Nontuberculous Mycobacterial Infections in Denmark: Characterization of an Emerging Disease

Background:
Nontuberculous mycobacteria (NTM) are a heterogeneous group of ubiquitous environmental bacteria with almost 200 species identified. Disease caused by NTM can be categorized as pulmonary NTM (P-NTM) or extrapulmonary NTM (E-NTM). P-NTM is most prevalent and often associated with previous lung disease, whereas E-NTM often affects immunocompromised patients.
Globally, the incidence and prevalence of NTM disease have been debated and are thought to be increasing, but the previous Danish data do not confirm this. Describing the incidence, prevalence, and sociodemographic factors of both P-NTM and E-NTM in Denmark will provide valuable insights that can aid in earlier patient identification and, thus, earlier treatment.
The literature has shown that P-NTM is associated with a high mortality rate but variability in reporting methods and case identification. E-NTM-related mortality has only rarely been reported and only with a short follow-up time. Very few studies report on the causes of death in P-NTM patients, and even fewer for E-NTM.
This underlines the importance of understanding the current mortality of both P-NTM and E-NTM, as well as investigating the causes of death.

Aim of thesis:
1.We aimed to determine the incidence and prevalence of P-NTM and E-NTM disease from 2000 to 2017 in Denmark and characterize the demography and social determinants of the patients.

2.We aimed to describe the mortality of P-NTM patients and evaluate the causes of death associated with the P-NTM diagnosis.

3.We aimed to investigate the mortality associated with an E-NTM diagnosis and the causes of death.

Methods:
This thesis is based on three nationwide retrospective register-based studies in Denmark. We established a nationwide NTM cohort of adult patients with a first-time ICD-10 code of NTM disease from 2000-2017, divided into P-NTM and E-NTM groups and matched 1:4 to controls.

Paper I: We estimated the incidence rates, prevalence and demographic variables for both E-NTM and P-NTM patients
Paper II and III: We analyzed causes of death as well as all-cause mortality for E-NTM and P-NTM.

Results:
Paper 1: We found 1,146 adults with an NTM diagnosis. P-NTM: 661 patients (median age, 66 years, 50.4% male). E-NTM: 485 patients (median age 57 years, 59.6% male). The yearly incidence rate (per 100,000) increased significantly from 2000-2017 for both P-NTM (0.4-1.3) and E-NTM (0.3-0.6). The incidence rate increased with patient age and was common in 70+ year-olds (19.3 per 100,000). Annual prevalence (per 100,000) increased as well for P-NTM (0.4-3.5) and E-NTM (0.3-1.0). E-NTM patients were more likely to be employed and had a higher educational level than P-NTM.
Paper 2: The 5-year mortality rate for the 661 P-NTM patients was 41% (95%Conficende interval (CI): 37-44%) compared to 11% (95%CI: 10-12%) for matched controls. P-NTM patients had a significantly higher crude hazard ratio of death at 3.1 (95%CI: 2.7-3.5), p<0.001. The hazard ratio of death remained higher when adjusting for the Charlson Comorbidity Index (CCI) at 1.9 (95%CI: 1.6-2.2) p<0.001. Death due to respiratory diseases was significantly more common for P-NTM, and mycobacterial infections were reported as the cause of death in 5.8%.
Paper 3: 5-year mortality for E-NTM patients was 27% (95%CI: 23-31%) compared to 11% (95%CI 10-12%) for controls. The crude hazard ratio of death was significantly higher for E-NTM patients (1.9; 95%CI: 1.58-2.29) p<0.001, which persisted after adjusting for CCI (1.3; 95%CI: 1.1-1.6), p<0.01. E-NTM patients had significantly higher CCI than controls (p<0.001). Cardiovascular disease and solid malignant neoplasms were the two most common causes of death for both E-NTM and controls, although the proportions were lower in the E-NTM group compared to controls. E-NTM died more due to hematological malignancies, other bacterial diseases, and HIV compared to controls.

Conclusion:
NTM incidence increased significantly from 2000 to 2017, with the highest burden in patients over 70 years. P-NTM and E-NTM patients are two distinct groups that differ in age, gender, education, and employment status.
A significantly higher mortality rate is found in P-NTM compared to controls. The causes of death are primarily respiratory and infectious diseases. E-NTM is also associated with an increased mortality compared to controls. The causes of death in E-NTM patients are very similar to controls, although with more deaths attributed to infections. Increased suspicion of P-NTM disease is warranted, especially in patients over 70 years.
This thesis underscores the emergence of NTM disease in Denmark and the high loss of life accompanying it. It stresses the need for improvement in patient identification and treatment.