Access the full text.
Sign up today, get DeepDyve free for 14 days.
Loading next page...
/lp/springer-journals/comparison-of-colistin-induced-nephrotoxicity-between-two-different-qKnVLJlt5w
- Publisher
- Springer Journals
- Copyright
- Copyright © The Author(s) 2021
- eISSN
- 2047-2994
- DOI
- 10.1186/s13756-021-00977-w
- Publisher site
- See Article on Publisher Site
Abstract
Background: Colistin is widely used in the treatment of nosocomial infections caused by carbapenem‑resistant gram‑negative bacilli (CR‑ GNB). Colistin‑induced nephrotoxicity is one of the major adverse reactions during colis‑ tin treatment. Comparisons of colistin‑induced nephrotoxicity between different formulations of colistin are rarely reported. Methods: In this retrospective cohort study, we enrolled intensive care unit–admitted patients if they had culture isolates of CR‑ GNB and underwent intravenous treatment with colistin. The occurrence of acute kidney injury (AKI) during intravenous treatment with colistin was recorded. The occurrence of colistin‑induced nephrotoxicity was ® ® compared between two formulations of colistin, Locolin , and Colimycin . Treatment outcomes associated with the occurrence of colistin‑induced nephrotoxicity were also investigated. ® ® Results: Among 195 patients, 95 who were treated with Locolin and 100 who were treated with Colimycin were included for analysis. Patients treated with Locolin had a higher rate of occurrence of stage 2 (46.3% vs. 32%, p = 0.040) and stage 3 (29.5% vs. 13%, p = 0.005) AKI than did those treated with Colimycin . In multivariate analysis, the presence of septic shock (adjusted odds ratio [aOR] 2.17, 95% confidence interval [CI] 1.10–4.26) and inappropri‑ ate colistin dosage (aOR 2.52, 95% CI 1.00–6.33) were clinical factors associated with colistin‑induced nephrotoxicity. Treatment with Colimycin was an independent factor associated with a lower risk of colistin‑induced nephrotoxicity (aOR 0.37, 95% CI 0.18–0.77). The mortality rate was comparable between patients with and without colistin‑induced nephrotoxicity. Conclusions: The risk of colistin‑induced nephrotoxicity significantly varied in different formulations of colistin in critically ill patients. Colistin‑induced nephrotoxicity was not associated with increased mortality rate. Keywords: Colistin, Nephrotoxicity, Acute kidney injury, Formulation, Mortality Background The emergence of multidrug-resistant organisms (MDRO) in nosocomial infections is a growing threat to the global health care system. Carbapenem-resistant *Correspondence: fdwang@vghtpe.gov.tw School of Medicine, National Yang Ming Chiao Tung University, No. 155, gram-negative bacilli (CR-GNB), such as CR-Acinetobac- Sec. 2, Linong Street, Taipei, Taiwan ter baumannii complex (CRAB), CR-Enterobacterales Full list of author information is available at the end of the article © The Author(s) 2021. Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http:// creat iveco mmons. org/ licen ses/ by/4. 0/. The Creative Commons Public Domain Dedication waiver (http:// creat iveco mmons. org/ publi cdoma in/ zero/1. 0/) applies to the data made available in this article, unless otherwise stated in a credit line to the data. Feng et al. Antimicrob Resist Infect Control (2021) 10:111 Page 2 of 12 (CRE), and CR-Pseudomonas aeruginosa (CRPA), are their clinical specimens and if they received intrave- common MDRO in nosocomial infections and are asso- nous colistin for ≥ 48 h. In patients receiving more than ciated with high morbidity and mortality rates [1–3]. one treatment course, only the first treatment course Colistin is one of the key agents used in the treatment of was included for analysis. Patients were excluded if they CR-GNB-induced nosocomial infections, especially hos- were aged < 20 years old, had a history of end-stage renal pital-acquired pneumonia [4]. disease (ESRD), were under regular dialysis during ini- Colistin is a polypeptide antibiotic and acts specifically tiation of intravenous colistin, missing data of baseline against gram-negative bacteria [5]. Despite the increasing creatinine and at least three follow-up creatinine, and use of colistin in the treatment of nosocomial infection, was treated with different brands of colistin in one treat - the use of intravenous colistin is frequently limited by its ment course. The study was approved by the Institutional adverse reactions, including nephrotoxicity and neuro- Review Board of Taipei Veterans General Hospital, and toxicity [6–9]. It is proposed that the interaction between the need for informed consent was waived (IRB No.: colistin and phospholipids in the cell membrane can lead 2019-11-009AC). to high membrane permeability of tubular epithelial cells and acute tubular necrosis [10]. Colistin-induced nephro- Data collection toxicity is characterized by decreased creatinine clear- Data of demographic characteristics (age, sex, body mass ance, proteinuria, cylindruria, or oliguria, and it usually index, and smoking status) and underlying comorbidities occurs in the first 5–7 days of treatment [11–14]. If dis - (diabetes, malignancies, renal insufficiency, chronic liver continued early, then acute kidney injury (AKI) is mostly disease, and heart failure) were obtained from hospital alleviated within 10 days from discontinuation [13]. chart review. Infection sources were determined accord- Several clinical factors such as old age, underlying ing to the site of specimen collection. In patients with comorbidities, concomitant nephrotoxins, higher colistin multiple samples containing CR-GNB, the sample that dose, and presence of septic shock have been proposed was obtained closest to the date of prescription of colistin to be related to colistin-induced nephrotoxicity [15–18]. was used to define infection source. Disease severity was However, the risk of nephrotoxicity based on different evaluated by the Acute Physiology and Chronic Health brands of colistin has rarely been compared thus far. Evaluation (APACHE) II score on the day of ICU admis- Previous in vitro studies have demonstrated that colis- sion. The presence of respiratory failure and septic shock timethate from different brands of colistin may have dif - (defined per SEPSIS-2 criteria) on the day of specimen ferent structural profiles and generates different range collection was also recorded. of exposure to colistin [19]. These differences may lead to variations in toxicodynamic effects in humans, which Carbapenem and colistin resistance determination is an important issue in critically ill patients. In the pre- Bacteria were phenotypically identified using the Vitek 2 sent study, we hypothesized that the risk of nephrotoxic- system (bioMérieux, Marcy l’Etoile, France). The mini - ity could be different based on different formulations of mum inhibitory concentration (MIC) of carbapenems colistin. To test our hypothesis, we conducted a retro- was determined using the broth microdilution method, spective study wherein we enrolled critically ill patients according to the CLSI guidelines [20]. Carbapenem who were treated with two different formulations of resistance was defined as resistance to imipenem or intravenous colistin. We aimed to compare differences meropenem (imipenem or meropenem MIC ≥ 4 mg/L in the occurrence of AKI between patients who received for Enterobacterales and MIC ≥ 8 mg/L for Pseudomonas two different formulations of intravenous colistin. The aeruginosa and Acinetobacter spp.) [20]. Colistin MICs clinical factors associated with AKI and the effect of were determined by broth microdilution as recom- AKI on treatment outcomes of critically ill patients were mended by the joint CLSI-EUCAST Polymyxin Break- investigated. We also performed high-performance liquid points Working Group [21]. chromatography (HPLC) to determine the difference in the composition of the two colistin formulations. Intravenous colistin administration and concomitant nephrotoxins Methods All the enrolled patients received intravenous colis- Patients and settings tin for ≥ 48 h. Intravenous colistin was initiated within This was a retrospective study conducted in a refer - seven days of CR-GNB cultured from clinical speci- ral medical center in Taiwan. Between January 2016 mens. For patients with multiple treatment courses of and October 2018, patients admitted in the inten- intravenous colistin, only the first treatment course was sive care unit (ICU) were enrolled if CR-GNB, includ- recorded. Two formulations of colistin, namely, Loco- ® ® ing CRAB, CRE, and CRPA, isolates was cultured from lin (Gentle, Taiwan) and Colimycin (T.T.Y., Taiwan), F eng et al. Antimicrob Resist Infect Control (2021) 10:111 Page 3 of 12 were available during the study period. Both the formu- Statistical analysis lations were supplied as 66.8 mg of colistin base activity Statistical analyses were performed using SPSS, version per vial, which was considered equivalent to 2 million 20.0, software (SPSS, Inc., Chicago, IL, USA). Partici- ® ® IU of sodium colistin methane sulfonate. The recom - pants were categorized into the Locolin and Colimycin mended loading and maintenance dosing of intrave- groups and analyzed accordingly. Continuous variables nous colistin was based on previous suggestion [22], such as APACHE II and hospital stays between sub- and adjusted based on body weight and renal function groups were compared using the Mann–Whitney U test, (Additional file 1: Table 1). Estimated glomerular filtra - and categorical variables were compared using Pearson’s tion rate (eGFR) was estimated using the Chronic Kid- chi-square or Fisher’s exact tests, as appropriate. We used ney Disease Epidemiology Collaboration equation [23]. mean imputation for missing data. Maintenance dose of intravenous colistin above the The occurrences of KDIGO stage 2, stage 3 AKI, and recommended dose was determined as inappropriate newly onset dialysis were compared between patients ® ® colistin dosage. treated with Locolin and those treated with Colimycin . Concomitant nephrotoxins, including aminoglycoside, Kaplan–Meier curves were constructed to evaluate the vancomycin, and intravenous contrast agent, that were difference in the occurrence of AKI between subgroups administered within 28 days of colistin treatment were of patients. Cox regression analysis was performed to also recorded. identify independent variables associated with KDIGO 3 AKI. Treatment outcomes including mechanical ventila- tor days, ICU stays, hospital stays, and all-cause mortality were also compared between patients with and without Analysis of colistin methane sulfonate (CMS) by HPLC ® ® the occurrence of KDIGO stage 3 AKI. All tests were The composition of Locolin and Colimycin was two-tailed, and a p-value of < 0.05 was considered statisti- analyzed by HPLC. The CMS reference standard was cally significant. obtained from U.S. Pharmacopeia (Rockville, MD, USA). The 600/717 HPLC system (Waters Corporation, Milford, Massachusetts, USA) consisted of a quaternary pump with an offline vacuum degasser and an autosampler with Results an injection capacity of 100 μL. Chromatographic sepa- Patient characteristics ration of Colistin A and Colistin B was achieved on the During the study period, the medical records of 195 ICU- ® ® C18 column (LiChrospher 100, LiChroCART 250–4, admitted patients who were administered with intrave- 150 mm × 4.6 mm i.d.). nous colistin were analyzed. Among these patients, 95 ® ® received Locolin and 100 received Colimycin . A flow - chart of the number of cases and reasons for exclusion Outcome definition is shown in Fig. 1. The demographic characteristics and The primary outcome evaluated in the present study disease severities of the enrolled patients are summa- was a comparison of the difference in the risk of newly rized in Table 1. Mean age of the enrolled patients was developed AKI between the two formulations of colistin 73.8 ± 13.5 years old and 70.8% (138/195) were male. Dia- during intravenous colistin treatment. Serum creatinine betes is the most common comorbidity (57/195, 29.2%) levels were recorded at baseline (day 1 of intravenous and 10.8% (21/195) of the enrolled cases had renal insuf- colistin administration) and thereafter until the end of ficiency. The median APACHE II score of the patients colistin treatment or death. AKI was determined based at ICU admission was 24 (IQR 19–30). At the time of on the definition of KDIGO recommendation by creati - colistin administration, 89.7% (175/195) had respiratory nine criteria [24]. The occurrence of stage 1 (with 1.5- to failure and 31.3% (61/195) had septic shock. Approxi- 1.9-fold increase or ≥ 0.3 mg/dL increase in the serum mately 80% of the CR-GNB isolates were cultured from creatinine level), stage 2 (with 2- to 2.9-fold increase in respiratory specimens. The proportions of CRAB, CRE, the serum creatinine level than that at baseline), and and CRPA were 72.3%, 23.1%, and 4.6%, respectively. stage 3 (≥ threefold increase in the creatinine level than The median daily colistin maintenance dose was 8 MIU that at baseline, or ≥ 4.0 mg/dL) AKI, and newly onset (IQR 4–10), and the median treatment duration of colis- dialysis was recorded. Other clinical factors associated tin was 7 days (IQR 4–12). Twenty-three patients (11.8%) with the development of AKI were also investigated. received colistin at an inappropriate maintenance dos- We also compared clinical outcomes, including age. Comparatively, patients who received Locolin mechanical ventilator days, ICU stays, hospital stays, and had a higher serum albumin level (2.8 g/dL vs. 2.6 g/ all-cause mortality at day 28 and on discharge, between dL, p = 0.017) and were more likely to receive concomi- patients with and without AKI development. tant aminoglycoside treatment (15.8% vs. 7%, p = 0.052) Feng et al. Antimicrob Resist Infect Control (2021) 10:111 Page 4 of 12 Table 1 Demographic characteristics of ICU‑admitted patients treated with two different formulations of intravenous colistin Overall patients Intravenous colistin P-value ® ® Locolin Colimycin No. of patients 195 95 100 Age, years (SD) 73.8 (13.5) 75.2 (12.3) 72.4 (14.6) 0.142 Sex 0.069 Male 138 (70.8%) 73 (76.8%) 65 (65%) Female 57 (29.2%) 22 (23.2%) 35 (35%) BMI (SD) 22.3 (4.6) 21.7 (4.8) 22.8 (4.3) 0.106 Smoking 78 (40%) 40 (42.1%) 38 (38%) 0.559 Comorbidities Diabetes 57 (29.2%) 28 (29.5%) 29 (29%) 0.942 Malignancies 33 (16.9%) 24 (25.3%) 9 (9%) 0.002 Renal insufficiency (CCr < 30) 21 (10.8%) 10 (10.5%) 11 (11%) 0.915 Chronic liver diseases 16 (8.2%) 3 (3.2%) 13 (13%) 0.017 Heart failure 17 (8.7%) 8 (8.4%) 9 (9%) 0.886 CR-GNB culture sources Respiratory specimens 157 (80.5%) 75 (78.9%) 82 (82%) 0.591 Urine 11 (5.6%) 6 (6.3%) 5 (5%) 0.691 Blood 10 (5.1%) 4 (4.2%) 6 (6%) 0.571 Others 17 (8.7%) 10 (10.5%) 7 (7%) 0.383 CR-GNB species CRAB 141 (72.3%) 64 (67.4%) 77 (77%) 0.270 CRE 45 (23.1%) 28(29.5%) 17 (17%) 0.051 CRPA 9 (4.6%) 3 (3.2%) 6 (6%) 0.344 Laboratory results (Mean, SD) Leukocytes (× 10 per L) 13.2 (9.0) 13.2 (8.3) 13.3 (9.7) 0.935 CRP (mg/dL) 10.0 (6.5) 10.3 (7.6) 9.6 (5.9) 0.464 Albumin level (median, IQR) 2.7 (2.4–3.0) 2.8 (2.5–3.1) 2.6 (2.3–2.9) 0.017 Concomitant nephrotoxins Aminoglycoside 22 (11.3%) 15 (15.8%) 7 (7%) 0.052 Vancomycin 20 (10.3%) 13 (13.7%) 7 (7%) 0.124 Intravenous contrast agent 39 (20.0%) 17 (17.9%) 22 (22%) 0.474 Disease severity APACHEII scores (median, IQR) 24 (19–30) 25 (20–30) 24 (19–30) 0.768 Respiratory failure 175 (89.7%) 83 (87.4%) 92 (92%) 0.287 Septic shock 61 (31.3%) 33 (34.7%) 28 (28%) 0.310 Intravenous colistin Daily dosage (MIU) (median, IQR) 8 (4–10) 8 (4–10) 8 (4–10) 0.814 Treatment duration (median, IQR) 7 (4–12) 7 (4–12) 7 (4–12) 0.782 Accumulated dosage (MIU) (median, IQR) 48 (28–80) 48 (28–64) 48 (24–88) 0.956 Inappropriate colistin dosage 23 (11.8%) 7 (7.4%) 16 (16%) 0.062 APACHE II Acute Physiology and Chronic Health Evaluation II, BMI body mass index, ICU intensive care unit, IQR interquartile range, CCr creatinine clearance, CR-GNB carbapenem-resistant gram-negative bacteria, SD standard deviation Data are presented as n (%) Including abscess, ascites, CSF, and pericardial effusion Evaluated on the day of ICU admission Present on the day of sample collection F eng et al. Antimicrob Resist Infect Control (2021) 10:111 Page 5 of 12 Fig. 1 Study flow diagram and reasons for patient exclusions. ESRD, end stage renal diseases; AKI, acute kidney injury and less likely to have colistin at an inappropriate dos- administration are shown in Fig. 2. Overall, the rate of age (7.4% vs. 16%, p = 0.062) than those who received occurrence of KDIGO stage 1, stage 2, stage 3 AKI, and Colimycin . Only few patients had concomitant non- newly initiated dialysis was 49.7%, 39%, 21%, and 9.2%, steroid anti-inflammatory drugs (n = 6), and no patient respectively. Comparatively, patients who received had concomitant amphotericin B (data not shown). Colimycin had a significantly lower rate of occurrence Otherwise, the two groups of patients had similar demo- of KDIGO stage 2 AKI (32% vs. 46.3%, p = 0.040) and graphic characteristics, underlying comorbidities, and KDIGO stage 3 AKI (13% vs. 29.5%, p = 0.005) than those disease severities. There were no significant differences who received Locolin . The rate of occurrence of newly in the dosage and duration of colistin treatment between initiated dialysis was comparable between the two groups the two groups of patients. of patients. Curves of Kaplan–Meier analysis of the occurrence Rate of occurrence of AKI of KDIGO stage 2 and stage 3 AKI in both groups of The rates of occurrence of various severities of AKI patients are shown in Fig. 3. Patients who received and newly initiated dialysis within 28 days after colistin Colimycin had a significantly lower rate of occurrence Feng et al. Antimicrob Resist Infect Control (2021) 10:111 Page 6 of 12 Fig. 2 Rate of occurrence of acute kidney injury in ICU‑admitted patients treated with intravenous colistin. KDIGO stage 1, stage 2, stage 3 AKI, and newly initiated dialysis are presented ® ® Fig. 3 Kaplan–Meier analysis of a stage 2 AKI and b stage 3 AKI between patients treated with Locolin and those treated with Colimycin . AKI, acute kidney injury of KDIGO 3 AKI than those who received Locolin (log recovery rate and AKI duration between the colistin rank p = 0.008). The curves separated early after the groups showed no significant differences. onset of colistin treatment. Among the patients with KDIGO stage 3 AKI, the Clinical factors associated with AKI complete or partial recovery rates was 53.6% (15/28) in Among the total enrolled patients, 41 (2%) had KDIGO ® ® the Locolin group and 53.8% (7/13) in the Colimycin stage 3 AKI. Comparisons of clinical factors between group. The median duration of AKI among patients patients with and without KDIGO stage 3 AKI are ® ® with Locolin and Colimycin was 9 (IQR 7–13) shown in Table 2. Patients with KDIGO stage 3 AKI were days and 10 (IQR 8–12) days, respectively. Both AKI more likely to have CR-GNB cultured from respiratory F eng et al. Antimicrob Resist Infect Control (2021) 10:111 Page 7 of 12 Table 2 Clinical characteristics of ICU patients with and without the occurrence of KDIGO 3 acute kidney injur y ≥ KDIGO 3 AKI P-value Yes, n = 41 No, n = 154 Age, years (SD) 71.4 (12.2) 74.4 (13.8) 0.138 Sex 0.421 Male 26 (63.4%) 112 (72.7%) Female 15 (36.6%) 42 (27.3%) BMI (SD) 22.2 (5.6) 22.3 (4.3) 0.788 Smoking 15 (36.6%) 63 (40.9%) 0.363 Comorbidities Diabetes 12 (29.5%) 45 (29.2%) 0.958 Malignancies 10 (24.4%) 23 (14.9%) 0.243 Renal insufficiency 6 (14.6%) 15 (9.7%) 0.211 Chronic liver diseases 3 (7.3%) 13 (8.4%) 1.000 Heart failure 3 (7.8%) 14 (9.1%) 0.767 CR-GNB culture sources 0.035 Respiratory specimens 39 (95.1%) 118 (76.6%) Urine 0 10 (6.5%) Blood 2 (4.9%) 9 (5.8%) Others 0 17 (11.0%) CR-GNB species 0.229 CRAB 34 (82.9) 107 (69.5%) CRE 6 (14.6%) 39 (25.3%) CRPA 1 (2.4%) 8 (5.2%) Laboratory results (Mean, SD) Leukocytes (× 10 per L) 11.5 (9.9) 13.7 (8.8) 0.183 CRP (mg/dL) 10.5 (10.0) 9.8 (5.6) 0.555 Albumin level 2.8 (2.5–3.1) 2.7 (2.4–3.0) 0.217 Concomitant nephrotoxins Aminoglycoside 5 (12.2%) 17 (11.0%) 0.835 Vancomycin 8 (19.5%) 12 (7.8%) 0.028 Intravenous contrast agent 9 (22%) 30 (19.5%) 0.725 Disease severities APACHEII scores (median, IQ R) 23 (18–31) 25 (20–30) 0.454 Respiratory failure 40 (97.6%) 135 (87.7%) 0.081 Septic shock 22 (53.7) 39 (25.3%) 0.001 Colistin treatment Daily dose (MIU) (median, IQR) 8 (7.4–10) 8 (4–10) 0.036 Treatment duration (days) (median, IQR) 7 (4–11) 7 (4–12) 0.403 Accumulated dose (MIU) (median, IQR) 46.5 (30.5–87.2) 48 (24–80) 0.594 Inappropriate colistin dose 9 (22.0%) 14 (9.1%) 0.023 APACHE II Acute Physiology and Chronic Health Evaluation II, BMI body mass index, ICU intensive care unit, IQR interquartile range, CCr creatinine clearance, CR-GNB carbapenem-resistant gram-negative bacteria, SD standard deviation Data are presented as n (%) Including abscess, ascites, CSF, and pericardial effusion Evaluated on the day of ICU admission Present on the day of sample collection specimens (95.1% vs. 76.6%, p = 0.035), concomitant p = 0.036), and inappropriate colistin maintenance dos- vancomycin treatment (19.5% vs. 7.8%, p = 0.028), septic age (22% vs. 9.1%, p = 0.023). shock (53.7% vs. 25.3%, p = 0.001), higher daily colistin Results of univariate and multivariate analyses of dose (8 MIU, IQR 7.4–10 MIU vs. 8 MIU, IQR 4–10 MIU, the clinical factors associated with the occurrence of Feng et al. Antimicrob Resist Infect Control (2021) 10:111 Page 8 of 12 Table 3 Univariate and multivariate Cox‑regression analysis of clinical factors associated with the occurrence of KDIGO 3 acute kidney injury Univariate analysis Multivariate analysis OR (95% CI) P-value aOR (95% CI) P-value Age 0.99 (0.97–1.01) 0.236 0.99 (0.96–1.02) 0.463 Sex Female 1.00 – 1.00 – Male 0.72 (0.38–1.36) 0.312 0.70 (0.34–1.46) 0.346 Baseline eGFR < 30 1.65 (0.69–3.91) 0.260 1.41 (0.48–4.14) 0.536 Diabetes 0.93 (0.47–1.82) 0.824 1.16 (0.56–2.41) 0.681 APACHII ≥ 25 0.64 (0.34–1.20) 0.162 0.53 (0.24–1.17) 0.118 Albumin ≤ 3 g/dL 0.81 (0.42–1.54) 0.522 0.66 (0.34–1.28) 0.221 CRAB 0.61 (0.29–1.29) 0.197 0.53 (0.24–1.17) 0.118 Aminoglycoside 1.01 (0.40–2.58) 0.982 0.99 (0.34–2.86) 0.979 Vancomycin 1.99 (0.92–4.3) 0.082 1.59 (0.65–3.89) 0.310 Intravenous contrast 1.07 (0.51–2.23) 0.865 0.89 (0.37–2.12) 0.785 Septic shock 2.79 (1.51–5.16) 0.001 2.17 (1.10–4.26) 0.025 Daily colistin > 10 MIU 1.75 (0.94–3.26) 0.077 1.32 (0.61–2.83) 0.482 Colistin treatment duration 0.98 (0.92–1.03) 0.411 0.99 (0.93–1.04) 0.617 Inappropriate colistin dosage 2.29 (1.09–4.80) 0.028 2.52 (1.00–6.33) 0.049 Colistin formulation Locolin 1.00 – 1.00 – Colimycin 0.42 (0.22–0.82) 0.011 0.37 (0.18–0.77) 0.008 Adjusted odds ratio (aOR) and 95%CI were derived from cox regression analysis Present on the day of sample collection KDIGO stage 3 AKI are shown in Table 3. In multivari- Eec ff t of AKI on treatment outcomes ate analysis, the independent factors associated with We further explored the impact of KDIGO stage 3 AKI KDIGO stage 3 AKI included septic shock (adjusted on treatment outcomes. As shown in Table 4, patients odds ratio [aOR] 2.17, 95% confidence interval [CI] with KDIGO stage 3 AKI had longer ventilator using days 1.10–4.26) and inappropriate colistin dosage (aOR (36 days, IQR 17–52 days vs. 22 days, IQR 13–42 days, 2.52, 95% CI 1.00–6.33). By contrast, when compared p = 0.036). Otherwise, patients with and without KDIGO ® ® to Locolin , Colimycin treatment was an independent stage 3 AKI had comparable hospital stays, ICU stays, factor associated with a lower risk of KDIGO stage 3 and all-cause mortality. Comparisons of treatment out- AKI (aOR 0.37, 95% CI 0.18–0.77). comes between patients receiving Locolin and those receiving Colimycin were also analyzed. As shown in Table 4 Treatment outcomes of patients with and without KDIGO 3 acute kidney injury, and with two different formulations of intravenous colistin KDIGO 3 AKI P-value Intravenous colistin ® ® Yes, n = 41 No, n = 154 Locolin , n = 95 Colimycin , n = 100 Ventilator days (Median, IQR) 36 (17–52) 22 (13–42) 0.036 28 (15–48) 24 (13–47) 0.048 ICU stay (days) (Median, IQR) 31 (19–47) 30 (20–51) 0.889 30 (19–47) 31 (21–54) 0.285 Hospital stay (days) (Median, IQR) 60 (38–94) 53 (31–80) 0.133 56 (35–87) 50 (29–76) 0.075 Mortality Day 28 14 (34.1%) 62 (40.3%) 0.476 34 (35.8%) 42 (42%) 0.374 Discharge 28 (68.3%) 91 (59.1%) 0.283 56 (58.9%) 63 (63%) 0.562 AKI acute kidney injury, ICU intensive care unit, IQR interquartile range F eng et al. Antimicrob Resist Infect Control (2021) 10:111 Page 9 of 12 Table 4, patients who were treated with Colimycin were [19]. However, the differences in nephrotoxicity between associated with shorter ventilator using days (28 days, various formulations of colistin remain unknown. IQR 15–48 days vs. 24 days, IQR 13–47 days, p = 0.048). Colistin is a multicomponent lipopeptide that contains ® ® Composition between Locolin and Colimycin CMS A and CMS B, which differ in the fatty acid chain ® ® The chromatographic profiles of Locolin , Colimycin , attached to the cyclic decapeptide moiety of the drug and reference standard are shown in Fig. 4. The propor - [29]. The proportion of CMS A and CMS B can have a ® ® tions of CMS A in Locolin , Colimycin , and reference large difference in commercial preparations of colistin standard were 15.2%, 10%, and 9.4%, respectively. [30]. Although there are comparable bactericidal effects between CMS A and CMS B, CMS A has been reported Discussion to have a higher nephrotoxic effect than CMS B in an ani - This retrospective study enrolled critically ill patients mal model study [29]. The different compositions of CMS who underwent intravenous colistin treatment for CR- A and CMS B in various formulations of colistin might GNB and evaluated the occurrence of AKI. During the lead to different risks of colistin-induced nephrotoxicity. colistin treatment period, KDIGO stage 1, stage 2, and In the present study, we demonstrated a significant dif - stage 3 AKI occurred in 49.7%, 39%, and 21% of patients, ference in the rate of occurrence of AKI between two dif- respectively. Meanwhile, 9.2% of the patients had newly ferent formulations of colistin. Furthermore, we reported initiated dialysis. Comparatively, patients who received that the proportion of CMS A in Locolin is 50% higher ® ® Colimycin had a lower rate of occurrence of KDIGO than that in Colimycin , which is in line with our clini- stage 2 and stage 3 AKI than those who received Loco- cal observation of an increased risk of nephrotoxicity ® ® lin . In multivariate analysis, we found that independ- in Locolin . To the best of our knowledge, this is the ent factors associated with KDIGO stage 3 AKI included first study to evaluate the occurrence of nephrotoxicity the presence of septic shock and inappropriate colistin between different formulations of colistin. Although the dosage. By contrast, Colimycin use was an independ- exact mechanisms remain uncertain, we speculate that ent factor associated with a lower rate of occurrence of the difference in the composition of CMS A and CMS KDIGO stage 3 AKI. We also found that the occurrence B in various colistin products could play a pivotal role. of KDIGO stage 3 AKI during colistin treatment was Clinicians should therefore be aware of the possible dif- associated with longer mechanical ventilator using days ference in the risks of nephrotoxicity among various for- but not related to higher all-cause mortality. mulations of colistin. Further studies are also warranted Nephrotoxicity and neurotoxicity are well-documented to verify our findings. adverse reactions associated with intravenous treatment We further evaluated the impact of the occurrence of with colistin. Nephrotoxicity in colistin is dose-depend- colistin-related AKI on treatment outcomes, which have ent and usually reversible [16, 18, 25, 26]. The nephro - rarely been evaluated thus far. We found that patients toxicity of colistin is mainly related to its D-aminobutyric with colistin-induced AKI may have prolonged depend- acid and fatty acid component. Apart from its bacteri- ence on mechanical ventilator; there were no differences cidal effects, colistin increases the membrane perme - in mortality and hospital stays between patients with and ability of tubular epithelial cells, which, in turn, leads to without colistin-related AKI. Our findings were consist - cell swelling and lysis [10]. Concerning toxicity, colistin ent with those of a previous study, which prospectively is administered as an inactive prodrug, CMS, which is enrolled patients infected by extensively drug-resistant the only parenteral form used clinically for colistin. CMS Acinetobacter baumannii and treated by colistin [31]. is off patent for many years, and several commercially However, ventilator dependence and hospital stays were available parenteral products of CMS are available on the noted in that study. A study on patients infected by drug- market. However, the number and location of methane resistant Pseudomonas aeruginosa reported the presence sulfonate groups attached on CMS vary widely in differ - of AKI as an independent factor associated with a high ent colistin products, which may, in turn, lead to differ - mortality rate [32]. Another study reported that patients ences in pharmacokinetics and pharmacodynamics of who experienced AKI had a higher mortality rate if kid- colistin in humans [27, 28]. Limited studies have been ney function failed to return to the baseline level [6]. performed to evaluate the differences between various Although the findings remain controversial, we believe brands of colistin. Li et al. reported a 20% difference in that close monitoring of renal function during colistin the colistin level in plasma between different formula - treatment and early discontinuation of colistin in patients tions of colistin [27]. He et al. reported distinct chroma- with AKI are the best ways to reduce the effect of AKI on tographic profiles of different colistin products [19]. They treatment outcomes in these critically ill patients. also demonstrated significantly different exposure to There are some limitations to this study. First, colistin between various brands of colistin in a rat model as this was a retrospective study, the demographic Feng et al. Antimicrob Resist Infect Control (2021) 10:111 Page 10 of 12 ® ® Fig. 4 HPLC analysis for a CMS reference standard of CMS, b Locolin , and c Colimycin . The compositions of CMS A and CMS B are presented separately. HPLC high‑performance liquid chromatography, CMS colistin methanesulfonate F eng et al. Antimicrob Resist Infect Control (2021) 10:111 Page 11 of 12 complex; CRE: Carbapenem‑resistant Enterobacterales; CRPA: Carbapenem‑ characteristics and disease severities were not equal resistant Pseudomonas aeruginosa; AKI: Acute kidney injury; ICU: Intensive between patients treated with Colimycin and those care units; APACHE: Acute Physiology and Chronic Health Evaluation; eGFR: treated with Locolin . Although we had performed mul- Estimated glomerular filtration rate; NAC: N‑acetylcysteine. tivariate analysis to adjust for the effects from clinical factors, our findings should be interpreted with caution. Supplementary Information Second, all the enrolled patients had CR-GNB isolated The online version contains supplementary material available at https:// doi. org/ 10. 1186/ s13756‑ 021‑ 00977‑w. from clinical specimens, and some of the patients may have colonization rather than true infection. However, Additional file 1. Recommended loading dose and daily maintenance the leukocyte count and the C-reactive protein level in doses of colistimethate. the enrolled patients were much above the upper normal limits, which indicated that most of the enrolled patients Acknowledgements had infection rather than colonization. Meanwhile, This work was supported, in part, by Taipei Veterans General Hospital ( V109E‑ the effect of colonization on our analysis was limited 004‑05, V108C‑020). because this study aimed to investigate colistin-induced Authors’ contributions nephrotoxicity, rather than treatment effectiveness. JYF, Y TL, and FDW conceptualize the study design. JYF and Y TL wrote the Third, exposure to concomitant nephrotoxins, including manuscript draft with input from all authors. JYF performed the data analysis. vancomycin, aminoglycosides, and contrast agent, was JYF, Y TL, SWP, KYY, YMC, DHTY, SYL, and FDW collected the data. All authors read and approved the final manuscript. not rare in our patients. Therefore, the risk of colistin- induced nephrotoxicity could be overestimated. Fourth, Funding we excluded patients with ESRD but included those with None. renal insufficiency; which may affect the development Availability of data and materials of AKI during colistin treatment. However, the propor- The datasets used and/or analyzed during the current study are available from tion of patients with renal insufficiency was comparable the corresponding author on reasonable request. between the two colistin groups, and we included renal insufficiency in our multivariate analysis. Finally, we Declarations enrolled critically ill patients who had ICU admission and Ethics approval and consent to participate high APACHE II scores. Most of them had respiratory The study was approved by local institutional review board. (IRB No.: failure, and nearly one-third of them received inotropic 2019‑11‑009AC). agents. Therefore, the findings obtained in our study may Consent for publications not be applicable to patients with low disease severities. Not applicable. Competing interests Conclusions The authors declare that they have no competing interests. This retrospective study involved critically ill patients Author details who were treated with intravenous colistin. We found Department of Chest Medicine, Taipei Veterans General Hospital, No. 201, significant differences in the rate of occurrence of colis - 2 Sec. 2, Shipai Road, Taipei, Taiwan. School of Medicine, National Yang Ming tin-induced nephrotoxicity between two formulations of Chiao Tung University, No. 155, Sec. 2, Linong Street, Taipei, Taiwan. I nstitute of Emergency and Critical Care Medicine, National Yang Ming Chiao Tung colistin. We also demonstrated different compositions University, No. 155, Sec. 2, Linong Street, Taipei, Taiwan. Department of Emer‑ in the two formulations of colistin. Other clinical factors gency Medicine, Taipei Veterans General Hospital, No. 201, Sec. 2, Shipai Road, associated with colistin-induced nephrotoxicity included Taipei, Taiwan. Institute of Public Health, National Yang Ming Chiao Tung University, No. 155, Sec. 2, Linong Street, Taipei, Taiwan. Division of Nephrol‑ septic shock and inappropriate colistin maintenance dos- ogy, Department of Medicine, Taipei Veterans General Hospital, No. 201, Sec. age. Our findings suggest that the risk of nephrotoxicity 7 2, Shipai Road, Taipei, Taiwan. Division of Infectious Diseases, Department in colistin could be different in various formulations of of Medicine, Taipei Veterans General Hospital, No. 201, Sec. 2, Shipai Road, Taipei 11217, Taiwan. colistin. The association between the risk of nephrotox - icity and the differences in the compositions of various Received: 10 February 2021 Accepted: 8 July 2021 colistin formulations deserves further studies for clarifi - cation. Meanwhile, close monitoring of renal function in high-risk populations and appropriate dosage adjustment during colistin treatment is crucial to decrease the risk of References 1. Watkins RR, Van Duin D. Current trends in the treatment of pneumonia colistin-induced nephrotoxicity in critically ill patients. due to multidrug‑resistant Gram‑negative bacteria. F1000Res 2019, 8. https:// doi. org/ 10. 12688/ f1000 resea rch. 16517.2. 2. Boyd DA, Mataseje LF, Pelude L, Mitchell R, Bryce E, Roscoe D et al. Results Abbreviations from the Canadian Nosocomial Infection Surveillance Program for MDRO: Multidrug‑resistant organisms; CR‑ GNB: Carbapenem‑resistant detection of carbapenemase‑producing Acinetobacter spp. in Canadian gram‑negative bacilli; CRAB: Carbapenem‑resistant ‑Acinetobacter baumannii Feng et al. Antimicrob Resist Infect Control (2021) 10:111 Page 12 of 12 hospitals, 2010–16. J Antimicrob Chemother 2019, 74:315–320. https:// 20. Weistein MP LJ, Bobenchik AM, Campeau S, Cullen SK, Fallas MF, Gold doi. org/ 10. 1093/ jac/ dky416. https:// doi. org/ 10. 1007/ s10096‑ 016‑ 2710‑0. H, Humphries RM, Kirn TJ. Clinical and Laboratory Standards Institute. 3. Wang Q, Zhang Y, Yao X, Xian H, Liu Y, Li H, et al. Risk factors and clinical Performance standards for antimicrobial susceptibility testing: twenty‑ outcomes for carbapenem‑resistant Enterobacteriaceae nosocomial ninth informational supplement. 30th edition. Clinical and Laboratory infections. Eur J Clin Microbiol Infect Dis. 2016;35:1679–89. Standards Institute Available at: https:// www. clsio rg/ media/ 3481/ m100e 4. Kalil AC, Metersky ML, Klompas M, Muscedere J, Sweeney DA, Palmer d30_ sampl epdf 2020. LB, et al. Management of adults with hospital‑acquired and ventilator ‑ 21. Testing TECoAS. Recommendations for MIC determination of colistin associated pneumonia: 2016 clinical practice guidelines by the Infectious (polymyxin E) As recommended by the joint CLSI‑EUCAST Polymyxin Diseases Society of America and the American Thoracic Society. Clin Breakpoints Working Group. Available at: https:// www. eucas torg/ filea Infect Dis. 2016;63:e61–111. https:// doi. org/ 10. 1093/ cid/ ciw353.dmin/ src/ media/ PDFs/ EUCAST_ files/ Gener al_ docum ents/ Recom menda 5. Trimble MJ, Mlynarcik P, Kolar M, Hancock RE. Polymyxin: alternative tions_ for_ MIC_ deter minat ion_ of_ colis tin_ March_ 2016p df 2016. mechanisms of action and resistance. Cold Spring Harb Perspect Med. 22. Garonzik SM, Li J, Thamlikitkul V, Paterson DL, Shoham S, Jacob J, et al. 2016. https:// doi. org/ 10. 1101/ cshpe rspect. a0252 88. Population pharmacokinetics of colistin methanesulfonate and formed 6. Miano TA, Lautenbach E, Wilson FP, Guo W, Borovskiy Y, Hennessy S. colistin in critically ill patients from a multicenter study provide dosing Attributable risk and time course of colistin‑associated acute kidney suggestions for various categories of patients. Antimicrob Agents Chem‑ injury. Clin J Am Soc Nephrol. 2018;13:542–50. https:// doi. org/ 10. 2215/ other. 2011;55:3284–94. https:// doi. org/ 10. 1128/ AAC. 01733‑ 10. CJN. 06980 717. 23. Levey AS, Stevens LA, Schmid CH, Zhang YL, Castro AF 3rd, Feldman HI, 7. Oliota AF, Penteado ST, Tonin FS, Fernandez‑Llimos F, Sanches AC. et al. A new equation to estimate glomerular filtration rate. Ann Intern Nephrotoxicity prevalence in patients treated with polymyxins: a system‑ Med. 2009;150:604–12. https:// doi. org/ 10. 7326/ 0003‑ 4819‑ 150‑9‑ 20090 atic review with meta‑analysis of observational studies. Diagn Microbiol 5050‑ 00006. Infect Dis. 2019;94:41–9. https:// doi. org/ 10. 1016/j. diagm icrob io. 2018. 11. 24. Kellum JA, Lameire N, Group KAGW. Diagnosis, evaluation, and man‑ 008. agement of acute kidney injury: a KDIGO summary (Part 1). Crit Care. 8. Honore PM, Jacobs R, Lochy S, De Waele E, Van Gorp V, De Regt J, et al. 2013;17:204. https:// doi. org/ 10. 1186/ cc114 54. Acute respiratory muscle weakness and apnea in a critically ill patient 25. Shields RK, Anand R, Clarke LG, Paronish JA, Weirich M, Perone H, et al. induced by colistin neurotoxicity: key potential role of hemoadsorption Defining the incidence and risk factors of colistin‑induced acute kidney elimination during continuous venovenous hemofiltration. Int J Nephrol injury by KDIGO criteria. PLoS ONE. 2017;12: e0173286. https:// doi. org/ 10. Renovasc Dis. 2013;6:107–11. https:// doi. org/ 10. 2147/ IJNRD. S42791.1371/ journ al. pone. 01732 86. 9. Nigam A, Kumari A, Jain R, Batra S. Colistin neurotoxicity: revisited. BMJ 26. Paul M, Bishara J, Levcovich A, Chowers M, Goldberg E, Singer P, et al. Case Rep. 2015. https:// doi. org/ 10. 1136/ bcr‑ 2015‑ 210787. Eec ff tiveness and safety of colistin: prospective comparative cohort 10. Ghlissi Z, Hakim A, Mnif H, Ayadi FM, Zeghal K, Rebai T, et al. Evaluation of study. J Antimicrob Chemother. 2010;65:1019–27. https:// doi. org/ 10. colistin nephrotoxicity administered at different doses in the rat model. 1093/ jac/ dkq069. Ren Fail. 2013;35:1130–5. https:// doi. org/ 10. 3109/ 08860 22X. 2013. 815091. 27. Li J, Milne RW, Nation RL, Turnidge JD, Coulthard K. Stability of colistin and 11. Florescu DF, Qiu F, McCartan MA, Mindru C, Fey PD, Kalil AC. What is the colistin methanesulfonate in aqueous media and plasma as determined efficacy and safety of colistin for the treatment of ventilator ‑associated by high‑performance liquid chromatography. Antimicrob Agents Chem‑ pneumonia? A systematic review and meta‑regression. Clin Infect Dis. other. 2003;47:1364–70. https:// doi. org/ 10. 1128/ aac. 47.4. 1364‑ 1370. 2003. 2012;54:670–80. https:// doi. org/ 10. 1093/ cid/ cir934. 28. Wallace SJ, Li J, Rayner CR, Coulthard K, Nation RL. Stability of colistin 12. Hartzell JD, Neff R, Ake J, Howard R, Olson S, Paolino K, et al. Nephrotoxic‑ methanesulfonate in pharmaceutical products and solutions for admin‑ ity associated with intravenous colistin (colistimethate sodium) treatment istration to patients. Antimicrob Agents Chemother. 2008;52:3047–51. at a tertiary care medical center. Clin Infect Dis. 2009;48:1724–8. https:// https:// doi. org/ 10. 1128/ AAC. 00103‑ 08. doi. org/ 10. 1086/ 599225. 29. Roberts KD, Azad MA, Wang J, Horne AS, Thompson PE, Nation RL, et al. 13. Dalfino L, Puntillo F, Mosca A, Monno R, Spada ML, Coppolecchia S, Antimicrobial activity and toxicity of the major lipopeptide components et al. High‑ dose, extended‑interval colistin administration in critically ill of polymyxin b and colistin: last‑line antibiotics against multidrug‑resist ‑ patients: is this the right dosing strategy? A preliminary study. Clin Infect ant gram‑negative bacteria. ACS Infect Dis. 2015;1:568–75. https:// doi. Dis. 2012;54:1720–6. https:// doi. org/ 10. 1093/ cid/ cis286.org/ 10. 1021/ acsin fecdis. 5b000 85. 14. Deryke CA, Crawford AJ, Uddin N, Wallace MR. Colistin dosing and 30. Brink AJ, Richards GA, Colombo G, Bortolotti F, Colombo P, Jehl F. Multi‑ nephrotoxicity in a large community teaching hospital. Antimicrob component antibiotic substances produced by fermentation: implica‑ Agents Chemother. 2010;54:4503–5. https:// doi. org/ 10. 1128/ AAC. tions for regulatory authorities, critically ill patients and generics. Int J 01707‑ 09. Antimicrob Agents. 2014;43:1–6. https:// doi. org/ 10. 1016/j. ijant imicag. 15. Lee YJ, Wi YM, Kwon YJ, Kim SR, Chang SH, Cho S. Association between 2013. 06. 013. colistin dose and development of nephrotoxicity. Crit Care Med. 31. Durante‑Mangoni E, Andini R, Signoriello S, Cavezza G, Murino P, Buono 2015;43:1187–93. https:// doi. org/ 10. 1097/ CCM. 00000 00000 000931. S, et al. Acute kidney injury during colistin therapy: a prospective study in 16. Kwon KH, Oh JY, Yoon YS, Jeong YJ, Kim KS, Shin SJ, et al. Colistin treat‑ patients with extensively‑ drug resistant Acinetobacter baumannii infec‑ ment in carbapenem‑resistant Acinetobacter baumannii pneumonia tions. Clin Microbiol Infect. 2016;22:984–9. https:// doi. org/ 10. 1016/j. cmi. patients: incidence of nephrotoxicity and outcomes. Int J Antimicrob 2016. 08. 004. Agents. 2015;45:605–9. https:// doi. org/ 10. 1016/j. ijant imicag. 2015. 01. 011. 32. Sorli L, Luque S, Segura C, Campillo N, Montero M, Esteve E, et al. Impact 17. Koksal I, Kaya S, Gencalioglu E, Yilmaz G. Evaluation of risk factors for intra‑ of colistin plasma levels on the clinical outcome of patients with infec‑ venous colistin use‑related nephrotoxicity. Oman Med J. 2016;31:318–21. tions caused by extremely drug‑resistant Pseudomonas aeruginosa. BMC https:// doi. org/ 10. 5001/ omj. 2016. 62. Infect Dis. 2017;17:11. https:// doi. org/ 10. 1186/ s12879‑ 016‑ 2117‑7. 18. Pogue JM, Lee J, Marchaim D, Yee V, Zhao JJ, Chopra T, et al. Incidence of and risk factors for colistin‑associated nephrotoxicity in a large academic Publisher’s Note health system. Clin Infect Dis. 2011;53:879–84. https:// doi. org/ 10. 1093/ Springer Nature remains neutral with regard to jurisdictional claims in pub‑ cid/ cir611. lished maps and institutional affiliations. 19. He H, Li JC, Nation RL, Jacob J, Chen G, Lee HJ, et al. Pharmacokinetics of four different brands of colistimethate and formed colistin in rats. J Anti‑ microb Chemother. 2013;68:2311–7. https:// doi. org/ 10. 1093/ jac/ dkt207.
Journal
Antimicrobial Resistance & Infection Control – Springer Journals
Published: Jul 30, 2021
Keywords: Colistin; Nephrotoxicity; Acute kidney injury; Formulation; Mortality