
Papua New Guinea (2010–2012)129	I: 9	M. avium (33.3) M. intracellulare (22.2); M. fortuitum (33.3)	M. terrae (22.2)	-	-	-

MAC – Mycobacterium avium complex

RGM – rapidly growing nontuberculous mycobacterial species

^*Included only patients whose radiographic disease pattern could be classified as cavitary, bronochiectatic, or consolidative

^**≥2 positive cultures or antimicrobial chemotherapy

^#M. avium and M. intracellulare concurrently isolated

^†Included patients with NTM isolated from extrapulmonary sites/specimens from extrapulmonary sites

^‡Sub-analysis of study population

⁺NTM-PD defined using ATS microbiologic criteria

In Florida, a study using the statewide clinical network and defining cases based on a single ICD code found an increasing disease prevalence, from 14.3/100,000 in 2012 to 22.6/100,000 in 2018 ⁵⁵. A second study in Florida at a single large academic medical center with NTM defined by pulmonary isolation found that M. abscessus was the predominant species, representing 39.1% of pulmonary isolates ⁵⁶ (Table 3). The finding of a predominance of M. abscessus in Florida is consistent with a recent study showing that the South had the highest proportion of M. abscessus isolates, and Florida had the highest 5-year period prevalence of M. abscessus (17%) ⁵⁷.

A population-based study in three counties of North Carolina (Durham, Wake, and Orange) estimated the prevalence of NTM isolation from 2006 through 2010 using primarily laboratory reports, for an average annual isolation prevalence of 9.4/100,000 (excluding M. gordonae; 11.5 including M. gordonae)⁵⁸. MAC comprised 50.9% of isolates, followed by M gordonae (20.4%) and M. abscessus complex (13.6) (Table 3) ⁵⁸. Cumulative incidence of NTM isolation was 8.8/100,000 across Durham County, Orange County, and Wake County (Table 2) ³⁸. A statewide study in Oregon from 2007 to 2012 used the ATS microbiologic criteria to define NTM-PD and found an average annual incidence of 5/100,000, ranging from 4.8/100,000 in 2007 to 5.6/100,000 in 2012 ¹⁰. The predominant species were MAC (85.7%), followed by rapidly growing mycobacteria (8.1%: M. abscessus/chelonae complex; M. chelonae; M. fortuitum complex) ¹⁰ (Table 2, Table 3).

Ontario, Canada

A prior report⁵⁹ showed a significant increase in NTM isolation and disease (ATS microbiologic criteria) from 1998–2010, based on analysis of isolates from the Public Health Ontario Laboratory, which represents approximately 95% of NTM isolates in Ontario. A more recent analysis in a subset of patients being treated for NTM-PD in a tertiary care center and who met the full ATS NTM disease diagnostic criteria found a significant increase in M. avium-PD between the periods 2009–2012 and 2015–2018 and no significant change in non-M. avium species in the same period, demonstrating that the increase in NTM-PD is being driven by M. avium ⁶⁰.

Central and South America

We identified six studies published between 2014 and 2022 from French Guiana ⁶¹, Mexico ⁶², Brazil ^63–65, and Panama ⁶⁶. All studies, apart from one, were single-center studies. The results of these studies may be impacted by referral bias. In addition, many studies lack a denominator in a defined population, which the limits the comparability of results.

In French Guiana, a retrospective, observational study of three hospitals during the period 2008–2018, identified 178 patients with NTM positive sputum cultures and 31 patients with NTM-PD. Incidence of NTM isolation and disease was 6.17/100,000 and 1.07/1000,000. This study defined disease using the full ATS diagnostic criteria. M. avium (52%) was the most common species among patients with NTM-PD, followed by M. intracellulare (29%), and M. abscessus (16%). Among patients with NTM isolates, M. fortuitum (23%) was most commonly identified, followed by M. intracellulare (17%), and M. avium (12%) ⁶¹.

A single center, retrospective study in Mexico City identified 158 patients that met the ATS criteria. The average annual isolation incidence increased for slow growing species from 0.6/1,000 admissions (2001–2011) to 1.9/1,000 admissions (2012–2017) and for rapidly growing species from 0.3/1,000 admissions (2001–2011) to 1.1/1,000 admissions (2012–2017) (Table 2). From these patients, the most common species were MAC (47%), M. abscessus (27.3%), and M. fortuitum (27.3%) (Table 3) ⁶².

Many studies on NTM in Brazil occur at the state-level and evaluate patient records at large, state referral hospitals that treat TB and NTM-related diseases. A retrospective, single institution study of patients in the state of Rio Grande do Sul identified 100 patients that met the full ATS criteria for NTM-PD (2003– 2013). The most common NTM species were M. avium (26%), M. kansasii (17%), M. abscessus (12%), and M. intracellulare (9%) (Table 3) ⁶³. In the state of São Paulo, 448 patients were identified as NTM-PD using the ATS microbiologic criteria. From these patients, M. kansasii (29.2%), M. avium (20.8%), M. intracellulare (19.6%), and M. abscessus (17.2%) were the most common species (Table 3) ⁶⁴. In the state of Ceará, NTM was isolated from pulmonary samples from 42 patients between 2005 and 2016. A high proportion of isolates were not sent to the National Reference Laboratory for identification (81%). Of the species that were identified, the most common were M. avium (4.8%), M. fortuitum (4.8%), and M. abscessus (4.8%) (Table 3) ⁶⁵.

A study at the national tuberculous laboratory in Montevideo, Uruguay identified 255 NTM isolates from 204 TB suspects during 2006–2018; 210 were collected from pulmonary samples. Most NTM isolates were identified as MAC species (57.6%), followed by M. kansasii (8.2%), M. gordonae (5.9%), and M. peregrinum (4.7%) (Table 3) ⁶⁷.

Europe

Studies from ten European countries published between 2014 and 2022 have calculated the incidence or prevalence of NTM isolation and/or NTM-PD (Table 2). Data sets, study populations and identification methods of NTM-PD patients differed significantly, thus comparison of the incidence or prevalence is difficult. Some of these studies have also calculated trends over the respective study period: either a stable (Denmark ⁶⁸, France ⁶⁹, The Netherlands ⁷⁰) or increasing NTM-isolation or NTM-PD incidence or prevalence (Germany ^{26, 71, 72}, Portugal ⁷³, Serbia ⁷⁴, Spain ^{75, 76}, UK ⁷⁷) was seen. Decreasing incidences for NTM isolation as reported in one study from Spain ⁷⁸ and one from UK ⁷⁹ are explained by the dominating decrease in M. kansasii isolation and PD in the Bilbao region ⁷⁸ and by the selection of the data source, e.g. primary care records, where the decrease of the incidence and prevalence of the strictly defined NTM-PD cases probably represented a shift of management of NTM-PD towards secondary care ⁷⁹. Species isolated from respiratory secretions or causing NTM-PD varies widely among countries and even within countries (Table 3); of note are the high percentages of NTM-PD caused by M. xenopi in Croatia, Czech Republic and Serbia, by M. kansasii in Poland and some Spanish regions, and by M. malmoense in Scotland and The Netherlands, resembling isolation data published by the NTM-NET ⁸⁰. A changing epidemiology of species causing NTM-PD over time is also noted in some regions ⁷⁸, such that analysis of aggregate trends may obscure species-specific changes. In Europe COPD often is a concomitant disease, with a mean age around 60 years and a generally similar overall male: female sex distribution with some exceptions (Table 4).

Table 4:

Age and Sex distribution of NTM isolations from respiratory specimen (I) and NTM pulmonary disease (D) by Region

North America
Location (dates)	N	Mean Age (years)	Female (%)	COPD (%)/ Bronchiectasis (%)/History of TB (%)	Other Pulmonary Disease (%)
Florida, U.S. (2011–2017)56	I: 271	60.5	28.5	20.8/2.2/2.2	Asthma (2.8), CF (11.1),ILD (3.6), Lung Cancer (7.5)
Florida, U.S. (2012–2018)55	D: 7963 (ICD-9/10)	-	54	27.7/ 30.7/10.6	CF (4.1)
Hawaii, U.S. (2005– 2013)52	I: 201	65.2	50	39.8/27.9/-	Lung Cancer (12)
	D: 254+	66	57	43/44/-	Lung Cancer (11)
Hawaii, U.S. (2005–2019)53	I: 739	Median: 63	54	-/-/-	-
Illinois, U.S. (2000–2012)135	I: 448	63.0	63	11–22#/-/5–9#	-
Iowa, U.S. (1996–2017)136	D: 185	Median: 63	55	30.3/26.5/-	CF (7.6), ILD (5.9), Structural Lung Disease (60.5)
North Carolina, U.S. (2006– 2010)58	I: 750	-	41.2	-/-/-	-
North Carolina, U.S. (2006– 2010)38	I: 507	60	51.8	-/-/-	-
Oregon, U.S. (2007–2012)10	I: 806	Median: 66	49.6	-/-/-	-
	D: 1146+	Median: 69	55.7	-/-/-	-
U.S. (2001– 2015)47	I: 4676 (ICD-9/ICD-10)	-	0.13	100/2.5/2.6	Asthma (12.2), ILD (12.8), Lung Cancer (3.1)
U.S. (2008– 2015)45	D: 6280 (ICD-9/ICD-10)	69	67.6	52.6/37/7	Aspergillosis (3.0), Asthma (23.2), CF (1.7)
U.S.-Affiliated Pacific Island Jurisdictions (2007–2011)54	I: 35‡	Median: 34.8	49.2	27.6/-/44.8	-
Central & South America
Location (dates)	N	Mean Age (years)	Female (%)	COPD (%)/ Bronchiectasis (%)/History of TB (%)	Other Pulmonary Disease (%)
Ceará, Brazil (2005–2016)65	I: 69†	38.6	26.1	11.9/-/73.8	All (13)*
French Guiana (2008–2018)61	178 (D: 31, I: 147)	49	39.3	17/5/16	Chronic Pulmonary Disease (33)
Mexico City, Mexico (2001–2017)62	D: 67	Median: 52–61##	48–50##	-/-/-	Any (13)*
Rio Grande do Sul, Brazil (2003–2013)63	D: 100	54.6	49	17/22/85	CF (1), Silicosis (1)
Europe
Location (dates)	N	Mean Age (years)	Female (%)	COPD (%)/ Bronchiectasis (%)/History of TB (%)	Other Pulmonary Disease (%)
Belgium (2010–2017)138	D: 165	Median: 64	43	41.8/20/7.3	Aspergillosis (4.2), CF (6.1), Prior NTM-PD (13.9)
Croatia (2006–2015)139	D: 137	Median: 66	47.5	45.3/29.2/27.7	Asthma (2.2), Prior NTM-PD (5.1)
France (2002–2013)140	D: 92	Median: 61.5	42.4	14.1/17.4/12	All (46.7)*
France (2009–2014)141	D: 477	Median: 65	29.6–56.2#	-/-/31.4	All (68)*
France (2010–2017)69	D: 5628 (ICD–10 or suggestive medication)	60.9	47.1	18.8/10.6/14.1	CF (3.2), Lung Cancer (5.7)
Germany (2009–2014)72	D: 85–126 per year (ICD-10) (2009–2014)	55–61 (2009–2014)	43.4–52.9 (2009–2014)	62.4–79.2 (2009–2014)/6.6–18.3 (2009–2014)/13.5–24 (2009–2014)	Asthma (20.8–29.2) (2009–2014), Lung Cancer (3.5–10.3) (2009–2014)
Germany (2010–2011)26	D: 125 (ICD–10)	49.8	50.4	-	-
Germany (2011–2016)71	D: 218 (ICD-10, German modification)	61.4	41.3	46.3/10.1/17	Asthma (25.2). ILD (3.7)
Greece (January 2007–May 2013)84	I: 62	–	–	42/35/13	Asthma (6.4) CF (1.6)
	D: 12	-	-	50/25/41	Asthma (8.3)
The Netherlands (2008–2013)144	D: 63	60.8	50.8	66.7–71.4#/-/10.5–14.3#	Asthma (1.6–15.8#), Lung Cancer (5.3–14.3#)
Poland (2010–2015)145	D: 36	58.5	83.3	19/33/28	Asthma (6), CF (8), ILD (28), Lung Cancer (16.7)
Portugal (2002–2012)73	D: 632 (NTM–PD treatment)†	Median: 54	39.7	6.3/-/-	ILD (3.3)
Serbia (2009–2016)148	D: 85†	59.2	34.1	40/10/13	-
Serbia (2010–2015)74	D: 126+	65.4	47.6	-	-
Spain (1997–2016)78	D: 327	56.8	28.8	30.1/29.3/-	Any (56)*
The United Kingdom (2007–2014)152	D: 112	65	51.8	44.5/38.4/9.8	Asthma (16.9), ILD (5.4)
East Asia
Location (dates)	N	Mean Age (years)	Female (%)	COPD (%)/ Bronchiectasis (%)/History of TB (%)	Other Pulmonary Disease (%)
China (2010–2015)120	D: 72	54.1	47.2	8.3/6.9/15.3	Silicosis (2.8)
China (2013–2016)157	D: 607	-	56.2	4.1/58.6/59.6	-
China (2014–2021)119	I: 1755	-	53.6	-/-/-	-
China (2015–2020)118	I: 789	Median: 36	39.0	-/-/22.4	-
China (2017–2018)158	D: 87	Median: 60	41.4	6.9/19.5/64.4	-
China (2019–2020)117	D: 458	-	34.9	9.2/26.2/45.6	Asthma (2.2)
Japan (2004–2006)23	D: 309 (ICD-8–10, estimated from Death Statistics)	67	64.7	3.2/4.2/19.1	Asthma (3.2), ILD (1.3), Lung Cancer (1.3)
Japan (1999–2010)165	D: 782	68.1	68.5	4.3/-/11.6	ILD (3.5)
Japan (2000–2013)108	D: 592+	66.0	61.1	6.4/-/9.5	Asthma (6.3), ILD (6.6)
Japan (2001–2009)132	D: 975	71.2	69.7	5.8/0.3/13.5	Asthma (1.8), ILD (4.0), Lung Cancer (5.5), Silicosis (1.3)
Japan (2006–2016)109	D: 2155+	Median: 69	66.5	-/-/-	-
Japan (2009–2014)107	D (Incident):11034 (2011) (ICD-10)	69.3	69.6	6.9/23.5 /6.3	Aspergillosis (6.1), ILD (9.9), Lung Cancer (5.3)
Japan (2009–2015)160	D: 114	Median: 74–78.5#	59–68#	(5–24)#/(11–15)#/(16–17)#	Asthma (8–17)#, ILD (5–8)#
Japan (2013–2015)166	D: 184++	69.5	75.5	-/5.4/8.2	-
Japan (2012–2013)105	D: 7167+	73.6**	65.5**	-/-/-	-
Japan (2014)167	D: 419 (ICD–10)	Median: 59	68	3.1/-/-	Aspergillosis (1.9), Asthma (18.3), DPB (1.6), ILD (6.0), Lung Cancer (4.5)
South Korea (2001–2015)104	D: 2329	Median: 60	59	-/-/-	-
South Korea (2003–2016)98	D: 46194 (ICD–10)	55.8	61.1	-/-/-	-
South Korea (2006–2016)100	D: 1017	62.7	58.8	13.5/83.9/48.3	Asthma (7.3), Lung Cancer (4.0)
South Korea (2007–2018)99	D: 45321 (ICD–10)	-	56.9	-/-/-	-
South Korea (2007–2019)102	I: 2984†	-	42.3	-/-/-	-
South Korea (2009–2015)133	D: 52551 (ICD-10)†	53.0	57.2	25.6/21.9/33.7	Asthma (33.2), ILD (2.6), Lung Cancer (5.8)
Taiwan (2000–2012)115	D: 3317	-	42.3	-/-/-	-
Taiwan (2003–2018)113	D(treatment case): 558 (ICD-9-CM)†	62.5	42.5	-/-/-	Chronic Lung Disease (42.8)
Taiwan (2005–2013)112	D: 450 (ICD-9-CM)†	-	38	-/-/-	-
Taiwan (2007–2010)114	I: 894	-	44.4	-/-/-	-
Taiwan (2010–2014)116	D: 1674	-	44.3	23.4/15.6/25.4	Asthma (5.2), ILD (6)
South-Southeast Asia
Location (dates)	N	Mean Age (years)	Female (%)	COPD (%)/ Bronchiectasis (%)/History of TB (%)	Other Pulmonary Disease (%)
India (1981–2020)161	D: 365	-	44	-/3/65	-
India (2013–2015)126	I: 263†	Median: 48	39.5	-/-/-	-
India (2015–2020)125	I: 105†	-	50.5	5.7/-/16.2	-
Pakistan (2016–2019)122	I: 169	Median: 45	39	-/-/65	-
Singapore (2011–2012)162	I: 485	Median: 70	37.9	14.2/28.7/34.4	Asthma (6.6); ILD (3.7)
Singapore (2012–2016)163	D: 352	Median: 67	48.3	13.1/36.4/27.6	Asthma (4.0)
Central & West Asia
Location (dates)	N	Mean Age (years)	Female (%)	COPD (%)/ Bronchiectasis (%)/History of TB (%)	Other Pulmonary Disease (%)
Iran (2016–2018)123	I: 95	47.4	42.1	-/-/-	-
Saudi Arabia (2006–2012)164	D: 40	54	42	7.5/15/12	Asthma (15), CF (2.5), ILD (10)
Oceania
Location (dates)	N	Mean Age (years)	Female (%)	COPD (%)/ Bronchiectasis (%)/History of TB (%)	Other Pulmonary Disease (%)
Australia (2001–2016)41	I: 12219†	Median: 66	50.1	-/-/-	-
Australia (2012–2015)3	I: 1222†	Median: 66	51	-/-/-	-
Papua New Guinea (2010–2012)129	I: 9	35.7	88.9	-/-/-	-

ICD-9 – International diagnostic classification of diseases – ninth revision

ICD-10 – International diagnostic classification of diseases – tenth revision

COPD – Chronic Obstructive Pulmonary Disease

TB – Tuberculosis

SGM – slow-growing nontuberculous mycobacteria

RGM – rapid-growing nontuberculous mycobacteria

ILD – Interstitial lung disease

DPB – Diffuse panbrochiolitis

CF – Cystic Fibrosis

All pulmonary comorbidities, including COPD, Bronchiectasis, History of TB

^**MAC patients only

^†Included patients with NTM isolated from extrapulmonary sites/specimens from extrapulmonary sites

⁺NTM-PD defined using ATS microbiologic criteria

⁺⁺NTM-PD defined using full ATS criteria or 2008 Japanese Society for Tuberculosis/Japanese Respiratory Society Guidelines for the diagnosis of NTM pulmonary infection

^#Range by NTM species

^##Range by RGM and SGM

^‡Sub-analysis of study population

Czech Republic

NTM isolates from the Public Health Institute Ostrava representing the Moravian and Silesian Regions from eastern Czech Republic were classified according to ATS diagnostic criteria and the incidence of each NTM species presented as average annual incidence in the studied period 2012–2018: The average incidence of NTM-PD patients was 1.10/100,000 (1.33/100,000 in men and 0.88/100,000 in women) during the study period, with MAC-PD patients predominantly women and M. kansasii-PD and M. xenopi-PD predominantly men, with distinct epidemiology in local districts ⁸¹.

Denmark

A retrospective nationwide cohort study including microbiological data from the Danish International Reference Laboratory (Staten Serum Institute) in Denmark from 1991–2015 reported an annual (isolation) incidence rate of NTM isolation from a pulmonary site of 2.14/100,000 in adults aged 15 and above and (using ATS microbiologic criteria only) an incidence of 1.10/100,000 for cases with definite and possible NTM disease. No significant trend towards an increase or decrease for definite NTM disease incidence (that included patients with definite NTM-PD as defined by the authors, which were not reported separately) was found ⁶⁸. However, a more recent study using ICD codes and representing the Central region of Denmark found an increasing trend in NTM-PD, suggesting a potential increase in clinically relevant disease ⁸²

France

The National Health System Database (Système National des Données de Santé -SNDS) which covers more than 99% of the French population was used to detect newly diagnosed patients with NTM-PD not previously treated (using outpatient drug consumptions) or hospitalized for NTM-PD in the last 3 years, using ICD-10 codes. A stable incidence of NTM-PD between 1.025/100,000 in 2010 and 1.096/100,000 in 2017 was reported. The NTM-PD prevalence was estimated at 5.92/100,000 inhabitants over 8 years ⁶⁹. The dataset did not include untreated NTM-PD patients (incidence and prevalence probably underestimated); however, the stable incidence is likely valid for France from 2010 through 2017 (Figure 4).

Figure 4.

Incidence of NTM-PD in France by treatment status from 2010–2017. From Veziris N, Andréjak C, Bouée S, Emery C, Obradovic M, Chiron R. Non-tuberculous mycobacterial pulmonary diseases in France: an 8 years nationwide study. BMC Infect Dis. Nov 17 2021;21(1):1165. doi:10.1186/s12879-021-06825-x; with permission. (Figure 1 in original)

Germany

In Germany, a population-based cohort study with a nested case–control design used ICD-10 codes from > 80 German company statutory health insurance datasets for the years 2010–2011 to define 125 patients with NTM-PD and to calculate yearly incidence rates of 1.12/100,000 in 2010 and 1.48/100,000 in 2011 ²⁶. Similarly, an anonymized German health claims database with an ICD-10 code 2011–2016, was analyzed to find 218 incident NTM-PD patients; prevalence of 3.79/100,000 was estimated. Using a prediction model; the prevalence of NTM-PD for both ICD-10-coded and non-coded individuals was five-fold higher at 19.05/100,000 ⁷¹. A third German study based on the Health Risk Institute health services research database (subset of ≈7 million persons) used ICD-10 codes to identify NTM-PD patients from 2009 to 2014. The data showed an increase in the annual overall prevalence rate from 2.3/100,000 to 3.3/100,000 ⁷². Using ICD-10 codes, it is probable that these studies underestimate the incidence and prevalence of NTM-PD as agreement of ICD-10 codes with the ATS diagnostic criteria has not been studied in Germany. The positive predictive value of ICD-9-CM codes ranged from 57–64%, and sensitivity for the diagnosis of NTM-PD to be low ranging from 21% to 26.9% ⁸³.

Greece

Adult in- and outpatients of the second largest tertiary referral hospital for patients with respiratory diseases in Athens, were retrospectively assessed for the presence of NTM isolation and NTM-PD using the ATS criteria. The reported incidence of pulmonary isolation (18.9/100,000 patients) and disease (8.8/100,000 patients) during the study period (2007–2013) was calculated as the total number of patients with respective isolation and disease divided by the total number of patients who attended this hospital as in- or outpatients, e.g. do not reflect the population of the area ⁸⁴.

Portugal

All 632 NTM patients treated at tuberculosis outpatient centres in Portugal (defined as NTM-PD patients) that were recorded in the electronic database (structured questionnaire and stored in the National Tuberculosis Surveillance System (SVIG-TB), Lisbon, Portugal) between 2002–2012 were retrospectively analyzed. The estimated annual incidence of treated NTM-PD was 0.54/100,000, during the study period, an increase of NTM-PD, mostly due to an increase in MAC-PD, was described ⁷³. The inclusion of only treated NTM-PD patients may have led to an underestimation; however, the fact that 41 of the treated patients had M. gordonae isolates leaves questions regarding the accuracy of the classification of NTM-PD in the patient cohort ⁸⁵.

Serbia

In Serbia, a retrospective and noninterventional study was performed by the national TB laboratory network representative for the whole country. Overall, 565 patients with 777 pulmonary NTM isolates collected between 2010 and 2015 were analysed for the presence of NTM-PD using the ATS microbiologic criteria. The annual isolation incidence of NTM increased from 0.9/100,000 to 1.6/100,000 without reaching statistical significance. In contrast, annual incidence rates of NTM-PD increased from 0.18/100,000 to 0.47/100,000. The most frequent NTM species in this country was M. xenopi (with annual incidence rates of 0.04/100,000–0.11/100,000), the annual incidence rate for MAC-PD was low (between 0.01/100,000 and 0.07/100,000) ⁷⁴.

Spain

In the Basque region in northern Spain, incidence rates of NTM-PD (as defined by ATS criteria) were calculated using the data from the Microbiology Laboratory and clinical records. Incidence of NTM-PD decreased from 1997–2016 significantly over the years from 10.6/100,000 and 8.8/100,000 in 2000 and 2001, respectively, to 1.8/100,000 in 2016. This decrease was solely explained by an annual decrease of 6.5% in M. kansasii-PD, whereas M. avium-PD had an annual rise of 3.3% ⁷⁸. In 13 municipalities in the Barcelona-South Health Region of Catalonia, a similar trend for the annual prevalence rates per 100,000 population for both M. kansasii and MAC isolation and pulmonary disease has been calculated for the period between 1994–2014, with similar high decreases of 9% for isolation of, and 11% for disease due to M. kansasii and increases of 10% for isolation of and 13% for disease due to MAC. Here overall pulmonary disease prevalence was calculated at 42.8/100,000 ⁷⁵. Rapid-growing mycobacteria isolated at a reference mycobacterial laboratory of the Catalan region similarly increased 5-fold from 2003 (0.34/100,000) to 2015 (1.73/100,000) ⁷⁶. The three studies performed in Spain during 1997 and 2016 show an increase of the incidence of MAC-isolation, of MAC-PD and of isolation of rapid growing mycobacteria, but a decrease in M. kansasii-isolation and M. kansasii–PD, supporting the conclusion that epidemiology of NTM should consider the different mycobacterial species separately.

The Netherlands

A study in the Netherlands used four separate databases, including two drug dispensing databases, an ICD-10 code database and a hospitalization database to estimate the prevalence of NTM-PD for the whole country between 2012 and 2019 by assigning the patients to “probable” NTM-PD patients and “possible” NTM-PD patients depending on the drug regimen used in the drug prescription databases, and to “confirmed NTM-PD patients” when the ICD-10 code A31.0 had been used. Prevalence estimates of the ICD-10 codes were corrected for the limited sensitivity (50%) for NTM-PD seen in previous studies and ranged in the different databases between 2.3/100,000 and 5.9/100,000 inhabitants, whereas a simultaneously performed survey amongst pulmonologists (likely over-) estimated the annual NTM-PD prevalence between 6.2/100,000 and 9.9/100,000. No increase was seen during the study period, in any of the study populations ⁷⁰ (Figure 5)

Figure 5.

a. Annual prevalence per 100,000 of NTM-PD in the Netherlands by database (2012–2019). From Schildkraut JA, Zweijpfenning SMH, Nap M, et al. The epidemiology of nontuberculous mycobacterial pulmonary disease in the Netherlands. ERJ Open Res. Jul 2021;7(3)doi:10.1183/23120541.00207-2021; with permission. 5b. Mean prevalence per 100,000 (2012–2019) of NTM-PD by database (Schildkraut et al., 2021); with permission. (Figure 1 in original). Reproduced with permission of the ERS 2023. ERJ Open Res 7: 00207–2021; DOI: 10.1183/23120541.00207-2021 Published 12 July 2021.

The United Kingdom

Primary care electronic healthcare records (Clinical Practice Research Datalink) representing 6.8% of the UK population were used to extract data for patients with NTM isolates during 2006 and 2016 and to classify patients as highly likely to have NTM-PD (‘strict cohort’, 85% of these being treated) and possible NTM-PD (‘expanded cohort’). A significant decrease in the incidence in the strict cohort from 2006 (3.85/100,000 person-years) to 2016 (1.28/100,000 person-years) was paralleled by a decrease of NTM-PD prevalence from 7.68/100,000 to 4.7/100,000 in primary care patients. This trend probably represents a shift of care of these (treated) patients towards secondary care ⁷⁹. This assessment is supported by a study from England, Wales and Northern Ireland that included all culture positive NTM isolates between 2007 and 2012 reported to Public Health England by five mycobacterial reference laboratories. The NTM isolation incidence in people with pulmonary isolates rose significantly from 4.0/100,000 to 6.1/100,000, driven by the MAC isolation incidence increasing from 1.3/100,000 in 2007 to 2.2/100,000 in 2012 ⁷⁷. Scottish isolates of NTM that had been submitted between 2000 and 2010 to the Scottish Mycobacteria Reference Laboratory were used to assign NTM-PD status to the respective patients using the ATS microbiologic criteria. A mean rate of 2.43 episodes of infection (that included extrapulmonary infections)/100,000 (range 2.06–2.71) was calculated, without significant trend over the study period. The incidence of NTM-PD was not calculated separately ⁸⁶.

African Region

Cohorts in African studies from 2014–2022 comprise patients with presumptive tuberculosis and/or patients with “chronic pulmonary TB” that were retrospectively analyzed. Data on NTM-PD were still rare and are mostly from studies from South Africa published before 2014 ⁸⁷, and thus not included in this review. Most of the studies have not performed chest imaging and seldom repetitive sputum isolation, so criteria for NTM-PD could not be applied. The data still suggest that a potentially significant proportion of African TB suspects may have NTM disease but may not be detected in routine care. The increasing availability of rapid TB-identification tests has decreased the number of studies reporting patients as being treated as MDR-TB while actually having NTM-PD. Good population-based studies to determine the incidence of NTM-PD in African countries are lacking; isolation prevalence in TB-suspects is reported here (Table 5). Similar to previous studies summarized by Okoi ⁸⁷ most studies report an isolation prevalence of below 10%. The ratio of growth of NTM to all mycobacterial growth varied significantly between 1% (Nigeria), 12% (Mali), 54% (Botswana) and 78% (Zambia), which may not be explained solely by different patient populations and detection methods (Table 5). The species isolated from respiratory secretions differ significantly within the African continent with MAC being the most frequent isolated species in most studies (Table 3). Maps of the continent with the distribution of the most frequent isolates from respiratory diseases and species causing NTM-PD have been published in a review covering African publications from 1940–2016 ⁸⁷.

Table 5:

Studies of Rates of Pulmonary NTM Isolates in Africa

Prevalence of Sputum Isolates
Location (dates)	Patient Cohort	Isolation Method	Ratio NTM/ Mycobacterial Growth (%)	Period Prevalence of NTM Isolation
				Period Duration (years)	Prevalence
Botswana (Aug 2012–Nov 2014)89	228 of 1940 symptomatic HIV infected patients with culture result	negative SD-Bioline Ag MPT64 assay (Abbott); GenoType CM and AS assays (Hain Lifescience)	53.4 (3.5 with two positive sputum cultures)	3	53.4% among culture positive specimen
Mali (2006–2013)92	41 of 439 patients suspected of having TB	negative MTBc Gen-Probe (AccuProbe); Nucleic acid probes for MAC, M. gordonae, M. kansasii (AccuProbe); gene sequencing	12.34	8	9.34% among individuals suspected of having TB
Nigeria, Cameroon, Ghana (Jan-Dec 2017)153	16 of 503 mycobacterial isolates from new and	negative IS6110 PCR; hsp65 PCR with sequencing	Nigeria: 1 Cameroon: 1.3 Ghana: 8.4	1	3.2% among isolates
	previously treated pulmonary TB patients
Sub-Saharan Africa (1940–2016)87	ATS criteria (systematic review) in 37 relevant studies	molecular techniques (n = 26), biochemical testing identification tools (n = 9); immunochromatographic assays (n = 2)		77	7.5% (isolation prevalence from all 37 papers reviewed)
Tunisia (2002–2016)88	60 of 1863 specimen from HIV-negative patients with presumptive clinical pulmonary TB with culture result	biochemical tests: niacin, nitrate, heat-resistant catalase and para-nitro benzoic acid; PCR targeting the recA intein.	3.2	15	isolation prevalence of 0.2/100,000 population of northern Tunisia.
Zambia (Aug 2013-July 2014)96	923 symptomatic patients with NTM-isolates among 6123 TB survey participants (1188 with mycobacterial growth)	negative capilia (TAUNS) lateral flow assay	77.7	2	1477/100,000 symptomatic NTM-patients among TB survey participants aged 15 years and above
Age and Sex Distribution and Pre-existing Pulmonary Diseases
Location (dates)	N	Mean Age (years)	Female (%)	COPD (%)/ Bronchiectasis (%)/TB (%)	Pre-existing Pulmonary Disease (%)
Botswana (Aug 2012 - Nov 2014)89	I: 228	NA; 45 patients > 50 years	53.90	-/-/12.4	-
Ethiopia (2017)93	I: 51	32.4 (of all TB-aNA NTM-pts)	51	-/-/-	-
Gabon (2018–2020)97	I: 1363	40	38.70	-/-/23.3	-
Ghana (2012–2014)90	I: 43	39.4	37.2	-/-/-	-
Ghana (Jan 2013-Mar 2014) 91	I: 473	39	68.4	-/-/-	-
Kenya (2020)94	I: 166	39	27	-/-/-	-
Mali (2006–2013)92	I: 41	35.8 years of all 439 pts	19.5	-/-/-	-
Nigeria, Cameroon, Ghana (Jan-Dec 2017)153	I: 503	45.14	35.30	-/-/-	-
Subsaharan Africa (1940–2016)87	D: 266 (from 37 studies)	35 years based on 17 of 37 studies with age data (isolation)	-	-/-/-	-
Tanziana (Nov 2019–Aug 2020)154	I: 24	47	33	-/-/-	Retreatment cases (60.9)
Tunisia (2002–2016)88	I: 60	-	30	-/-/-	-
Zambia (Aug 2013–July 2014)96	I: 923	NA; age groups 15– 24: 8.7%; 65+: 28.6%	51.70	-/7.15/-	Chest X-Ray: Nodules (22), Cavitation ( 13.8), Fibrosis (15)

Sub-Saharan Africa

A systematic review published in 2017 ⁸⁷ retrieved 37 out of 373 articles published from 1940–2016 from which data on pulmonary NTM isolation in patient cohorts from sub-Saharan Africa were extracted. The calculated prevalence of pulmonary NTM isolation was 7.5% and 16.5% (512 of 3,096) in those previously treated for tuberculosis. MAC was the most frequently isolated NTM in 19 of 37 studies with a large variation of MAC isolation prevalence from 15.0% in Tanzania to 57.8% in Mali. NTM-PD as defined in the ATS criteria was described in 7 of the 37 studies: 266 (27.7%) of 962 patients with sufficient data had NTM-PD. M. kansasii (69.2%) was the most common species amongst the isolates ⁸⁷.

North Africa

A study over a long period (2002–2016) looked retrospectively at sputum collected from HIV-negative Tunisian patients with presumptive pulmonary TB. 60 of 1,864 (3.2%) specimens grew NTM; from 30 isolates 7 were M. kansasii, none belonged to the MAC group, possibly indicating problems in the selection and identification process. The estimated NTM isolation prevalence of 0.2/100,000 population calculated for the regions of Bizerte and Zaghouan during the period 2002–2011, should thus be quoted with caution ⁸⁸.

Southern Africa

In Botswana, a 2012–2014 substudy of the Xpert evaluation study for TB detection performed in 22 clinical sites (representing 12 out of the 28 districts and included HIV patients). 228/427 (53.4%) of patients with mycobacterial growth had NTM (114 MAC, (109 M. intracellulare), 16 M. gordonae, 9 M. malmoense, 8 M. simiae ), only 8 of these (3.5%) had two positive sputum cultures and presumptive NTM-disease ⁸⁹.

West Africa

Two studies from 2012–2014 were reported from Ghana looking at 1,755 mycobacterial isolates from smear positive sputum samples ⁹⁰ in a HIV-positive cohort of 571 patients ⁹¹. The NTM isolation rate in smear positive sputum was 2.5%, identified by hsp65 gene PCR as M. avium subs paratuberculosis (30.2%), M. intracellulare (41.9), and M. abscessus (11.3%) ⁹⁰. In the HIV cohort, the NTM isolation rate was 51% (50/98 patients with mycobacterial growth). 38 NTM isolates were speciated: nine were MAC (23.7%), 3 were M. simiae (7.9%) and 3 were M. chelonae complex (7.9%). Sixty-five percent of the patients with NTM isolates had a CD4-count <100, 24% had died during the 6 months of follow up, similar to 30% of TB patients ⁹¹. In a retrospective single-center study from Mali 2006–2013, NTM were detected in 41/439 (9.34%) TB suspects and 41/332 patients with mycobacterial isolations (12.3% isolation). Of the 20 species identified from patients; 50% were M. avium ⁹².

East Africa

In Ethiopia in 2017, 51 of 697 (7.3%) patients with presumptive pulmonary TB had NTM isolates, 34 had been misclassified as TB (14 new cases, 12 relapses, 7 treatment failures); species were identified in 35 isolates (42.9% M. simiae, 14.3% M. abscessus) ⁹³. A study from Kenya in symptomatic patients detected 166 sputum isolates that were TB negative; of 146 identified, 122 were NTM, mostly MAC (31%), M. fortuitum complex (20%) and M. abscessus complex (14%) ⁹⁴. In 2012–2013 in 744 sputum samples from 372 TB suspects in Tanzania, mycobacterial cultures were positive in 121 (32.5%) patients, 36 (9.7%) were NTM. The most frequent species of the 28 identified were M. gordonae and M. interjectum (each 6/28) ⁹⁵. Our search retrieved only one population-based Zambian study ⁹⁶. 923 participants in a 1-year (Aug 2013-July 2014) national TB prevalence survey, NTM positive sputum cultures and negative TB test were defined as NTM patients without further species differentiation. In this cohort, 923/1,188 (78%) of positive cultures were defined as NTM; 265/1,188 (12%) were MTB. Of the participants with NTM, 71% were symptomatic. The prevalence of symptomatic NTM was estimated at 1,477/100,000⁹⁶.

Central Africa

In 1,363 sputum samples collected from presumptive TB patients in Gabon (2018–2020) in a mixed retro- and prospective cross-sectional study, 26.6% (137/515) patients with mycobacterial isolates had NTM, most were MAC (61.4%, 81/132) with M. intracellulare dominating (74/81). The next most common species were M. fortuitum (21.9%) and M. abscessus (6.6%)⁹⁷.

Asian Region

Population-based epidemiological data have been reported from Taiwan, Korea, and Japan, using primarily insurance claims data since 2014. These reports showed a consistent upward trend in these countries. Importantly, the incidence has surpassed that of TB in Japan, similar to Western countries, and the estimated prevalence reached approximately 150/100,000 in 2020. The species distribution shows M. intracellulare predominance in China and Korea, but M. avium is common in Japan. Interestingly, the proportions of cases with M. abscessus species are higher in the southern regions of China, Taiwan, and Japan. The proportion of NTM is also increasing in culture-positive specimens in China. Middle-aged female predominance is reported in most countries. However, patients with underlying lung diseases, such as COPD and previous TB, seem more common in Taiwan and other countries than in Korea and Japan. More epidemiological reports from Asian countries are warranted.

South Korea

During 2014–2022, population-based data on the incidence and prevalence of NTM-PD using the National Health Insurance Service (NHIS) database were reported in South Korea ^{98, 99}. Park et al reported that the incidence rate increased from 1.0/100,000 in 2002 to 17.9/100,000 in 2016; the prevalence rate was 1.2/100,000 in 2003 and increased to 33.3/100,000 in 2016 ⁹⁸ (Figure 6). The 5-year mortality rate was 17.8% (2003–2016), higher among men than among women (28.3% vs. 9.9%). Summarizing population-based and single-center studies, the proportion of NTM-PD patients who were women in South Korea was 56.9–61.1%, with a mean age of 53.0–62.7 years ^98–100. Species in the M. avium complex (63.6–75%) were most commonly identified, followed by subspecies of M. abscessus (10.8–22%). M. intracellulare (38.3–50.6%) was isolated more frequently than M. avium (17.5–23.1%), in most regions, although there were exceptions ^100–104.

Figure 6.

a. Age-adjusted incidence of NTM infection per 100,000 population by sex (2003–2016). From Park SC, Kang MJ, Han CH, et al. Prevalence, incidence, and mortality of nontuberculous mycobacterial infection in Korea: a nationwide population-based study. BMC Pulm Med. Aug 1 2019;19(1):140. doi:10.1186/s12890-019-0901-z; with permission. 6b. Incidence of NTM infection per 100,000 population by age group (2003–2016) (Park et al., 2019); with permission. (Figure 2 in original)

Japan

A nationwide laboratory data-based study analyzing 7,167 NTM-PD using the ATS microbiologic criteria found a period prevalence of 24 during 2012–2013 ¹⁰⁵. The most prevalent species was MAC (97.5%), followed by M. kansasii and M. abscessus (2–3%). In contrast to South Korea, the proportion of M. avium was higher than that of M. intracellulare (65.1% vs. 32.4%). Interestingly, the proportion of M. intracellulare gradually increased from the eastern to the western part of Japan and further increased to South Korea ¹⁰⁶(Figure 7). Furthermore, the proportion of M. abscessus was higher in the Kyushu-Okinawa regions close to Korea and Taiwan. An insurance claims data analysis showed that the incidence and prevalence rates were 8.6/100,000 treated individuals and 29.0/100,0000 treated individuals, respectively¹⁰⁷. Because this study used ICD codes to define disease, prevalence and incidence are likely underestimated. Patient characteristics were similar in South Korea, with an older female predominance^{105, 108, 109}, but lung disease complications were less common compared to South Korea. However, although bronchiectasis was common in females, COPD, sequelae of TB, and interstitial pneumonia were higher in males with age. Analysis of mortality data found that the crude mortality rate increased from 0.003/100,000 to 1.93/100,000 from 1970 to 2016 ^{23, 110}. In Japan, annual health exams with regular chest imaging are optional for citizens over 40 and mandated for all workers over age 35. In South Korea, these exams are biannual for persons over age 40, with high compliance rates. These annual exams may lead to detection of milder cases, but are unlikely to explain recent increasing trends, as they have been in place for more than 30 years ¹¹¹.

Figure 7.

a. Incidence of M. intracellulare-PD per 100,000 population in 344 medical regions of Japan (2011–2013). From Morimoto K, Ato M, Hasegawa N, Mitarai S. Population-Based Distribution of Mycobacterium avium and Mycobacterium intracellulare in Japan. Microbiology Research. 2021;12(3):739–743; with permission. 7b. Incidence of M. avium-PD per 100,000 population in 344 medical regions of Japan (2011–2013) (Morimoto et al., 2021); with permission. (Figure 1 in original)

Taiwan

Well-analyzed studies using several methodologies have been published ^112–116. The age-adjusted incidence rate increased from 5.3/100,000 in 2005 to 14.8/100,000 in 2013 ¹¹². The data among treated cases, an incidence and prevalence of 0.54/100,000–3.35/100,000 and 0.68/100,000–7.17/100,000, respectively, during the study period from 2003 to 2018 ¹¹³. MAC was predominant (34.4–41.5%) ^114–116, but the proportion of M. abscessus-chelonae complex species (24.3–30.8%) was higher than that for South Korea and Japan. A multicenter study reported that the proportion of M. abscessus was higher in southern Taiwan, while MAC was predominant in the northern part of the country ¹¹⁶. Male predominance was reported in most of the reports, which differed from South Korea and Japan ^112–116.

China

No population-based data were found in China or other Asian regions. NTM represented between 4.8–8.5% of all mycobacteria isolated and between 6.6–15.5% of all pulmonary mycobacterial disease cases., with an upward trend ^117–119. The isolates and disease proportion were higher in the south coastal area. MAC was predominant, followed by M. abscessus and M. kansasii. The proportion of M. intracellulare (28.2–70.1-%) was higher than that of M. avium (2.8–16.3%) ¹²⁰, similar to South Korea. The proportion of RGM was higher in the southern region than in the other regions.

Southeast to Western Asia

The proportion of NTM in mycobacterial isolates in Turkey and Pakistan was 5.5–16.2% ^{121, 122}. Single-center studies were reported in Iran^{123, 124} and India^125–127, but the clinical trends and species need to be clarified in the future.

Oceania

In Australia, epidemiologic data are based primarily on notifications from the state of Queensland, where notification of mycobacterial infections is mandated. The number of reported cases in 2015 was 1,222 (25.9/100,000)³, showing a consistently upward trend. Nearly 40% of cases were M. intracellulare, followed by M. avium at 10% and M. abscessus at 8.5%. An isolated strain analysis was reported from French Polynesia and Papua New Guinea ^{128, 129}.

Summary and Research Gaps

NTM-PD and NTM pulmonary isolation are increasing in most countries and regions globally. Isolate-based analysis in several countries indicate distinct trends by NTM species, highlighting the need for more species-specific analyses in NTM epidemiology. Available data regarding species indicates that MAC continues to predominate in most regions, although in Ontario, Canada and distinct European countries M. kansasii, M. xenopi, and M. malmoense are found with greater frequency. Variation in methods and case definitions limit comparisons across countries and regions. Population-based data including incidence and/or prevalence are available from North America, Europe, East Asia (Japan, South Korea, and Taiwan), and Queensland, Australia. Studies from Central and South America, Africa, China, the Middle East, and South Asia (India) among patients with suspected TB and MDR-TB are important to document the burden of undiagnosed NTM in this population and ensure appropriate treatment. This highlights the need for enhanced mycobacterial laboratory capacity to discriminate between cases of TB and NTM by routine speciation of mycobacterial isolates. Clinical data available from several regions have allowed better characterization of the affected populations. Across regions, the majority of cases occur among persons aged >50 years. In North America and East Asia, the proportion with COPD is typically less than 20%, whereas in Europe the proportion with COPD is higher, above 30% in 9 of 13 studies with this information.

Implementation of surveillance systems with standard case definitions is needed to allow for global comparison of prevalence and trends. Current data where NTM-PD is a notifiable condition have shown the importance and utility of these surveillance data. In particular, timely notification data have allowed inference of the incubation period for infection and disease and have further allowed identification of specific high risk environmental niches and environmental risk factors.

Key points.

1. Population-based data from North America, Europe, and East Asia (primarily Japan, South Korea, and Taiwan) demonstrate continued increases in NTM isolation and disease across regions and in most countries.

2. In countries and regions where population-based data are not available, including countries within the African continent, regions in China, as well as most of Central or South America, NTM species identification among persons screened for TB provides a measure of the unrecognized burden of NTM.

3. NTM-PD incidence and prevalence remain generally higher in East Asia (based on data primarily from South Korea, Japan, and Taiwan), North America, and Australia than in Europe

4. Species-specific differences exist in prevalence, trends, and distribution within countries and regions, and should be considered when analyzing trends data: in most geographic areas globally, increases are driven by MAC or M. avium disease.

5. Cumulative exposure to soil and water aerosols increases the risk of infection and disease in high-risk populations.

6. Water quality and composition influences the risk for NTM infection, as has been shown in a study analyzing the concentrations of vanadium and molybdenum in source water for municipal water systems which are associated with an increased risk of MAC and M. abscessus complex in the US.

Synopsis.

NTM isolation and pulmonary disease (NTM-PD) have continued to increase in most regions of the world, driven mainly by M. avium. Single-center studies also support increasing trends as well as a persistent burden of undiagnosed NTM among persons suspected of having tuberculosis (TB), in countries with moderate to high TB prevalence. Cumulative exposure to water and soil presents an increased risk to susceptible hosts, and trace metals in water supply are recently recognised risk factors. Establishing standard case definitions for subnational and national surveillance systems with mandatory notification of NTM-PD are needed to allow comparisons within and across countries and regions.

Clinics Care Points:

When physiicians or other health care providers suspect NTM pumonary infection or disease, they should collect appropriate samples including 3 samples for AFB culture, in addition to chest X rays or High Resolution CT (HRCT).