Q J Med 2002; 95: 291-297
© 2002 Association of Physicians
Risk factors for nephrotoxicity in elderly patients receiving once-daily aminoglycosides
1 From the Infectious Diseases Unit, 2 Department of Geriatrics, and 4 Clinical Microbiology Laboratory, Shaare Zedek Medical Center, Jerusalem, affiliated with the Faculty of Health Sciences, Ben Gurion University of the Negev, Be'er-Sheba, and 3 Department of Public Health, Hebrew University Hadassah Medical School, Jerusalem, Israel
Received 30 August 2001 and in revised form 28 January 2002
 |
Summary |
|---|
 Top Summary IntroductionMethodsResultsDiscussionReferences |
|---|
Background: There remain concerns about the safety of once-daily dosing of aminoglycosides (AGs) in the elderly.
Aim: To assess the safety of once-daily AGs in elderly patients and evaluate possible risk factors for nephrotoxicity.
Design: Prospective, non-interventional surveillance study.
Methods: All patients receiving AGs were monitored over 4 months. Clinicians determined the AG dose for each patient after estimating patient weight and calculating creatinine clearance (CrCl) using the Cockcroft-Gault formula. Parallel figures were calculated by the investigators using measured weight. Clinicians obtained an AG trough level 24 h after initiation of treatment, and, if non-toxic, every 5–7 days thereafter. AG toxicity was defined as an increase in serum creatinine of 
50%.
Results: In the study period, 249 consecutive patients received an AG: 116 (47%) males, mean±SD age 75±16 years. Forty-two (17%) received amikacin and 207 (83%) gentamicin. An increase of 50% in serum creatinine was detected in 31/249 (12.4%); maximal creatinine was 
177 µmol/l in 16/249 (6.4%), 186–265 µmol/l in nine (3.6%), and >265 µmol/l in six (2.4%). None developed oliguric renal failure. Renal damage correlated with a high AG trough level (>1.1 µg/ml) (p<0.001); haemoglobin level <10 g/dl (p<0.05); hospital admission >7 days prior to AG treatment (p<0.005); and AG treatment 11 days (p<0.05). Mean CrCl based on estimated weight was 52±18 ml/min; that based on actual weight was 71±37 ml/min. Despite this, mean AG dose was 1.3±0.6 higher than optimal.
Conclusions: Oliguric and/or lasting renal toxicity is rare in elderly patients receiving once-daily aminoglycosides for <11 days, if regular trough drug levels are monitored.
|
Introduction |
|---|
|
TopSummaryIntroductionMethodsResultsDiscussionReferences |
|---|
Aminoglycosides are the mainstay of antimicrobial treatment of infections caused by Gram-negative bacilli, including intra-abdominal and urinary tract infections, and diverse hospital-acquired infections.1 Although various side-effects have been reported, including ototoxicity and neuromuscular junction blockade, the major drawback of aminoglycoside usage is nephrotoxicity.1,2 Several risk factors for development of this complication have been identified, including diabetes mellitus, duration of treatment, dehydration, and advanced age.1–6 However, there continues to be debate in the literature about whether age is a risk factor for aminoglycoside-induced nephrotoxicity and to what extent.1
Several publications in the last decade have demonstrated the efficacy of once-daily aminoglycoside therapy as opposed to that of traditional twice or three times daily dosing.7–13 This regimen is essentially based on the post-antibiotic effect of aminoglycosides, a property absent in ß-lactam antimicrobials.14 Comparative studies have shown that once-daily dosing is not generally more nephrotoxic than conventional dosing.3 However, there remain concerns about the safety of once-daily dosing in several subgroups of patients, e.g. neonates, pregnant women and elderly patients. We were interested in assessing the safety of once-daily aminoglycoside treatment in the elderly, because a large percentage of patients admitted to our hospital are older than 65 years, and many of these patients suffer from infections that may be treated with aminoglycosides. In addition, we wished to further identify risk factors for aminoglycoside nephrotoxicity, including deviation from ideal body weight (obese and cachectic patients), anaemia and hypoalbuminaemia. In a prospective study of elderly patients receiving aminoglycoside treatment, we analysed these risk factors, along with established risk factors such as diabetes mellitus, length of aminoglycoside treatment, length of hospital stay and dehydration.
|
Methods |
|---|
|
TopSummaryIntroductionMethodsResultsDiscussionReferences |
|---|
The study was conducted in Shaare Zedek Medical Center, a 550-bed university-affiliated institution. The hospital contains all major services. The hospital serves the general population of Jerusalem; a substantial proportion of the patients are elderly, and many are referred from nursing homes. The infectious diseases unit has issued protocols for empiric antimicrobial use for the most frequently encountered conditions, including treatment and prophylaxis. As a matter of policy, aminoglycosides are the drugs of choice for empiric treatment of infections caused by Gram-negative bacilli, e.g. intra-abdominal and urinary tract infections and most nosocomial infections. Gentamicin is the most-commonly used aminoglycoside; amikacin is reserved for serious nosocomial infections in which resistance to gentamicin is anticipated, especially in the intensive care unit or for patients who have received multiple courses of broad-spectrum antimicrobials.15 In instances where organisms other than Gram-negative bacilli are expected to be involved, another agent is added to cover the additional organisms.
Prior to administration of aminoglycosides, clinicians are expected to estimate the creatinine clearance in ml/min, using the Cockcroft-Gault formula, which uses age in years, ideal body weight in kg (see below), and serum creatinine:16,17 the appropriate aminoglycoside dose for each individual patient is determined with dosing tables.1 When the estimated creatinine clearance is <30 ml/min, the clinicians are advised to use an alternative agent to aminoglycosides, such as fluoroquinolones, third or fourth generation cephalosporins, or piperacillin-tazobactam, aztreonam or a carbapenem. Our hospital revised aminoglycoside therapy from multiple-daily dosing to once-daily dosing in 1996. Clinicians are expected to obtain a serum sample for trough level testing 24 h after the initial dose has been administered, just prior to the second dose. If the level is <1 µg/ml, repeat trough levels are determined every 5 days, or sooner if an increase in creatinine or a decrease in urine output is detected. If the trough level is above the threshold of toxicity, the interval between doses is increased and a repeat trough level is obtained to ensure safety of treatment.1 Aminoglycoside drug levels are determined with fluorescence polarization immunoassay technology and microparticle enzyme immunoassay technology (Abbott GmbH Diagnostika).
In this prospective, non-interventional study, we screened all in-patients receiving aminoglycoside treatment (gentamicin or amikacin). The parameters recorded in the questionnaire were: patient age; gender; weight—actual, ideal, and weight index (actual weight/ideal weight); height; body mass index (weight/height2); and creatinine clearance (CrCl) as estimated by the clinicians using the Cockcroft-Gault formula,16,17 mostly using an eye-estimation of body weight; creatinine clearance as calculated by the investigators after actually weighing the patients, using Cockcroft's formula; and creatinine clearance as calculated by the investigators, using the ideal body weight with Cockcroft's formula. The latter formula measures CrCl in ml/min as follows: [(140-age in years)xideal body weight in kg]/[serum creatinine in µmol/lx72]. Additional parameters that were assessed include the following: admitting department; infectious diagnosis; presence of diabetes mellitus; type of aminoglycoside used; duration of aminoglycoside treatment; trough levels; all medications the patient received; initial haemoglobin; initial albumin; level of dehydration on admission (as defined by the BUN/creatinine ratio); total days of hospital stay up to initiation of aminoglycoside treatment; initial blood pressure; body temperature on admission; arterial pH on admission; initial creatinine; creatinine ratio (ratio between the highest serum creatinine and initial serum creatinine level); and lastly ratio of aminoglycoside dose actually given divided by the optimal dose (if the weight index was >1, the ideal dose was calculated by the ideal body weight, whereas if the weight index was 1, the ideal dose was calculated by actual body weight). The ideal body weight for men was 50 kg, and for women 45.5 kg, to which 0.9 kg was added for each cm over 150 cm height;18 the investigators calculated the ideal body weight for each patient after measuring their height. We also evaluated the usefulness of Levey's new formula for calculation of the glomerular filtration rate,19 as compared to that of Cockcroft's widely used formula.16,17 A significant deterioration in renal function was defined as an increase in serum creatinine of more than 50% from prior-to-treatment to maximal, post-initiation-of-treatment creatinine levels.1,7,9–12
We used the Epi Info 6.04c program (CDC) for data entry and statistical analysis. The statistical tests employed were Student's t test, 
2 and linear regression. For logistic regression analysis, we used the SPSS-X program.
|
Results |
|---|
|
TopSummaryIntroductionMethodsResultsDiscussionReferences |
|---|
During the 4-month study period, 249 consecutive patients receiving aminoglycosides were enrolled for data collection and analysis, of whom 116 (47%) were male and 133 (53%) female. Mean±SD age was 75±16 (range 21–102); 205 patients (82%) were >65 years old. Forty-two patients (17%) received amikacin, 207 gentamicin (83%). The majority of patients (237, 95%) were admitted to the medical and geriatric departments, and the remaining 12 (5%) to surgical departments. Baseline demographic, clinical and laboratory data are presented in Table 1
.
|
A significant deterioration in renal function, defined as an increase of
50% in serum creatinine,1,7,9–11 was detected in 31 of the 249 patients (12.4%). The maximal creatinine was 177 µmol/l in 16 of the 249 patients (6.4%), 186–265 µmol/l in nine patients (3.6%), and >265 µmol/l (327–769 µmol/l) in six patients (2.4%). None of the patients developed oliguric renal failure; most subsequently experienced significant reduction of serum creatinine prior to discharge from hospital. In the group of 31 patients who experienced an increase of 50% in creatinine, the mean±SD initial creatinine (in µmol/l) was 79±24 (44–133), the mean maximal creatinine was 222±179 (71–875), and the mean creatinine ratio was 234±156 (137–767).
Tables 2 and 3 show, respectively, the univariate analysis of clinical and laboratory risk factors for deterioration of renal function, defined by increase of 50% in serum creatinine. Concomitant drugs were grouped in 12 categories: additional antibiotics, diuretics, angiotensin converting enzyme inhibitors, beta-blockers, calcium antagonists, serum-cholesterol-reducing agents, warfarin, glucose-reducing agents, digoxin, non-steroidal anti-inflammatory agents, nitrates, minor tranquillizers, and miscellaneous drugs. Only diuretics, length of hospital admission prior to aminoglycoside treatment and total duration of aminoglycoside treatment 11 days, were statistically linked to aminoglycoside-associated nephrotoxicity (Table 2). Table 3 demonstrates the very significant correlation between renal damage and a high aminoglycoside trough level, and a weaker, but still significant correlation with anaemia. Mean duration of treatment was 4.4 days, with 72 (29%) of patients receiving >6 days of aminoglycosides. No particular distinguishing features of the latter group could be detected.
|
|
Linear regressions between the glomerular filtration rate, calculated by Levey's formula,19 and our estimations of creatinine clearance using Cockroft's formula, are shown in Table 4
. From our study we cannot conclude which formula yields the most accurate result. However, it is clear that the department estimations were based, in many instances, on estimated rather than on actually measured weight; the department estimates were far below the calculated ones, by either method, and their correlation was poor. The creatinine clearance as calculated by Cockroft's formula had much higher correlation with Levey's formula, especially when it was calculated with the ideal body weight.
|
Logistic regression analysis of our data used various statistical models and logarithmic transformations. Several models were limited due to selections and/or lack of sufficient observations, e.g. arterial pH was obtained from 60 patients only. From the various models the following picture emerged: a maximal aminoglycoside trough level >1.1 µg/ml was far ahead in the regression, and in most models a baseline pH <7.38 was closely associated with subsequent renal failure. Following these two variables were a haemoglobin level <10 g/dl, a diastolic blood pressure <70 mmHg and total admission >7 days prior to aminoglycoside treatment.
|
Discussion |
|---|
|
TopSummaryIntroductionMethodsResultsDiscussionReferences |
|---|
Worldwide, elderly patients account for an increasing percentage of admitted patients to most hospitals. Many of these patients are treated for urinary tract infections or hospital-acquired infections, often caused by Gram-negative bacilli. In our hospital, as in most, empiric treatment of these infections is with an aminoglycoside. Only if the estimated creatinine clearance is calculated to be relatively low, e.g. <25–30 ml/min, are clinicians advised to use alternative agents such as fluoroquinolones, third or fourth generation cephalosporins, or piperacillin-tazobactam, aztreonam or a carbapenem. Use of aminoglycosides, when possible, is still recommended despite potential side effects, in order to limit the emergence of multidrug-resistant organisms and to curtail the rising cost of antimicrobial therapy. Continuing concern, however, about the safety of this policy, especially in elderly patients, led to the present study. We attempted to establish the safety of once-daily aminoglycosides in the elderly, and to determine risk factors for development of nephrotoxicity.
We prospectively assessed the safety of once-daily aminoglycosides in 249 elderly patients (mean age 75±16) over a 4-month period. An increase of 50% in serum creatinine was detected in 31 of the 249 patients (12.4%), a figure very similar to that reported from similar series, including younger patients, receiving aminoglycoside therapy.1–6 The maximal creatinine was 177 µmol/l in 16/249 patients (6.4%), while in 15 (6%) it was >186 µmol/l. None of the patients developed oliguric renal failure, and all subsequently experienced significant reduction of serum creatinine prior to discharge from hospital. Univariate and logistic regression analysis revealed a very significant correlation between renal damage and a high aminoglycoside trough level (>1.1 µg/ml) (p<0.001), and a weaker, but still significant correlation with a haemoglobin level <10 g/dl (p<0.05), duration of treatment 11 days (p<0.05), and length of hospital admission of >7 days prior to aminoglycoside treatment (p<0.05). The latter factor probably reflects the complexity and fragility of patients staying longer in the hospital. It therefore appears that oliguric and/or lasting renal toxicity is uncommon in elderly patients receiving once-daily aminoglycosides. Age, diabetes mellitus, initial creatinine and dehydration were not associated with toxicity. None of the 15 patients who experienced an increase of >50% in serum creatinine levels and whose highest creatinine level was >177 µmol/l had diabetes mellitus.
During the study, we realized that most of the patients were not weighed routinely, and weight was determined by the clinicians by eye-estimation only. The results of the clinicians' calculations were compared with those based upon actual weight as measured by the investigators. Despite the apparent underestimation of the patients' weight and, therefore, creatinine clearance by the clinicians, a mean aminoglycoside dose that was 1.3±0.6 higher than optimal was prescribed (Table 1
). This apparent contradiction suggests that clinicians often do not properly use creatinine clearance for calculation of optimal aminoglycoside dosing. The simple and time-honoured method of weighing patients (possibly by using special devices for bedridden patients), or alternatively, of measuring the patient's height and thereby calculating ideal body weight, plus proper usage of derived creatinine clearance for calculation of aminoglycoside dosing, might lead to more appropriate therapy and thereby diminish the risk of nephrotoxicity.
We recently conducted a prospective study of 158 consecutive patients receiving aminoglycosides during a 2-month period, comparing results of patients <65 years old with those of older patients.21 Nephrotoxicity, defined as an increase of 50% in creatinine levels during and up to 48 h post-treatment, occurred in 4/66 patients (6%) <65 years of age, compared to 10/92 patients (11%) aged 65 years or older (NS). Overall, nephrotoxicity occurred in 8.9% of patients, an incidence very similar to that reported in the literature2,7–11,22 and found in the current study; none developed irreversible renal failure. Univariate analysis showed that an aminoglycoside trough level of >2.5 µg/ml (p<0.001), presence of diabetes mellitus (p<0.01) and pneumonia (p<0.01) were risk factors for aminoglycoside-associated nephrotoxicity.21 Age was not found to be an independent risk factor for aminoglycoside-induced nephrotoxicity. Additional factors may determine nephrotoxicity in the elderly, including duration of treatment, presence of diabetes mellitus, liver disease and hypokalaemia.4–6,23,24
Advanced age has been associated with an increased incidence of aminoglycoside nephrotoxicity, especially in rats.25–28 Subclinical evidence of impaired renal function was also demonstrated in humans.29 Studies of risk factors for aminoglycoside nephrotoxicity in humans concluded that age was a risk factor, although statistical significance was found only in complex multivariate analyses.4–6 The correlation of increased risk of toxicity with age and/or pre-existing renal disease may be misleading, and it is unclear whether an increased risk exists when the dosing regimen is adjusted for pre-existing decrease in glomerular filtration rate.1
In summary, in this study of 249 mostly elderly patients receiving once-daily aminoglycosides, only 31 patients (12.4%) developed nephrotoxicity as defined by a >50% increase in serum creatinine level, and in only six patients (2.4%) did the creatinine reach >265 µmol/l. None of these patients developed oliguric and/or persisting renal disease. Trough drug levels must be measured. Estimation of creatinine clearance with Cockcroft's formula should be done after actually weighing the patients, rather than estimating their weight, with greater adherence to dosage tables, when calculating the appropriate aminoglycoside dose. These actions may render the use of once-daily treatment with aminoglycosides safe even in the elderly.
|
Notes |
|---|
Address correspondence to Dr A.M. Yinnon, Infectious Diseases Unit, Shaare Zedek Medical Center, PO Box 3235, Jerusalem 91031, Israel. e-mail: Yinnon{at}SZMC.org.il
|
References |
|---|
|
TopSummaryIntroductionMethodsResultsDiscussionReferences |
|---|
1. Gilbert DN. Aminoglycosides. In: Mandell GL, Bennett JE, Dolin R, eds. Principles and Practice of Infectious Diseases, 5th edn. New York, Churchill Livingstone, 2000:307–36.
2.
Mingeot-Leclercq MP, Tulkens PM. Aminoglycosides: nephrotoxicity. Antimicrob Ag Chemother1999; 43:1003–12.
3. Ali MZ, Goetz MB. A meta-analysis of the relative efficacy and toxicity of single daily dosing versus multiple daily dosing of aminoglycosides. Clin Infect Dis1997; 24:796–809.[Web of Science][Medline]
4. Moore RD, Smith CR, Lipsky JJ, Mellits ED, Lietman PS. Risk factors for nephrotoxicity in patients treated with aminoglycosides. Ann Intern Med1984; 100:352–7.
5. Sawyers CL, Moore RD, Lerner SA, Smith CR. A model for predicting nephrotoxicity in patients treated with aminoglycosides. J Infect Dis1986; 153:1062–8.[Web of Science][Medline]
6. Smith CR, Moore RD, Lietman PS. Studies of risk factors for aminoglycoside nephrotoxicity. Am J Kidney Dis1986; 8:308–13.[Web of Science][Medline]
7. Prins JM, Bueller HR, Kuijper EJ, Tange RA, Speelman P. Once versus thrice daily gentamicin in patients with serious infections. Lancet1993; 341:335–9.[Web of Science][Medline]
8. Sturm AW. Netilmicin in the treatment of gram-negative bacteremia: single daily versus multiple daily dosage. J Infect Dis1989; 159:931–7.[Web of Science][Medline]
9.
Hatala R, Dinh T, Cook DJ. Once-daily aminoglycoside dosing in immunocompetent adults: a meta-analysis. Ann Intern Med1996; 124:717–25.
10.
Barza M, Ioannidis JPA, Cappelleri JC, Lau J. Single or multiple daily doses of aminoglycosides: a meta-analysis. Br Med J1996; 312:338–45.
11.
Gilbert DN. Once-daily aminoglycoside therapy. Antimicrob Ag Chemother1991; 35:399–405.
12.
Freeman CD, Nicolau DP, Belliveau PP, Nightingale CH. Once-daily dosing of aminoglycosides: review and recommendations for clinical practice. J Antimicrob Chemother1997; 39:677–86.
13. Bertino JS, Rotschafer JC. Editorial response: single daily dosing of aminoglycosides: a concept whose time has not yet come. Clin Infect Dis1997; 24:820–3.[Web of Science][Medline]
14. Vogelman BS, Craig WA. Postantibiotic effects. J Antimicrob Chemother1985; 15(Suppl. A):37–46.
15. Yinnon AM, Schlesinger Y, Gabbay D, Rudensky B. Analysis of 5 years of bacteraemias: importance of stratification of microbial susceptibilities by source of patient. J Infect1997; 35:17–23.[Web of Science][Medline]
16. Cockcroft DW, Gault MH. Prediction of creatinine clearance from serum creatinine. Nephron1976; 16:31–41.[Web of Science][Medline]
17. Gault MH, Longerich LL, Harnett JD, Wesolowski C. Predicting glomerular function from adjusted serum creatinine. Nephron1992; 62:249–56.[Web of Science][Medline]
18. Halsted CH. Malnutrition and nutritional assessment. In: Braunwald E, Fauci AS, Kasper DL, Hauser SL, Longo DL, Jameson JL, eds. Harrison's Principles of Internal Medicine, 15th edn. New York, McGraw-Hill, 2001:458.
19.
Levey AS, Bosch JP, Breyer Lewis J, Greene T, Rogers N, Roth D. A more accurate method to estimate glomerular filtration rate from serum creatinine: a new prediction equation. Ann Intern Med1999; 130:461–70.
20. Walker RJ, Duggin GG. Drug nephrotoxicity. Annu Rev Pharmacol Toxicol1988; 28:331–45.[Web of Science][Medline]
21. Susser Z, Raveh D, Yinnon AM. Safety of once-daily aminoglycosides in the elderly. J Am Geriatr Soc2000; 48:857–9.[Web of Science][Medline]
22. Lietman PS, Smith CR. Aminoglycoside nephrotoxicity in humans. Rev Infect Dis1983; 5(Suppl. 2):S284–93.
23. Lietman PS. Liver disease, aminoglycoside antibiotics, and renal dysfunction. Hepatology1988; 4:966–8.
24. Thompson JR, Simonsen R, Spindler MA, Southern PM, Cronin RE. Protective effect of KCl loading in gentamicin nephrotoxicity. Am J Kidney Dis1990; 15:583–91.[Web of Science][Medline]
25. Marre R, Tarara N, Louton T, Sack K. Age-dependent nephrotoxicity and the pharmacokinetics of gentamicin in rats. Eur J Pediatr1980; 133:25–9.[Web of Science][Medline]
26. Provoost AP, Adejuyigbe O, Wolff ED. Nephrotoxicity of aminoglycosides in young and adult rats. Pediatr Res1985; 19:1191–6.[Medline]
27.
Beauchamp D, Gourde P, Thereault G, Bergeron MG. Age-dependent gentamicin experimental nephrotoxicity. J Pharmacol Exp Ther1992; 260:444–9.
28. McMartin DN, Engel SG. Effect of aging on gentamicin nephrotoxicity in rats. Res Commun Chem Pathol Pharmacol1982; 38:193–207.[Web of Science][Medline]
29. Fujita K, Sayama T, Abe S, Murayama T, Tashiro H. Age-dependent aminoglycoside nephrotoxicity. J Urol1985; 134:596–7.[Web of Science][Medline]
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  N. M. Selby, S. Shaw, N. Woodier, R. J. Fluck, and N. V. Kolhe Gentamicin-associated acute kidney injury QJM, December 1, 2009; 102(12): 873 - 880. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 L. Leibovici, L. Vidal, and M. Paul Aminoglycoside drugs in clinical practice: an evidence-based approach J. Antimicrob. Chemother., February 1, 2009; 63(2): 246 - 251. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Freundlich, R. W. Thomsen, L. Pedersen, H. West, and H. C. Schonheyder Aminoglycoside treatment and mortality after bacteraemia in patients given appropriate empirical therapy: a Danish hospital-based cohort study J. Antimicrob. Chemother., November 1, 2007; 60(5): 1115 - 1123. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||