Next Article in Journal
Synthesis of Degraded Limonoid Analogs as New Antibacterial Scaffolds against Staphylococcus aureus
Next Article in Special Issue
Clinical and Economic Burden of Carbapenem-Resistant Infection or Colonization Caused by Klebsiella pneumoniae, Pseudomonas aeruginosa, Acinetobacter baumannii: A Multicenter Study in China
Previous Article in Journal
Update on Acute Bone and Joint Infections in Paediatrics: A Narrative Review on the Most Recent Evidence-Based Recommendations and Appropriate Antinfective Therapy
Previous Article in Special Issue
Antimicrobial Resistance Patterns in Clostridioides difficile Strains Isolated from Neonates in Germany
 
 
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

Impact of Colistin Dosing on the Incidence of Nephrotoxicity in a Tertiary Care Hospital in Saudi Arabia

1
Pharmaceutical Care Department, Ministry of National Guard Health Affairs, King Abdullah International Medical Research Center, King Saud Bin Abdulaziz University for Health Sciences, Jeddah 21423, Saudi Arabia
2
Pharmacy Department, Beth Israel Deaconess Medical Center, Boston, MA 02115, USA
3
Infection Prevention and Control Department, Ministry of National Guard Health Affairs, King Abdullah International Medical Research Center, King Abdulaziz Medical City, Riyadh 11426, Saudi Arabia
4
University of Toronto, University Health Network, Division of Nephrology, Toronto, ON M5G 2N2, Canada
*
Author to whom correspondence should be addressed.
Submission received: 6 July 2020 / Revised: 30 July 2020 / Accepted: 4 August 2020 / Published: 6 August 2020
(This article belongs to the Special Issue Antibiotics Use and Antimicrobial Resistance in Hospital)

Abstract

:
Colistin therapy is associated with the development of nephrotoxicity. We examined the incidence and risk factors of nephrotoxicity associated with colistin dosing. We included adult hospitalized patients who received intravenous (IV) colistin for >72 h between January 2014 and December 2015. The primary endpoint was the incidence of colistin-associated acute kidney injury (AKI). The secondary analyses were predictors of nephrotoxicity, proportions of patients inappropriately dosed with colistin according to the Food and Drug Administration (FDA), European Medicines Agency (EMA), and Garonzik formula and clinical cure rate. We enrolled 198 patients with a mean age of 55.67 ± 19.35 years, 62% were men, and 60% were infected with multidrug-resistant organisms. AKI occurred in 44.4% (95% CI: 37.4–51.7). Multivariable analysis demonstrated that daily colistin dose per body weight (kg) was associated with AKI (OR: 1.57, 95% CI: 1.08–2.30; p = 0.02). Other significant predictors included serum albumin level, body mass index (BMI), and severity of illness. None of the patients received loading doses, however FDA-recommended dosing was achieved in 70.2% and the clinical cure rate was 13%. The incidence of colistin-associated AKI is high. Daily colistin dose, BMI, serum albumin level, and severity of illness are independent predictors of nephrotoxicity.

1. Introduction

The emergence of multidrug-resistant (MDR) Gram-negative microorganisms is a worldwide health concern [1]. Infections involving these organisms are associated with delayed initiation of the appropriate antimicrobial therapy and increased health-care costs and mortality [2]. Colistin, an antibiotic discovered in the late 1940s and approved by the FDA in 1962, exhibits activity against a wide range of Gram-negative pathogens, including MDR organisms [3,4]. Although effective, its use was abandoned because of a high incidence of associated nephrotoxicity. Nevertheless, it has recently been reintroduced into practice because of the inadequacy of other antimicrobial agents against MDR organisms. Despite the fact that colistin has been available for over 50 years, the optimal dosing regimen has not been defined.
Colistin is formulated as the prodrug colistimethate sodium. Two main colistimethate sodium products are available: one approved by the FDA and the other by the EMA. In recent years, the pharmacokinetics and pharmacodynamics of colistin have been further elucidated [5]. This prompted the FDA and EMA to update their dosing recommendations in 2013 and 2014, respectively [3,6]. Although both updates were as a result of the same body of evidence, significant variation exists in their dosing recommendations. A recent study that compared both dosing regimens found that the EMA regimen, which includes higher overall maintenance doses, is more likely to achieve the target steady-state concentration of colistin than the FDA recommendations [4,7]. Additionally, only the EMA advocates the use of a loading dose as proposed by an earlier pharmacokinetic study in critically ill patients [8]. The study also provided a formula to calculate the optimal loading and maintenance doses of colistin based on serum colistin concentrations.
One of the most significant drawbacks of the use of colistin is the high incidence of associated nephrotoxicity. However, it remains controversial whether this adverse effect is dose-dependent, time-dependent, or both [9,10]. According to several studies, the incidence of colistin-induced nephrotoxicity ranges from 31 to 60% [11,12,13,14]. These studies evaluated different dosing regimens and found different rates of nephrotoxicity in their study populations. Several risk factors can increase a patient’s risk of developing colistin-induced nephrotoxicity, including chronic kidney disease, diabetes mellitus, cancer, anemia, and exposure to multiple events harmful to the kidneys (e.g., critical illness, sepsis, shock, burns, trauma, major surgery, and nephrotoxin exposure) [15,16]. However, colistin-induced nephrotoxicity is reversible and rarely results in permanent kidney damage [17,18]. The risk of nephrotoxicity must be weighed against the consequences of inadequately treating a life-threatening infection [17,18].
As there was no clear guidance for dosing colistin in our hospital and the presence of evolving updates in the dosing following Garonzik and EMA recommendations at the time we conducted the study, there was a lack of data on the development of colistin-associated AKI in our setting and its association with dosing. Given the high frequency of colistin use at our institution, we examined the incidence of colistin-induced nephrotoxicity in our patient population and evaluated the association between colistin dosing and the incidence of nephrotoxicity, predictors of colistin-associated nephrotoxicity and clinical cure rate associated with our current dosing regimen.

2. Materials and Methods

2.1. Study Design and Setting

We conducted a retrospective chart review study at the King Khalid Hospital, King Abdulaziz Medical City, Jeddah, Saudi Arabia. The Informatics Department generated a list of all patients who received IV colistin between January 2014 and December 2015.

2.2. Study Participants

Patients were included in the study if they were aged 18 years or older and had received IV colistin therapy for at least 72 h for treatment of infections as per the discretion of the treating physician. Patients were excluded if they had AKI at the time of initiation of colistin therapy, as defined by an increase in the serum creatinine concentration of ≥1.5 times of the baseline value as per Kidney Diseases Initiative Global Outcomes (KDIGO) guidelines for AKI [19]; lacked data on baseline serum creatinine concentration; required renal replacement therapy before colistin initiation; received only inhaled colistin; or were pregnant. For patients who received more than one course of colistin, only the first course was included. All data were collected using the electronic health records (EHRs) system and patients’ charts.

2.3. Study Outcomes

The primary outcome was the incidence of AKI, which was defined as an increase in serum creatinine concentration of ≥1.5 times the baseline value during the treatment course of colistin as defined by KDIGO [19]. Secondary analyses included the stage of AKI as defined by KDIGO [19], predictors of AKI, and impact of colistin dosing on the incidence of nephrotoxicity. Additionally, we assessed the proportions of patients inappropriately dosed according to dosing recommendations by the FDA [4], EMA [6], and Garonzik et al. [8] and determined the clinical cure rate associated with our current dosing regimen.

2.4. Data Collection

The data collected included baseline demographic information; colistin dosing and duration of therapy; concomitant antibiotic administration; and laboratory findings such as albumin, hemoglobin, white blood cell, and serum creatinine levels (evaluated at baseline, peak (if AKI occurred), and the end of colistin treatment). We also recorded isolates of the bacterial cultures and susceptibility patterns to various antimicrobial agents and defined MDR organisms for Enterobacterales, A. baumannii complex, and P. aeruginosa as the resistance to three or more classes of antimicrobials [20]. Patients who developed AKI were followed up for 3 months after the end of colistin treatment to evaluate its reversibility. To assess the severity of illness and risk of mortality in our patient population, we calculated the Charlson Comorbidity Index (CCI) [21,22] and Acute Physiologic Assessment and Chronic Health Evaluation (APACHE) II scores of ICU patients [22]. Additionally, we recorded concomitant use of nephrotoxic drugs such as vancomycin, amphotericin-B, aminoglycosides, non-steroidal anti-inflammatory drugs, angiotensin-converting enzyme inhibitors, angiotensin-II receptor blockers, contrast media, acyclovir, diuretics, and vasopressors. We evaluated the clinical cure rate of those who completed at least 7 days of colistin therapy, which we defined as the resolution of leukocytosis and fever within 48 h of day 7 of colistin therapy as per previous studies [10,23].

2.5. Sample Size

A sample of 194 patients was calculated to provide a 95% confidence limits of ± 7% around an assumed incidence of AKI of 45% [11,12,13,14,24].

2.6. Statistical Analysis

Descriptive statistics such as means ± standard deviation, medians, and interquartile ranges (IQRs) and percentages were used to present baseline demographics. Exact binomial calculations were used to estimate the incidence of AKI and 95% Confidence intervals (CIs). The chi-squared test was used to compare the proportions of baseline demographic characteristics and the Student’s t and Mann–Whitney U tests were used to compare continuous baseline characteristics with normal and non-normal distributions, respectively. The chi-squared test was also used to compare the proportions of patients who received appropriate colistin dosing or new recommendations. Logistic regression was used to perform a crude analysis of AKI incidence and its association with daily colistin dose. Multivariable logistic regression was used to adjust for potential confounders and identify predictors associated with the development of nephrotoxicity. The variables included in the model were selected based on prior knowledge, our data whenever feasible, and included age, BMI, baseline creatinine clearance, total daily colistin dose (mg/kg/day), use of vasopressor agents, use of diuretics, ICU admission, baseline serum albumin concentration, CCI, and APACHE II score. We assessed the model fitness using Hosmer Lemeshow’s goodness-of-fit test and the area under the receiver operator characteristic (ROC) curve. As some of these variables were missing in our data, we addressed the missingness of important variables which would impact the multivariate analysis. We used simple regression imputation [25,26] to handle missing baseline albumin values by developing a prediction model for missing baseline albumin values, which included the following predictors; gender, sex, ICU admission, CCI, and the total daily colistin dose (mg/kg/day). We used an indicator variable method [27] to handle missing non-applicable APACHE II scores of non-ICU patients. A two-sided p-value of <0.05 was considered to indicate statistical significance in all analyses. Data analysis was performed using STATA 14 (StataCorp LP, College Station, TX, USA).

2.7. Ethics

The study received approval from the institutional review board of the King Abdullah International Medical Research Center, Ministry of National Guard Health Affairs, Riyadh, Saudi Arabia (RJ16/048/J) in July 2016. The research was conducted according to the Declaration of Helsinki and national and institutional standards. We did not obtain informed consent from the participants given the retrospective nature of the study.

3. Results

3.1. Patients

In total, 467 patients were screened, of whom 198 met the inclusion criteria. The reasons for exclusion of the remaining 269 participants are demonstrated in Figure 1. Enrollment and screening flow diagram.
The mean age of the participants was 55.7 ± 19.36 years, 62% were men, the mean baseline creatinine clearance was 108.54 ± 46.26 mL/min, and the mean cumulative dose of colistin was 3.33 ± 0.99 mg/kg/day. Regarding baseline cultures, 60% (119/198) contained MDR organisms, 24.8% (49/198) contained no organisms, and 15.5% (30/198) contained non-MDR organisms. However, only 54% (107/198) of these isolates demonstrated sensitivity to colistin, 5.6% (11/198) were resistant to colistin, and 15.7% (31/198) did not have a reported sensitivity in the EHRs. The mean cumulative dose of colistin among patients who developed acute kidney injury was 195.54 ± 61.60 mg/day vs.198.33 ± 66.87 in those who did not develop AKI (p = 0.763). The baseline characteristics of the patients were stratified based on the development of AKI and are presented in Table 1.

3.2. Outcomes

3.2.1. Primary Outcome

At the primary end point, AKI occurred in 88 of the 198 patients (44.4%; 95% CI: 37.4–51.7). Of those 88 patients, 39 (44.3%) had reversible AKI and their serum creatinine returned to baseline levels, whereas the serum creatinine of 44 (50%) patients did not revert to baseline levels and the recovery serum creatinine data of five patients were missing. The median time to AKI development was 6 days (IQR: 3.5–11 days).

3.2.2. Secondary Outcomes

We noticed 38/198 missing values for serum albumin which had a normal distribution, so we used the following multiple imputation model to impute for missing serum albumin (Y).
Serum albumin (Y) = β0 + β1 Sex + β2 Age+ β3 Admission to ICU+ β4 Acute kidney injury + β5 CCI + β6 average colistin dose per weight per day
β0: regression coefficient constant, β1, β2 β3 β4, β5, β6, β7 are the regression coefficients of the model variables (sex, age, admission to ICU, development of acute kidney injury, CCI, and average colistin dose per weight per day (mg/kg/day), respectively).
Subsequently, we replaced the missing values of the serum albumin at baseline with the results of the regression model for predicted albumin stated above.
As our study included 131 patients admitted to ICU, we had to impute for missing non-applicable APCAHE-II for non-ICU patients, so we used indicator variable method for a missing APCHE-II.
A univariate analysis revealed that age, BMI, baseline albumin concentration, baseline creatinine clearance, admission to ICU, CCI, APACHE II score, diuretic use, and vasopressor use were associated with colistin-induced AKI. Univariate analysis of the association between AKI development and cumulative daily colistin dosing (mg/kg/day) revealed no statistical significance. However, multivariate analysis revealed that, for each 1 mg/kg/day of colistin, the odds of developing AKI increased by ~1.6 times, which was statistically significant after adjustment for potential confounders as demonstrated in Table 2.
Furthermore, multivariate analysis demonstrated that only baseline serum albumin concentration, BMI, and CCI were the other predictors associated with AKI development (p = 0.004, 0.02 and 0.04, respectively), as reported in Table 2. The model demonstrated a Hosmer Lemeshow’s goodness-of-fit of 211.63, p = 0.10, and an area under the ROC curve of 0.80. Details for multivariate analysis are reported in Table 2.
More than 75% of the patients had Stage-1 or Stage-2 AKI. Table 3 illustrates the secondary outcomes, including AKI classifications of the stages.
Clinical cure occurred in 13% of only 70% (138/198) of our participants; we were unable to assess the clinical cure rate of the whole cohort at Day 7 because data were missing. Among the patients who did not achieve clinical cure, the infections of only 10 participants exhibited resistance to colistin. Other secondary outcomes, such as 30-day mortality and adherence to the dosing recommendations of the FDA [4], the EMA [6], and Garonzik et al. [8], are presented in Table 3.

4. Discussion

In this study, we observed a 44.4% incidence of colistin-associated AKI. Although this is consistent with the range previously reported for colistin-induced AKI, it is high given that the average daily colistin dose administered to our patients was lower than the recommended dose by FDA, EMA, and the most recent consensus guidelines for optimal use of the polymyxins published in 2019 [8,28,29,30,31]. In addition, a study reported an incidence of colistin-induced nephrotoxicity of 33.9%, despite the fact that the participants received average daily doses of 4.6 mg/kg/day (based on ideal body weight (IBW)) [32]. However, in that study, the Risk, Injury, Failure, Loss of Kidney Function, and End-Stage Kidney Disease criteria were used for the assessment of AKI and APACHE II scores and CCI were not reported, so we were unable to compare the severity of illness with that of our patients [32].
Another study evaluated patients receiving a similar dosing regimen but shorter courses of colistin (8 days; IQR: 5–14) compared with our cohort [33]. The study reported AKI in 29% of the participants at 7 days, despite the fact that 50% of the patients received a loading dose compared with none in our study [33]. It is unclear whether use of a loading dose and higher colistin maintenance doses are associated with an increased incidence of nephrotoxicity. The evidence supporting the use of a loading dose is lacking appropriate clinical outcomes; however, current data do not show a higher risk of nephrotoxicity with use of a loading dose [10,34]. Furthermore, the consensus guidelines for optimal use of the polymyxins published in 2019 acknowledged this gap of evidence, advocated the use of loading doses as the need to achieve rapid therapeutic colistin concentrations outweighs the possible risk of AKI development which may be associated with the use of loading doses [30].
In our study, after adjustment for potential confounders, the odds of nephrotoxicity increased by ~1.6 times for each 1 mg/kg/day of colistin. A previous study reported an association between cumulative colistin doses of ≥5 mg/day (based on IBW) and nephrotoxicity with an OR of 23.4 (95% CI: 5.3–103.6) [13]. Other predictors associated with AKI identified in our study included the severity of illness (characterized by a high CCI) and low baseline serum albumin concentration, consistent with previous literature [14,15,35,36,37]. Furthermore, almost all of our patients received nephrotoxic agents, which have been identified as predictors of nephrotoxicity [13,36].
Our study demonstrated an alarmingly low clinical cure rate, which should be interpreted with caution. First, our patients were underdosed with colistin according to the new consensus guidelines for the appropriate use of polymyxins in 2019 as per their baseline kidney function (mean baseline creatinine clearance was 108.54 ± 46.26 mL/min in our cohort) and none of our patients received a loading dose [30]. A previous study reported that a colistin dose of >4.4 mg/kg/day (based on IBW) was an independent predictor of global clinical cure rate and achieved a higher clinical cure rate at Day 7 (40%) than lower doses (22%; p = 0.03) [23]. Loading and higher maintenance doses of colistin may achieve a therapeutic concentration more rapidly. This new regimen may increase the higher clinical cure rate and reduce the duration of therapy [30,34]. Second, our cohort included severely ill patients, with an average APACHE II score of 20 (which is highly predictive of severity of illness) and a 40% mortality rate. Third, respiratory isolates were the most common sources of infection (47.5%); IV administration of colistin may have lower efficacy against respiratory infections because of its limited penetration into the pulmonary epithelial lining fluid [38]. Additionally, only 11.6% of our patients received both inhaled and IV colistin. This practice is not consistent with the 2016 Infectious Disease Society of America guidelines for the management of adults with hospital-acquired and ventilator-associated pneumonia caused by MDR and extensively drug-resistant strains of Acinetobacter baumannii, Pseudomonas aeruginosa, and Enterobacteriaceae published during the study period which recommended the use of adjunctive inhaled colistin in addition to IV colistin [39]. Furthermore, this recommendation is consistent with the most recent consensus guidelines for optimal use of the polymyxins published in 2019 [30]. Fourth, we cannot explain our low clinical cure rate in terms of the presence of colistin-resistant isolates (5%). However, the resistance rate may have been inaccurate because of use of the gradient test at our institution at the time of the study, which underestimates colistin MIC values and resistance as per the European Committee on Antimicrobial Susceptibility Testing (EUCAST) and Clinical and Laboratory Standards Institute (CLSI) statement [40]. Finally, our lab did not test for the genetic mutation for isolates of Klebsiella pneumonia at that time and ceftazidime-avibactam was not approved by Saudi Food and Drug Authorities until 2018.
Our study has several limitations. First, because of its retrospective design, we were unable to assess the appropriateness of indication for the use of colistin therapy given that it was started as per the discretion of the treating physician and there was no standard definition for an infection requiring the use of colistin. This may have influenced the clinical cure rates if colistin use was not the most suitable choice. Additionally, data were missing for variables such as baseline serum albumin concentration, vital signs and complete blood counts on day 7 post-treatment. These missing variables limited our ability to assess the clinical cure rate of the whole cohort. We believe that these data were missing at random or data on worst-case scenarios were missing completely at random. For example, all missing baseline serum albumin values were for non-ICU patients, who were expected to be haemodynamically stable and less critically ill than ICU patients. Although we attempted to overcome the missing values for serum albumin concentration by regression imputation, as it is an important variable which would impact the adjusted analysis, this may have resulted in residual confounding. Second, selection bias resulting from the inclusion of patients with complete data for the outcomes of interest may have caused us to overestimate the overall rate of AKI in our cohort. Third, we were unable to assess the association between various regulatory recommendations as we initially planned and the incidence of nephrotoxicity because of suboptimal dosing and the lack of loading doses as per the recommendations of Garonzik et al. [8] and the EMA [26]. However, our data can serve as historical comparison data to assess the implementation of the new consensus of polymyxin use as we integrated into our practice [28]. Fourth, our study has limited generalizability to ICU patients, who represent two-thirds of our cohort, and to patients with a high CCI. Finally, as we had 88 events of AKI, we had limitations in the number of variables that we could include in the model to avoid over fitness for the adjusted analysis.
Our study has several strengths. We adjusted for potential confounders based on prior knowledge and our data, and the model demonstrated a good fitness and a ROC curve. Additionally, we were able to follow a large number of patients during the course of colistin therapy through EHRs. Finally, our study used the total cumulative daily dose of colistin per kilogram of body weight as a novel, pragmatic way of expressing the association between colistin dosing and nephrotoxicity.
Use of colistin for the eradication of MDR organisms has multiple dimensions. Therefore, future studies should adopt a comprehensive approach that aims to improve clinical cure and minimize nephrotoxicity. Research should prospectively individualize colistin-dosing regimens, loading doses, and duration of treatment based on clinical infectious syndromes and therapeutic drug monitoring and aims to assess new dosing recommendations in a pragmatic setting [30].

5. Conclusions

In conclusion, our study demonstrated that colistin-associated nephrotoxicity was high in our population. Cumulative daily dose of colistin per body weight, BMI, serum albumin concentration, and severity of illness were independent predictors of nephrotoxicity. Future studies must assess new dosing of colistin regimens to provide insights on strategies that maximize target therapeutic outcomes and optimize patient safety.

Author Contributions

Conceptualization, R.A., W.A., P.E. and S.I.; methodology, R.A., W.A., P.E., N.S. and S.I.; validation, R.A., W.A., A.N. and S.I.; formal analysis, S.I.; investigation, R.A., W.A., A.N., P.E., N.S., M.Q. and S.I.; resources, R.A., W.A., A.N., P.E. and S.I.; data curation, R.A., W.A., A.N., P.E. and S.I.; writing—original draft preparation, R.A., W.A. and A.N.; writing—review and editing, R.A., W.A., A.N., P.E., N.S., M.Q. and S.I.; visualization, S.I.; supervision, P.E. and S.I.; project administration, S.I. All authors have read and agreed to the published version of the manuscript.

Funding

This research received no external funding. The study was approved by IRB of the King Abdullah International Medical Research Center, Ministry of National Guard Health Affairs, Riyadh, Saudi Arabia

Acknowledgments

We presented the preliminary findings of this study as a poster presentation at the following conferences; (1) the Patient Safety Forum from 20th–23rd March 2017, Ministry of National Guard Health Affairs, King Saud Bin Abdulaziz University for Health Sciences, Riyadh, Saudi Arabia; (2) The Second International conference on Exploring New Horizons of Pharmacy Practice, 16th–18th April 2017 Intercontinental Hotel, Jeddah, Saudi Arabia; and (3) Antimicrobial Awareness Week, King Abdulaziz Medical City, Jeddah, Saudi Arabia, 16th November 2017, where it was awarded second prize for best poster presentation. We extend our appreciation to E. John Orav, Associate Professor of Biostatistics, Harvard T. H. Chan School of Public Health, Associate Professor of Medicine, Harvard Medical School and Associate Biostatistician, Division of General Internal Medicine and Primary Care, Brigham and Women’s Hospital for his invaluable input on statistical analyses and review of the statistical analysis of the manuscript.

Conflicts of Interest

The authors declare no conflict of interest.

References

  1. Media Center. Who Publishes a List of Bacteria for Which New Antibiotics Are Urgently Needed. Available online: http://www.who.int/mediacentre/news/releases/2017/bacteria-antibiotics-needed/en/ (accessed on 2 October 2017).
  2. MacVane, S.H. Antimicrobial Resistance in the Intensive Care Unit: A Focus on Gram-Negative Bacterial Infections. J. Intensive Care Med. 2017, 32, 25–37. [Google Scholar] [CrossRef] [PubMed]
  3. Lexi-Comp., Inc. Lexi-Drugs®; Lexi-Comp, Inc.: Cleveland, OH, USA, 2017. [Google Scholar]
  4. Food and Drug Administration. Coly-Mycin® M Parenteral (Colistimethate for Injection, USP). Available online: https://www.accessdata.fda.gov/drugsatfda_docs/label/2009/050108s026lbl (accessed on 29 July 2020).
  5. Bergen, P.J.; Li, J.; Nation, R.L. Dosing of colistin-back to basic PK/PD. Curr. Opin. Pharmacol. 2011, 11, 464–469. [Google Scholar] [CrossRef] [PubMed] [Green Version]
  6. Colomycin Injection. Available online: https://www.medicines.org.uk/emc/medicine/1590 (accessed on 19 August 2017).
  7. Nation, R.L.; Garonzik, S.; Li, J.; Thamlikitkul, V.; Giamarellos-Bourboulis, E.J.; Paterson, D.L.; Turnidge, J.D.; Forrest, A.; Silveira, F.P. Updated US and European Dose Recommendations for Intravenous Colistin: How Do They Perform? Clin. Infect. Dis. 2016, 62, 552–558. [Google Scholar] [CrossRef] [PubMed] [Green Version]
  8. Garonzik, S.M.; Li, J.; Thamlikitkul, V.; Paterson, D.L.; Shoham, S.; Jacob, J.; Silveira, F.P.; Forrest, A.; Nation, R.L. Population pharmacokinetics of colistin methanesulfonate and formed colistin in critically ill patients from a multicenter study provide dosing suggestions for various categories of patients. Antimicrob. Agents Chemother. 2011, 55, 3284–3294. [Google Scholar] [CrossRef] [PubMed] [Green Version]
  9. Dalfino, L.; Puntillo, F.; Mosca, A.; Monno, R.; Spada, M.L.; Coppolecchia, S.; Giuseppe, M.; Bruno, F.; Brienza, N. High-dose, extended-interval colistin administration in critically ill patients: Is this the right dosing strategy? A preliminary study. Clin. Infect. Dis. 2012, 54, 1720–1726. [Google Scholar] [CrossRef]
  10. Elefritz, J.L.; Bauer, K.A.; Jones, C.; Mangino, J.E.; Porter, K.; Murphy, C.V. Efficacy and Safety of a Colistin Loading Dose, High-Dose Maintenance Regimen in Critically Ill. Patients With Multidrug-Resistant Gram-Negative Pneumonia. J. Intensive Care Med. 2017, 32, 487–493. [Google Scholar] [CrossRef]
  11. Akajagbor, D.S.; Wilson, S.L.; Shere-Wolfe, K.D.; Dakum, P.; Charurat, M.E.; Gilliam, B.L. Higher incidence of acute kidney injury with intravenous colistimethate sodium compared with polymyxin B in critically ill patients at a tertiary care medical center. Clin. Infect. Dis. 2013, 57, 1300–1303. [Google Scholar] [CrossRef] [Green Version]
  12. Gauthier, T.P.; Wolowich, W.R.; Reddy, A.; Cano, E.; Abbo, L.; Smith, L.B. Incidence and predictors of nephrotoxicity associated with intravenous colistin in overweight and obese patients. Antimicrob. Agents Chemother. 2012, 56, 2392–2396. [Google Scholar] [CrossRef] [Green Version]
  13. Pogue, J.M.; Lee, J.; Marchaim, D.; Yee, V.; Zhao, J.J.; Chopra, T.; Lephart, P.; Kaye, K.S. Incidence of and risk factors for colistin-associated nephrotoxicity in a large academic health system. Clin. Infect. Dis. 2011, 53, 879–884. [Google Scholar] [CrossRef] [Green Version]
  14. Rocco, M.; Montini, L.; Alessandri, E.; Venditti, M.; Laderchi, A.; De Pascale, G.; Raponi, G.; Vitale, M.; Pietropaoli, P.; Antonelli, M. Risk factors for acute kidney injury in critically ill patients receiving high intravenous doses of colistin methanesulfonate and/or other nephrotoxic antibiotics: A retrospective cohort study. Crit. Care 2013, 17, R174. [Google Scholar] [CrossRef] [Green Version]
  15. Kwon, J.A.; Lee, J.E.; Huh, W.; Peck, K.R.; Kim, Y.G.; Kim, D.J.; Oh, H.Y. Predictors of acute kidney injury associated with intravenous colistin treatment. Int. J. Antimicrob. Agent 2010, 35, 473–477. [Google Scholar] [CrossRef] [PubMed]
  16. Tuon, F.F.; Rigatto, M.H.; Lopes, C.K.; Kamei, L.K.; Rocha, J.L.; Zavascki, A. Risk factors for acute kidney injury in patients treated with polymyxin B or colistin methanesulfonate sodium. Int. J. Antimicrob. Agents 2014, 43, 349–452. [Google Scholar] [CrossRef] [PubMed]
  17. Hartzell, J.D.; Neff, R.; Ake, J.; Howard, R.; Olson, S.; Paolino, K.; Vishnepolsky, M.; Weintrob, A.; Wortmann, G. Nephrotoxicity associated with intravenous colistin (colistimethate sodium) treatment at a tertiary care medical center. Clin. Infect. Dis. 2009, 48, 1724–1728. [Google Scholar] [CrossRef] [PubMed]
  18. Spapen, H.; Jacobs, R.; Van Gorp, V.; Troubleyn, J.; Honore, P.M. Renal and neurological side effects of colistin in critically ill patients. Ann. Intensive Care 2011, 1, 14. [Google Scholar] [CrossRef] [PubMed] [Green Version]
  19. Kidney Disease: Improving Global Outcomes (KDIGO) Acute Kidney Injury Work Group. KDIGO Clinical Practice Guideline for Acute Kidney Injury. Kidney Int. 2012, 2 (Suppl. 1), 1–138. [Google Scholar]
  20. Wolfensberger, A.; Kuster, S.P.; Marchesi, M.; Zbinden, R.; Hombach, M. The effect of varying multidrug-resistence (MDR) definitions on rates of MDR gram-negative rods. Antimicrob. Resist. Infect. Control 2019, 8, 193. [Google Scholar] [CrossRef] [Green Version]
  21. Charlson, M.E.; Pompei, P.; Ales, K.L.; MacKenzie, C.R. A new method of classifying prognostic comorbidity in longitudinal studies: Development and validation. J. Chronic Dis. 1987, 40, 373–383. [Google Scholar] [CrossRef]
  22. Knaus, W.A.; Draper, E.A.; Wagner, D.P.; Zimmerman, J.E. APACHE II: A severity of disease classification system. Crit. Care Med. 1985, 13, 818–829. [Google Scholar] [CrossRef]
  23. Gibson, G.A.; Bauer, S.R.; Neuner, E.A.; Bass, S.N.; Lam, S.W. Influence of Colistin Dose on Global Cure in Patients with Bacteremia Due to Carbapenem-Resistant Gram-Negative Bacilli. Antimicrob. Agents Chemother. 2015, 60, 431–436. [Google Scholar] [CrossRef] [Green Version]
  24. OpenEpi. Sample Size for Frequency in a Population. Available online: http://www.openepi.com/SampleSize/SSPropor.htm (accessed on 12 May 2016).
  25. Sterne, J.A.C.; White, I.R.; Carlin, J.B.; Spratt, M.; Royston, P.; Kenward, M.G.; Wood, A.M.; Carpenter, J.R. Multiple imputation for missing data in epidemiological and clinical research: Potential and pitfalls. BMJ 2009, 338, b2393. [Google Scholar] [CrossRef]
  26. Pedersen, A.B.; Mikkelsen, E.M.; Cronin-Fenton, D.; Kristensen, N.R.; Pham, T.M.; Pedersen, L.; Petersen, I. Missing data and multiple imputation in clinical epidemiological research. Clin. Epidemiol. 2017, 9, 157–166. [Google Scholar] [CrossRef] [PubMed] [Green Version]
  27. Groenwold, R.H.H.; White, I.R.; Donders, A.R.T.; Carpenter, J.R.; Altman, D.G.; Moons, K.G.M. Missing covariate data in clinical research: When and when not to use the missing-indicator method for analysis. Can. Med. Assoc. J. J. Med. Can. 2012, 184, 1265–1269. [Google Scholar] [CrossRef] [PubMed] [Green Version]
  28. European Medicines Agency. European Medicines Agency Completes Review of Polymyxin-Based Medicines. Recommendations Issued for Safe Use in Patients with Serious Infections Resistant to Standard Antibiotics. Available online: http://www.ema.europa.eu/docs/en_GB/document_library/Press_release/2014/10/WC500176334 (accessed on 21 July 2017).
  29. Food and Drug Administration. Coly-Mycin® M Parenteral (Colistimethate for Injection, USP). Available online: https://www.accessdata.fda.gov/drugsatfda_docs/label/2017/050108s033lbl (accessed on 29 July 2020).
  30. Tsuji, B.T.; Pogue, J.M.; Zavascki, A.P.; Paul, M.; Daikos, G.L.; Forrest, A.; Giacobbe, D.R.; Viscoli, C.; Giamarellou, H.; Karaiskos, I.; et al. International Consensus Guidelines for the Optimal Use of the Polymyxins: Endorsed by the American College of Clinical Pharmacy (ACCP), European Society of Clinical Microbiology and Infectious Diseases (ESCMID), Infectious Diseases Society of America (IDSA), International Society for Anti-infective Pharmacology (ISAP), Society of Critical Care Medicine (SCCM), and Society of Infectious Diseases Pharmacists (SIDP). Pharmacotherapy 2019, 39, 10–39. [Google Scholar] [PubMed]
  31. Chien, H.T.; Lin, Y.C.; Sheu, C.C.; Hsieh, K.P.; Chang, J.S. Is colistin-associated acute kidney injury clinically important in adults? A systematic review and meta-analysis. Int. J. Antimicrob. Agents 2020, 55, 105889. [Google Scholar] [CrossRef]
  32. Phe, K.; Lee, Y.; McDaneld, P.M.; Prasad, N.; Yin, T.; Figueroa, D.A.; Musick, W.L.; Cottreau, J.M.; Hu, M.; Tam, V.H. In vitro assessment and multicenter cohort study of comparative nephrotoxicity rates associated with colistimethate versus polymyxin B therapy. Antimicrob. Agents Chemother. 2014, 58, 2740–2746. [Google Scholar] [CrossRef] [Green Version]
  33. Shields, R.K.; Anand, R.; Clarke, L.G.; Paronish, J.A.; Weirich, M.; Perone, H.; Kieserman, J.; Freedy, H.; Andrzejewski, C.; Bonilla, H. Defining the incidence and risk factors of colistin-induced acute kidney injury by KDIGO criteria. PLoS ONE 2017, 12, e0173286. [Google Scholar] [CrossRef]
  34. Vardakas, K.Z.; Rellos, K.; Triarides, N.A.; Falagas, M.E. Colistin loading dose: Evaluation of the published pharmacokinetic and clinical data. Int J. Antimicrob. Agents 2016, 48, 475–484. [Google Scholar] [CrossRef]
  35. Omrani, A.S.; Alfahad, W.A.; Shoukri, M.M.; Baadani, A.M.; Aldalbahi, S.; Almitwazi, A.A.; Albarrak, A.M. High dose intravenous colistin methanesulfonate therapy is associated with high rates of nephrotoxicity; a prospective cohort study from Saudi Arabia. Ann. Clin. Microbiol. Antimicrob. 2015, 14, 3. [Google Scholar] [CrossRef] [Green Version]
  36. Sorli, L.; Luque, S.; Grau, S.; Berenguer, N.; Segura, C.; Montero, M.M.; Alvarez-Lerma, F.; Knobel, H.; Benito, N.; Horcajada, J.P. Trough colistin plasma level is an independent risk factor for nephrotoxicity: A prospective observational cohort study. BMC Infect. Dis. 2013, 13, 380. [Google Scholar] [CrossRef] [Green Version]
  37. Giacobbe, D.R.; di Masi, A.; Leboffe, L.; Del Bono, V.; Rossi, M.; Cappiello, D.; Coppo, E.; Marchese, A.; Casulli, A.; Signori, A.; et al. Hypoalbuminemia as a predictor of acute kidney injury during colistin treatment. Sci. Rep. 2018, 8, 11968. [Google Scholar] [CrossRef]
  38. Yapa, S.; Li, J.; Patel, K.; Wilson, J.W.; Dooley, M.J.; George, J.; Clark, D.; Poole, S.; Williams, E.; Porter, C.J.; et al. Pulmonary and systemic pharmacokinetics of inhaled and intravenous colistin methanesulfonate in cystic fibrosis patients: Targeting advantage of inhalational administration. Antimicrob. Agents Chemother. 2014, 58, 2570–2579. [Google Scholar] [CrossRef] [PubMed] [Green Version]
  39. Kalil, A.C.; Metersky, M.L.; Klompas, M.; Muscedere, J.; Sweeney, D.A.; Palmer, L.B.; Napolitano, L.M.; O’Grady, N.P.; Bartlett, J.G.; Carratala, J.; et al. Executive Summary: Management of Adults With Hospital-acquired and Ventilator-associated Pneumonia: 2016 Clinical Practice Guidelines by the Infectious Diseases Society of America and the American Thoracic Society. Clin. Infect. Dis. 2016, 63, 575–582. [Google Scholar] [CrossRef] [PubMed]
  40. The European Committee on Antimicrobial Susceptibility Testing.Eucast Warnings Concerning Antimicrobial Susceptibility Testing Products or Procedures. 20 June 2017. Available online: http://www.eucast.org/ast_of_bacteria/warnings/ (accessed on 10 January 2018).
Figure 1. Enrollment and screening flow diagram.
Figure 1. Enrollment and screening flow diagram.
Antibiotics 09 00485 g001
Table 1. Baseline characteristics.
Table 1. Baseline characteristics.
Baseline CharacteristicsAKI 1
(n = 88)
No AKI 1
(n = 110)
p-Value 2
Age (years)59.5 ± 17.752.62 ± 20.110.011
Sex (male)50 (56.8)72 (65.5)0.214
Body mass index (kg/m2)28 (24–34.2)25.1 (21.5–28.7)0.003
Baseline CrCL3 (mL/min)99.79 ± 45.44115.53 ± 45.910.017
ICU admission65 (73.9)66 (60)0.04
ICU stay prior to colistin therapy (days)3 (0–11)6 (0–16)0.079
Charlson Comorbidity Index (CCI)5.5 (3–7)3 (0–6)<0.001
APACHE II score 422.93 ± 7.1919.72 ± 6.910.013
Anaemia84 (95.45)102 (92.73)0.424
Serum albumin concentration (g/L) 524.53 ± 4.927.85 ± 5.19<0.001
Colistin
Dosing body weight, DBW (kg) 658.64 ± 12.8960.9 ± 13.450.392
Duration of colistin therapy (days)11 (6–16.5)12 (8–16)0.203
Colistin dose (mg/kg/day) 73.29 ± 13.24 ± 0.970.702
Causative organism 0.182
Acinetobacter baumannii28 (31.82)43 (39.09)-
Klebsiella pneumoniae17 (19.32)9 (8.2)-
Pseudomonas aeruginosa7 (8)17 (15.5)-
Site of infection 0.111
Respiratory36 (40.9) 58 (52.7)-
Blood25 (28.4) 19 (17.3) -
Urine13 (14.8)22 (20)-
Other 814 (15.9) 11 (10)-
Antibiotic therapy 0.667
Combined use of antimicrobial agents76 (86.4)94 (85.5)-
Combined systemic, nebulizer use of colistin and other antibiotics9 (10.2)14 (12.7)-
Other 93 (3.4) 2 (1.82)
Combined use with nephrotoxic agents87 (98.9)106 (96.4)0.265
ACEIs 10/ARBs 1122 (25)21 (19.1)0.316
Intravenous contrast media39 (44.3)43 (39.1)0.458
Diuretics67 (60.9)72 (81.8)0.001
Vasopressors59 (67.1)46 (41.8)0.001
Aminoglycosides15 (17.1)11 (10)0.145
Amphotericin10 (11.4)2 (1.82)0.005
Vancomycin78 (88.6)86 (78.2)0.053
Acyclovir11 (12.5)13 (11.8)0.884
Non-steroidal anti-inflammatory drugs2 (2.3)4 (3.6)0.578
1 Data are reported as means ± standard deviation for continuous variables or medians (interquartile ranges) and numbers (%) as appropriate. 2 p-value based on the chi-squared test or Fisher’s exact test or the t-test and Mann–Whitney U test as appropriate, with p < 0.05 indicating significance (in bold). 3 Baseline CrCL (Creatinine Clearance) based on the Cockcroft–Gault equation. 4 APACHE II score for the prediction of ICU mortality calculated within 24 h of ICU admission, a value missing for all non-ICU patients. 5 Serum albumin concentration at baseline, including imputed values for 38 patients. 6 DBW based on actual body weight or ideal body weight (IBW) if actual body weight is ≥130% of the IBW. 7 Colistin dosing per body weight. 8 Other infections, including complicated skin and soft tissue infections, intra-abdominal infections, and catheter tip infections. 9 Other combinations of systemic and inhaled colistin and monotherapy. 10 ACEIs: angiotensin-converting enzyme inhibitors. 11 ARBs: angiotensin receptor blockers.
Table 2. Univariate and multivariate analysis for predictors of acute kidney injury (AKI).
Table 2. Univariate and multivariate analysis for predictors of acute kidney injury (AKI).
VariablesUnivariate Analysis
Odd Ratio (OR),
95% CI, p-Value (p)
Multivariate Analysis
OR,
95% CI, p-Value
Colistin dose (mg/kg/day)1.09,
95% CI (0.83–1.46), p = 0.52
OR: 1.57,
95% CI (1.08–2.30), p = 0.02 1
Age (years)1.02,
95% CI (1.003–1.035), p = 0.01
0.98,
95% CI (0.96–1.01), p = 0.26
Sex (female)1.44,
95% CI (0.81–2.56), p = 0.215
NA 2
BMI (kg/m2)1.07,
95% CI (1.03–1.12), p < 0.001
1.06,
95% CI (1.01–1.11), p = 0.02
Creatinine clearance (m//min)0.99,
95% CI (0.986–0.998), p = 0.02
0.99,
95% CI (0.98–1), p = 0.09
Admission to ICU1.88,
95% CI (1.02–3.47), p = 0.042
0.25,
95% CI (0.03–1.90), p = 0.18
Hospitalization in ICU prior to colistin use (days)0.99,
95% CI (0.99–1.004), p = 0.24
NA 2
Charlson Comorbidity Index1.20,
95% CI (1.09–1.33), p = <0.001
1.16,
95% CI (1–1.34), p = 0.04
APCHAE-II score1.04,
95% CI (1.02–1.07), p = <0.001
1.03,
95% CI (0.97–1.10), p = 0.31
Serum albumin (g/L)0.87,
95% CI (0.81–0.93), p = <0.001
0.89,
95% CI (0.83–0.96), p = 0.004
Use of nephrotoxic agent3.28,
95% CI (0.36–29.91), p = 0.29
NA 3
Diuretic use2.89,
95% CI (1.49–5.60), p = 0.002
1.01,
95% CI (0.43–2.36), p = 0.99
Vasopressor use2.83,
95% CI (1.58–5.08), p < 0.001
2.1,
95% CI (0.90–4.91), p = 0.09
1 Adjusted for age (years), BMI (kg/m2), baseline creatinine clearance (mL/min), admission to ICU, Charleston Comorbidity Index, baseline APACHE II score, serum albumin concentration (g/L), diuretic and vasopressor use. 2 NA: Not applicable as sex and hospitalization in ICU prior to colistin use (days) were not significantly associated with AKI in the univariate analysis so were excluded from the multivariate analysis. 3 NA: Not applicable as concomitant use of nephrotoxic medications with colistin was dropped from the model because almost all patients received one nephrotoxic agent so it cannot be a predictor.
Table 3. Secondary end points.
Table 3. Secondary end points.
Secondary End PointsAKI
88 (44.4)
n (%)
No AKI
110 (55.5)
n (%)
p-Value
AKI classifications 1
   Stage I29 (33)--
   Stage II39 (44.3)--
   Stage III20 (22.7)--
Clinical cure 25 (5.7)13 (11.8)0.136
Mortality at 30 days42 (47.7)24 (21.8)<0.001
Dosing appropriateness
   FDA 363 (72)74 (67.3)0.513
   EMA
      Loading doseNoneNone-
      Maintenance dose5 (5.7)10 (9.1)0.368
   Garonzik et al. [8]
      Loading doseNoneNone-
      Maintenance dose5 (5.7)0 (0)-
1 Acute kidney injury (AKI) classification according to the Kidney International global outcomes criteria for the staging of AKI. 2 Clinical cure: 60 patients were missing (comprising 40 patients who received colistin for <7 days and 20 patients missing White Blood Cells data on Day 7). 3 Calculated based on FDA recommendations and categories of creatinine clearance at baseline.

Share and Cite

MDPI and ACS Style

Almutairy, R.; Aljrarri, W.; Noor, A.; Elsamadisi, P.; Shamas, N.; Qureshi, M.; Ismail, S. Impact of Colistin Dosing on the Incidence of Nephrotoxicity in a Tertiary Care Hospital in Saudi Arabia. Antibiotics 2020, 9, 485. https://0-doi-org.brum.beds.ac.uk/10.3390/antibiotics9080485

AMA Style

Almutairy R, Aljrarri W, Noor A, Elsamadisi P, Shamas N, Qureshi M, Ismail S. Impact of Colistin Dosing on the Incidence of Nephrotoxicity in a Tertiary Care Hospital in Saudi Arabia. Antibiotics. 2020; 9(8):485. https://0-doi-org.brum.beds.ac.uk/10.3390/antibiotics9080485

Chicago/Turabian Style

Almutairy, Reem, Waad Aljrarri, Afnan Noor, Pansy Elsamadisi, Nour Shamas, Mohammad Qureshi, and Sherine Ismail. 2020. "Impact of Colistin Dosing on the Incidence of Nephrotoxicity in a Tertiary Care Hospital in Saudi Arabia" Antibiotics 9, no. 8: 485. https://0-doi-org.brum.beds.ac.uk/10.3390/antibiotics9080485

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Metrics

Back to TopTop