Journal of Oncology Pharmacy Practice http://opp.sagepub.com/
Retrospective review of vancomycin-induced nephrotoxicity in patients with leukemia Amber Fullmer, Deborah McCue and Chun Feng J Oncol Pharm Pract published online 31 October 2013 DOI: 10.1177/1078155213509847 The online version of this article can be found at: http://opp.sagepub.com/content/early/2013/10/30/1078155213509847
Published by: http://www.sagepublications.com
Additional services and information for Journal of Oncology Pharmacy Practice can be found at: Email Alerts: http://opp.sagepub.com/cgi/alerts Subscriptions: http://opp.sagepub.com/subscriptions Reprints: http://www.sagepub.com/journalsReprints.nav Permissions: http://www.sagepub.com/journalsPermissions.nav
>> OnlineFirst Version of Record - Oct 31, 2013 What is This?
Downloaded from opp.sagepub.com at Maastricht University on July 8, 2014
XML Template (2013) [29.10.2013–7:22am] //blrnas3/cenpro/ApplicationFiles/Journals/SAGE/3B2/OPPJ/Vol00000/130140/APPFile/SG-OPPJ130140.3d
(OPP)
[1–6] [INVALID Stage]
Journal of
Oncology Pharmacy Practice
Original Article
Retrospective review of vancomycininduced nephrotoxicity in patients with leukemia
J Oncol Pharm Practice 0(0) 1–6 ! The Author(s) 2013 Reprints and permissions: sagepub.co.uk/journalsPermissions.nav DOI: 10.1177/1078155213509847 opp.sagepub.com
Amber Fullmer, Deborah McCue and Chun Feng
Abstract Background: The occurrence of nephrotoxicity with vancomycin is approximately 17%, but can increase to 35% when combined with other nephrotoxic agents. Patients with hematologic malignancies may be at greater risk for vancomycininduced nephrotoxicity due to nephrotoxic chemotherapy and tumor lysis syndrome. Objective: The primary objective of this study was to determine the occurrence of nephrotoxicity in adult patients with leukemia receiving vancomycin. Methods: A retrospective review approved by the Institutional Review Board was conducted on adult patients with leukemia who received at least one dose of vancomycin during hospital admission between 1 January 2009 and 30 April 2009. Results: Forty patients had an occurrence of nephrotoxicity (16%) while 210 patients did not have an occurrence of nephrotoxicity. In multivariate analysis, variables significantly associated with development of nephrotoxicity included active disease status (odds ratio, 4.38 [95% CI 1.1–29.4], p ¼ 0.0291), concomitant intravenous acyclovir administration (odds ratio, 3.83 [95% CI, 1.6–8.9]; p ¼ 0.0022), and concomitant amphotericin administration (odds ratio, 4.26 [95% CI, 1.9–9.4]; p ¼ 0.0004). Conclusion: The occurrence of nephrotoxicity in patients with leukemia treated with vancomycin was 16% in our study, similar to previously published reports. Active disease status and concomitant use of intravenous acyclovir and amphotericin were identified as significant risk factors for development of nephrotoxicity. The presence of risk factors for vancomycin nephrotoxicity should be evaluated prior to initiation of therapy to determine appropriateness of use.
Keywords Vancomycin, leukemia, nephrotoxicity
Introduction Vancomycin, a glycopeptide antibiotic, is indicated for the treatment of suspected or proven infections with Gram-positive bacteria, such as methicillin-resistant Staphylococcus aureus (MRSA). In the cancer population, the Infectious Diseases Society of America (IDSA) recommends vancomycin as part of the management of neutropenic fever in patients with suspected catheterrelated infections, skin or soft-tissue infection, pneumonia, or hemodynamic instability.1 Cancer patients have an increased risk of developing Gram-positive bacteremia as a result of indwelling intravascular catheters, prolonged neutropenia, mucositis due to
chemotherapy, and fluoroquinolone prophylaxis.2,3 The optimal ratio of the area under the concentration-time curve (AUC) to the minimum inhibitory concentration (MIC) of 400 has been recommended, as well as target trough levels in the range 15–20 mg/mL for the treatment of bacteremia, endocarditis,
Division of Pharmacy, The University of Texas MD Anderson Cancer Center, Houston, TX, USA Corresponding author: Amber Fullmer, Division of Pharmacy, The University of Texas MD Anderson Cancer Center, 1515 Holcombe Blvd, Unit 0377, Houston, TX 77030, USA. Email: [email protected]
Downloaded from opp.sagepub.com at Maastricht University on July 8, 2014
XML Template (2013) [29.10.2013–7:22am] //blrnas3/cenpro/ApplicationFiles/Journals/SAGE/3B2/OPPJ/Vol00000/130140/APPFile/SG-OPPJ130140.3d
(OPP)
[1–6] [INVALID Stage]
2
Journal of Oncology Pharmacy Practice 0(0)
osteomyelitis, meningitis, and hospital acquired pneumonia caused by Gram-positive organisms.4 Early preparations of vancomycin were associated with a high occurrence of infusion-related events, ototoxicity, and nephrotoxicity, likely due to impurities in the formulation.5–7 The currently available product, which is a purified preparation of vancomycin, has been reported to induce nephrotoxicity in up to 17% of patients when used alone and up to 35% of patients when used in combination with aminoglycosides.4,6 Nephrotoxicity has commonly been defined as an increase of 0.5 mg/dL or an increase of 50% in serum creatinine from baseline.4,8–11 The RIFLE criteria further classifies nephrotoxicity into three levels based on a percentage increase of serum creatinine from baseline, with RISK defined as >50% rise in serum creatinine, INJURY defined as >100% rise in serum creatinine, and FAILURE defined as >200% rise in serum creatinine.12 Nephrotoxicity with vancomycin has been correlated with doses greater than 4 g per day, high trough levels, extended treatment duration, older age, compromised kidney function, and concomitant nephrotoxic agents.4,5,13,14 It has been suggested that patients with hematologic malignancies are at greater risk for vancomycin-induced nephrotoxicity.5,15 Chemotherapy agents, as well as tumor lysis syndrome may compound the above risk factors in this patient population. Our retrospective review was conducted to determine the incidence of nephrotoxicity and to identify potential risk factors in patients with leukemia treated with vancomycin.
Objectives The primary objective of this retrospective review was to determine the incidence of nephrotoxicity in adult patients with leukemia treated with vancomycin. The secondary objectives were to categorize the occurrence of nephrotoxicity based on serum creatinine elevations according to the RIFLE criteria and identify risk factors for the development of nephrotoxicity.
Methods A retrospective review, approved by the Institutional Review Board, was conducted on patients who were admitted to the inpatient leukemia service and received vancomycin from 1 January 2009 through 30 April 2009. An initial list of patients was generated by the Division of Pharmacy databases based on admission to the inpatient leukemia service and the presence of orders for vancomycin. Patients over 18 years of age were included if they were admitted to the adult leukemia service, had a leukemia diagnosis of acute myeloid
leukemia (AML), acute promyelocytic leukemia (APL), biphenotypic leukemia, acute lymphoblastic leukemia (ALL), Burkitt’s leukemia, chronic myeloid leukemia (CML), chronic lymphocytic leukemia (CLL), hairy cell leukemia, myelodysplastic syndrome (MDS), myelofibrosis, or chronic myelomonocytic leukemia (CMML) and received at least one dose of vancomycin as documented on the medication administration record. Patients with leukemia diagnoses who displayed similar disease course and treatment were grouped together for analysis. Patients were excluded for baseline serum creatinine above the upper limit of normal as defined by the laboratory standard for the institution. During the study period, females were excluded for baseline serum creatinine >1 mg/dL and males were excluded for baseline serum creatinine >1.3 mg/dL. Patients receiving dialysis prior to vancomycin administration were also excluded. In addition, patients with vancomycin initiated before admission to the institution were excluded. Data were collected from inpatient admissions using the institutional electronic health record system. Demographic data including age, sex, specific leukemia diagnosis, disease status (i.e. active leukemia vs. leukemia in remission), and comorbid conditions including hypertension and diabetes were collected for all patients. Baseline serum creatinine was defined as the serum creatinine on the date of vancomycin initiation or within two days prior. Peak serum creatinine was defined as the highest serum creatinine during treatment with vancomycin and within seven days following vancomycin discontinuation. Nephrotoxicity was defined as a 50% rise in serum creatinine and was further classified based on the RIFLE criteria: RISK defined as baseline serum creatinine 1.5, INJURY defined as baseline serum creatinine 2, and FAILURE defined as baseline serum creatinine 3. In patients who developed nephrotoxicity, additional data were collected up to seven days following discontinuation of vancomycin. This included dialysis initiation, return of serum creatinine to normal range or to baseline in those patients who never exceeded the upper limits of the normal range, nephrotoxicity that did not resolve, and occurrence of death with nephrotoxicity. Concomitant nephrotoxic medications within two days before vancomycin initiation and throughout the seven days following vancomcyin discontinuation were collected for all patients. These medications included intravenous (IV) contrast, angiotensin-converting enzyme inhibitors or angiotensin receptor blockers, IV acyclovir, aminoglycosides, amphotericin, clofarabine, IV cytarabine, foscarnet, furosemide, IV methotrexate, and IV and oral sulfamethoxazole/trimethoprim. Concomitant rasburicase administration was also collected as an assessment of potential risk
Downloaded from opp.sagepub.com at Maastricht University on July 8, 2014
XML Template (2013) [29.10.2013–7:22am] //blrnas3/cenpro/ApplicationFiles/Journals/SAGE/3B2/OPPJ/Vol00000/130140/APPFile/SG-OPPJ130140.3d
(OPP)
[1–6] [INVALID Stage]
Fullmer et al.
3
of tumor lysis syndrome. Vancomycin trough levels, as labeled in the electronic health record system, were collected if at least three doses were administered on the medication administration record prior to obtaining the level.
Statistical analysis Statistics were calculated using JMP 10 (SAS Institute Inc, Cary, North Carolina). Descriptive statistics were used to summarize the study population. Parametric and non-parametric tests including 2, Fisher exact test, Student t test, and Wilcoxon Rank Sum test were used to evaluate variables of interest when appropriate. All comparisons were unpaired, and all tests of significance were two-tailed. A p value 0.05 was considered statistically significant. To identify factors independently associated with nephrotoxicity, multivariate logistic regression analyses were employed. Variables associated with nephrotoxicity in the univariate analysis were considered for inclusion in the models.
Results Two hundred and ninety patients admitted to the inpatient leukemia service starting 1 January 2009 were screened in order of admission date until a total of
250 patients met inclusion for review. The number of patients included was capped due to manpower/time constraints. Of the 40 patients not included in the analysis, the majority were excluded for baseline serum creatinine above the upper limit of normal (n ¼ 30). Additional patients were excluded for vancomycin administration prior to arrival at the institution (n ¼ 4), having a diagnosis not meeting inclusion criteria (n ¼ 1), lack of baseline creatinine reported prior to vancomycin initiation (n ¼ 1), and lack of documentation of vancomycin administration on medication administration record (n ¼ 4). Forty patients (16%) developed nephrotoxicity while receiving vancomycin therapy. Demographic information between patients who had an occurrence of nephrotoxicity (n ¼ 40) and those who did not (n ¼ 210) were similar (Table 1). However, active leukemia, as compared with leukemia in remission, was more common in patients who developed nephrotoxicity (p ¼ 0.005). The median peak white blood cell (WBC) count during treatment with vancomycin and within seven days following discontinuation was significantly higher in patients who developed nephrotoxicity compared to patients who did not develop nephrotoxicity (5.2 cells/mL vs.1.7 cells/mL, p ¼ 0.01). The type of leukemia did not affect the occurrence of nephrotoxicity (p ¼ 0.4459).
Table 1. Demographics. Characteristic Age (years), mean (SD) Male, no. (%) Newly diagnosed, no. (%) Active disease, no. (%) Type of leukemia AML/APL/Biphenotypic, no (%) ALL/Burkitt’s, no (%) CML, no (%) CLL/Hairy cell, no (%) MDS/Myelofibrosis/CMML, no (%) Comorbid condition Diabetes, no (%) Hypertension, no. (%) Peak WBC (cells/mL), median (IQR) Baseline serum creatinine, mean (SD) Length of therapy (days), median (IQR) Total vancomycin doses, mean (SD)
Nephrotoxicity (n ¼ 40)
No nephrotoxicity (n ¼ 210)
55.4 20 18 38
(18) (50) (45) (95)
53.1 116 95 159
(16) (55.2) (45.2) (75.7)
0.41 0.6 1 0.005
27 5 1 2 5
(67.5) (12.5) (2.5) (5) (12.5)
124 43 11 13 19
(59.1) (20.5) (5.2) (6.2) (9.1)
0.32 0.24 0.46 0.77 0.5
2 14 5.2 0.72 7.5 14.9
(5) (35) (18.8) (0.19) (7) (13.1)
25 73 1.7 0.77 3 9
(11.9) (34.8) (5.4) (0.2) (4) (8.8)
0.27 1 0.01 0.12
Retrospective review of vancomycin-induced nephrotoxicity in patients with leukemia Amber Fullmer, Deborah McCue and Chun Feng J Oncol Pharm Pract published online 31 October 2013 DOI: 10.1177/1078155213509847 The online version of this article can be found at: http://opp.sagepub.com/content/early/2013/10/30/1078155213509847
Published by: http://www.sagepublications.com
Additional services and information for Journal of Oncology Pharmacy Practice can be found at: Email Alerts: http://opp.sagepub.com/cgi/alerts Subscriptions: http://opp.sagepub.com/subscriptions Reprints: http://www.sagepub.com/journalsReprints.nav Permissions: http://www.sagepub.com/journalsPermissions.nav
>> OnlineFirst Version of Record - Oct 31, 2013 What is This?
Downloaded from opp.sagepub.com at Maastricht University on July 8, 2014
XML Template (2013) [29.10.2013–7:22am] //blrnas3/cenpro/ApplicationFiles/Journals/SAGE/3B2/OPPJ/Vol00000/130140/APPFile/SG-OPPJ130140.3d
(OPP)
[1–6] [INVALID Stage]
Journal of
Oncology Pharmacy Practice
Original Article
Retrospective review of vancomycininduced nephrotoxicity in patients with leukemia
J Oncol Pharm Practice 0(0) 1–6 ! The Author(s) 2013 Reprints and permissions: sagepub.co.uk/journalsPermissions.nav DOI: 10.1177/1078155213509847 opp.sagepub.com
Amber Fullmer, Deborah McCue and Chun Feng
Abstract Background: The occurrence of nephrotoxicity with vancomycin is approximately 17%, but can increase to 35% when combined with other nephrotoxic agents. Patients with hematologic malignancies may be at greater risk for vancomycininduced nephrotoxicity due to nephrotoxic chemotherapy and tumor lysis syndrome. Objective: The primary objective of this study was to determine the occurrence of nephrotoxicity in adult patients with leukemia receiving vancomycin. Methods: A retrospective review approved by the Institutional Review Board was conducted on adult patients with leukemia who received at least one dose of vancomycin during hospital admission between 1 January 2009 and 30 April 2009. Results: Forty patients had an occurrence of nephrotoxicity (16%) while 210 patients did not have an occurrence of nephrotoxicity. In multivariate analysis, variables significantly associated with development of nephrotoxicity included active disease status (odds ratio, 4.38 [95% CI 1.1–29.4], p ¼ 0.0291), concomitant intravenous acyclovir administration (odds ratio, 3.83 [95% CI, 1.6–8.9]; p ¼ 0.0022), and concomitant amphotericin administration (odds ratio, 4.26 [95% CI, 1.9–9.4]; p ¼ 0.0004). Conclusion: The occurrence of nephrotoxicity in patients with leukemia treated with vancomycin was 16% in our study, similar to previously published reports. Active disease status and concomitant use of intravenous acyclovir and amphotericin were identified as significant risk factors for development of nephrotoxicity. The presence of risk factors for vancomycin nephrotoxicity should be evaluated prior to initiation of therapy to determine appropriateness of use.
Keywords Vancomycin, leukemia, nephrotoxicity
Introduction Vancomycin, a glycopeptide antibiotic, is indicated for the treatment of suspected or proven infections with Gram-positive bacteria, such as methicillin-resistant Staphylococcus aureus (MRSA). In the cancer population, the Infectious Diseases Society of America (IDSA) recommends vancomycin as part of the management of neutropenic fever in patients with suspected catheterrelated infections, skin or soft-tissue infection, pneumonia, or hemodynamic instability.1 Cancer patients have an increased risk of developing Gram-positive bacteremia as a result of indwelling intravascular catheters, prolonged neutropenia, mucositis due to
chemotherapy, and fluoroquinolone prophylaxis.2,3 The optimal ratio of the area under the concentration-time curve (AUC) to the minimum inhibitory concentration (MIC) of 400 has been recommended, as well as target trough levels in the range 15–20 mg/mL for the treatment of bacteremia, endocarditis,
Division of Pharmacy, The University of Texas MD Anderson Cancer Center, Houston, TX, USA Corresponding author: Amber Fullmer, Division of Pharmacy, The University of Texas MD Anderson Cancer Center, 1515 Holcombe Blvd, Unit 0377, Houston, TX 77030, USA. Email: [email protected]
Downloaded from opp.sagepub.com at Maastricht University on July 8, 2014
XML Template (2013) [29.10.2013–7:22am] //blrnas3/cenpro/ApplicationFiles/Journals/SAGE/3B2/OPPJ/Vol00000/130140/APPFile/SG-OPPJ130140.3d
(OPP)
[1–6] [INVALID Stage]
2
Journal of Oncology Pharmacy Practice 0(0)
osteomyelitis, meningitis, and hospital acquired pneumonia caused by Gram-positive organisms.4 Early preparations of vancomycin were associated with a high occurrence of infusion-related events, ototoxicity, and nephrotoxicity, likely due to impurities in the formulation.5–7 The currently available product, which is a purified preparation of vancomycin, has been reported to induce nephrotoxicity in up to 17% of patients when used alone and up to 35% of patients when used in combination with aminoglycosides.4,6 Nephrotoxicity has commonly been defined as an increase of 0.5 mg/dL or an increase of 50% in serum creatinine from baseline.4,8–11 The RIFLE criteria further classifies nephrotoxicity into three levels based on a percentage increase of serum creatinine from baseline, with RISK defined as >50% rise in serum creatinine, INJURY defined as >100% rise in serum creatinine, and FAILURE defined as >200% rise in serum creatinine.12 Nephrotoxicity with vancomycin has been correlated with doses greater than 4 g per day, high trough levels, extended treatment duration, older age, compromised kidney function, and concomitant nephrotoxic agents.4,5,13,14 It has been suggested that patients with hematologic malignancies are at greater risk for vancomycin-induced nephrotoxicity.5,15 Chemotherapy agents, as well as tumor lysis syndrome may compound the above risk factors in this patient population. Our retrospective review was conducted to determine the incidence of nephrotoxicity and to identify potential risk factors in patients with leukemia treated with vancomycin.
Objectives The primary objective of this retrospective review was to determine the incidence of nephrotoxicity in adult patients with leukemia treated with vancomycin. The secondary objectives were to categorize the occurrence of nephrotoxicity based on serum creatinine elevations according to the RIFLE criteria and identify risk factors for the development of nephrotoxicity.
Methods A retrospective review, approved by the Institutional Review Board, was conducted on patients who were admitted to the inpatient leukemia service and received vancomycin from 1 January 2009 through 30 April 2009. An initial list of patients was generated by the Division of Pharmacy databases based on admission to the inpatient leukemia service and the presence of orders for vancomycin. Patients over 18 years of age were included if they were admitted to the adult leukemia service, had a leukemia diagnosis of acute myeloid
leukemia (AML), acute promyelocytic leukemia (APL), biphenotypic leukemia, acute lymphoblastic leukemia (ALL), Burkitt’s leukemia, chronic myeloid leukemia (CML), chronic lymphocytic leukemia (CLL), hairy cell leukemia, myelodysplastic syndrome (MDS), myelofibrosis, or chronic myelomonocytic leukemia (CMML) and received at least one dose of vancomycin as documented on the medication administration record. Patients with leukemia diagnoses who displayed similar disease course and treatment were grouped together for analysis. Patients were excluded for baseline serum creatinine above the upper limit of normal as defined by the laboratory standard for the institution. During the study period, females were excluded for baseline serum creatinine >1 mg/dL and males were excluded for baseline serum creatinine >1.3 mg/dL. Patients receiving dialysis prior to vancomycin administration were also excluded. In addition, patients with vancomycin initiated before admission to the institution were excluded. Data were collected from inpatient admissions using the institutional electronic health record system. Demographic data including age, sex, specific leukemia diagnosis, disease status (i.e. active leukemia vs. leukemia in remission), and comorbid conditions including hypertension and diabetes were collected for all patients. Baseline serum creatinine was defined as the serum creatinine on the date of vancomycin initiation or within two days prior. Peak serum creatinine was defined as the highest serum creatinine during treatment with vancomycin and within seven days following vancomycin discontinuation. Nephrotoxicity was defined as a 50% rise in serum creatinine and was further classified based on the RIFLE criteria: RISK defined as baseline serum creatinine 1.5, INJURY defined as baseline serum creatinine 2, and FAILURE defined as baseline serum creatinine 3. In patients who developed nephrotoxicity, additional data were collected up to seven days following discontinuation of vancomycin. This included dialysis initiation, return of serum creatinine to normal range or to baseline in those patients who never exceeded the upper limits of the normal range, nephrotoxicity that did not resolve, and occurrence of death with nephrotoxicity. Concomitant nephrotoxic medications within two days before vancomycin initiation and throughout the seven days following vancomcyin discontinuation were collected for all patients. These medications included intravenous (IV) contrast, angiotensin-converting enzyme inhibitors or angiotensin receptor blockers, IV acyclovir, aminoglycosides, amphotericin, clofarabine, IV cytarabine, foscarnet, furosemide, IV methotrexate, and IV and oral sulfamethoxazole/trimethoprim. Concomitant rasburicase administration was also collected as an assessment of potential risk
Downloaded from opp.sagepub.com at Maastricht University on July 8, 2014
XML Template (2013) [29.10.2013–7:22am] //blrnas3/cenpro/ApplicationFiles/Journals/SAGE/3B2/OPPJ/Vol00000/130140/APPFile/SG-OPPJ130140.3d
(OPP)
[1–6] [INVALID Stage]
Fullmer et al.
3
of tumor lysis syndrome. Vancomycin trough levels, as labeled in the electronic health record system, were collected if at least three doses were administered on the medication administration record prior to obtaining the level.
Statistical analysis Statistics were calculated using JMP 10 (SAS Institute Inc, Cary, North Carolina). Descriptive statistics were used to summarize the study population. Parametric and non-parametric tests including 2, Fisher exact test, Student t test, and Wilcoxon Rank Sum test were used to evaluate variables of interest when appropriate. All comparisons were unpaired, and all tests of significance were two-tailed. A p value 0.05 was considered statistically significant. To identify factors independently associated with nephrotoxicity, multivariate logistic regression analyses were employed. Variables associated with nephrotoxicity in the univariate analysis were considered for inclusion in the models.
Results Two hundred and ninety patients admitted to the inpatient leukemia service starting 1 January 2009 were screened in order of admission date until a total of
250 patients met inclusion for review. The number of patients included was capped due to manpower/time constraints. Of the 40 patients not included in the analysis, the majority were excluded for baseline serum creatinine above the upper limit of normal (n ¼ 30). Additional patients were excluded for vancomycin administration prior to arrival at the institution (n ¼ 4), having a diagnosis not meeting inclusion criteria (n ¼ 1), lack of baseline creatinine reported prior to vancomycin initiation (n ¼ 1), and lack of documentation of vancomycin administration on medication administration record (n ¼ 4). Forty patients (16%) developed nephrotoxicity while receiving vancomycin therapy. Demographic information between patients who had an occurrence of nephrotoxicity (n ¼ 40) and those who did not (n ¼ 210) were similar (Table 1). However, active leukemia, as compared with leukemia in remission, was more common in patients who developed nephrotoxicity (p ¼ 0.005). The median peak white blood cell (WBC) count during treatment with vancomycin and within seven days following discontinuation was significantly higher in patients who developed nephrotoxicity compared to patients who did not develop nephrotoxicity (5.2 cells/mL vs.1.7 cells/mL, p ¼ 0.01). The type of leukemia did not affect the occurrence of nephrotoxicity (p ¼ 0.4459).
Table 1. Demographics. Characteristic Age (years), mean (SD) Male, no. (%) Newly diagnosed, no. (%) Active disease, no. (%) Type of leukemia AML/APL/Biphenotypic, no (%) ALL/Burkitt’s, no (%) CML, no (%) CLL/Hairy cell, no (%) MDS/Myelofibrosis/CMML, no (%) Comorbid condition Diabetes, no (%) Hypertension, no. (%) Peak WBC (cells/mL), median (IQR) Baseline serum creatinine, mean (SD) Length of therapy (days), median (IQR) Total vancomycin doses, mean (SD)
Nephrotoxicity (n ¼ 40)
No nephrotoxicity (n ¼ 210)
55.4 20 18 38
(18) (50) (45) (95)
53.1 116 95 159
(16) (55.2) (45.2) (75.7)
0.41 0.6 1 0.005
27 5 1 2 5
(67.5) (12.5) (2.5) (5) (12.5)
124 43 11 13 19
(59.1) (20.5) (5.2) (6.2) (9.1)
0.32 0.24 0.46 0.77 0.5
2 14 5.2 0.72 7.5 14.9
(5) (35) (18.8) (0.19) (7) (13.1)
25 73 1.7 0.77 3 9
(11.9) (34.8) (5.4) (0.2) (4) (8.8)
0.27 1 0.01 0.12
