Changing Epidemiology of Pulmonary Nontuberculous Mycobacteria Infections

Nontuberculous mycobacteria (NTM) disease is a notifiable condition in Queensland, Australia. Mycobacterial isolates that require species identification are forwarded to the Queensland Mycobacterial Reference Laboratory, providing a central opportunity to capture statewide data on the epidemiology of NTM disease. We compared isolates obtained in 1999 and 2005 and used data from the Queensland notification scheme to report the clinical relevance of these isolates. The incidence of notified cases of clinically significant pulmonary disease rose from 2.2 (1999) to 3.2 (2005) per 100,000 population. The pattern of disease has changed from predominantly cavitary disease in middle-aged men who smoke to fibronodular disease in elderly women. Mycobacterium intracellulare is the main pathogen associated with the increase in isolates speciated in Queensland.

NTM disease is not a reportable condition in most countries because no evidence of human-to-human transmission exists; therefore, it is not considered a public health concern. However, the organisms are ubiquitous in the environment, and substantial evidence shows that the environmental niche for Mycobacterium intracellulare (the most common pulmonary pathogen) is in biofi lms lining suburban water pipes. Many NTM pathogens have been isolated from drinking water (7). Some clinicians believe the condition should be classifi ed as an environmental health concern, similar to that caused by Legionella spp.
Globally, geographic variability in environmental exposure and prevalence of NTM disease is signifi cant (8). Without detailed clinical information, differentiating between contamination of specimens, colonization/infection, and disease is diffi cult; laboratory reports of isolates do not always refl ect the true incidence of disease. To determine if disease is present, sputum specimens and often a bronchoscopic sample of a patient's lower respiratory tract must be collected. In addition, computed tomography scanning and clinical appointments with primary care providers and specialists are needed. Because this investigative process is costly to the healthcare system and the patient, accurate epidemiologic data on this condition should be of interest to public health experts.
Studies of avian versus human Mantoux testing in schoolchildren have shown that exposure to NTM organisms is common in Queensland (9)(10)(11). Therefore, since the introduction of TB control services, disease caused by NTM in Queensland has been notifi able. This practice has been continued primarily to avoid confounding of smearpositive cases with TB. The notifi cation process provides a unique opportunity to study the clinical signifi cance of isolates positive for NTM and the features such as age and sex, symptoms, underlying conditions, and radiology results of patients with disease, avoiding the inherent bias that occurs in case series that are reported by tertiary and quaternary referral centers. The incidence of clinically reported cases of pulmonary disease caused by M. avium complex (MAC) in Queensland has been increasing (1985, 0.63/100,000 population; 1994, 1.21/100,000; and 1999, 2.2/100,000). In 2005, the Queensland Tuberculosis Control Centre (QT-BCC) revised the notifi cation process to ensure collection of meaningful clinical data and follow-up of clinically signifi cant NTM cases.

Methods
At the time of this study, public hospitals requesting mycobacterial cultures forwarded all specimens to the Queensland Mycobacterial Reference Laboratory (QMRL). Requests in the private sector are processed by 2 main laboratories and a few smaller laboratories. The 2 pri-vate laboratories (representing >90% of the state's private mycobacterial pathology service) identify mycobacteria in specimens and report as "atypical mycobacteria, not further specifi ed." After culture, not all isolates are forwarded to the QMRL for species identifi cation. However, isolates are forwarded if the specimen is smear positive (to exclude TB), was obtained from a bronchial wash or a site other than sputum, and if >1 specimen is positive for that patient or the treating clinician requests it. In 1999, a specifi c research project was undertaken to capture all unspeciated isolates from private laboratories and better assess the true impact of mycobacteriology in the state. This study was approved by the Research Ethics Committee of the Princess Alexandra Hospital (138/98) in accordance with the Australian National Health and Medical Research Council guidelines. Because it proved quite labor intensive for all isolate reports to be forwarded for this study, it did not continue after 1999. Reports of all positive cultures identifi ed by QMRL are forwarded to the QTBCC, which sends questionnaires to clinicians asking for the clinical signifi cance of all mycobacterial isolates. Clinicians are provided with guidelines (as per American Thoracic Society/Infectious Disease Society of America [ATS/IDSA]) to assist in the diagnosis of signifi cant disease.
More detailed clinical information is requested for patients who have disease, including details on predisposing conditions, radiologic and bacteriologic test results, and treatment received. Provision of this information is voluntary. Follow-up surveys evaluate any changes in regimen brought about by side effects or intolerance to drugs or failure of the starting regimen. Bacteriology is monitored for success of treatment. An update on the patient's status is requested yearly for 3 years after treatment has stopped. All case fi les notifi ed as clinically signifi cant disease are reviewed by QTBCC clinicians to confi rm concordance with ATS/IDSA criteria for disease (12).
Most nonpulmonary isolates grown by private laboratories are sent to the QMRL for speciation. For patients with nonpulmonary NTM isolates, the clinical signifi cance of the isolate is sought by QTBCC; species of organism and site of infection is recorded. However, details on individual treatment regimens and outcomes are not collected.
Statistical calculations were performed by using Ope-nEpi: Open Source Epidemiologic Statistics for Public Health, version 2.3. (www.openepi.com). Population statistics were obtained from the Australian Bureau of Statistics (13,14), including population data for different age categories. The rate of positive NTM isolates is reported as number per 100,000 population, as are age-specifi c and gender-specifi c rates of pulmonary disease. Comparisons of statistics for 1999 and 2005 were made by using χ 2 2 × 2 tables in OpenEpi.

Results
A more aggressive capture of single, positive, unspeciated isolates occurred in 1999; the total number of NTM isolates was 14.8/100,000 population. Because the number of speciated pulmonary isolates in each year (1999 and 2005) should not have been affected by the increase in data collection in 1999, the increase in speciated isolates for these 2 years most likely refl ects the true increase in signifi cant isolates (9.1-13.6/100,000 population). The total number of isolates in 2005 was 15.8/100,000 population. Assuming the proportion of unspeciated isolates remained constant, and on the basis of the increase in speciated isolates, we estimated the total number of isolates in 2005 to be 22.1/100,000 population.
The difference between the 2 years was due to the higher number of pulmonary isolates as demonstrated in Table 1. Speciated pulmonary isolates rose signifi cantly, from 5.5 to 10.2/100,000 population.
There were differences between the species of NTM The notifi ed signifi cance of pulmonary and extrapulmonary isolates is shown in Table 2. Of 488 patients with pulmonary NTM isolates, only 26.6% had signifi cant disease overall. Percentages were slightly higher for patients with M. intracellulare isolates (39.4%), M. avium (33.3%), M. kansasii (52.6%), and much lower for patients infected with species traditionally thought more likely to be contaminants, such as M. gordonae (11.1%). Other rarer species were not thought to be associated with pulmonary disease (Table 3).
Of the isolates from nonpulmonary sites, the most common species was M. fortuitum (33/143), followed by M. intracellulare (12), M. abscessus (11), and M. chelonae (11). Overall, 68.5% of nonpulmonary isolates were felt to be associated with clinically signifi cant NTM disease. Table 4 shows the statistical signifi cance of these isolates according to species.
The age and sex distribution of patients with significant pulmonary disease differed in 1999 and 2005. In 1999, the absolute number of men with disease outnumbered women in all age groups with the exception of the >75-year  age group. In 2005, women having NTM disease outnumbered men in all age groups >30 years; the difference was most marked in elderly persons. When adjusted for population within each age bracket and represented as rates per 100,000 (Figure 3), a signifi cant increase was seen in women with disease who were 60-74 years of age (p = 0.0005), a decrease in men 30-44 years of age (p = 0.030) There was a combined overall increase in patients 60-74 years of age (p = 0.002) and a decrease in those 30-44 years of age (p = 0.013).

Discussion
In 2005, NTM were isolated from 15.6 persons per 100,000 population; however, the true number of isolates from that year is estimated at 22.1/100,000, a signifi cant increase from 1999. Little change has been recorded in the isolation of NTM from nonpulmonary sites. However, the incidence of isolation of NTM from human pulmonary specimens has increased to 12.1/100,000 population. The incidence of notifi ed signifi cant pulmonary disease has increased from 2.2 to 3.2/100,000 population per year. This increase is largely due to an increase in the number of elderly women with noncavitary disease.
The observed changes in speciation are interesting epidemiologic observations, but their clinical relevance is yet to be defi ned. M. intracellulare is the main pathogen associated with the increase in disease, followed by M. abscessus, and M. kansasii, all clinically signifi cant pulmonary pathogens. We have observed a change in the age and gender ratio. NTM disease has increased in elderly female patients who have predominantly nodular bronchiectasis, but disease has decreased in middle-aged men.
The epidemiology of NTM disease is diffi cult to study and equally diffi cult to report in a meaningful way. We have reported the incidence of new cases of clinically signifi cant disease notifi ed by clinicians that met the 1997 ATS/ISDA criteria (12) for disease. The true number of cases of disease may be an underestimate because the experience and attitudes of clinicians with regard to assessing the clinical signifi cance of isolates is varied. It is likely that other cases of signifi cant disease were not notifi ed. The more recent ATS/IDSA standards of 2007 (15) have more relaxed microbiological criteria, and it is likely more case-patients would now be considered to have disease. Because NTM disease is a chronic disease, often over many years, the prevalence of disease will be much higher and more likely will refl ect the impact on the healthcare system. Even with the current notifi cation process in Queensland, we failed to capture all single, positive, unspeciated isolates.
A rising body of literature supports the observation by clinicians that the incidence and prevalence of disease is increasing. Other studies in Australia have also shown an increase over earlier time periods (16)(17)(18). The most recent report is from the Northern Territory. All isolates from 1989-1997 were examined (16). Disease incidence increased from 2.7 to 4.7 per 100,000 population between the fi rst and second halves of this time period. Pang re-    creased use of computed tomography scans, better laboratory techniques, or greater clinician awareness. Improved detection methods, such as the introduction of broth culture (MGIT BACTEC 960; Becton Dickinson, Franklin Lakes, NJ, USA) before the data collection of 1999, also cannot explain our observations. An increase in environmental exposure may have contributed to the increasing incidence of disease. The main environmental niche for M. intracellulare appears to be in biofi lms lining drinking water pipes (7). M. intracellulare has been isolated from drinking water in Brisbane, as have M. abscessus and M. kansasii (R. Thomson, unpub. data). During the study period, Queensland had been affected considerably by drought, and restrictions to domestic water use had been enforced. The reduction in fl ow within distribution systems may have led to an increased chlorine degradation time and, hence, lower disinfectant levels at point of use. Previously, M. scrofulaceum was the main pathogen associated with childhood lymphadenitis. A change from chloramination to chlorination of water led to a reduction in the isolation of this species. The resurgence of this species in 2005 supports the notion of increased chlorine degradation, brought about by fl ow reductions but potentially also by temperature increases associated with climate change.
Changes in behavior associated with water restrictions may also have been a factor in resurgence of disease. The changes in turbulence that occur with the turning on and off of taps lead to disruptions in biofi lms and release of mycobacteria into the water. However, some authors have suggested that patients may inhale aerosolized mycobacteria while showering (23,24). Recently, Feazel et al. (25) found mycobacterial DNA concentrated in the biofi lms of showerheads in the United States. Queensland implemented a strong public awareness campaign promoting 4-minute showers to reduce water usage, which in theory should reduce exposure to mycobacteria through this route.
Several case series have reported gender imbalances in this disease. Earlier studies have demonstrated a male predominance with cavitary disease (8,16,26). More recent studies have shown a female predominance (3,(27)(28)(29) of older patients with less preexisting lung disease. The differences between these groups of studies may be attributed to sampling bias and population differences. We have been able to demonstrate that, in fact, the pattern of disease is changing, from cavitary disease in middle-  aged men who smoke to the nodular bronchiectatic form of disease in elderly women. Queensland government statistics suggest that the median age of the population is predicted to rise from 36.3 years in 2006 to 45 years in 2051, and the number of persons >65 years of age will increase 4-fold by 2051. As the vulnerable population increases, NTM disease is also likely to increase. Therefore, the impact of this disease on health budgets is worthy of further attention.

Acknowledgments
The author thanks Robyn Carter and staff at the QMRL for the provision of laboratory isolates and Christopher Macdermott, Hiranthi Walpola, Christine Logan, Mark Stickley, and Anastasios Konstantinos for assistance in maintaining the NTM database. Thanks is also extended to the physicians in Queensland who volunteered information regarding individual patients for NTM data collection and the private laboratories (Sullivan and Nicolaides Pathology and Queensland Medical Laboratory) for additional information provided in 1999.
In 1999 R.T. was supported by a research scholarship from the Princess Alexandra Hospital Research Foundation. Dr Thomson is a thoracic physician at the QTBCC, The Prince Charles Hospital, and Greenslopes Private Hospital. She is also a PhD student researching environmental aspects of nontuberculous mycobacteria related to pulmonary infections.

Earning CME Credit
To obtain credit, you should fi rst read the journal article. After reading the article, you should be able to answer the following, related, multiple-choice questions. To complete the questions and earn continuing medical education (CME) credit, please go to http://www.medscapecme.com/journal/eid. Credit cannot be obtained for tests completed on paper, although you may use the worksheet below to keep a record of your answers. You must be a registered user on Medscape. com. If you are not registered on Medscape.com, please click on the New Users: Free Registration link on the left hand side of the website to register. Only one answer is correct for each question. Once you successfully answer all post-test questions you will be able to view and/or print your certifi cate. For questions regarding the content of this activity, contact the accredited provider, CME@medscape.net. For technical assistance, contact CME@webmd.net. American Medical Association's Physician's Recognition Award (AMA PRA) credits are accepted in the US as evidence of participation in CME activities. For further information on this award, please refer to http://www.ama-assn.org/ama/pub/category/2922.html. The AMA has determined that physicians not licensed in the US who participate in this CME activity are eligible for AMA PRA Category 1 Credits™. Through agreements that the AMA has made with agencies in some countries, AMA PRA credit is acceptable as evidence of participation in CME activities. If you are not licensed in the US and want to obtain an AMA PRA CME credit, please complete the questions online, print the certifi cate and present it to your national medical association.