Outcomes of Resectional Thoracic Surgery for the Treatment of Pulmonary Nontuberculous Mycobacterial Disease in 105 Patients in the United States

Abstract

Nontuberculous mycobacterial disease (NTM) is increasing in the United States. NTM can cause chronic, debilitating pulmonary disease primarily in older individuals, often requiring long-term, aggressive multi-drug antibiotic therapy.1Griffith D.E. Aksamit T. Brown-Elliott B.A. et al.An official ATS/IDSA statement: diagnosis, treatment, and prevention of nontuberculous mycobacterial diseases.Am J Respir Crit Care Med. 2007; 175: 367-416Google Scholar NTM treatment guidelines therefore have proposed adjuvant surgical resection in addition to a multidrug regimen for selected patients.1Griffith D.E. Aksamit T. Brown-Elliott B.A. et al.An official ATS/IDSA statement: diagnosis, treatment, and prevention of nontuberculous mycobacterial diseases.Am J Respir Crit Care Med. 2007; 175: 367-416Google Scholar,2Daley C.L. Iaccarino J.M. Lange C. et al.Treatment of nontuberculous mycobacterial pulmonary disease: an official ATS/ERS/ESCMID/IDSA clinical practice guideline.Eur Respir J. 2020; 56: 2000535Google Scholar Surgical intervention is intended to remove areas of parenchymal disease that are poorly penetrated by antibiotics.1Griffith D.E. Aksamit T. Brown-Elliott B.A. et al.An official ATS/IDSA statement: diagnosis, treatment, and prevention of nontuberculous mycobacterial diseases.Am J Respir Crit Care Med. 2007; 175: 367-416Google Scholar Surgical resection has been advocated in the setting of localized disease, relapse after treatment discontinuation, refractory disease, macrolide-resistant disease, or intolerance to a multidrug regimen.3Lang-Lazdunski L. Offredo C. Le Pimpec-Barthes F. et al.Pulmonary resection for Mycobacterium xenopi pulmonary infection.Ann Thorac Surg. 2001; 72: 1877-1882Google Scholar Video-assisted thoracoscopic surgery (VATS) has been reported to be associated with fewer complications than traditional thoracotomy.4Morino A. Murase K. Yamada K. Complications after video-assisted thoracic surgery in patients with pulmonary nontuberculous mycobacterial lung disease who underwent preoperative pulmonary rehabilitation.J Phys Ther Sci. 2015; 27: 2541-2544Google Scholar To date, only limited data exist on pulmonary resection for pulmonary NTM disease.3Lang-Lazdunski L. Offredo C. Le Pimpec-Barthes F. et al.Pulmonary resection for Mycobacterium xenopi pulmonary infection.Ann Thorac Surg. 2001; 72: 1877-1882Google Scholar, 4Morino A. Murase K. Yamada K. Complications after video-assisted thoracic surgery in patients with pulmonary nontuberculous mycobacterial lung disease who underwent preoperative pulmonary rehabilitation.J Phys Ther Sci. 2015; 27: 2541-2544Google Scholar, 5Koh W.J. Kim Y.H. Kwon O.J. et al.Surgical treatment of pulmonary diseases due to nontuberculous mycobacteria.J Korean Med Sci. 2008; 23: 397-401Google Scholar, 6Asakura T. Hayakawa N. Hasegawa N. et al.Long-term outcome of pulmonary resection for nontuberculous mycobacterial pulmonary disease.Clin Infect Dis. 2017; 65: 244-251Google Scholar, 7Aznar M.L. Zubrinic M. Siemienowicz M. et al.Adjuvant lung resection in the management of nontuberculous mycobacterial lung infection: a retrospective matched cohort study.Respir Med. 2018; 142: 1-6Google Scholar, 8Nelson K.G. Griffith D.E. Brown B.A. et al.Results of operation in Mycobacterium avium-intracellulare lung disease.Ann Thorac Surg. 1998; 66: 325-330Google Scholar Accordingly, we examined surgical outcomes in patients who received pulmonary resections at two large NTM treatment centers in the United States. We performed an electronic medical record search to identify patients who received cardiothoracic surgery and who received a cross diagnosis of NTM disease (International Classification of Diseases-9th edition code, 031.0, and 10th edition code, A31.0) at Oregon Health & Science University and the University of Texas Health Science Center at Tyler. We identified 105 patients who met the diagnosis criteria1Griffith D.E. Aksamit T. Brown-Elliott B.A. et al.An official ATS/IDSA statement: diagnosis, treatment, and prevention of nontuberculous mycobacterial diseases.Am J Respir Crit Care Med. 2007; 175: 367-416Google Scholar and who received pulmonary resection between May 2010 and September 2019. Those with cystic fibrosis or HIV were excluded. From the 105 patients, we collected the most recent demographic and presurgical data within 90 days before surgery and postsurgical data within 90 days of discharge. Logistic regression analyses were performed to evaluate risk factors of 90-day postsurgical adverse events. This study was approved by the institutional review board at both study centers. Most of the 105 patients were white (90%) and female (75%), with a mean age of 65 years (SD, 11; range, 36 to 84 years old) (Table 1). Mycobacterium avium complex (MAC) was the most common species on surgical culture (41%). Before surgery, a majority of the patients (98%) received a multidrug antibiotic regimen for NTM disease, which most commonly included a macrolide (73%), ethambutol (68%) and a rifamycin (62%), for a mean 25 months (SD, 18; range, 1 to 84 mo). Fifty-eight patients (55%) were treated with IV amikacin for a mean 9 weeks (SD, 22) preoperatively. There were no intraoperative deaths.Table 1Demographic and Clinical Characteristics of 105 PatientsaForty-seven from University of Texas and 58 from Oregon Health & Science University. in the United States Who Received Resectional Surgery for the Treatment of Pulmonary Nontuberculous Mycobacterial DiseaseCharacteristicsMeasureDemographic Age, mean ± SD, y65.4 ± 11.4 Female, No. (%)79 (75.2) Race, No. (%)White94 (89.5)Asian5 (4.8)Black1 (1.0)American Indian or Alaska Native1 (1.0)Other2 (1.9)Unknown2 (1.9) Smoking status, No. (%)Never smoked58 (55.2)Former smoker36 (34.2)Current smoker7 (6.7)Unknown4 (3.8)Preoperative Duration of IV amikacin use, mean ± SD/median (interquartile range), wks9.0 ± 22.6/3.0 (1) Coinfection before surgery, No. (%)Pseudomonas17 (16.2)Aspergillus9 (8.6)Staphylococcus aureus2 (1.9)Strenotrophomonas2 (1.9)Unknown16 (15.2) Albumin, mean ± SD, g/L4.0 ± 0.5 FEV1, mean ± SD, percent predicted58 ± 20 Received multi-antibiotic treatments for nontuberculous mycobacteria before surgery, No. (%)103 (98) No. of nontuberculous mycobacteria treatment episodesbOne treatment episode is defined as multi-drug antibiotic regimen that targeted pulmonary nontuberculous mycobacteria disease for ≥1 year. before surgery, No. (%)150 (47.6)226 (24.8)37 (6.7)42 (1.9) Antibiotics used for nontuberculous mycobacteria therapy before surgery, No. (%)Ethambutol71 (67.6)Azithromycin66 (62.9)IV amikacin58 (55.2)Rifampin50 (47.6)Inhaled amikacin20 (19.1)Rifabutin15 (14.3)Clarithromycin10 (9.5)Moxifloxicin5 (4.8)OthercIncludes clofazimine (n = 4), streptomycin (n = 4), imipenem (n = 3), linezolid (n = 3), amoxicillin/clavulanic acid (n = 1), cefoxitin (n = 1), isoniazid (n = 1), tigecycline (n = 1).18 (17.1) Duration of multi-antibiotic therapy before surgery, mean ± SD/median (interquartile range), mo24.5 ± 18.0/21.5 (24.0)Perioperative BMI at the time of surgery, mean ± SD, kg/ cm221.9 ± 3.5 Primary nontuberculous mycobacteria species on surgical culture, No. (%)dIdentified by one or more positive respiratory cultures.M avium-intracellulare complex43 (41.0)M abscessus6 (5.7)M simiae1 (1.0)Other5 (4.8)Negative49 (46.7) Acid-fast bacilli smear positivity on surgical culture, No. (%)10 (17.9) Positive for aspergillus on surgical culture, No. (%)25 (23.8) Cavitary disease, No. (%)65 (64.4) Bronchiectasis, No. (%)87 (82.9) COPD/emphysema, No. (%)19 (18.1) Bilateral surgery, No. (%)15 (14.3) Video-assisted thoracic surgery, No. (%)69 (65.7) Surgery type, No. (%)Lobectomy53 (50.5)Wedge resection36 (34.3)Lobectomy + segmentectomy6 (5.7)Segmentectomy4 (3.8)Bilobectomy + segmentectomy3 (2.9)Pneumonectomy2 (1.9)Completion pneumonectomy1 (1.0)a Forty-seven from University of Texas and 58 from Oregon Health & Science University.b One treatment episode is defined as multi-drug antibiotic regimen that targeted pulmonary nontuberculous mycobacteria disease for ≥1 year.c Includes clofazimine (n = 4), streptomycin (n = 4), imipenem (n = 3), linezolid (n = 3), amoxicillin/clavulanic acid (n = 1), cefoxitin (n = 1), isoniazid (n = 1), tigecycline (n = 1).d Identified by one or more positive respiratory cultures. Open table in a new tab Within 90 days of discharge, complications occurred in 27 patients (26%) (Table 2). Perioperative IV amikacin administration (OR, 0.27; 95% CI, 0.11-0.67), female sex (OR, 0.96; 95% CI, 0.93-1.00), and VATS (OR, 0.15; 95% CI, 0.06-0.39 [reference: open surgery]) were associated with a decreased odds of experiencing an adverse event within 90 days of discharge. Bacterial coinfection (OR, 5.33; 95% CI, 1.97-14.40), cavitary disease (OR, 10.62; 95% CI, 2.34-48.15 [reference: nodular disease]) and lobectomy/segmentectomy (OR, 5.50; 95% CI, 1.52-19.94 [reference: wedge resection]) were associated with an increased odds of adverse outcomes. The odds of an adverse event were higher for M abscessus when compared with MAC (OR, 2.11; 95% CI, 0.57-7.84), although not statistically significant. Among potential risk factors for postsurgical adverse events, concurrent bacterial infection and lobectomy/segmentectomy remained significant after multivariable analysis.Table 2Postsurgical Clinical Characteristics After Resectional Surgery for Pulmonary Nontuberculous Mycobacterial Disease in 105 Patients in the United StatesPostsurgical Clinical CharacteristicsMeasureDuration of multi-drug regimen after surgery, mean ± SD, mo10.7 ± 7.8Adverse events within 30 days of discharge, No. (%) Additional surgery due to postsurgical complications6 (5.8) Cardiac complications6 (5.8) Pneumothorax4 (3.8) Persistent surgical pain that required therapy2 (1.9) Worsening of nontuberculous mycobacterial disease (required restart or intensification of therapy)2 (1.9) Pleural effusion5 (4.8)Serious adverse events within 30 days of discharge, No. (%) Death3 (2.9) Bronchopleural fistula0Adverse events within 90 days of discharge, No. (%) Additional surgery due to postsurgical complications10 (9.8) Cardiac complicationsaArrhythmia/atrial fibrillation (n = 4); myocardial infarction (n = 1); sinus tachycardia (n = 1); other (n = 2).8 (7.8) Pneumothorax7 (6.9) Persistent surgical pain that required therapy5 (4.9) Worsening of nontuberculous mycobacterial disease (required restart or intensification of therapy)5 (4.9) Pleural effusion5 (4.9)Serious adverse events within 90 days of discharge, No. (%) DeathbAlso had pneumothorax (n = 3), pleural effusion (n = 1), and cardiac complications (n = 1; atrial fibrillation and arrhythmia). Cause of death: acute respiratory distress syndrome (n = 1), presumed aspiration pneumonia (n = 1), metastatic lung cancer (n = 1), unknown (n = 2).5 (4.9) Bronchopleural fistula4 (3.9)Change in cough compared with before surgery as reported by treating physician within 90 days of discharge, No. (%) Improved46 (45.1) Unchanged30 (29.4) Worse20 (19.6) Unknown9 (8.8)Albumin, mean ± SD, g/L 3 mo after surgery3.9 ± 0.8 6 mo after surgery4.2 ± 1.0FEV1, mean ± SD, % predicted 3 mo after surgery61 ± 18 6 mo after surgery58 ± 18Culture conversion within 12 mo after surgery, No. (%) Converted negative31 (47.0) Persistent negative17 (25.8) Persistent positive12 (18.2) Recurrence6 (9.1) No postsurgical cultures availablecNo records found of postsurgical cultures.39 (37.1)a Arrhythmia/atrial fibrillation (n = 4); myocardial infarction (n = 1); sinus tachycardia (n = 1); other (n = 2).b Also had pneumothorax (n = 3), pleural effusion (n = 1), and cardiac complications (n = 1; atrial fibrillation and arrhythmia). Cause of death: acute respiratory distress syndrome (n = 1), presumed aspiration pneumonia (n = 1), metastatic lung cancer (n = 1), unknown (n = 2).c No records found of postsurgical cultures. Open table in a new tab Postoperatively, 56 patients (53%) used IV amikacin for a mean of 6 weeks (SD, 6). Of the 66 patients (63%) with postsurgical mycobacterial cultures, 17 patients (26%) remained persistently negative from a negative culture status before surgery, and 12 patients (18%) remained persistently positive (Table 2). Although no follow-up cultures were available for some patients, most (73%) remained culture-negative or converted to negative after surgery. Thirty-one patients (47%) converted their mycobacterial cultures from persistent preoperative positive to negative within 12 months after surgery, of which 29 patients (94%) had cavitary disease compared with nonconverters of whom only 19 patients (56%) had cavitary disease (P = .001). Six patients (9%) experienced a recurrence (culture negative before surgery and positive after surgery), of which one-half isolated the same NTM species as initial NTM diagnosis. Ninety-four patients (90%) received multidrug therapy after surgery for a mean of 11 months (SD, 8). We report on 105 patients who received a pulmonary resection for pulmonary NTM disease. A majority of the patients (98%) were treated with a multidrug regimen before surgery. IV amikacin was administered before and after surgery to reduce bacterial burden in approximately one-half of the patients. Previous studies have demonstrated favorable outcomes in patients with localized NTM disease after pulmonary resection in combination with multidrug antimicrobial therapy. These studies showed high rates of sputum conversion and low postoperative mortality rates, with moderate complication rates.3Lang-Lazdunski L. Offredo C. Le Pimpec-Barthes F. et al.Pulmonary resection for Mycobacterium xenopi pulmonary infection.Ann Thorac Surg. 2001; 72: 1877-1882Google Scholar, 4Morino A. Murase K. Yamada K. Complications after video-assisted thoracic surgery in patients with pulmonary nontuberculous mycobacterial lung disease who underwent preoperative pulmonary rehabilitation.J Phys Ther Sci. 2015; 27: 2541-2544Google Scholar, 5Koh W.J. Kim Y.H. Kwon O.J. et al.Surgical treatment of pulmonary diseases due to nontuberculous mycobacteria.J Korean Med Sci. 2008; 23: 397-401Google Scholar, 6Asakura T. Hayakawa N. Hasegawa N. et al.Long-term outcome of pulmonary resection for nontuberculous mycobacterial pulmonary disease.Clin Infect Dis. 2017; 65: 244-251Google Scholar, 7Aznar M.L. Zubrinic M. Siemienowicz M. et al.Adjuvant lung resection in the management of nontuberculous mycobacterial lung infection: a retrospective matched cohort study.Respir Med. 2018; 142: 1-6Google Scholar,9Yu J.A. Pomerantz M. Bishop A. et al.Lady Windermere revisited: treatment with thoracoscopic lobectomy/segmentectomy for right middle lobe and lingular bronchiectasis associated with non-tuberculous mycobacterial disease.Eur J Cardiothorac Surg. 2011; 40: 671-675Google Scholar Others have reported higher postoperative complication and mortality rates, but with favorable outcomes overall.9Yu J.A. Pomerantz M. Bishop A. et al.Lady Windermere revisited: treatment with thoracoscopic lobectomy/segmentectomy for right middle lobe and lingular bronchiectasis associated with non-tuberculous mycobacterial disease.Eur J Cardiothorac Surg. 2011; 40: 671-675Google Scholar,10Shiraishi Y. Nakajima Y. Katsuragi N. et al.Pneumonectomy for nontuberculous mycobacterial infections.Ann Thorac Surg. 2004; 78: 399-403Google Scholar The overall complication rate at 3 months after surgery was 26% in our study with a death rate of 5%, which was comparable with previous studies.6Asakura T. Hayakawa N. Hasegawa N. et al.Long-term outcome of pulmonary resection for nontuberculous mycobacterial pulmonary disease.Clin Infect Dis. 2017; 65: 244-251Google Scholar,7Aznar M.L. Zubrinic M. Siemienowicz M. et al.Adjuvant lung resection in the management of nontuberculous mycobacterial lung infection: a retrospective matched cohort study.Respir Med. 2018; 142: 1-6Google Scholar,10Shiraishi Y. Nakajima Y. Katsuragi N. et al.Pneumonectomy for nontuberculous mycobacterial infections.Ann Thorac Surg. 2004; 78: 399-403Google Scholar However, a majority of the patients remained on antibiotic therapy through most of the follow-up period. Also, the advantage of VATS over open approaches is an important finding because VATS may be a preferred option for pulmonary NTM cases at specialized centers with surgeons experienced in this technique. Our study had limitations. Because of the retrospective nature of the study design, accuracy and completeness of data were less than achievable in a prospective study. Subjects were selected from two tertiary referral centers, both with expertise in surgical management of pulmonary NTM disease, which may have led to systematically different outcomes compared with community settings and which may have limited the generalizability. The timing of surgery in relation to disease severity or diagnosis could not be assessed because only 3 months of presurgical data were collected. Our data suggest that more complicated surgical approaches (eg, pneumonectomy, open surgery) that are associated typically with more severe disease are associated with worse outcomes. Finally, we did not evaluate factors that affected the decision-making by the treating physicians regarding the timing and type of pulmonary resection and therapy. Our study provides important data on the outcomes of resection for the treatment of pulmonary NTM disease in US settings. Because treatment paradigms for pulmonary NTM disease are better established, VATS for resection surgery may provide an effective treatment option for select patients, especially when accompanied by perioperative IV aminoglycoside therapy. To determine the optimal timing, perioperative antimicrobial therapy, and selection criteria for surgery, prospective multicenter cohort studies with comparison groups with long-term outcome data with comprehensive microbiologic and laboratory data are needed.