Aznar et al. BMC Pharmacology and Toxicology (2019) 20:37 https://doi.org/10.1186/s40360-019-0302-1 RESEARCH ARTICLE Open Access Safety and effectiveness of low-dose amikacin in nontuberculous mycobacterial pulmonary disease treated in Toronto, Canada Maria Luisa Aznar1,2* , Theodore K. Marras1, Ahmed Said Elshal1,3, Mahtab Mehrabi1 and Sarah K. Brode1,4 Abstract Background: Treatment guidelines suggest either a low-dose or high-dose approach when prescribing amikacin for nontuberculous mycobacterial pulmonary disease (NTM PD), but data supporting the low-dose approach are limited. The purpose of this study was to describe the safety and efficacy of the use of a low-dose of intravenous amikacin in a cohort of patients with NTM PD. Methods: We retrospectively reviewed all patients with NTM PD who received amikacin at our institution between July 1, 2003 and February 28, 2017. Demographics, clinical, microbiological and radiological data, indication and dose of amikacin, and adverse drug effects were recorded. Results: A total of 107 patients received a regimen containing amikacin for a median (IQR) of 7 (4–11) months. Seventy (65.4%) were female and the mean age (SD) was 58.3 (14.9) years. Amikacin was started at a median dose of 9.9 (2.5) mg/kg/day. Ototoxicity was observed in 30/77 (39%) patients and it was related to female sex (OR 4.96, 95%CI 1.24–19.87), and total dose of amikacin per bodyweight (OR 1.62, 95%CI 1.08–2.43). Patients of East Asian ethnicity were less likely to develop ototoxicity (0.24, 95%CI 0.06–0.95). Out of 96 patients who received amikacin for more than 3 months, 65 (67.7%) experienced symptom improvement and 30/62 (49.2%) converted their sputum to culture negative within a year. Conclusions: Patients with NTM PD treated with low-dose intravenous amikacin frequently developed ototoxicity, which was associated with female sex, and total dose of amikacin per bodyweight. Physicians should carefully consider dose, treatment duration, and long term prognosis in balancing risks and benefits of intravenous amikacin in NTM PD. Keywords: Amikacin, NTM lung disease, Nontuberculous mycobacteria Background a long course of multidrug therapy [1]. Despite treatment, Nontuberculous mycobacteria (NTM) are ubiquitous mi- prognosis is poor, and mortality remains high [1–4]. croorganisms that may cause human infections, mainly af- Use of injectable aminoglycosides has been associated fecting the lungs. Nontuberculous mycobacteria with higher rates of sputum conversion compared with pulmonary disease (NTM PD) is often severe and requires multidrug regimens not containing aminoglycosides in Mycobacterium avium complex (MAC) [5] and Myco- bacterium abscessus (M. abscessus) pulmonary disease * Correspondence: maznarru@gmail.com [6]. Guidelines from the American Thoracic Society / 1 Joint Division of Respirology, Department of Medicine, University Health Infectious Diseases Society of America (ATS/IDSA) rec- Network and Sinai Health System, 399 Bathurst Street, Toronto M5T 2S8, ON, Canada ommend amikacin use as first line therapy against M. 2 Medicine Department, Vall d’Hebron University Hospital, PROSICS Barcelona, abscessus pulmonary disease, advanced (especially fibro- Universitat Autònoma de Barcelona, Barcelona, Spain cavitary) or previously treated MAC pulmonary disease, Full list of author information is available at the end of the article © The Author(s). 2019 Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated. Aznar et al. BMC Pharmacology and Toxicology (2019) 20:37 Page 2 of 9 and as second line therapy against some other NTM in- weekly (but ranging from 2 to 7 days per week), with the fections [1]. The well described ototoxicity, vestibular dose adjusted to target a peak level of 20–25 mg/ml [1]. toxicity and renal toxicity due to amikacin limit the use During our study period, the ATS/IDSA guidelines did of this drug for prolonged periods of time. Moreover, not significantly change with respect to amikacin dosing the correlation between clinical outcomes and amikacin recommendations. The 1997 guidelines did not provide susceptibility testing break-points in patients with NTM dosing recommendations for MAC, but for M. abscessus PD are not firmly established. Data suggest that a high they recommend either a low or higher dose approach, Minimum Inhibitory Concentration (MIC) for amikacin either once or twice a day [11]. is associated with worse clinical outcomes in MAC-PD All patients who received amikacin for any duration [7, 8], although formal recommendations have not yet were reviewed for toxicities. Blood work including serum been developed for MAC-PD. There are no data regard- creatinine was routinely performed every month while ing the association between MIC for amikacin and clin- on amikacin. Nephrotoxicity was defined as increase in ical outcomes in M. abscessus-PD, although M. the absolute serum creatinine level of > 20 umol/L from abscessus is considered susceptible to amikacin at a MIC baseline or > 20% change in subsequent testing as in pre- ≤16 mcg/ml [9]. Furthermore, although treatment guide- vious studies [12]. Audiograms during amikacin therapy lines suggest either a low-dose or high-dose approach (advised to be done routinely) were recorded and com- when prescribing amikacin for NTM PD, data support- pared with audiograms performed before starting amika- ing the low-dose approach are scarce. Despite of all cin. Bilateral hearing was tested at 250, 500, 1000, 2000, these limitations, amikacin remains one of the most po- 4000, and 8000 Hz. Patients whose hearing threshold tent drugs for the treatment of NTM PD, and therefore, level were greater than 40 dB before starting amikacin data on its risks and benefits are critical. were considered as having disabling baseline hearing The objective of this study was to describe the safety impairment [13]. Ototoxicity was defined as a > 20 dB and efficacy of the use of low-dose intravenous amikacin neurosensory hearing loss from baseline in either ear at in a cohort of patients with NTM PD. any single frequency [14], seen on at least two consecu- tive audiograms, unless only one audiogram was per- Methods formed after amikacin initiation. Persistent ototoxicity All patients with NTM-PD treated in the Toronto was defined if neurosensory hearing loss was observed Western Hospital NTM clinic between July 1, 2003 and after discontinuation of amikacin. Subjective symptoms February 28, 2017 were retrospectively reviewed. Inclu- of ototoxicity were also recorded. Vestibular toxicity was sion criteria for the study were: a) diagnosis of NTM-PD defined as having occurred if the patient reported com- defined according to the ATS/IDSA criteria [1], and b) patible symptoms during a clinic visit, ie vertigo, dis- treatment regimen that included intravenous amikacin equilibrium, or ataxia, +/− nausea/vomiting or physical for any period of time. The study was approved by the examination suggestive of vestibulotoxicity. University Health Network Research Ethics Board. Base- line demographic, clinical, microbiological and radio- Sputum culture and drug susceptibility testing logical data were collected. Specimen culture was performed using the Bactec 460 The predominant radiological pattern was classified as TB system (Becton Dickinson Microbiology Systems, nodular bronchiectatic, fibrocavitary, consolidation, ran- USA) until May 2000 and, thereafter, with the Bactec dom nodules and unclassifiable based on computed MGIT 960 system (Becton Dickinson Microbiology tomography (CT) [10]. Indication, dose and duration of Systems). Specimens were processed and analyzed as amikacin were also recorded. described by Hanna et al. [15]. NTM cultures were speciated using DNA probes (AccuProbe, Gen-Probe, Amikacin administration and toxicity USA) for MAC and Mycobacterium gordonae, and by A Peripherally Inserted Central Catheter (PICC) was high-performance liquid chromatography for other spe- inserted and the first dose of amikacin was administered cies [16]. Drug Susceptibility Testing was performed at the hospital and, if no adverse reactions, the following using the CLSI-recommended method of broth microdi- doses were administered at an outpatient setting (infu- lution, using commercially prepared microtiter dilution sion clinic or patient residence), generally administered plates (Thermo Fisher, Cleveland, OH) [9]. Concentra- during a 30-min infusion. Routine amikacin peak levels tions of amikacin tested ranged from ≤1 to 64 μg/ml. were performed 30 to 45 min after amikacin infusion Isolates with a MIC ≤16 were considered susceptible, every month, with the subsequent doses adjusted ac- those with a MIC of 32 or 64 were considered inter- cordingly. Amikacin dosing followed the low-dose mediate and those with a MIC > 64 were considered approach suggested in the ATS/IDSA guidelines; the resistant [8, 16]. Since amikacin Drug Susceptibility starting dose was 8–10 mg/kg, typically given thrice Testing requires special request, it was not always Aznar et al. BMC Pharmacology and Toxicology (2019) 20:37 Page 3 of 9 requested in the early study period because of a lack of Table 1 Baseline characteristics of patients treated with data regarding clinical correlation. amikacin Baseline characteristics N = 107 (%) Clinical and microbiological outcomes Sex, Female 70 (65.4) Criteria for clinical outcomes and culture conversion Age (Years), Mean (SD) 58.3 (14.9) have been previously defined [17]. Characteristics of pa- Race tients who converted their sputum were compared with Caucasian 72 (67.3) those of patients who continued to have positive cultures after starting amikacin. South Asian 4 (3.7) East Asian 28 (26.2) Statistical analysis African 3 (2.8) Descriptive statistics are presented as number (percent) Comorbidities and median (Interquartile range, IQR) or mean (Stand- COPD/Emphysema 19 (17.7) ard deviation, SD) depending on variable normality. The Previous TB 13 (12.1) normal distribution of quantitative variables was tested through the Shapiro–Wilk test. Categoric variables were Cystic fibrosis 6 (5.6) compared using chi-square tests or Fisher’s exact tests Asthma 5 (4.7) and continuous variables were compared using t-tests or Autoimmune disorders 5 (4.7) Mann-Whitney tests. Demographics, dose and duration Alpha 1 antitrypsin deficiency 2 (1.9) of amikacin, repeated exposure to amikacin, baseline Sarcoidosis 2 (1.9) weight, baseline serum creatinine and baseline disabling ABPA 2 (1.9) hearing impairment were selected as potential risk fac- tors of ototoxicity. Demographics, NTM species, CT Allogenic stem cell transplant 2 (1.9) pattern, dose and duration of amikacin, and surgical HIV infection 1 (0.9) management were selected as potential prognostic vari- Species ables for culture conversion. We developed two models MACa 69 (64.5) using a multivariate logistic regression analysis including M. abscessus 21 (19.6) ototoxicity and culture conversion as dependent variables. M. xenopi 11 (10.3) Variables with a p-value < 0.2 in the univariate analysis, and variables considered a priori to be clinically important, MAC and M. xenopi 4 (3.7) were included as independent variables. Multicollinearity M. shimodei 1 (0.9) was explored using variance inflation factor (VIF). A M. fortuitum 1 (0.9) two-sided P value of 0.05 was used as the cutoff for statis- AFB smear positive ever 100 (93.5) tical significance. All analyses were performed using SPSS Aspergillus isolation 39 (36.4) statistical software (IBM SPSS version 23, Armonk, N.Y.). CT pattern Results Nodular bronchiectatic 55 (51.4) One hundred and seven patients were treated with a Fibrocavitary 40 (37.4) regimen containing intravenous amikacin. Median dur- Random nodules 2 (1.9) ation of amikacin (IQR) was 7 (4–11) months. Table 1 Consolidation 0 (0) shows baseline characteristics of patients treated with Unclassifiable 10 (9.3) amikacin. Forty-four (41.1%) patients had previously Cavity 66 (61.7) been treated for NTM PD and one patient had received amikacin previously at another institution. Amikacin Indication for treatment was started at a median (IQR) dose of 9.5 (8.3–10.4) mg/ First line 39 (36.4) kg/day, thrice weekly in 77 (72%) patients, 5 days weekly Second line 68 (63.6) in 20 (18.7%) patients, daily in 5 (4.7%) patients, twice SD Standard deviation, COPD Chronic Obstructive Pulmonary Disease, TB weekly in 4 (3.7%) patients, and every 48 h in 1 (0.9%) Tuberculosis, SLE Systemic Lupus Erythematous, ABPA Allergic Bronchopulmonary Aspergillosis, HIV Human Immunodeficiency Virus, MAC patient. Amikacin was prescribed in combination with a Mycobacterium avium complex, AFB Acid Fast Bacilli, CT multidrug oral antibiotic regimen in all patients, and 33 Computed Tomography a (30.8%) patients underwent lung surgery (30 patients Five patients had Clarithromycin resistant MAC; MAC isolates were not identified to the species level prior to 2010, and so are reported as underwent lung resection, 1 patient underwent lung MAC herein transplant, 1 patient underwent cardiopulmonary trans- plant and 1 patient underwent a Clagett window). Aznar et al. BMC Pharmacology and Toxicology (2019) 20:37 Page 4 of 9 Twenty-four patients (22.4%) received a second course Table 2 Adverse effects associated with amikacin of amikacin for a median (IQR) of 4 (3–10) months, in- Adverse effects N (%) cluding 9 (37.5%) patients who restarted all antibiotics Ototoxicity due to an infection recurrence and 15 (62.5%) who never Subjective symptoms 43/107 (40.2%) stopped the oral antibiotics and were re-started on ami- Hearing impairment seen in audiograms 30/77 (39.0%) kacin due to worsening of their symptoms. Nine patients (8.4%) received a third course of amikacin for a median Nephrotoxicity 6/96 (6.2%) (IQR) of 26 (4.5–35) months. Median (IQR) total dose Vestibular toxicity 4/107 (3.7%) of amikacin was 50.5 g (27.3–106.7). PICC related complications The first course of amikacin was discontinued in 34 Venous thromboembolism 7/107 (6.5%) (31.8%) patients due to an adverse effect (17 (50%) oto- Bloodstream infection 1/107 (0.9%) toxicity, 6 (17.6%) fatigue, 4 (11.8%) renal insufficiency, 5 Fatigue 6/107 (5.6%) (14.7%) generalized skin rash, 1 (2.9%) paresthesias, and 1 (2.9%) restless extremities), 4 (3.7%) patients discontin- Skin rash 5/107 (4.7%) ued treatment due to poor improvement of symptoms Paresthesias 1/107 (0.9%) and 4 (3.7%) discontinued amikacin for other reasons Restless extremities 1/107 (0.9%) (need to travel in 1 case, need to treat another infection PICC Peripherally Inserted Central Catheter in 1 case, and poor adherence in 2 cases). Ten (9.3%) pa- tients died while on amikacin. Causes of death were symptoms. Audiograms were available in 77 (72%) NTM progression in 6 patients, postoperative complica- patients. Five (6.5%) patients only had one follow up tion after adjuvant lung resection in 2 patients, aspergil- audiogram. Overall, there was objective ototoxicity in 30 losis in 1 patient and cardiac arrest in 1 patient. Out of (39%) patients; 16 (53.3%) of these 30 patients reported 24 patients who received a second course of amikacin, subjective ototoxicity. Out of 34 patients with subjective 10 (41.7%) stopped the amikacin due to an adverse effect ototoxicity who had an audiogram, 16 (47.1%) had evi- (5 ototoxicity, 2 skin rash, 2 PICC infection and 1 vestibu- dence of objective ototoxicity. Ototoxicity appeared a lar toxicity), 1 stopped amikacin due to poor clinical re- median of 5.5 (3–11) months after the initiation of ami- sponse, and 2 stopped amikacin due to a need to treat kacin. Although high frequency hearing loss (>2000 Hz) other medical conditions. Out of 9 patients who received was seen most often, 3 (10%) of the 30 patients showed a third course of amikacin, 2 were still on amikacin at the deficits in both high and low frequencies. end of the study, 1 completed the treatment course, 4 de- Patients developing ototoxicity were more likely to be veloped an adverse effect (acute renal insufficiency in 1 female (OR 4.96, 95%CI 1.24–19.87, p = 0.024), received patient, ototoxicity in 1 patient, skin rash in 1 patient and a higher total dose of amikacin per baseline bodyweight fatigue in 1 patient), and 2 died due to NTM progression. (g/kg) (OR 1.62, 95%CI 1.08–2.43, p = 0.020) and were less likely to be of East Asian ethnicity (OR 0.24, 95%CI Adverse effects 0.06–0.95, p = 0.042) than patients who did not develop Four (3.7%) patients developed vestibular toxicity after ototoxicity. This information is summarized in Table 3. starting amikacin. Blood work was available in 96 patients, and it showed elevated serum creatinine con- Amikacin MIC centrations in 6 (6.2%) patients. Renal insufficiency was Twenty-six patients (14 infected by MAC and 12 infected reversible in 5 patients after stopping amikacin and in 1 by M. abscessus) had amikacin Drug Susceptibility Testing after decreasing the dose of amikacin. Seven (6.5%) before initiating this agent. Out of 14 MAC isolates tested, patients experienced a PICC associated thrombosis, one 11 (78.6%) showed susceptibility to amikacin, 2 (14.3%) of them was diagnosed of a factor V Leyden deficiency showed intermediate resistance and 1 (7.1%) was resistant thereafter, and one patient developed a Staphylococcus to amikacin. Eleven (91.7%) M abscessus isolates were aureus bacteremia. Five patients developed a skin rash susceptible to amikacin, and 1 (8.3%) was resistant. Three that disappeared after stopping the amikacin. Six out of 4 patients whose isolates were either intermediate patients experienced severe fatigue while on amikacin or resistant to amikacin had not been previously exposed that lead to discontinuation of treatment. Table 2 to aminoglycosides. Thirteen patients had repeat Drug summarizes the adverse effects associated with amikacin. Susceptibility Testing after receiving amikacin and one of them developed amikacin resistance. Ototoxicity Forty-three (40.2%) patients developed subjective oto- Microbiological and clinical outcomes toxicity while on amikacin; 32 reported tinnitus, 7 Sixty-five (67.7%) of the 96 patients who received a first reported subjective hearing loss and 4 reported both course of amikacin for ≥3 months experienced Aznar et al. BMC Pharmacology and Toxicology (2019) 20:37 Page 5 of 9 Table 3 Comparison of demographic and clinical characteristics in patients with and without audiographically demonstrated ototoxicity Ototoxicity No ototoxicity Univariate model Multivariate modela N = 30 N = 47 OR (95%CI) P value OR (95%CI) P value Sex, female 24 (80%) 29 (61.7%) 2.48 (0.85–7.24) 0.096 4.96 (1.24–19.87) 0.024 Age (years), median (IQR) 61 (43–65) 65 (51–72) 0.98 (0.95–1.01) 0.247 0.98 (0.94–1.02) 0.297 Race Caucasian 22 (73.3%) 29 (61.7%) 1.71 (0.63–4.64) 0.295 South Asian 1 (3.3%) 1 (2.1%) 1.59 (0.09–26.36) 0.748 East Asian 6 (20%) 16 (34%) 0.48 (0.16–1.42) 0.188 0.24 (0.06–0.95) 0.042 African 1 (3.3%) 1 (2.1) 1.59 (0.09–26.36) 0.748 Weekly dose AK per bodyweight 29 (23.7–41.3) 25.2 (21.8–29.4) 1.05 (1.00–1.10) 0.047 1.03 (0.98–1.09) 0.248 (mg/kg/week), median (IQR)b Months on AK, median (IQR) 11.5 (7–18.25) 7 (4–10) 1.04 (1.00–1.08) 0.052 Total dose AK per bodyweight (g/kg), 1.81 (0.85–3.13) 0.74 (0.51–1.27) 1.59 (1.08–2.33) 0.018 1.62 (1.08–2.43) 0.020 Median (IQR)c Total dose AK (g), Median (IQR)d 77.3 (52.9–196.3) 41.7 (25.6–67.5) 1.01 (1.00–1.02) 0.011 Repeated exposure to amikacin 9 (30%) 10 (21.3%) 1.59 (0.56–1.52) 0.388 Baseline weight (kg), Median (IQR) 51.5 (41–66.5) 55 (48–64) 0.99 (0.96–1.03) 0.712 Initial serum creatinine (umol/L), 66.0 (56.7–74.0) 63 (53–75) 1.01 (0.97–1.04) 0.656 mean (SD) Baseline disabling hearing impairment 8 (26.7%) 24 (51.1%) 0.35 (0.13–0.94) 0.037 0.54 (0.17–1.74) 0.305 Current treatment with macrolides 29 (96.7%) 45 (95.7%) 1.23 (0.11–14.87) 0.839 Nephrotoxicity 1 (3.6%) 4 (8.5%) 0.38 (0.04–3.58) 0.398 IQR Interquartile Range, AK Amikacin a Variables with a p-value < 0.2 in the univariate analysis, and variables considered a priori to be clinically important, were included in the multivariate model, wherein multicollinearity was explored using variance inflation factor (VIF). Months on amikacin and total dose amikacin per bodyweight both had VIFs> 10, and so were eliminated b OR represents risk associated with each increase of 1 mg/kg/week c OR represents risk associated with each increase of 1 g/kg total dose d OR represents risk associated with each increase of 1 g total dose Total dose of amikacin and months on amikacin were excluded from the multivariate analysis due to collinearity improvement in their symptoms after starting amikacin associated with sputum culture conversion are shown in and 31 (45.6%) out of 68 patients who had a previous Table 4. positive sputum culture before starting amikacin and fol- low up sputum after amikacin initiation converted their Discussion sputum within a year following the initiation of amika- We observed that ototoxicity was common, and it was cin. Twenty-six patients experienced persistent culture associated with female sex, and total dose of amikacin conversion. None of the 15 patients who received a re- per bodyweight in patients with NTM PD. Vestibular current (second or third) course of amikacin for ≥3 toxicity and nephrotoxicity were not frequently ob- months and who had a positive culture before starting served. Most patients had symptomatic improvement amikacin achieved sputum culture conversion. after starting amikacin, however, only half of them Patients who achieved sputum culture conversion achieved sputum culture conversion. underwent surgery more often than those who did not Amikacin has been shown to be an effective drug in achieve sputum culture conversion (61.3% vs 5.4%, p < the management of NTM PD, especially in patients with 0.001). Excluding patients who underwent surgery, 12/ cavitary disease and in those whose isolate is macrolide 47 (25.5%) of patients achieved sputum culture conver- resistant MAC or M. abscessus [6, 18, 19]. However, the sion. One of the 2 patients with a MIC for amikacin > 64 potential adverse effects due to aminoglycosides limit achieved sputum culture conversion. In the multivariate the use of these drugs. The most common adverse effects logistic regression model, surgical intervention was the due to aminoglycosides are ototoxicity, vestibular toxicity only variable associated with sputum culture conversion and nephrotoxicity [20, 21]. Previous studies describing (OR 33.01, 95%CI 4.79–22.70, p < 0.001). Factors patients with NTM pulmonary disease treated with Aznar et al. BMC Pharmacology and Toxicology (2019) 20:37 Page 6 of 9 Table 4 Patient characteristics and their association with sputum culture conversion Sputum culture No sputum Univariate model Multivariate model conversion culture OR (95%CI) P value OR (95%CI) P value N = 31 conversion N = 37 Sex, female 23 (74.2%) 23 (62.2%) 1.75 (0.62–4.97) 0.293 Age (years), median (IQR) 58 (50–67) 61 (44–47.5) 1.00 (0.97–1.03) 0.953 Species/complex MAC 17 (54.8%) 20 (54.1%) 1.02 (0.38–2.73) 0.969 M. abscessus 3 (9.7%) 10 (27%) 0.28 (0.76–22.23) 0.073 0.87 (0.18–4.28) 0.868 M. xenopi 6 (19.4%) 2 (5.4%) 4.12 (0.76–22.23) 0.099 9.50 (0.68–131.82) 0.094 MAC/M. xenopi 1 (3.2%) 2 (5.4%) Macrolide resistant MAC 2 (6.5%) 3 (8.1%) 0.76 (0.12–4.89) 0.774 M. simiae 1 (3.2%) 0 M. fortuitum 1 (3.2%) 0 FC pattern 13 (35.1%) 13 (41.9%) 1.33 (0.50–3.56) 0.566 Cavity 21 (67.7%) 23 (62.2%) 1.28 (0.47–3.49) 0.632 AFB positive ever 30 (96.8%) 37 (100%) 3.69 (0.15–93.81) 0.791 Total dose AK per bodyweight 0.79 (0.55–2.29) 1.20 (0.74–2.09) 0.81 (0.55–1.18) 0.268 0.54 (0.26–1.10) 0.089 (g/kg), median (IQR)b Weekly dose AK per bodyweight 26.09 (23.61–33.87) 27.03 (22.6–40.3) 0.98 (0.94–1.03) 0.438 (mg/kg/week), median (IQR)a Months on AK, median (IQR) 8 (4–18) 8 (6–15) 0.98 (0.94–1.02) 0.356 Total dose AK (g), Median (IQR)c 46.8 (29.4–130.3) 60.3 (37.8–118.0) 1.00 (0.99–1.01) 0.725 Surgery 19 (61.3%) 2 (5.4%) 27.7 (5.61–136.9) < 0.001 33.01 (4.79–227.27) < 0.001 FC Fibrocavitary pattern, IQR Interquartile Range, MAC Mycobacterium avium Complex, AK Amikacin a OR represents risk associated with each increase of 1 mg/kg/week b OR represents risk associated with each increase of 1 g/kg total dose c OR represents risk associated with each increase of 1 g total dose regimens containing aminoglycosides have reported a be calculated that patients received approximately 462 variable incidence of adverse effects, although doses and mg/kg and 540 mg/kg cumulative doses respectively; the duration of amikacin varied among studies [6, 12, 14]. lower total dose per bodyweight used in these studies In our study, we observed that 39% of patients experi- may explain the lower rates of ototoxicity observed, enced ototoxicity after a median of 5.5 (3–11) months of compared with the Peloquin study and our study. amikacin. Peloquin and colleagues observed a similar in- We observed that most patients with ototoxicity expe- cidence of ototoxicity (37%). Although the doses used in rienced hearing impairment in higher frequencies. our study are lower than the doses used in the Peloquin Aminoglycosides cause degeneration of hair cells in the study (median dose of 9.5 mg/kg thrice weekly vs. 15 organ of Corti, predominantly in the basal turn which is mg/kg daily or 25 mg/kg thrice weekly, respectively), the required to sense high-frequency sounds, so total duration of amikacin used in our study was longer high-frequency hearing loss is more common in patients (median 33 vs. 15 weeks), so the total dose of amikacin developing ototoxicity related with aminoglycosides [23]. per kilogram bodyweight was similar (909 mg/kg in our Half of our patients with subjective symptoms of ototox- study vs. 1012 mg/kg in the daily dose group and 967 icity had a normal audiogram. The absence of objective mg/kg in the thrice-weekly dosing group in the Peloquin hearing changes in patients reporting symptoms of oto- study). Lee et al. observed a 25% rate of ototoxicity in toxicity have been previously reported [14] and it might patients infected by M. abscessus receiving a median be due to different mechanisms of ototoxicity. Some of dose of amikacin of 19.3 mg/kg/day during 4 weeks [22]. the subjective symptoms could be related to an acute Ellender et al. observed a 18% rate of ototoxicity in pa- and not permanent toxicity due to high dose aminogly- tients with NTM PD receiving a median dose of 22 mg/ cosides and may not be reflected in audiograms [24]. In kg/day for a median of 8.64 weeks [6, 12]. Although the addition, we did not review changes in speech recogni- total dose of amikacin per bodyweight was not provided tion thresholds, which may potentially vary from from the studies by Ellender et al. and Lee et al., it can changes in the pure-tone audiogram thresholds. Aznar et al. BMC Pharmacology and Toxicology (2019) 20:37 Page 7 of 9 Different factors have been associated with an in- exclusively with antibiotics). Ellender at al. had similar creased risk of ototoxicity. Age, male sex, low body results; 76% of patients achieved a symptomatic im- weight, total dose of amikacin and duration of amikacin provement and 38% achieved persistent sputum culture have been reported as risk factors for ototoxicity and conversion [12]. Although the proportion achieving nephrotoxicity [14, 25–29]. In our study, we observed sputum culture conversion was somewhat higher in our that females and patients who received a higher total study, 21 (20%) of our patients underwent surgical resec- dose of amikacin per bodyweight were more likely to tion, compared with no patients in their study. experience ototoxicity. We also observed that East Asian Namkoong et al. presented a small number of patients patients were less likely to develop ototoxicity than with M. abscessus pulmonary disease, and 8/13 (61.5%) patients from other ethnicities. The explanation of this achieved persistent sputum culture conversion; they also finding is not clear. Although some mitochondrial muta- used a higher daily dose and shorter duration of amika- tions have been linked to ototoxic responses to amino- cin than in our study, but did not report total dose [6]. glycosides, it is not clear if there are differences in the Other studies describing amikacin administration proto- prevalence of these mutations along ethnic groups [30]. cols and adverse events in NTM-PD did not present Macrolides have been also associated with ototoxicity data on clinical outcomes [14, 37]. Given the limited [31, 32]. We did not observe differences regarding studies describing amikacin use in NTM-PD, and the macrolide use between patients who developed ototox- heterogeneity in the patients studied in terms of NTM icity and those who did not, but since almost all of our species treated, severity of disease, companion drugs, patients received macrolides, our dataset lacked ad- and use of surgery, it is difficult to draw conclusions equate variability to address this question. regarding the optimal amikacin protocol to maximize Vestibular toxicity observed in our study was 3.7%, clinical outcomes and minimize toxicity. The ATS/IDSA lower than in previous reports (7.4–9%) [14, 33]. How- guidelines describe either low-dose or high-dose ap- ever, our data were based on patient self-reported symp- proaches, noting that high-dose is not likely to be toler- toms rather than physical examination focused on ated for long durations that may be required to control vestibular toxicity, so we may have missed some cases of NTM-PD. It is clear that the cumulative dose of amika- toxicity. Six patients experienced nephrotoxicity, and cin is reliably associated with ototoxicity. In our clinical serum creatinine returned to normal values in all of experience the large majority of NTM-PD patients pre- them after either stopping or decreasing the dose of fer to withhold amikacin (and experience worsening amikacin. Nephrotoxicity is another concerning adverse NTM-PD) rather than experience severe hearing impair- effect associated with the use of aminoglycosides. How- ment. We think it is important to consider both of the ever, it is more often observed in patients receiving sev- above when crafting a regimen for individual patients. In eral doses a day and those who already have baseline situations where definitive therapy is possible, like surgi- renal impairment [34]. Previous series reporting patients cal resection of limited disease, an intensive high-dose, with NTM PD treated with intravenous amikacin have short-term regimen may be preferable, assuming the not observed nephrotoxicity as a common adverse effect likelihood of recurrence is low. In situations where [35]. We also observed that 6 patients developed a PICC disease suppression is the goal, where it is likely that associated thrombosis and one patient developed a PICC amikacin will be useful for a prolonged period or in associated systemic infection, similar to rates previously recurrent courses, then a low-dose regimen may be reported [36]. preferred. Twenty-six patients had amikacin Drug Susceptibility We also observed that none of the patients who re- Testing performed before starting this agent and two of ceived a subsequent course of amikacin achieved sputum them (1 MAC and 1 M. abscessus) had an amikacin MIC culture conversion. Although no previous studies have >64. Although correlation between break points and reported specific data about subsequent courses of ami- clinical outcomes are not clear, it seems that patients in- kacin, patients previously treated for NTM lung disease fected by MAC whose isolates have a MIC for amikacin are less likely to achieve sputum culture conversion than > 64 have a poor prognosis compared with those who patients not previously treated [38]. We observed that have lower MICs [7, 8]. In our study, one of the two pa- patients who underwent lung resection for the manage- tients with an isolate with a MIC >64 achieved sputum ment of NTM pulmonary disease were more likely to conversion. Interestingly, neither of these patients was achieve sputum culture conversion than patients treated known to have previously received aminoglycosides. exclusively with antibiotics. This finding has been previ- Symptomatic improvement after taking amikacin was ously reported for patients with an infection caused by observed in 68% of patients, and persistent sputum cul- M. abscessus and macrolide resistant MAC [4, 19]. ture conversion within a year after the initiation of ami- Our study has several limitations. First, some of the kacin was observed in 46% of patients (25% treated adverse effects attributed to amikacin may have been Aznar et al. BMC Pharmacology and Toxicology (2019) 20:37 Page 8 of 9 due to another of the multiple accompanying antibiotics SKB; Original draft preparation: MLA, TKM, SKB; Review and editing MLA, TKM, that patients received for the treatment of NTM-PD. SA, MM, SKB. All the authors read and approved the final manuscript. However, the adverse effects reported in the study were Ethics approval and consent to participate thought to be related to amikacin. Second, the study was The study was approved by the University Health Network Research Ethics conducted retrospectively, so information was not col- Board, with a waived requirement for informed consent due to the retrospective nature of the analysis. lected in a standardized fashion. This is particularly rele- vant in that not all patients had audiograms performed. Consent for publication Although we recommend baseline and routine follow-up Not applicable. audiometry for all amikacin-treated patients, patients with more risk factors or subjective symptoms of ototox- Competing interests SB and TM report that their institution has been paid by Insmed for their icity may have been more adherent to this recommenda- involvement in a randomized clinical trial of inhaled amikacin for NTM lung tion. Accordingly, the risk of ototoxicity may be disease, outside the submitted work. TM reports that his institution has been overestimated. Third, there may have been differences in paid for his consultative and/or educational services from Insmed, Astra- Zeneca, RedHill and Horizon, all outside the submitted work. SB reports that management decisions during the long time period of her institution has been paid for her consultative and/or educational services our study. However, as noted above, the ATS/IDSA from Boehringer Ingelheim, outside the submitted work. guidelines did not significantly change with respect to amikacin dosing recommendations during our study Publisher’s Note period. Likewise our approach did not significantly Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. change, typically following a lower dose strategy because of our patients’ age range and strong aversion to the risk Author details 1 of ototoxicity as well as the high rates of recurrence of Joint Division of Respirology, Department of Medicine, University Health Network and Sinai Health System, 399 Bathurst Street, Toronto M5T 2S8, ON, NTM lung disease with the potential need for retreat- Canada. 2Medicine Department, Vall d’Hebron University Hospital, PROSICS ment with amikacin. Barcelona, Universitat Autònoma de Barcelona, Barcelona, Spain. 3 Gastroenterology Department, National Hepatology and Tropical Medicine Institute, Cairo, Egypt. 4West Park Healthcare Centre, Toronto, Canada. Conclusions We observed that patients with NTM-PD treated with Received: 18 January 2019 Accepted: 17 April 2019 low-dose intravenous amikacin frequently developed ototoxicity, and it was associated with female sex, and References total dose of amikacin per bodyweight. The symptoms of 1. Griffith DE, Aksamit T, Brown-Elliott BA, Catanzaro A, Daley C, Gordin F, et al. most patients improved after starting amikacin, and half An official ATS/IDSA statement: diagnosis, treatment, and prevention of nontuberculous mycobacterial diseases. Am J Respir Crit Care Med. 2007; of them achieved sputum culture conversion. More 175(4):367–416. research is needed regarding the role of amikacin in 2. Field SK, Fisher D, Cowie RL. Mycobacterium avium complex pulmonary NTM-PD, and the optimal protocol to achieve the most disease in patients without HIV infection. Chest. 2004;126(2):566–81. 3. Fleshner M, Olivier KN, Shaw PA, Adjemian J, Strollo S, Claypool RJ, et al. benefit with the least adverse effects. Mortality among patients with pulmonary non-tuberculous mycobacteria disease. Int J Tuberc Lung Dis. 2016;20(5):582–7. Abbreviations 4. Jarand J, Levin A, Zhang L, Huitt G, Mitchell JD, Daley CL. Clinical and ATS/IDSA: American Thoracic Society / Infectious Diseases Society of microbiologic outcomes in patients receiving treatment for Mycobacterium America; CT: Computed tomography; dB: Decibels; IQR: Interquartile range; abscessus pulmonary disease. Clin Infect Dis. 2011;52(5):565–71. MAC: Mycobacterium avium complex; MIC: Minimum Inhibitory 5. Kobashi Y, Matsushima T, Oka M. A double-blind randomized study of Concentration; NTM PD: NTM pulmonary disease; NTM: Nontuberculous aminoglycoside infusion with combined therapy for pulmonary mycobacterial; PICC: Peripherally Inserted Central Catheter; SD: Standard Mycobacterium avium complex disease. Respir Med. 2007;101(1):130–8. deviation; VIF: Variance inflation factor 6. Namkoong H, Morimoto K, Nishimura T, Tanaka H, Sugiura H, Yamada Y, et al. Clinical efficacy and safety of multidrug therapy including thrice weekly Acknowledgements intravenous amikacin administration for Mycobacterium abscessus None pulmonary disease in outpatient settings: a case series. BMC Infect Dis. 2016; 16:396. 7. Brown-Elliott BA, Iakhiaeva E, Griffith DE, Woods GL, Stout JE, Wolfe CR, et al. Funding In vitro activity of amikacin against isolates of Mycobacterium avium No funding was received for this study. complex with proposed MIC breakpoints and finding of a 16S rRNA gene mutation in treated isolates. J Clin Microbiol. 2013;51(10):3389–94. Availability of data and materials 8. Olivier KN, Griffith DE, Eagle G, McGinnis Ii JP, Micioni L, Liu K, et al. The data that support the findings of this study are available from the first Randomized trial of liposomal amikacin for inhalation in nontuberculous author but restrictions apply to the availability of these data, which were mycobacterial lung disease. Am J Respir Crit Care Med. 2017;195(6):814–23. used under license for the current study, and so are not publicly available. 9. Clinical Laboratory Standards Institute. Susceptibility testing of Data are however available from the authors upon reasonable request and mycobacteria, nocardiae, and other aerobic actinomycetes; approved with permission of the University Health Network Research Ethics Board. standard. 2nd ed. CLSI document No. M24-A2. Wayne: CLSI; 2011. 10. Carrillo MC, Patsios D, Wagnetz U, Jamieson F, Marras TK. Comparison of the Authors’ contributions spectrum of radiologic and clinical manifestations of pulmonary disease Conceptualization: MLA, TKM, SKB; Formal analysis: MLA, TKM, SKB caused by Mycobacterium avium complex and Mycobacterium xenopi. Can Methodology: MLA, TKM, SA, MM, SKB; Data collection: MLA, TKM, SA, MM, Assoc Radiol J. 2014;65(3):207–13. Aznar et al. BMC Pharmacology and Toxicology (2019) 20:37 Page 9 of 9 11. Diagnosis and treatment of disease caused by nontuberculous mycobacteria. 34. Wargo KA, Edwards JD. Aminoglycoside-induced nephrotoxicity. J Pharm This official statement of the American Thoracic Society was approved by the Pract. 2014;27(6):573–7. Board of Directors, March 1997. Medical section of the American Lung 35. Lyu J, Jang HJ, Song JW, Choi CM, Oh YM, Lee SD, et al. Outcomes in Association. Am J Respir Crit Care Med. 1997;156(2 Pt 2):S1–25. patients with Mycobacterium abscessus pulmonary disease treated with 12. Ellender CM, Law DB, Thomson RM, Eather GW. Safety of IV amikacin in the long-term injectable drugs. Respir Med. 2011;105(5):781–7. treatment of pulmonary non-tuberculous mycobacterial disease. 36. Chopra V, Montoya A, Joshi D, Becker C, Brant A, McGuirk H, et al. Respirology. 2016;21(2):357–62. Peripherally inserted central catheter use in skilled nursing facilities: a pilot 13. WHO. Future programme developments for prevention of deafness and study. J Am Geriatr Soc. 2015;63(9):1894–9. hearing impairment: report of the first informal consultation, Geneva, 1997 37. Lee H, Sohn YM, Ko JY, Lee SY, Jhun BW, Park HY, et al. Once-daily dosing http://whqlibdoc.who.int/hq/1997/WHO_PDH_97.3.pdf . of amikacin for treatment of. Int J Tuberc Lung Dis. 2017;21(7):818–24. 14. Peloquin CA, Berning SE, Nitta AT, Simone PM, Goble M, Huitt GA, et al. 38. Tanaka E, Kimoto T, Tsuyuguchi K, Watanabe I, Matsumoto H, Niimi A, et al. Aminoglycoside toxicity: daily versus thrice-weekly dosing for treatment of Effect of clarithromycin regimen for Mycobacterium avium complex mycobacterial diseases. Clin Infect Dis. 2004;38(11):1538–44. pulmonary disease. Am J Respir Crit Care Med. 1999;160(3):866–72. 15. Hanna BA, Ebrahimzadeh A, Elliott LB, Morgan MA, Novak SM, Rusch-Gerdes S, et al. Multicenter evaluation of the BACTEC MGIT 960 system for recovery of mycobacteria. J Clin Microbiol. 1999;37(3):748–52. 16. Butler WR, Guthertz LS. Mycolic acid analysis by high-performance liquid chromatography for identification of Mycobacterium species. Clin Microbiol Rev. 2001;14(4):704–26. 17. Aznar ML, Brode SK, Mehrabi M, Marras TK. Safety and effectiveness of clofazimine in nontuberculous mycobacterial lung disease. Can J Respir Crit Care Sleep Med. 2018;2(2):72 77. https://doi.org/10.1080/24745332.2017. 1410455. 18. Griffith DE, Brown-Elliott BA, Langsjoen B, 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(8):928–34. 19. 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. Ann Am Thorac Soc. 2016;13(11):1904–11. 20. Kahlmeter G, Dahlager JI. Aminoglycoside toxicity - a review of clinical studies published between 1975 and 1982. J Antimicrob Chemother. 1984; 13(Suppl A):9–22. 21. Brummett RE, Fox KE. Aminoglycoside-induced hearing loss in humans. Antimicrob Agents Chemother. 1989;33(6):797–800. 22. Lee H, Sohn YM, Ko JY, Lee SY, Jhun BW, Park HY, et al. Once-daily dosing of amikacin for treatment of Mycobacterium abscessus lung disease. Int J Tuberc Lung Dis. 2017;21(7):818–24. 23. Zheng Y, Schachern PA, Sone M, Papapella MM. Aminoglycoside ototoxicity. Otol Neurotol. 2001;22(2):266–8. 24. Barclay ML, Kirkpatrick CM, Begg EJ. Once daily aminoglycoside therapy. Is it less toxic than multiple daily doses and how should it be monitored? Clin Pharmacokinet. 1999;36(2):89–98. 25. Gatell JM, Ferran F, Araujo V, Bonet M, Soriano E, Traserra J, et al. Univariate and multivariate analyses of risk factors predisposing to auditory toxicity in patients receiving aminoglycosides. Antimicrob Agents Chemother. 1987; 31(9):1383–7. 26. Modongo C, Pasipanodya JG, Zetola NM, Williams SM, Sirugo G, Gumbo T. Amikacin concentrations predictive of ototoxicity in multidrug-resistant tuberculosis patients. Antimicrob Agents Chemother. 2015;59(10):6337–43. 27. Javadi MR, Abtahi B, Gholami K, Safari Moghadam B, Tabarsi P, Salamzadeh J. The incidence of amikacin ototoxicity in multidrug-ResistantTuberculosis patients. Iran J Pharm Res. 2011;10(4):905–11. 28. Sturdy A, Goodman A, José RJ, Loyse A, O'Donoghue M, Kon OM, et al. Multidrug-resistant tuberculosis (MDR-TB) treatment in the UK: a study of injectable use and toxicity in practice. J Antimicrob Chemother. 2011;66(8): 1815–20. 29. Sagwa EL, Ruswa N, Mavhunga F, Rennie T, Leufkens HG, Mantel-Teeuwisse AK. Comparing amikacin and kanamycin-induced hearing loss in multidrug- resistant tuberculosis treatment under programmatic conditions in a Namibian retrospective cohort. BMC Pharmacol Toxicol. 2015;16:36. 30. Jing W, Zongjie H, Denggang F, Na H, Bin Z, Aifen Z, et al. Mitochondrial mutations associated with aminoglycoside ototoxicity and hearing loss susceptibility identified by meta-analysis. J Med Genet. 2015;52(2):95–103. 31. Tseng AL, Dolovich L, Salit IE. Azithromycin-related ototoxicity in patients infected with human immunodeficiency virus. Clin Infect Dis. 1997;24(1):76–7. 32. Swanson DJ, Sung RJ, Fine MJ, Orloff JJ, Chu SY, Yu VL. Erythromycin ototoxicity: prospective assessment with serum concentrations and audiograms in a study of patients with pneumonia. Am J Med. 1992;92(1):61–8. 33. Ariano RE, Zelenitsky SA, Kassum DA. Aminoglycoside-induced vestibular injury: maintaining a sense of balance. Ann Pharmacother. 2008;42(9):1282–9.