Prophylactic Gentamicin Is Not Associated with Acute Kidney Injury in Patients with Open Fractures

Abstract

Background:

Data on antimicrobial prophylaxis for open fractures is limited, with many protocols based on expert recommendations. These protocols include aminoglycosides (AGs) for fractures with significant soft tissue injury, but these drugs are associated with acute kidney injury (AKI) in other settings; this risk has not been defined for open fracture prophylaxis.

Methods:

We performed a retrospective study from May 2012 to October 2014 at our Level 1 trauma center. Patients with open fractures were evaluated for demographics, location/type of fracture, injury severity, and receipt of an AG. Outcomes included rates of AKI, infection, and mortality.

Results:

There were 167 patients with open fractures during the study period (119 males, mean age 42 ± 17 [standard deviation] years), with 80 (48%) receiving prophylactic gentamicin (AG+ group). The AG+ and AG- patients had similar fracture sites and Injury Severity Scores (ISSs) (12.6 ± 9.9 AG+ vs. 15.9 ± 13.2 AG-) but were more likely to have sustained blunt trauma (96% AG+ vs. 77%; p < 0.001) or received intravenous contrast medium ≤48 h from admission (75% AG+ vs. 56% AG-; p = 0.01). Gentamicin was not associated with AKI (odds ratio [OR] 0.22; 95% confidence interval [CI] 0.020–2.44; p = 0.22), whereas hypotension on admission (OR 10.7; 95% CI 1.42–80.93; p = 0.02) and ISS (OR 1.1; 95% CI 1.01–1.20; p = 0.02) were both associated with AKI. Only four fracture site infections were identified, three in the AG+ group and one in the AG- group (3.8% vs. 1.1%; p = 0.27). The mortality rate was greater in the AG- group (3.8% vs. 12.6%; p = 0.04).

Conclusions:

Prophylactic gentamicin is not associated with AKI, whereas hypotension on admission and higher ISS were. The use of nephrotoxic agents, including aminoglycosides, should be restricted in open fracture patients presenting with hypotension or a high ISS.

Current Eastern Association for the Surgery of Trauma (EAST) guidelines addressing antibiotic prophylaxis for patients suffering open bone fractures recommend the use of an aminoglycoside (AG) in addition to an agent directed at gram-positive bacteria for patients with Gustilo type III open fractures [1]. A major clinical concern with the use of AGs as prophylactic agents is the well-described nephrotoxicity associated with this class of antibiotics. Nephrotoxicity complicates 10%–25% of AG therapeutic courses, and the risk for nephrotoxicity is increased by several factors, including the specific agent (e.g., gentamicin), greater patient age, long duration of therapy, higher AG dose, and concurrent use of other nephrotoxic agents (e.g., vancomycin, cephalosporins, contrast medium) [2]. The short duration of AG exposure recommended for patients with open fractures (<72 h) should favor a reduced risk of AG-associated nephrotoxicity, but these patients may have additional risk factors (e.g., contrast medium, cephalosporin, or vancomycin exposure) that negate the benefits of reduced AG exposure.

The few clinical trials that have examined the prophylactic use of AG in patients with open fractures have not focused on the potential nephrotoxicity of these agents, so clinicians caring for these patients have no evidence-based foundation to rely on when balancing the risks and benefits of this prophylactic tactic [3,4]. We performed a retrospective study to examine the possible nephrotoxicity associated with prophylactic AG use in patients with open fractures at a Level 1 trauma center.

Patients and Methods

This study was designed as a retrospective, single-center protocol in a 537-bed county hospital serving as the regional Level 1 trauma center. Consecutive patients with open fractures evaluated at this facility between May 2012 and October 2014 were identified from the trauma registry and included in the study. Data pertaining to demographics, injury type/severity, medications, laboratory/radiographic findings, and clinician assessments for fracture site infections were extracted from the trauma registry and electronic medical record. Acute kidney injury (AKI) was defined as any increase in the serum creatinine concentration ≥50% above the patient's admission value. Statistical analysis was done with SAS software version 12.1 (SAS Institute, Inc., Cary, NC.). Categorical variables were analyzed with χ² tests, and continuous variables were compared through Student t-tests. Logistic regression analysis was done with AKI as the dependent variable and receipt of AG (yes, no), hypotension at admission (systolic blood pressure <100 mm Hg), and Injury Severity Score (ISS) as independent variables.

Results

A total of 167 patients were included in this study, and Table 1 outlines the baseline demographics and additional clinical variables based on whether the patients received any dose of AG during the incident admission. The only AG used during this study period was gentamicin. Eighty patients received at least one dose of gentamicin, and the mean daily dose was 346.4 mg (range 80–640 mg), with a mean duration of 3.5 d (range 1–18 d). The AG+ and AG- cohorts had similar baseline clinical characteristics (Table 1), with a few exceptions: (1) The AG+ group had a higher percentage of patients with penetrating trauma (23% vs. 3.8%); (2) there was a difference in the admission systolic blood pressure between the groups; (3) a higher percentage of AG+ patients received intravenous (IV) contrast medium; and (4) there was a higher mortality rate in the AG- group. We identified four fracture site infections, three in the AG+ group and one in the AG- group (3.8% vs. 1.1%; p = 0.27). Acute kidney injury developed in six of the patients (3.6%), with one patient in the AG+ cohort (1.25%) and five in the AG- cohort (5.75%; p = 0.10 for the comparison).

Table 1.

Aminoglycoside (AG) Use for Prophylaxis in Patients with Open Fractures

	Total (n = 167)	AG- (n = 87)	AG+ (n = 80)	p^a
Average age (y) ± SD	42.1 ± 17.3	42.7 ± 19.0	41.4 ± 15.4	0.62
≤17	6	4	2
18–64	143	70	73
65+	18	13	5
Gender
Male (%)	119 (71.3)	64 (73.6)	55 (68.8)	0.49
Female (%)	48 (28.7)	23 (26.4)	25 (31.2)
Fracture (%)
Lower extremity	134 (80.2)	67 (77)	67 (83.8)	0.27
Upper extremity	27 (16.2)	17 (19.6)	10 (12.5)
Both	6 ( 3.6)	3 ( 3.4)	3 ( 3.8)
Type of injury (%)
Blunt	144 (86.2)	67 (77)	77 (96.3)	<0.001
Penetrating	23 (13.8)	20 (23)	3 ( 3.8)
Injury Severity Score
<15	112	54	58	0.10
15–24	18	8	10
25–75	37	25	12
Abbreviated Injury Scale, Extremities/Pelvis	2.52	2.51	2.54	0.73
Average systolic blood pressure ± SD	129.3 ± 30.1	124.7 ± 33.8	134.0 ± 25.0	0.05
Average admission serum creatinine concentration (±SD)	1.0 ± 0.3	1.1 ± 0.4	1.0 ± 0.3	0.48
Hypotension (%)	13.8	16.0	11.5	0.41
IV contrast medium (%)	65.3	56.3	75.0	0.01
Co-morbidity (%)	37.1	41.4	32.5	0.24
Fracture site infection (%)	2.4	1.1	3.8	0.27
Mortality rate (%)	8.4	12.6	3.8	0.04

T statistic or χ² statistic computed for differences. Factors with statistically significant differences are shown in boldface type.

Table 2 displays the results of the logistic regression analysis. Receipt of gentamicin was not associated with AKI. On the other hand, hypotension was strongly associated with AKI, and the ISS demonstrated a modest association with AKI. All of the patients with AKI had received IV contrast medium, but we were unable to resolve the regression analysis utilizing IV contrast as an independent variable despite the clear association with AKI.

Table 2.

Multivariable Odds Ratios for Predicting Acute Kidney Infection (AKI) (C = 0.92) ^a

Variable	Odds Ratio	p	95%CI
AG	0.22	0.22	0.020– 2.436
Hypotension	10.73	0.02	1.423–80.932
ISS	1.10	0.02	1.014– 1.198

Logistic regression model with dependent variable AKI and independent variables AG, hypotension, and ISS.

AG = aminoglycoside; AKI = acute kidney injury; CI = confidence interval; ISS = Injury Severity Score.

Discussion

Nephrotoxicity associated with AG is a well-described adverse event and common barrier to broader use of this class of antibiotics. Aminoglycoside-induced proximal tubular damage in the kidney is closely tied to drug accumulation in the proximal tubular cells; and a host cation transporter, megalin/cubilin, mediates this accumulation via endocytosis [2]. Downstream events that follow AG endocytosis include both apoptosis of tubular epithelial cells, as well increased local inflammation and subsequent inflammatory damage, leading to overt tubular dysfunction and AKI. The decision to use AG therefore must weigh the risks of drug toxicity, mainly in the forms of nephrotoxicity and ototoxicity, with the benefits afforded by these broadly active antibiotics. We performed this retrospective study of patients with open fractures specifically to help inform clinicians of the potential risk of nephrotoxicity associated with the prophylactic use of AG for this patient population.

Our study did not find an association between gentamicin, the only AG used during the study period at our hospital, and AKI in patients with open fractures. The extent of AG exposure in our study was modest, with an average daily dose of ∼350 mg and a mean duration of 3.5 days. These figures are consistent with the recommended use of AG for prophylaxis of open fractures proposed in the EAST guidelines [1]. There was a statistically significant difference between the admission systolic blood pressure of the AG+ and AG- groups, but the clinical significance of the absolute mean difference (9.3 mm Hg) is unclear. However, we did find a strong association between hypotension at the time of admission and AKI. Hypotension has been described as a risk factor for AKI among diverse populations, including trauma patients, with kidney hypoperfusion and acute tubular necrosis as the primary mechanisms [5,6]. We also found a more modest association between ISS and AKI in this study, an association that has been variably described in the literature [7,8]. Intravenous contrast medium administration clearly was associated with AKI, but we were unable to assess the statistical impact of this well-described risk factor because of an unsolvable regression equation created by the small sample size of this study.

Alternative prophylactic regimens for open fractures have been investigated to a limited degree. Patzakis et al. performed a randomized, double-blind study that compared ciprofloxacin with the combination of cefamandole plus gentamicin for the prophylaxis of open fractures [3]. They found a much higher, although not significantly so, infection rate for patients with Gustilo class III fractures receiving ciprofloxacin compared with the combination (31% vs. 7.7%; p = 0.079). This study did not specifically delineate AKI as a clinical outcome but reported no antibiotic-related toxicities in the cefamandole/gentamicin group. Sorger et al. performed a randomized trial of once-daily high-dose (6 mg/kg) gentamicin compared with conventionally dosed gentamicin (2.5 mg/kg twice daily) for prophylaxis in patients with Gustilo class II and III open fractures [4]. The primary outcome of interest was the efficacy of each regimen in infection prevention, and the authors found no difference between them. None of the patients in this study developed AKI, although the authors did note a “mild increase” in the blood urea nitrogen/creatinine ratios after 7–10 d of twice-daily gentamicin exposure in two patients; these ratios normalized after discontinuation of gentamicin prophylaxis. Rodriguez et al. performed a retrospective study comparing clinical outcomes in trauma patients at a single institution before and after the implementation of an open fracture prophylaxis protocol that excluded the use of aminoglycosides, vancomycin, and penicillin [9]. The new protocol included the following antimicrobial agents for 48 h according to Gustilo grade: Patients with fractures of grades I/II received cefazolin (clindamycin if allergic); patients with grade III fractures received ceftriaxone (clindamycin and aztreonam if allergic). The authors did not find a difference in the incidence of skin/soft tissue infections between the pre-protocol and post-protocol cohorts, but they did observe a significant decrease in the use of aminoglycosides and vancomycin (53.5% pre-protocol vs. 16.4% post-protocol; p = 0.0001). This study supports the use of alternative antimicrobial agents, particularly cephalosporins, for open fracture prophylaxis but is limited by the single-center design and the evaluation of groups of patients from different time periods.

The limitations of our study include the single-center, retrospective design; the small number of patients with the outcome of interest (six cases of AKI); and the lack of matching between the AG+ and AG- cohorts. However, the cohorts were similar with respect to demographic traits, frequency of co-morbidities, and injury severity. The higher rate of contrast medium exposure among the AG+ patients would be expected to favor more AKI in this group, as additional nephrotoxins are associated with a greater risk of AG-associated nephrotoxicity. The lack of nephrotoxicity in the AG+ group supports the current EAST guideline recommendations for the prophylactic use of AG in open fracture protocols.

Footnotes

Author Disclosure Statement

The authors have no conflicts of interest with regard to this manuscript.

References

Hoff

, Bonadies

, Cachecho

, Dorlac

. East Practice Management Guidelines Work Group: Update to practice management guidelines for prophylactic antibiotic use in open fractures. J Trauma, 2011; 70:751–754.

Lopez-Novoa

, Quiros

, Vicente

, et al. New insights into the mechanism of aminoglycoside nephrotoxicity: An integrative point of view. Kidney Int, 2011; 79:33–45.

Patzakis

, Bains

, Lee

, et al. Prospective, randomized, double-blind study comparing single-agent antibiotic therapy, ciprofloxacin, to combination antibiotic therapy in open fracture wounds. J Orthop Trauma, 2000; 14:529–533.

Sorger

, Kirk

, Ruhnke

, et al. Once daily, high dose versus divided, low dose gentamicin for open fractures. Clin Orthop Relat Res, 1999;(366):197–204.

Liu

, Prowle

, Licari

, et al. Changes in blood pressure before the development of nosocomial acute kidney injury. Nephrol Dial Transplant, 2009; 24:504–511.

Panwar

, Lanyon

, Davies

, et al. Mean perfusion pressure deficit during the initial management of shock: An observational cohort study. J Crit Care, 2013; 28:816–824.

Bagshaw

, George

, Gibney

, Bellomo

. A multi-center evaluation of early acute kidney injury in critically ill trauma patients. Ren Fail, 2008; 30:581–589.

Skinner

, Hardcastle

, Rodseth

, Muckart

. The incidence and outcomes of acute kidney injury amongst patients admitted to a level I trauma unit. Injury, 2014; 45:259–264.

Rodriguez

, Jung

, Goulet

, et al. Evidence-based protocol for prophylactic antibiotics in open fractures: Improved antibiotic stewardship with no increase in infection rates. J Trauma Acute Care Surg, 2014; 77:400–408.