- Email: [email protected]
A retrospective observational study of antimicrobial treatment for non-tuberculous mycobacteria disease using a nationwide claims database in Japan
J Infect Chemother xxx (xxxx) xxx
Contents lists available at ScienceDirect
Journal of Infection and Chemotherapy journal homepage: http://www.elsevier.com/locate/jic
Original Article
A retrospective observational study of antimicrobial treatment for non-tuberculous mycobacteria disease using a nationwide claims database in Japan* Hidetoshi Igari a, *, Kazutaka Yamagishi a, Shingo Yamazaki b, Shota Murata c, Misuzu Yahaba a, Shin Takayanagi a, Yohei Kawasaki d, Toshibumi Taniguchi a a
Division of Infection Control, Chiba University Hospital, 1-8-1 Inohana Chuo-ku, Chiba, 260-8677, Japan Division of Pharmacy, Chiba University Hospital, 1-8-1 Inohana Chuo-Ku, Chiba, 260-8677, Japan Department of Clinical Laboratory, Chiba University Hospital, 1-8-1 Inohana Chuo-Ku, Chiba, 260-8677, Japan d Biostatistics Section, Clinical Research Center, Chiba University Hospital, 1-8-1 Inohana Chuo-Ku, Chiba, 260-8677, Japan b c
a r t i c l e i n f o
a b s t r a c t
Article history: Received 12 July 2019 Received in revised form 26 August 2019 Accepted 21 October 2019 Available online xxx
The Macrolides (MLs), clarithromycin and azithromycin, are key drugs for non-tuberculous mycobacteria (NTM) diseases treatment. A three antibiotics regimen including MLs, rifampicin (RFP) and ethambutol (EB) has been recommended for the treatment of NTM diseases in ATS/IDSA guideline. However, antibiotics are not necessarily prescribed in compliance with the guideline. Inappropriate regimens are risk of introducing MLs resistance. Therefore, we planned this study to evaluate the current Japanese NTM diseases treatment conditions. We used the national database (NDB) from 2011 to 2014. A total of 183 patients were entered into the study. The patients number increased at an accelerating rate in patients aged 55 years. Patients aged 55 years made up 91.3% of the total NTM diseases. Male and female patients were 61 and 122, respectively, a female/male ratio of 2.00. Clarithromycin, RFP, EB and fluoroquinolones were frequently prescribed, with the numbers of prescriptions being 125, 66, 57 and 45, respectively. The regimen of MLs, RFP and EB recommend by ATS/IDSA guideline 2007 was only followed by 25.1% of the patients. MLs monotherapy was as high as 30.6% of NTM diseases and would be a risk factor leading to an increase of MLs resistance and poor treatment outcome. Without effective NTM disease therapy, the increase of MLs-resistant NTM diseases would be a burden for Japanese health care facilities. © 2019 Japanese Society of Chemotherapy and The Japanese Association for Infectious Diseases. Published by Elsevier Ltd. All rights reserved.
Keywords: Non-tuberculous mycobacteria disease National data base Treatment Epidemiology
1. Introduction The incidence of non-tuberculous mycobacteria (NTM) disease has increased and even overwhelmed that of tuberculosis in Japan [1,2]. The new macrolides (MLs), clarithromycin (CAM) and azithromycin (AZM) have been introduced for NTM treatment.
* All authors meet the ICMJE authorship criteria. All authors have seen and approved the manuscript, contributed significantly to the work. * Corresponding author. E-mail addresses: [email protected] (H. Igari), [email protected] (K. Yamagishi), [email protected] (S. Yamazaki), [email protected] (S. Murata), [email protected] (M. Yahaba), [email protected] (S. Takayanagi), [email protected] (Y. Kawasaki), [email protected] (T. Taniguchi).
Usually, a three-drug regimen consisting of the new MLs, rifampicin (RFP) and ethambutol (EB) has been recommended for the treatment of NTM diseases according to ATS/IDSA guideline [3]. However, antibiotics are not necessarily prescribed in compliance with the guideline owing to adverse events [4]. Adherence to the 2007 ATS/IDSA guideline for NTM disease has been poor. Suboptimal or potentially harmful antibiotic regimens are commonly prescribed in the United States [5]. Inappropriate regimens, such as MLs monotherapy and ML-including regimens without adequate companion drugs are shown to be a risk of introducing MLs resistance [6]. Thus, MLs-resistant NTM disease is difficult to control because of the lack of effective drugs [7]. Inadequate prescriptions, not recommended in the ATS/IDSA guideline, are common world-wide [4]. The prescription tendencies of Japanese physicians are also included in this study. Then, we
https://doi.org/10.1016/j.jiac.2019.10.009 1341-321X/© 2019 Japanese Society of Chemotherapy and The Japanese Association for Infectious Diseases. Published by Elsevier Ltd. All rights reserved.
Please cite this article as: Igari H et al., A retrospective observational study of antimicrobial treatment for non-tuberculous mycobacteria disease using a nationwide claims database in Japan, J Infect Chemother, https://doi.org/10.1016/j.jiac.2019.10.009
2
H. Igari et al. / J Infect Chemother xxx (xxxx) xxx
planned this study to evaluate the current Japanese NTM disease treatment conditions. Medical services are provided under the universal health insurance system in Japan. The Ministry of Health, Labour and Welfare (MHLW) in Japan provides the national database (NDB) for research purposes. We therefore conducted a retrospective study based on the national database (NDB) from 2011 to 2014. This database study would reveal the NTM diseases treatment in Japanese health care facilities. Our objectives are to estimate the antibiotics regimen for NTM diseases, with strong focus on the adherence to the guideline, and to describe the epidemiology of NTM diseases. 2. Materials and methods 2.1. Study design This is a retrospective cohort study using samples of the NDB, which randomly sampled 1% of the claim information associated with outpatient medical service data from every October from 2011 through 2014. Usually, we should adopt clinical diagnoses on International Classification of Diseases, 10th version (ICD-10) codes. Unfortunately, ICD-10 codes are not used for outpatient claim information. Then, we selected patients with their diagnosis containing pulmonary NTM diseases equivalent to ICD-10 codes corresponding to A31 (infection due to other mycobacteria, excluding leprosy and tuberculosis). However, ICD-10 A31 is consisted of the following three subgroups; A310 (pulmonary mycobacterial infection due to mycobacterium), A311 (cutaneous mycobacterial infection), and A318 (other mycobacterial infections) and A319 (mycobacterial infection, unspecified). Then, the patients corresponding to A318 and A319 were excluded. We further selected patients who were prescribed the following anti-NTM drugs; CAM, AZM, RFP, rifabutin (RBT), EB, aminoglycosides (AGs) and fluoroquinolones (FQs). And we checked the prescription of imipenem/cilastatin as is administered for NTM disease caused by M. abscessus. We used the following information: age, sex and drugs for NTM diseases treatment. The primary endpoint is to reveal the antibiotics prescribed for NTM diseases. We also analyzed combinations of antibiotics including MLs monotherapy. The secondary endpoint is to reveal the NTM disease frequency according to age group and sex. This study was approved by the Research Ethics Committee of the Graduate School of Medicine and School of Medicine, Chiba University, and was conducted in accordance with the principles of the Declaration of Helsinki. This retrospective cohort study was conducted in accordance with the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) guidelines. 2.2. Statistical analysis We calculated the number of patients and proportions for categorical data. The differences in baseline characteristics of subjects were compared by Chi-square test or Fisher's exact test for categorical variables. A two-sided P-value of <0.05 was considered statistically significant, and all analysis results were calculated by SAS Version 9.4 for Windows and JMP® Pro 13.0.0 (SAS Institute Inc., Cary, NC, USA). 3. Results A total of 183 patients were entered. Their distributions are shown by age groups and sex (Fig. 1). Median age was 70 e 74 years, as age information was classified in 5-year steps from age years in our database. The number of patients increased at an accelerating rate with age from 55 years. Patients aged 55 years totaled as
many as 91.3% of all NTM diseases patients. The numbers of male and female patients were 61 and 122, respectively, and the female/ male ratio was 2.00. No patients were being prescribed anti-HIV drugs. The prescribed anti-NTM drugs are shown in Table 1. CAM, RFP, EB, FQs were frequently prescribed, and the numbers of prescriptions were 125, 66, 57 and 45, respectively. The combinations of anti-NTM drugs are shown in Table 2. The prescriptions that included MLs (CAM or AZM) were for 126 patients. The regimen of MLs, EB and RFP (or RBT), as is recommend in ATS/IDSA guideline 2007, was for 46 patients. Only 25.1% of antibiotic regimens met ATS/IDSA guideline. MLs monotherapy was for 56 patients, 30.6% of the NTM diseases. Anti-NTM therapies without MLs were prescribed for 57 patients, representing 31.1% of the NTM disease patients. The therapies without CAM vary. FQs or AGs are frequently observed. The combination therapy by one to three drugs consisting of the antituberculosis drugs isoniazid (INH), RFP and EB were also counted. The details of anti-NTM therapies were stratified according to the three age groups of <65 years, 65 e 74 years and 75 years (Table 3). The recommended regimen consisting of MLs, EB and RFP (or RBT) did not reveal significant differences among the three age groups. The ratios of the regimen including MLs at age <65 years, 65 e 74 years and 75 years were 56.6%, 69.0% and 77.8%, respectively, and the differences among these three age groups were significant. (P < 0.05) However, the frequency of CAM monotherapy also increased in elderly patients. The percentages of MLs, in which CAM was mainly used as monotherapy, were 17.0%, 31.0% and 40.3% for <65 years, 65 e 74 years and 75 years, respectively, and were significantly different among these three age groups. (p < 0.05) Thus, the increase of MLs prescriptions in elderly patients resulted in the increase of monotherapy. 4. Discussion This study analyzed the current NTM disease treatment condition in Japan. At first, we demonstrated the age distribution of NTM diseases and the male/female ratio (Fig. 1). These results were almost the same trend as was shown by Izumi K et al. [2] Our sampled claim database could be used for the analysis for the antibiotics usage in NTM diseases patients. Our current study revealed that only 25% of NTM diseases patients received antibiotics therapy by the standard regimen recommended by ATS/IDSA guideline [3]. The other NTM diseases patients, 75% of total, are treated by other therapies. ATS/IDSA guideline 2007 recommends the combination therapy including CAM (or AZM), EB and RFP (or RBT), occasionally adding AGs [3]. Although MLs are key antibiotics for NTM disease treatment, the therapy including MLs remained as low as 68.9% in this study. Especially, MLs monotherapy counted as high as 30% of NTM disease patients. Morimoto et al. demonstrated that inappropriate NTM disease therapy lead to MLs-resistant NTM diseases [7]. Their study analyzed MLs-resistant pulmonary Mycobacterium avium complex (MAC) diseases, and revealed that about 30% of MLsresistant pulmonary MAC disease patients received MLs monotherapy. They concluded that CAM monotherapy was a risk factor for MLs resistance [3]. When MLs resistance is detected in NTM diseases, effective NTM treatment is restricted [6,8]. The current NTM disease treatment condition has a risk of an increase in MLs-resistant NTM diseases. It is obvious that further improvement for appropriate anti-NTM treatments is needed. The urgency of providing better education of appropriate NTM treatment for physicians caring for NTM disease patients is clear.
Please cite this article as: Igari H et al., A retrospective observational study of antimicrobial treatment for non-tuberculous mycobacteria disease using a nationwide claims database in Japan, J Infect Chemother, https://doi.org/10.1016/j.jiac.2019.10.009
H. Igari et al. / J Infect Chemother xxx (xxxx) xxx
3
Fig. 1. Numbers of nontuberculous mycobacteria disease patients by age groups. The numbers of non-tuberculous mycobacteria (NTM) diseases are calculated by age groups. Our data base is a national one from 2011 to 2014 reflecting the medical service provided under universal health insurance in Japan. NTM disease patients increase in elderly people aged 55 years or more.
We further analyzed NTM treatment regimen stratified according to age classifications. The percentage of the recommended regimen of CAM (or AZM), EB and RFP (RBT) with adherence to guideline was around 30% in this study, as in the above discussion, and these figures did not show significant differences among age groups (Table 3). The MLs prescription rate at age of <65 years was 56.6%, and this figure significantly increased in older patients. Nonetheless, this trend did not contribute to an increase in the recommended therapy based on the guideline, and resulted in MLs monotherapy. We understand that some patients discontinue the recommended treatment owing to adverse events, but a large majority of patients are prescribed antibiotic regimens that do not meet ATS/IDSA guideline 2007. NTM diseases become common diseases in consideration of NDB analysis by Namkoong et al. [1,2] Japanese physicians would have opportunities to treat NTM diseases even if their specialty is not respiratory disease or infectious disease. It is evident that we need further tactics for improving the adherence to guideline. In addition to the education for instruction of NTM treatment, we also anticipate progress in the research for developing more effective therapies and providing alternative therapies for patients with adverse events. We used NDB to assess the current NTM disease treatment in outpatients’ clinics across Japan. The number of NTM disease Table 1 Prescribed anti-NTM drugs (total patient number (n) ¼ 183). anti-NTM drug
n
(%)
clarithromycin azithromycin rifampicin rifabutin ethambutol fluoroquinolones aminoglycosides isoniazid azithromycin imipenem/cilastatin
125 1 66 2 57 45 17 11 1 0
(68.3) (0.5) (36.1) (1.1) (31.1) (24.6) (9.3) (6.0) (0.5) (0.0)
NTM: non tuberculosis mycobacteria.
patients has increased at an accelerating rate among elderly patients aged 55 years, and the patients of this age group consist of 91.3% of the total. Morimoto et al. analyzed the Japanese NTM disease bases on microbiology laboratory results, and patients with NTM increased among elderly patients aged 50 years, with a female/male ratio of 2.0 [2]. Our results also support their study results [9]. This study has several limitations. First, it was based on an analysis of NDB, and the details of every patient's information, especially causative microorganism, were none or restricted. Then, we could not sufficiently confirm the diagnostic background Table 2 Combination of anti-NTM drugs (total patient number (n) ¼ 183). Combination of anti-NTM drugs
n
(%)
including macrolide CAM, EB and RFP (or RBT) AZM, EB and RFP CAM, EB CAM, RFP (RBT) CAM, RFP and FQs CAM and FQs CAM monotherapy not including macrolide FQs alone AGs alone INH alone INH, RFP and EB INH and RFP RFP (RBT) and FQs RFP, EB and FQs INH and EB EB and FQs
126 45 1 5 8 5 6 56 57 28 14 4 3 3 2 1 1 1
(68.9) (24.6) (0.5) (2.7) (4.4) (2.7) (3.3) (30.6) (31.1) (15.3) (7.7) (2.2) (1.6) (1.6) (1.1) (0.5) (0.5) (0.5)
additional drugs þFQ 1, þAG 2 þFQ 1 þAG 1
NTM: non tuberculosis mycobacterium. CAM: clarithromycin. EB: ethambutol. RFP: rifampicin. RBT: rifabutin. FQ: fluoroquinolone. AG: aminoglycoside. AZM: azithromycin.
Please cite this article as: Igari H et al., A retrospective observational study of antimicrobial treatment for non-tuberculous mycobacteria disease using a nationwide claims database in Japan, J Infect Chemother, https://doi.org/10.1016/j.jiac.2019.10.009
4
H. Igari et al. / J Infect Chemother xxx (xxxx) xxx
Table 3 Anti-NTM drugs stratified according to three age classes.
Therapy including CAM or AZM CAM (or AZM), EB and RFP (or RBT) CAM monotherapy
<65 year (n ¼ 53) (%)
65e74 year (n ¼ 58) (%)
75 year (n ¼ 72) (%)
p value
30 15 9
40 16 18
56 15 29
<0.05 n.s. <0.05
(56.6) (28.3) (17.0)
(69.0) (27.6) (31.0)
(77.8) (20.8) (40.3)
NTM: non tuberculosis mycobacterium. CAM: clarithromycin. AZM: azithromycin. EB: ethambutol. RFP: rifampicin. RBT: rifabutin. n.s.: not significant.
supporting the probability of NTM diagnosis. The reliability of NTM disease might be secured on the prescription of antibiotics for the treatment of NTM diseases. MLs have anti-inflammatory effect other than antibiotics effect, and then some physicians prescribe MLs (usually CAM) for chronic bronchial inflammatory diseases like diffuse pan-bronchitis and chronic bronchitis. These uses are also prescribed for longer durations and would possibly have resulted in increases of resistant bacteria. Second, this study was conducted on sampling data consisting of just a one-month duration. The number of NTM disease patients was limited, and these data were limited to single-month periods, each October of 2011e14. NTM diseases are chronic diseases, and they do not resolve in a short period. We cannot confirm the outcome of longer treatment, more than onemonth, for NTM diseases. The yearly trend of NTM disease should be analyzed. Third, the species of NTM is not confirmed in this database. The most frequently detected NTMs are Mycobacterium avium and Mycobacterium intracellulare, occupying 83% in Japan [10]. Recently, NTM disease caused by Mycobacterium abscessus (M.abscessus) has increased. The treatment of M. abscessus disease consists of combination therapies based on MLs. Fourth, assessment of the severity of NTM diseases was not performed in this study. NTM diseases are not necessarily treated by antibiotics. There might be some patients under observation owing to their stable conditions. We cannot estimate these patients without antibiotics prescription. Fifth, the antibiotics for NTM diseases are used for infectious diseases other than NTM diseases. FQs and MLs are often used for respiratory infectious diseases. INH monotherapy is used for latent tuberculosis infection. The antibiotics listed in Tables 1 and 2 might be prescribed for infectious diseases other than NTM diseases. This study revealed the age class epidemiology of NTM diseases, confirming the previous study. The current conditions of NTM diseases therapy is not sufficient, as the frequency of NTM treatment based on guideline is as low as about 25%, and the MLs monotherapy is as high as 30%, which would be a risk factor leading to an increase of MLs resistance and poor treatment outcome. Without effective NTM disease therapy, the increase of MLs resistant NTM diseases is sure to be a burden for Japanese health care facilities. Authors’ contributions Hidetoshi - Igari Study planning, data analysis and draft writing.
Shin Takayanagi - Study planning. Kazutaka Yamagishi - Data analysis. Shingo Yamazaki - Data analysis. Shota Murata - Data analysis. Misuzu Yahaba - Study planning. Yohei Kawasaki - Data analysis. Toshibumi Taniguchi - Study planning, data analysis. Declaration of Competing Interest None declared. Acknowledgement This study was conducted without funding. References [1] Namkoong H, Kurashima A, Morimoto K, Hoshino Y, Hasegawa N, Ato M, et al. Epidemiology of pulmonary nontuberculous mycobacterial disease. Japan. Emerg Infect Dis 2016;22:1116e7. [2] Izumi K, Morimoto K, Hasegawa N, Uchimura K, Kawatsu L, Ato M, et al. Epidemiology of adults and children treated for nontuberculous mycobacterial pulmonary disease in Japan. Ann Assoc Am Thorac Soc 2019;16:341e7. [3] Griffith DE, Aksamit T, Brown-Elliott BA, Catanzaro A, Daley C, Gordin F, et al. ATS Mycobacterial Diseases Subcommittee; American Thoracic Society; Infectious Disease Society of America. An official ATS/IDSA statement: diagnosis, treatment, and prevention of nontuberculous mycobacterial diseases. Am J Respir Crit Care Med 2007;175:367e416. [4] van Ingen J, Wagner D, Gallagher J, Morimoto K, Lange C, Haworth CS, et al. NTM-NET. Poor adherence to management guidelines in nontuberculous mycobacterial pulmonary diseases. Eur Respir J 2017;15:49. [5] Adjemian J, Prevots DR, Gallagher J, Heap K, Gupta R, Griffith D. Lack of adherence to evidence-based treatment guidelines for nontuberculous mycobacterial lung disease. Ann Assoc Am Thorac Soc 2014;11:9e16. [6] Griffith DE, Brown-Elliott BA, Langsjoen B, Zhang Y, Pan X, Girard W, et al. Clinical and molecular analysis of macrolide resistance in Mycobacterium avium complex lung disease. Am J Respir Crit Care Med 2006;174:928e34. [7] Morimoto K, Namkoong H, Hasegawa N, Nakagawa T, Morino E, Shiraishi Y, et al. Macrolide-resistant Mycobacterium avium complex lung disease: analysis of 102 consecutive cases.; nontuberculous mycobacteriosis Japan research consortium. Ann Assoc Am Thorac Soc 2016;13:1904e11. [8] Griffith DE, Aksamit TR. Therapy of refractory nontuberculous mycobacterial lung disease. Curr Opin Infect Dis 2012;25:218e27. [9] Morimoto K, Hasegawa N, Izumi K, Namkoong H, Uchimura K, Yoshiyama T, et al. A laboratory-based analysis of nontuberculous mycobacterial lung disease in Japan from 2012 to 2013. Ann Assoc Am Thorac Soc 2017;14:49e56. [10] Sakatani M. Non-tuberculous mycobacterium disease. Kekkaku 2005;1:25e30 [in Japanese].
Please cite this article as: Igari H et al., A retrospective observational study of antimicrobial treatment for non-tuberculous mycobacteria disease using a nationwide claims database in Japan, J Infect Chemother, https://doi.org/10.1016/j.jiac.2019.10.009
Contents lists available at ScienceDirect
Journal of Infection and Chemotherapy journal homepage: http://www.elsevier.com/locate/jic
Original Article
A retrospective observational study of antimicrobial treatment for non-tuberculous mycobacteria disease using a nationwide claims database in Japan* Hidetoshi Igari a, *, Kazutaka Yamagishi a, Shingo Yamazaki b, Shota Murata c, Misuzu Yahaba a, Shin Takayanagi a, Yohei Kawasaki d, Toshibumi Taniguchi a a
Division of Infection Control, Chiba University Hospital, 1-8-1 Inohana Chuo-ku, Chiba, 260-8677, Japan Division of Pharmacy, Chiba University Hospital, 1-8-1 Inohana Chuo-Ku, Chiba, 260-8677, Japan Department of Clinical Laboratory, Chiba University Hospital, 1-8-1 Inohana Chuo-Ku, Chiba, 260-8677, Japan d Biostatistics Section, Clinical Research Center, Chiba University Hospital, 1-8-1 Inohana Chuo-Ku, Chiba, 260-8677, Japan b c
a r t i c l e i n f o
a b s t r a c t
Article history: Received 12 July 2019 Received in revised form 26 August 2019 Accepted 21 October 2019 Available online xxx
The Macrolides (MLs), clarithromycin and azithromycin, are key drugs for non-tuberculous mycobacteria (NTM) diseases treatment. A three antibiotics regimen including MLs, rifampicin (RFP) and ethambutol (EB) has been recommended for the treatment of NTM diseases in ATS/IDSA guideline. However, antibiotics are not necessarily prescribed in compliance with the guideline. Inappropriate regimens are risk of introducing MLs resistance. Therefore, we planned this study to evaluate the current Japanese NTM diseases treatment conditions. We used the national database (NDB) from 2011 to 2014. A total of 183 patients were entered into the study. The patients number increased at an accelerating rate in patients aged 55 years. Patients aged 55 years made up 91.3% of the total NTM diseases. Male and female patients were 61 and 122, respectively, a female/male ratio of 2.00. Clarithromycin, RFP, EB and fluoroquinolones were frequently prescribed, with the numbers of prescriptions being 125, 66, 57 and 45, respectively. The regimen of MLs, RFP and EB recommend by ATS/IDSA guideline 2007 was only followed by 25.1% of the patients. MLs monotherapy was as high as 30.6% of NTM diseases and would be a risk factor leading to an increase of MLs resistance and poor treatment outcome. Without effective NTM disease therapy, the increase of MLs-resistant NTM diseases would be a burden for Japanese health care facilities. © 2019 Japanese Society of Chemotherapy and The Japanese Association for Infectious Diseases. Published by Elsevier Ltd. All rights reserved.
Keywords: Non-tuberculous mycobacteria disease National data base Treatment Epidemiology
1. Introduction The incidence of non-tuberculous mycobacteria (NTM) disease has increased and even overwhelmed that of tuberculosis in Japan [1,2]. The new macrolides (MLs), clarithromycin (CAM) and azithromycin (AZM) have been introduced for NTM treatment.
* All authors meet the ICMJE authorship criteria. All authors have seen and approved the manuscript, contributed significantly to the work. * Corresponding author. E-mail addresses: [email protected] (H. Igari), [email protected] (K. Yamagishi), [email protected] (S. Yamazaki), [email protected] (S. Murata), [email protected] (M. Yahaba), [email protected] (S. Takayanagi), [email protected] (Y. Kawasaki), [email protected] (T. Taniguchi).
Usually, a three-drug regimen consisting of the new MLs, rifampicin (RFP) and ethambutol (EB) has been recommended for the treatment of NTM diseases according to ATS/IDSA guideline [3]. However, antibiotics are not necessarily prescribed in compliance with the guideline owing to adverse events [4]. Adherence to the 2007 ATS/IDSA guideline for NTM disease has been poor. Suboptimal or potentially harmful antibiotic regimens are commonly prescribed in the United States [5]. Inappropriate regimens, such as MLs monotherapy and ML-including regimens without adequate companion drugs are shown to be a risk of introducing MLs resistance [6]. Thus, MLs-resistant NTM disease is difficult to control because of the lack of effective drugs [7]. Inadequate prescriptions, not recommended in the ATS/IDSA guideline, are common world-wide [4]. The prescription tendencies of Japanese physicians are also included in this study. Then, we
https://doi.org/10.1016/j.jiac.2019.10.009 1341-321X/© 2019 Japanese Society of Chemotherapy and The Japanese Association for Infectious Diseases. Published by Elsevier Ltd. All rights reserved.
Please cite this article as: Igari H et al., A retrospective observational study of antimicrobial treatment for non-tuberculous mycobacteria disease using a nationwide claims database in Japan, J Infect Chemother, https://doi.org/10.1016/j.jiac.2019.10.009
2
H. Igari et al. / J Infect Chemother xxx (xxxx) xxx
planned this study to evaluate the current Japanese NTM disease treatment conditions. Medical services are provided under the universal health insurance system in Japan. The Ministry of Health, Labour and Welfare (MHLW) in Japan provides the national database (NDB) for research purposes. We therefore conducted a retrospective study based on the national database (NDB) from 2011 to 2014. This database study would reveal the NTM diseases treatment in Japanese health care facilities. Our objectives are to estimate the antibiotics regimen for NTM diseases, with strong focus on the adherence to the guideline, and to describe the epidemiology of NTM diseases. 2. Materials and methods 2.1. Study design This is a retrospective cohort study using samples of the NDB, which randomly sampled 1% of the claim information associated with outpatient medical service data from every October from 2011 through 2014. Usually, we should adopt clinical diagnoses on International Classification of Diseases, 10th version (ICD-10) codes. Unfortunately, ICD-10 codes are not used for outpatient claim information. Then, we selected patients with their diagnosis containing pulmonary NTM diseases equivalent to ICD-10 codes corresponding to A31 (infection due to other mycobacteria, excluding leprosy and tuberculosis). However, ICD-10 A31 is consisted of the following three subgroups; A310 (pulmonary mycobacterial infection due to mycobacterium), A311 (cutaneous mycobacterial infection), and A318 (other mycobacterial infections) and A319 (mycobacterial infection, unspecified). Then, the patients corresponding to A318 and A319 were excluded. We further selected patients who were prescribed the following anti-NTM drugs; CAM, AZM, RFP, rifabutin (RBT), EB, aminoglycosides (AGs) and fluoroquinolones (FQs). And we checked the prescription of imipenem/cilastatin as is administered for NTM disease caused by M. abscessus. We used the following information: age, sex and drugs for NTM diseases treatment. The primary endpoint is to reveal the antibiotics prescribed for NTM diseases. We also analyzed combinations of antibiotics including MLs monotherapy. The secondary endpoint is to reveal the NTM disease frequency according to age group and sex. This study was approved by the Research Ethics Committee of the Graduate School of Medicine and School of Medicine, Chiba University, and was conducted in accordance with the principles of the Declaration of Helsinki. This retrospective cohort study was conducted in accordance with the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) guidelines. 2.2. Statistical analysis We calculated the number of patients and proportions for categorical data. The differences in baseline characteristics of subjects were compared by Chi-square test or Fisher's exact test for categorical variables. A two-sided P-value of <0.05 was considered statistically significant, and all analysis results were calculated by SAS Version 9.4 for Windows and JMP® Pro 13.0.0 (SAS Institute Inc., Cary, NC, USA). 3. Results A total of 183 patients were entered. Their distributions are shown by age groups and sex (Fig. 1). Median age was 70 e 74 years, as age information was classified in 5-year steps from age years in our database. The number of patients increased at an accelerating rate with age from 55 years. Patients aged 55 years totaled as
many as 91.3% of all NTM diseases patients. The numbers of male and female patients were 61 and 122, respectively, and the female/ male ratio was 2.00. No patients were being prescribed anti-HIV drugs. The prescribed anti-NTM drugs are shown in Table 1. CAM, RFP, EB, FQs were frequently prescribed, and the numbers of prescriptions were 125, 66, 57 and 45, respectively. The combinations of anti-NTM drugs are shown in Table 2. The prescriptions that included MLs (CAM or AZM) were for 126 patients. The regimen of MLs, EB and RFP (or RBT), as is recommend in ATS/IDSA guideline 2007, was for 46 patients. Only 25.1% of antibiotic regimens met ATS/IDSA guideline. MLs monotherapy was for 56 patients, 30.6% of the NTM diseases. Anti-NTM therapies without MLs were prescribed for 57 patients, representing 31.1% of the NTM disease patients. The therapies without CAM vary. FQs or AGs are frequently observed. The combination therapy by one to three drugs consisting of the antituberculosis drugs isoniazid (INH), RFP and EB were also counted. The details of anti-NTM therapies were stratified according to the three age groups of <65 years, 65 e 74 years and 75 years (Table 3). The recommended regimen consisting of MLs, EB and RFP (or RBT) did not reveal significant differences among the three age groups. The ratios of the regimen including MLs at age <65 years, 65 e 74 years and 75 years were 56.6%, 69.0% and 77.8%, respectively, and the differences among these three age groups were significant. (P < 0.05) However, the frequency of CAM monotherapy also increased in elderly patients. The percentages of MLs, in which CAM was mainly used as monotherapy, were 17.0%, 31.0% and 40.3% for <65 years, 65 e 74 years and 75 years, respectively, and were significantly different among these three age groups. (p < 0.05) Thus, the increase of MLs prescriptions in elderly patients resulted in the increase of monotherapy. 4. Discussion This study analyzed the current NTM disease treatment condition in Japan. At first, we demonstrated the age distribution of NTM diseases and the male/female ratio (Fig. 1). These results were almost the same trend as was shown by Izumi K et al. [2] Our sampled claim database could be used for the analysis for the antibiotics usage in NTM diseases patients. Our current study revealed that only 25% of NTM diseases patients received antibiotics therapy by the standard regimen recommended by ATS/IDSA guideline [3]. The other NTM diseases patients, 75% of total, are treated by other therapies. ATS/IDSA guideline 2007 recommends the combination therapy including CAM (or AZM), EB and RFP (or RBT), occasionally adding AGs [3]. Although MLs are key antibiotics for NTM disease treatment, the therapy including MLs remained as low as 68.9% in this study. Especially, MLs monotherapy counted as high as 30% of NTM disease patients. Morimoto et al. demonstrated that inappropriate NTM disease therapy lead to MLs-resistant NTM diseases [7]. Their study analyzed MLs-resistant pulmonary Mycobacterium avium complex (MAC) diseases, and revealed that about 30% of MLsresistant pulmonary MAC disease patients received MLs monotherapy. They concluded that CAM monotherapy was a risk factor for MLs resistance [3]. When MLs resistance is detected in NTM diseases, effective NTM treatment is restricted [6,8]. The current NTM disease treatment condition has a risk of an increase in MLs-resistant NTM diseases. It is obvious that further improvement for appropriate anti-NTM treatments is needed. The urgency of providing better education of appropriate NTM treatment for physicians caring for NTM disease patients is clear.
Please cite this article as: Igari H et al., A retrospective observational study of antimicrobial treatment for non-tuberculous mycobacteria disease using a nationwide claims database in Japan, J Infect Chemother, https://doi.org/10.1016/j.jiac.2019.10.009
H. Igari et al. / J Infect Chemother xxx (xxxx) xxx
3
Fig. 1. Numbers of nontuberculous mycobacteria disease patients by age groups. The numbers of non-tuberculous mycobacteria (NTM) diseases are calculated by age groups. Our data base is a national one from 2011 to 2014 reflecting the medical service provided under universal health insurance in Japan. NTM disease patients increase in elderly people aged 55 years or more.
We further analyzed NTM treatment regimen stratified according to age classifications. The percentage of the recommended regimen of CAM (or AZM), EB and RFP (RBT) with adherence to guideline was around 30% in this study, as in the above discussion, and these figures did not show significant differences among age groups (Table 3). The MLs prescription rate at age of <65 years was 56.6%, and this figure significantly increased in older patients. Nonetheless, this trend did not contribute to an increase in the recommended therapy based on the guideline, and resulted in MLs monotherapy. We understand that some patients discontinue the recommended treatment owing to adverse events, but a large majority of patients are prescribed antibiotic regimens that do not meet ATS/IDSA guideline 2007. NTM diseases become common diseases in consideration of NDB analysis by Namkoong et al. [1,2] Japanese physicians would have opportunities to treat NTM diseases even if their specialty is not respiratory disease or infectious disease. It is evident that we need further tactics for improving the adherence to guideline. In addition to the education for instruction of NTM treatment, we also anticipate progress in the research for developing more effective therapies and providing alternative therapies for patients with adverse events. We used NDB to assess the current NTM disease treatment in outpatients’ clinics across Japan. The number of NTM disease Table 1 Prescribed anti-NTM drugs (total patient number (n) ¼ 183). anti-NTM drug
n
(%)
clarithromycin azithromycin rifampicin rifabutin ethambutol fluoroquinolones aminoglycosides isoniazid azithromycin imipenem/cilastatin
125 1 66 2 57 45 17 11 1 0
(68.3) (0.5) (36.1) (1.1) (31.1) (24.6) (9.3) (6.0) (0.5) (0.0)
NTM: non tuberculosis mycobacteria.
patients has increased at an accelerating rate among elderly patients aged 55 years, and the patients of this age group consist of 91.3% of the total. Morimoto et al. analyzed the Japanese NTM disease bases on microbiology laboratory results, and patients with NTM increased among elderly patients aged 50 years, with a female/male ratio of 2.0 [2]. Our results also support their study results [9]. This study has several limitations. First, it was based on an analysis of NDB, and the details of every patient's information, especially causative microorganism, were none or restricted. Then, we could not sufficiently confirm the diagnostic background Table 2 Combination of anti-NTM drugs (total patient number (n) ¼ 183). Combination of anti-NTM drugs
n
(%)
including macrolide CAM, EB and RFP (or RBT) AZM, EB and RFP CAM, EB CAM, RFP (RBT) CAM, RFP and FQs CAM and FQs CAM monotherapy not including macrolide FQs alone AGs alone INH alone INH, RFP and EB INH and RFP RFP (RBT) and FQs RFP, EB and FQs INH and EB EB and FQs
126 45 1 5 8 5 6 56 57 28 14 4 3 3 2 1 1 1
(68.9) (24.6) (0.5) (2.7) (4.4) (2.7) (3.3) (30.6) (31.1) (15.3) (7.7) (2.2) (1.6) (1.6) (1.1) (0.5) (0.5) (0.5)
additional drugs þFQ 1, þAG 2 þFQ 1 þAG 1
NTM: non tuberculosis mycobacterium. CAM: clarithromycin. EB: ethambutol. RFP: rifampicin. RBT: rifabutin. FQ: fluoroquinolone. AG: aminoglycoside. AZM: azithromycin.
Please cite this article as: Igari H et al., A retrospective observational study of antimicrobial treatment for non-tuberculous mycobacteria disease using a nationwide claims database in Japan, J Infect Chemother, https://doi.org/10.1016/j.jiac.2019.10.009
4
H. Igari et al. / J Infect Chemother xxx (xxxx) xxx
Table 3 Anti-NTM drugs stratified according to three age classes.
Therapy including CAM or AZM CAM (or AZM), EB and RFP (or RBT) CAM monotherapy
<65 year (n ¼ 53) (%)
65e74 year (n ¼ 58) (%)
75 year (n ¼ 72) (%)
p value
30 15 9
40 16 18
56 15 29
<0.05 n.s. <0.05
(56.6) (28.3) (17.0)
(69.0) (27.6) (31.0)
(77.8) (20.8) (40.3)
NTM: non tuberculosis mycobacterium. CAM: clarithromycin. AZM: azithromycin. EB: ethambutol. RFP: rifampicin. RBT: rifabutin. n.s.: not significant.
supporting the probability of NTM diagnosis. The reliability of NTM disease might be secured on the prescription of antibiotics for the treatment of NTM diseases. MLs have anti-inflammatory effect other than antibiotics effect, and then some physicians prescribe MLs (usually CAM) for chronic bronchial inflammatory diseases like diffuse pan-bronchitis and chronic bronchitis. These uses are also prescribed for longer durations and would possibly have resulted in increases of resistant bacteria. Second, this study was conducted on sampling data consisting of just a one-month duration. The number of NTM disease patients was limited, and these data were limited to single-month periods, each October of 2011e14. NTM diseases are chronic diseases, and they do not resolve in a short period. We cannot confirm the outcome of longer treatment, more than onemonth, for NTM diseases. The yearly trend of NTM disease should be analyzed. Third, the species of NTM is not confirmed in this database. The most frequently detected NTMs are Mycobacterium avium and Mycobacterium intracellulare, occupying 83% in Japan [10]. Recently, NTM disease caused by Mycobacterium abscessus (M.abscessus) has increased. The treatment of M. abscessus disease consists of combination therapies based on MLs. Fourth, assessment of the severity of NTM diseases was not performed in this study. NTM diseases are not necessarily treated by antibiotics. There might be some patients under observation owing to their stable conditions. We cannot estimate these patients without antibiotics prescription. Fifth, the antibiotics for NTM diseases are used for infectious diseases other than NTM diseases. FQs and MLs are often used for respiratory infectious diseases. INH monotherapy is used for latent tuberculosis infection. The antibiotics listed in Tables 1 and 2 might be prescribed for infectious diseases other than NTM diseases. This study revealed the age class epidemiology of NTM diseases, confirming the previous study. The current conditions of NTM diseases therapy is not sufficient, as the frequency of NTM treatment based on guideline is as low as about 25%, and the MLs monotherapy is as high as 30%, which would be a risk factor leading to an increase of MLs resistance and poor treatment outcome. Without effective NTM disease therapy, the increase of MLs resistant NTM diseases is sure to be a burden for Japanese health care facilities. Authors’ contributions Hidetoshi - Igari Study planning, data analysis and draft writing.
Shin Takayanagi - Study planning. Kazutaka Yamagishi - Data analysis. Shingo Yamazaki - Data analysis. Shota Murata - Data analysis. Misuzu Yahaba - Study planning. Yohei Kawasaki - Data analysis. Toshibumi Taniguchi - Study planning, data analysis. Declaration of Competing Interest None declared. Acknowledgement This study was conducted without funding. References [1] Namkoong H, Kurashima A, Morimoto K, Hoshino Y, Hasegawa N, Ato M, et al. Epidemiology of pulmonary nontuberculous mycobacterial disease. Japan. Emerg Infect Dis 2016;22:1116e7. [2] Izumi K, Morimoto K, Hasegawa N, Uchimura K, Kawatsu L, Ato M, et al. Epidemiology of adults and children treated for nontuberculous mycobacterial pulmonary disease in Japan. Ann Assoc Am Thorac Soc 2019;16:341e7. [3] Griffith DE, Aksamit T, Brown-Elliott BA, Catanzaro A, Daley C, Gordin F, et al. ATS Mycobacterial Diseases Subcommittee; American Thoracic Society; Infectious Disease Society of America. An official ATS/IDSA statement: diagnosis, treatment, and prevention of nontuberculous mycobacterial diseases. Am J Respir Crit Care Med 2007;175:367e416. [4] van Ingen J, Wagner D, Gallagher J, Morimoto K, Lange C, Haworth CS, et al. NTM-NET. Poor adherence to management guidelines in nontuberculous mycobacterial pulmonary diseases. Eur Respir J 2017;15:49. [5] Adjemian J, Prevots DR, Gallagher J, Heap K, Gupta R, Griffith D. Lack of adherence to evidence-based treatment guidelines for nontuberculous mycobacterial lung disease. Ann Assoc Am Thorac Soc 2014;11:9e16. [6] Griffith DE, Brown-Elliott BA, Langsjoen B, Zhang Y, Pan X, Girard W, et al. Clinical and molecular analysis of macrolide resistance in Mycobacterium avium complex lung disease. Am J Respir Crit Care Med 2006;174:928e34. [7] Morimoto K, Namkoong H, Hasegawa N, Nakagawa T, Morino E, Shiraishi Y, et al. Macrolide-resistant Mycobacterium avium complex lung disease: analysis of 102 consecutive cases.; nontuberculous mycobacteriosis Japan research consortium. Ann Assoc Am Thorac Soc 2016;13:1904e11. [8] Griffith DE, Aksamit TR. Therapy of refractory nontuberculous mycobacterial lung disease. Curr Opin Infect Dis 2012;25:218e27. [9] Morimoto K, Hasegawa N, Izumi K, Namkoong H, Uchimura K, Yoshiyama T, et al. A laboratory-based analysis of nontuberculous mycobacterial lung disease in Japan from 2012 to 2013. Ann Assoc Am Thorac Soc 2017;14:49e56. [10] Sakatani M. Non-tuberculous mycobacterium disease. Kekkaku 2005;1:25e30 [in Japanese].
Please cite this article as: Igari H et al., A retrospective observational study of antimicrobial treatment for non-tuberculous mycobacteria disease using a nationwide claims database in Japan, J Infect Chemother, https://doi.org/10.1016/j.jiac.2019.10.009