Mycobacterium Gordonae from a Patient with Pulmonary Disease

Int J Clin Exp Pathol 2016;9(7):7733-7738 www.ijcep.com /ISSN:1936-2625/IJCEP0024186

Case Report Who is the criminal? Isolates of Mycobacterium tuberculosis, Mycobacterium abscessus and Mycobacterium gordonae from a patient with pulmonary disease

Jianhao Wei1,2*, Yi Jiang1,3*, Guilian Li1,3*, Qian Guo1,2, Haican Liu1,3, Donglei Xu1,3, Jianghua Zheng2, Kanglin Wan1,3

1State Key Laboratory for Infectious Disease Prevention and Control, National Institute for Communicable Disease Control and Prevention, Chinese Center for Disease Control and Prevention, Beijing, P. R. China; 2Department of Clinical Laboratory, Shanghai Public Health Clinical Center, Shanghai, P. R. China; 3Collaborative Innovation Center for Diagnosis and Treatment of Infectious Diseases, Hangzhou, P. R. China. *Equal contributors. Received January 16, 2016; Accepted March 26, 2016; Epub July 1, 2016; Published July 15, 2016

Abstract: Few reports have described pulmonary disease caused by multi-pathogen infections with mycobacteria. We report a well-documented case of a unique pulmonary disease that resulted from infections with M. tuberculosis, M. abscessus and M. gordonae in an adult woman who was not immunodeficient. Combining her medical history and the drug sensitivities of the M. abscessus and M. gordonae isolates, the patient was treated with amikacin, cefoxitin and clarithromycin. Her symptoms and physical status were relieved up to February 2015.

Keywords: Mycobacterium tuberculosis, mycobacterium abscessus, mycobacterium gordonae, multiple infection, pulmonary disease

Introduction Previous case history: The patient had no family history of any lung diseases and never smoked, Non-tuberculous mycobacteria (NTM) diseases but had been diagnosed with tuberculous pleu- are likened to the Sword of Damocles, no one ritis in 2006. She said she had received regular knows why, when or where, the unfortunate treatment for more than 12 mouths until cured. infection will happen, nor who the unfortunate She worked as a teacher in a middle school of person that succumbs to the misery will be. Beijing and was not an immunocompromised Because there are many different species of patient. NTM, standardized treatments of NTM infections may not be sticky enough. An even worse In December 2012, the patient suffered fever, scenario comprises the existence of multi- rigor, tussiculation and nausea, with no obvious infections. Here, we report a well-documented causes, for 6 days and was diagnosed with case of a unique pulmonary disease, caused by pneumonia at Beijing Shijitan Hospital. Based an infection with Mycobacterium (M.) tubercu- on this diagnosis, she was treated with moxi- losis, M. abscessus and M. gordonae. The floxacin and ambroxol hydrochloride until she patient was an adult woman who was not was discharged from the hospital. immunodeficient, but who had a history of On September 2013, the patient was admitted tuberculous pleuritis. We present the develop- to the 309th hospital of the Chinese People’s ments of the infection and the treatment of the Liberation Army, because of cough, sputum and patient. dyspnea, for 3 weeks. Scattered high-density Case presentation nodules and calcifications in both of her lungs and bronchiectasis in the lingual segment of A 37-year-old woman was referred to our labo- the upper lobe of her left lung were observed ratory in November 2014 because of the unsat- using chest computed tomography (CT) (Figure isfactory treatment of her pulmonary disease. 1A). Sputum samples were found positive for Mycobacterium and pulmonary disease

Figure 1. A 37-year-old woman with a multi-infection of M. tuberculosis, M. abscessus and M. gordonae. A: Comput- ed tomography (CT) in September 2013 showing the scattered high-density nodules and calcifications in the lungs and bronchiectasis in the lingual segment of the upper lobe of her left lung. Sputum samples were found positive for M. tuberculosis. B: CT in May 2014 showing some improvement in the left lung. Some patchy shadows had dissolved, but a new cavity had appeared in the right lung, which was previously a tubercle shadow. M. abscessus was first isolated in March 2014, and after two months of therapy with pasiniazid, linezolid, pyrazinamide and rifapentine, M. abscessus was still found positive in the patient’s sputum samples. C: CT in October 2014 showing the new lesions and localized pleural thickening adhesion that appeared in the patient’s left lung, some previously patchy shadows had dissolved, the diameter of the cavity in the right lung had become smaller and the bronchiectasis in the lingual segment of the upper lobe of the patient’s left lung remained. Both M. abscessus and M. gordonae were isolated from the patient at this time. acid-fast bacilli (AFB), and a sample obtained the patient went to the hospital several times by bronchoscopy was similarly positive. Isolates as an outpatient and sputum smears tested for from the samples were identified as M. tuber- AFB were still found positive. No data of any culosis. Routine laboratory testing revealed her other tests performed at this stage were pro- white-cell count, CD4+ T cells and hemoglobin vided to us. levels were normal, the C-reactive protein level was 21.47 mg/L, and she tested negative for During March 2014, the patient was admitted the antibodies against the human immunodefi- to the Beijing Chest Hospital (BCH) for treat- ciency virus. In combination with her history of ment of the stubborn disease. M. abscessus tuberculous pleuritis in 2006, the patient was was cultured from the bronchial lavage fluid diagnosed with secondary (reactivation) tuber- and there was no evidence of M. tuberculosis, culosis and bronchiectasis, but the data of the so her treatment was changed to pasiniazide, treatment given at this stage were not avail- linezolid, pyrazinamide and rifapentine. On May able. From December 2013 to February 2014, 2014, she was reexamined using CT; most of

7734 Int J Clin Exp Pathol 2016;9(7):7733-7738 Mycobacterium and pulmonary disease the characteristic lesions in her left lung had We also used monoclonal isolates to confirm improved, some patchy shadows had dissolved, the pathogen (s), because M. tuberculosis and and the tubercle shadow in her right lung had M. gordonae had previously been isolated from become a new cavity (Figure 1B). Isolates from the patient. Subcultures were prepared on her sputum were identified as M. abscessus 7H10-OADC plates to separate the single colo- again, so a new therapy with clarithromycin, nies. Three different orange colonies and 144 ethambutol, levofloxacin, and rifapentine was cream-smooth morphological colonies were initiated. The patient’s condition continued to found in the subcultures of the sample from 13 improve, except for intermittent tussiculation. November after incubation for 12 days. The molecular tests were then repeated on the Unfortunately the patient’s disease relapsed in monoclonal isolates. The cream-smooth mor- October 2014. Due to the tussiculation and phological and the orange colonies were identi- weight loss (nearly 10 kg over the previous fied asM. abscessus and M. gordonae, respec- eight months), she was admitted to the BCH. tively. The presence of M. tuberculosis was fur- The CT showed new lesions and localized pleu- ther excluded via an additional PCR-IS6110 ral thickening adhesion had appeared in her test [5]. left lung (Figure 1C). Surprisingly, of the six positive cultures from the induced sputum taken The susceptibilities of the isolates to thirty-six between 4 October and 30 October, most were antimicrobial agents (purchased from Sigma- identified as M. abscessus and two contained Aldrich) were tested in accordance with the both M. abscessus and M. gordonae. In addi- guidelines from the Clinical and Laboratory tion, the cultures grown from another bronchial Standards Institute (CLSI) [6]. Table 1 shows lavage fluid sample (of the lower lobe of her the minimum inhibitory concentrations (MICs) right lung) on 17 October were identified as M. of each drug against the isolates of M. absces- gordonae. The patient was considered to have sus and M. gordonae, respectively. Based on a multi-infection with M. abscessus and M. gor- the drug susceptibility tests, amikacin was donae. Rifapentine was replaced with cefoxitin included in the therapeutic regimen for the in her therapeutic regimen. Then, her doctor patient; until February 2015, she was treated contacted us and referred her to our laboratory with intravenous (IV) amikacin (400 mg qd), IV for further examination. cefoxitin (60 mg/kg bid), and oral clarithromycin (300 mg tid). The patient’s symptoms Assessment and treatment at our laboratory: showed some improvement and the last three During November 2014, we collected ten spu- sputum cultures were negative for AFB. tum samples by sputum induction; samples from four of the days (5, 6, 13 and 17 November) Discussion were found positive for AFB via sputum smear With the development of molecular biological microscopy. After four days of incubation by the techniques, more than 150 NTM species have BACTEC MGIT 960 system, three of the sam- been described [7]. Some NTM species are ples (collected 5, 13, and 17 November, respec- known pathogens in patients with pulmonary tively) were found positive for culture; these diseases [8], including M. abscessus and M. isolates were tolerant of 10 mg/ml 2-thiophen- gordonae, although they are rare. Such NTM ecarboxylic acid hydrazide and 500 mg/ml species are increasingly being recognized as p-nitro benzoic acid. The isolates were deter- pathogens responsible for respiratory tract mined to be mycobacterial species using 16S infections as well. Multi-infections with M. rRNA gene PCR-sequencing. The sequences tuberculosis, M. abscessus and M. gordonae showed 100% similarity with M. abscessus, have not been described before. In this report, according to the basic local alignment search the patient met the diagnostic criteria of the tool (BLAST) online database. As we knew that American Thoracic Society and the Infectious the 16S rRNA gene is not a sufficient criterion Disease Society of America, published for NTM for the identification of species in theM. chelo- infections [9], so a multi-infection comprising nae-abscessus complex [1, 2], further species M. tuberculosis, M. abscessus and M. gordo- identification was performed using partial nae was confirmed. hsp65 and rpoB genes [3, 4]. We detected 99% identity with M. abscessus in both the hsp65 The process of multi-infections is chronic and and the rpoB genes. complex. Previous studies have suggested that

7735 Int J Clin Exp Pathol 2016;9(7):7733-7738 Mycobacterium and pulmonary disease

Table 1. The minimum inhibitory concentrations (MIC) of each drug during the drug-susceptibility tests performed with the M. abscessus and M. gordonae isolates M. abscessus M. gordonae Drug MIC (µg/ml) Breakpoint (µg/ml) MIC (µg/ml) Breakpoint (µg/ml) 1 Streptomymicn 8~16 10 0.5~1 Report 2 Amikacin 2~4 16~64 1~2 32 3 Kanamycin 4~8 16~64 2~4 32 4 Tobramycin 1~2 2~8 0.5~1 2~8 5 Azithromycin 16~32 2~8 1~2 16 6 Erythrocin 0.25~0.5 2~8 < 0.5 16 7 Roxithromycin 2~4 2~8 < 0.5 16 8 Clarithromycin 1~2 2~8 < 0.125 16 9 Moxifloxacin 0.5~1 1~4 < 0.125 2 10 Ofloxacin 1~2 1~4 0.5~1 1~4 11 Levofloxacin 0.125~0.25 1~4 0.125~0.25 2 12 Sparfloxacin 0.5~1 1~4 < 0.125 1~4 13 Gatifloxacin 0.5~1 1~4 < 0.125 1~4 14 Ciprofloxacin 0.25~0.5 1~4 < 0.125 2 15 Rifampin 16~32 1 < 0.5 1 16 Rifabutin 0.125~0.25 2 0.25~0.5 1 17 Rifapentine 8~16 2 0.25~0.5 1 18 Tetracycline > 128 1~8 2~4 4 19 Doxycycline > 128 1~8 < 0.5 4 20 Minocycline 16~32 1~8 0.5~1 4 21 Cefepime 2~4 16~128 - - 22 Cefmetazole 1~2 16~128 - - 23 Cefoperazone > 128 16~128 - - 24 Cefoxitin 4~8 16~128 - - 25 Isoniazid > 128 5 2~4 Report 26 Ethambutol 16~32 4 1~2 4 27 Cycloserine > 128 32 64~128 32 28 Ethionamide 64~128 2~8 2~4 8~32 29 Cotrimoxazole 16~32 38 2~4 38 30 Linezolid 1~2 8~32 0.125~0.25 16 31 Capreomycin 16~32 16~64 2~4 32 32 Clofazimine 1~2 0.25 < 0.5 0.25 33 Dapsone > 128 4 4~8 4 34 Meropenem 64~128 4~32 8~16 4~32 35 Thiacetazone 16~32 8 > 128 8 36 Trimethoprim-sulfamethoxazole 16~32 38 2~4 38 a. “-” means data deficient. patients with underlying conditions, such as tuberculosis, M. abscessus, and M. gordonae. cystic fibrosis, chronic obstructive pulmonary NTM lung diseases and bronchiectasis are disease, bronchiectasis, malignancy, and im- inextricably linked pathophysiologically. It has munosuppressive diseases (especially AIDS) been debated whether bronchiectasis is a are more susceptible to NTM diseases [9]. In cause, or consequence (or both), of NTM pul- this study, the patient had been diagnosed with monary diseases [10]. There is accumulating tuberculosis and bronchiectasis in 2013, and evidence and clinical experience that supports then suffered from a multi-infection, with M. both standpoints [10, 11]. Besides this, a study

7736 Int J Clin Exp Pathol 2016;9(7):7733-7738 Mycobacterium and pulmonary disease has indicated that women and the elderly of cific antibiotics and in vitro susceptibility is specific morphotypes are more vulnerable to unclear [14]. For our isolates, the MICs of thirty- pulmonary NTM diseases, although the mecha- six antimicrobial agents were tested. In the lit- nisms underlying this phenomenon are still erature, we found that the presence of an being investigated [12]. The preexisting tuber- inducible erythromycin methylase (erm) gene culosis in the patient from our study was anoth- may confer macrolide resistance to M. absces- er risk factor for her to develop an NTM respira- sus [14]. Therefore, a prolonged (14 day) sus- tory infection; we speculated that postinflam- ceptibility-test of the macrolide antibiotics was matory bronchiectasis probably contributed to performed, according to CLSI guidelines [6]. this association [10]. However, it is unclear Ultimately, we found that the M. abscessus iso- whether prior infection with M. abscessus is an late was sensitive to 9 drugs include amikacin, independent risk factor for M. gordonae infec- clarithromycin and cefoxitin. The M. gordonae tion. Although NTM colonized in the lower respi- isolate was sensitive to 12 drugs, including ratory tract, it may also reduce the ciliary beat amikacin and clarithromycin. These results frequency and caused epithelial damage [13]. were comparable to related studies [15]. Thus, Overall, it appears that a specific woman’s mor- the different treatment responses to these photype, preexisting bronchiectasis and tuber- drugs by M. abscessus and M. gordonae could culosis may lead to multi-infections with M. be predicted. In further analysis of the break- tuberculosis, M. abscessus and M. gordonae. points reported in recent research [14-16], both of the isolates were found sensitive to In this case, microscopic and microbiological amikacin, kanamycin, tobramycin, erythrocin, tests were performed repeatedly to examine clarithromycin, moxifloxacin, levofloxacin, spar- the pathogens present, but who was the crimi- floxacin, gatifloxacin, ciprofloxacin, rifabutin, nal? The most typical and common pathogen in linezolid and trimethoprim-sulfamethoxazole. human tuberculosis is M. tuberculosis and the Therefore, according to the susceptibility patient had a history of secondary tuberculosis; results, the patient was treated with a combi- unfortunately it appears the regular anti-tuber- nation of amikacin, clarithromycin and cefoxi- culosis treatment did not actually cure the tin, and her symptoms showed some improve- patient. However, as a result of the improvement after eight weeks. ment of the patient’s treatment, to include macrolides, M. tuberculosis could be consid- In conclusion, a well-documented case of a ered the least suspicious for the prolonged unique pulmonary disease, caused by multi- infection. The most frequent and enduring infection with M. tuberculosis, M. abscessus pathogen isolated from the patient was M. and M. gordonae, in an adult woman was abscessus, which was not only reported as the reported. The optimal therapies for such myco- third most common respiratory pathogen bacterial multi-infections require more antimi- among the NTM infections in North America crobial agents and more cautious treatment [12], but also the most difficult to treat [14]. The methods; this could be a new challenge. other pathogen isolated was M. gordonae, an occasional and hard-to-isolate suspect. The Acknowledgements perplexity is that M. gordonae is frequently isolated [7], but infrequently reported as an infec- We thank the Beijing Chest Hospital staff for tion in literature [9]. The reason that M. gordo- their excellent contribution to this study. This nae isolates are rarely reported as infections work was funded by projects 2013ZX10003- might be clinical, due to a lack of proper testing 006-002-001 and 2013ZX10004-001 of the methods; so patients are diagnosed and treat- National Key Program of Mega Infectious ed inaccurately. Therefore, we performed the Disease. The funders had no role in the study treatment cautiously, against both M. absces- design, data collection and analysis, manu- sus and M. gordonae. script preparation, or decision to publish. The study was approved by the Ethics Committee Currently, clinicians still use in vitro susceptibil- of the National Institute for Communicable ity reports for NTM isolates, which list the anti- Disease Control and Prevention, Chinese microbial agents as either susceptible or resis- Center for Disease Control and Prevention. tant, based on the MICs, even though the cor- Written informed consent was obtained from relation between the in vivo response for spe- the patient for publication of this report and the

7737 Int J Clin Exp Pathol 2016;9(7):7733-7738 Mycobacterium and pulmonary disease accompanying images. A copy of the written [8] Gommans EP, Even P, Linssen CF, van Dessel consent is available for review by the editor of H, van Haren E, de Vries GJ, Dingemans AM, this journal. Kotz D and Rohde GG. Risk factors for mortali- ty in patients with pulmonary infections with Disclosure of conflict of interest non-tuberculous mycobacteria: a retrospective cohort study. Respir Med 2015; 109: 137-145. None. [9] Griffith DE, Aksamit T, Brown-Elliott BA, Catanzaro A, Daley C, Gordin F, Holland SM, Address correspondence to: Kanglin Wan, National Horsburgh R, Huitt G, Iademarco MF, Iseman M, Olivier K, Ruoss S, von Reyn CF, Wallace RJ Institute for Communicable Disease Control and Jr, Winthrop K; ATS Mycobacterial Diseases Prevention, Chinese Center for Disease Control and Subcommittee; American Thoracic Society; Prevention, P.O. Box 5, Changping, Beijing 102206, Infectious Disease Society of America. An offi- P. R. China. Tel: 0086 10 58900779; Fax: 0086 10 cial ATS/IDSA statement: diagnosis, treatment, 58900779; E-mail: [email protected] and prevention of nontuberculous mycobacterial diseases. Am J Respir Crit Care Med 2007; References 175: 367-416. [10] Griffith DE and Aksamit TR. Bronchiectasis and [1] Drancourt M, Bollet C, Carlioz A, Martelin R, nontuberculous mycobacterial disease. Clin Gayral JP and Raoult D. 16S ribosomal DNA Chest Med 2012; 33: 283-295. sequence analysis of a large collection of envi- [11] Aksamit TR, Philley JV and Griffith DE. Nontu- ronmental and clinical unidentifiable bacterial berculous mycobacterial (NTM) lung disease: isolates. J Clin Microbiol 2000; 38: 3623- the top ten essentials. Respir Med 2014; 108: 3630. 417-425. [2] Fox GE, Wisotzkey JD and Jurtshuk P Jr. How [12] Prevots DR and Marras TK. Epidemiology of close is close: 16S rRNA sequence identity human pulmonary infection with nontubercu- may not be sufficient to guarantee species lous mycobacteria: a review. Clin Chest Med identity. Int J Syst Bacteriol 1992; 42: 166- 2015; 36: 13-34. 170. [13] Fowler CJ, Olivier KN, Leung JM, Smith CC, [3] Kim H, Kim SH, Shim TS, Kim MN, Bai GH, Park Huth AG, Root H, Kuhns DB, Logun C, Zelazny YG, Lee SH, Chae GT, Cha CY, Kook YH and Kim A, Frein CA, Daub J, Haney C, Shelhamer JH, BJ. Differentiation of Mycobacterium species Bryant CE and Holland SM. Abnormal nasal ni- by analysis of the heat-shock protein 65 gene tric oxide production, ciliary beat frequency, (hsp65). Int J Syst Evol Microbiol 2005; 55: and Toll-like receptor response in pulmonary 1649-1656. nontuberculous mycobacterial disease epithe- [4] Adekambi T, Colson P and Drancourt M. rpoB- lium. Am J Respir Crit Care Med 2013; 187: based identification of nonpigmented and late- 1374-1381. pigmenting rapidly growing mycobacteria. J [14] Kasperbauer SH and De Groote MA. The treat- Clin Microbiol 2003; 41: 5699-5708. ment of rapidly growing mycobacterial infec- [5] Savelkoul PH, Catsburg A, Mulder S, Oosten- tions. Clin Chest Med 2015; 36: 67-78. dorp L, Schirm J, Wilke H, van der Zanden AG [15] Li G, Lian LL, Wan L, Zhang J, Zhao X, Jiang Y, and Noordhoek GT. Detection of Mycobacterium Zhao LL, Liu H and Wan K. Antimicrobial sus- tuberculosis complex with Real Time PCR: ceptibility of standard strains of nontubercu- comparison of different primer-probe sets lous mycobacteria by microplate Alamar Blue based on the IS6110 element. J Microbiol assay. PLoS One 2013; 8: e84065. Methods 2006; 66: 177-180. [16] Pfyffer GE, Bonato DA, Ebrahimzadeh A, Gross [6] CLSI. Susceptibility Testing of Mycobacteria, W, Hotaling J, Kornblum J, Laszlo A, Roberts Nocardiae, and Other Aerobic Actinomycetes; G, Salfinger M, Wittwer F and Siddiqi S. Approved Standard-Second Edition. CLSI docu- Multicenter laboratory validation of suscepti- ment M24-A2. Wayne, PA: Clinical and Labo- bility testing of Mycobacterium tuberculosis ratory Standards Institute; 2011. against classical second-line and newer anti- [7] van der Werf MJ, Kodmon C, Katalinic-Jankovic microbial drugs by using the radiometric V, Kummik T, Soini H, Richter E, Papaventsis D, BACTEC 460 technique and the proportion Tortoli E, Perrin M, van Soolingen D, Zolnir- method with solid media. J Clin Microbiol Dovc M and Ostergaard Thomsen V. Inventory 1999; 37: 3179-3186. study of non-tuberculous mycobacteria in the European Union. BMC Infect Dis 2014; 14: 62.

7738 Int J Clin Exp Pathol 2016;9(7):7733-7738