Cochleotoxicity and acute kidney injury secondary to parenteral gentamicin administration in a dog

Abstract

A 5-year-old neutered male Labrador retriever dog was presented on referral for anuric acute kidney injury (AKI) presumed secondary to parenteral gentamicin administration. Initial management of AKI included a high dose of furosemide for 16 hours which may have contributed to the renal and cochlear damage. The dog received a total of 8 intermittent hemodialysis sessions during hospitalization. While in hospital, the dog became deaf, and brainstem auditory evoked response testing revealed absence of waveforms in both ears, consistent with bilateral deafness due to severe cochleotoxicity. After 33 days of hospitalization, the dog was discharged with persistent deafness, confirmed by a second brainstem auditory evoked response.

Aminoglycosides are concentration dependent, bactericidal antibiotics which inhibit bacterial protein synthesis. They bind the prokaryotic 30S ribosomal subunit, leading to misreading of mRNA and production of incorrect proteins (1). This effect is specific to prokaryotic RNA, although there are similar subunits in mammalian mitochrondrial RNA which are a potential target for toxicity (2). Parenteral administration of aminoglycosides has been documented to cause nephrotoxicity and ototoxicity in both humans and animals (3–5). The parenteral use of these drugs is generally limited to serious Gram-negative bacterial infections for which other antimicrobials are not effective or cases in which susceptibility cannot be established before treatment. Gentamicin, however, is commonly used for topical treatment of otitis externa in companion animals (6). The incidence of ototoxicity with gentamicin therapy is not well-documented in veterinary patients, although both vestibular and cochlear effects have been described in cats receiving parenteral gentamicin (7).

Accumulation of aminoglycosides in the renal tubular cells can cause proximal tubular necrosis (8). Severe toxicosis can lead to persistent renal anuria, the prognosis of which is poor without renal replacement therapy (9). The use of hemodialysis in companion animals is expanding; however, its availability remains limited. The primary indications for hemodialysis in domestic animals include refractory azotemia, severe electrolyte disturbances, volume overload, and certain acute toxicoses (9). Furosemide is frequently used in veterinary medicine for management of acute kidney injury (AKI) not responsive to IV fluid therapy alone, despite lack of evidence supporting this approach. In humans, the use of furosemide in AKI has not been shown to improve mortality rates but may be indicated in cases of AKI and concurrent acute lung injury (10). The combination of furosemide therapy with aminoglycosides can exacerbate the damage to the vestibulocochlear system (5). This report details an unusual case of combined cochleotoxicity and acute kidney injury secondary to parenteral gentamicin administration in a dog.

Case description

A 5-year-old neutered male Labrador retriever dog weighing 41.3 kg was presented on referral for AKI, International Renal Interest Society grade V, subgrade oligo-anuric, and requiring renal replacement therapy (11). The dog was diagnosed with otitis externa of the left ear that had been unresponsive to topical and empiric enteral antibiotic therapy for several months. An aerobic culture from the left ear canal identified a beta hemolytic Streptococcus. Based on susceptibility results, gentamicin therapy was initiated at a dose of 12 mg/kg body weight (BW) per day subcutaneously for 13 d. Ten days into the course of treatment, the dog was lethargic and hyporexic, and progressed to anorexia and vomiting. The dog was evaluated by his primary care veterinarian who conducted blood analysis which revealed a severe azotemia with a creatinine of 1105 μmol/L [reference range (RR): 44.2 to 159.1 μmol/L] and a blood urea nitrogen (BUN) > 46 mmol/L (RR: 2.5 to 9.6 mmol/L), and hyperphosphatemia (3.8 mmol/L, RR: 0.9 to 1.8 mmol/L). Urinalysis revealed a urine specific gravity of 1.007, hematuria (50 cells/μL), pyuria (100 cells/μL), proteinuria (1 g/L), and glucosuria (5.55 mmol/L). Initial treatment included crystalloid therapy (intravenous Ringer’s lactate solution, 50 mL/kg BW bolus, then 8.5 mL/kg BW per hour), maropitant (1 mg/kg BW, SC, once), and famotidine (1 mg/kg BW, IV, once).

The dog was transferred to a local referral hospital in the afternoon. On presentation, the bladder was small and an indwelling urinary catheter was placed to quantify urine output. Intravenous crystalloids were continued at 7.5 mL/kg BW per hour. A continuous rate infusion (CRI) of furosemide at 1 mg/kg BW per hour, IV, for 16 h was started shortly after admission. After 7 h of hospitalization, the dog remained anuric. Crystalloid (13 mL/kg BW, IV, once) and mannitol (2 g/kg BW, IV, once) boluses were given, and a mannitol CRI (60 mg/kg BW per hour, IV, for 6.5 h) and dolasetron (0.7 mg/kg BW, IV, once) were added. A limited abdominal ultrasound that night revealed moderate peritoneal effusion and left-sided retroperitoneal effusion. The dog was mildly hypertensive (systolic blood pressure: 168 to 178 mmHg), and buprenorphine (0.008 mg/kg BW, IV, once) and hydralazine (0.5 mg/kg BW, PO, once) were given.

The next morning (15 h after presentation to the referral hospital), the dog had produced less than 0.05 mL/kg BW per hour of urine, gained 2.4 kg, and developed peripheral edema. Intravenous crystalloid therapy was decreased overnight as the dog remained anuric, then was discontinued after 16 h of hospitalization. Recheck blood analysis that morning revealed persistent azotemia with a creatinine of 1069.9 μmol/L (RR: 44.2 to 159.1 μmol/L) and BUN of 51 mmol/L (RR: 2.5 to 9.6 mmol/L). Referral to a tertiary facility was recommended for hemodialysis due to AKI and volume overload.

On presentation to our facility, the dog’s physical examination revealed obtundation and there was mild vestibular ataxia noted by the critical care service while the dog was hospitalized. A consultation with the neurologist was performed 48 h after admission; at this time the signs of ataxia were no longer present. Minimal urine in the indwelling urinary catheter collection set, and moderate generalized peripheral edema were also identified at presentation. These findings, in addition to severe renal azotemia, were consistent with acute kidney injury and fluid overload.

Serial chemistries were performed during hospitalization. The progression of azotemia in relationship to the dog’s fluid requirements is summarized in Figure 1. Leptospira serology was negative for 7 tested serovars. Thoracic radiographs were taken to confirm nasogastric tube and esophageal tube placement on days 7 and 21 from admission to tertiary referral. Both series were interpreted as unremarkable. On day 18, thoracic radiographs were taken when the dog began coughing acutely and had an episode of hematemesis. Radiographs revealed pulmonary vascular distension, pleural and peritoneal effusion, with evidence of pulmonary edema. The dog was receiving intravenous crystalloid therapy at 23 mL/kg BW per hour and there was suspicion of iatrogenic fluid overload. An echocardiogram on day 19 showed no structural cardiac disease. Repeat aerobic culture from the left ear taken at discharge revealed no growth.

Figure 1.

Plasma creatinine values and fluid requirements during the time of hospitalization (31 d), and days when hemodialysis was performed for the management of anuric acute kidney injury.

As the dog’s mentation improved, there was increasing concern for hearing loss. The dog had decreased response to sound beginning around day 10 of hospitalization. A brainstem auditory evoked response (BAER) test was performed by a Board-certified veterinary neurologist on day 15 using an EMG/NCV/EP electrodiagnostic unit (Sierra Summit Electrodiagnostic Unit; Cadwell Industries, Kennewick, Washington, USA). Subdermal electrodes were used with the reference electrode placed rostral to the contralateral tragus and the ground electrode placed at the occiput. The recording electrode was placed just rostral to the tragus of the ear being tested. Tubal inserts were used to deliver alternating polarity clicks of 100 ms duration and rate of 11.1/s at 90 dB in the evaluated ear with masking noise in the non-evaluated ear at 30 dB. A total of 500 responses were averaged. Results showed no response to sound stimulation confirming deafness in the patient, likely sensorineural deafness in the context of a normal aural examination and no structural disease. Results of the remainder of the neurologic examination were unremarkable except for a mild proprioceptive deficit in the left hindlimb. Repeat BAER evaluation on day 27 found no change in evoked response.

Due to anuria and fluid overload, fluid therapy was discontinued. A hemodialysis catheter (Hemo-Cath silicone double lumen catheter, 11.5F × 24 cm; MedCOMP, Harleysville, Pennsylvania, USA) was placed at admission, followed immediately by the dog’s first hemodialysis session. The first attempt at placement of the hemodialysis catheter in the right jugular vein failed due to previous venipuncture and hematoma formation, but placement in the left jugular vein was successful. Each session was conducted on an intermittent hemodialysis (IHD) platform (Gambro Phoenix, Lakewood, Colorado, USA), using neonatal and pediatric dialyzers (F3 Hemoflow, F160; Optiflux Fresenius, Waltham, Massachusetts, USA) and blood sets (Gambro Cartridge blood sets, 40 mL and 75 mL; Gambro Pheonix), and variable rates of heparin therapy (Heparin Sodium Injection, 1000 USP units/mL; Aurobindo Pharma, Telangana, India) to achieve an activated clotting time (ACT) of 200 to 240 s (RR: 80 to 120 s).

The dog received IHD daily for 3 d, then once every 1 to 3 d for a total of 8 sessions during the first 14 d of hospitalization (Figure 1). He received 2 packed red blood cell transfusions (after the first and seventh IHD sessions) and a single fresh frozen plasma transfusion (after the first IHD session) to treat post-dialysis hypovolemia secondary to excessive hemorrhage at the right jugular site where the first hemodialysis catheter placement had been attempted. No ultrafiltration was done to remove volume overload during the first dialysis session due to excessive hemorrhage and acute volume loss. As the dog became polyuric and no longer dialysis-dependent, intravenous fluids (Lactated Ringer’s Injection USP, 1000 mL; Baxter Healthcare, Deerfield, Illinois, USA and Plasmalyte 56 and 5% dextrose Injection USP, 1000 mL; Baxter Healthcare) were correspondingly increased for 22 d (Figure 1). He received intravenous isotonic crystalloid solution (Plasmalyte 56 and 5% dextrose Injection USP, 1000 mL; Baxter Healthcare) starting on day 12. From days 16 to 26, a combination of crystalloids (Lactated Ringer’s Injection USP, 1000 mL; Baxter Healthcare and Plasmalyte 56 and 5% dextrose Injection USP, 1000 mL; Baxter Healthcare) was used. An esophageal feeding tube was placed on day 21 of hospitalization, and on day 26 of hospitalization the dog was transitioned completely to enteral fluid administration.

Facial edema developed initially, likely due to partial occlusion of both jugular veins and fluid overload. The facial edema resolved as the hematoma formed in the right jugular vein resolved and the fluid overload improved. A nasogastric tube was placed on day 7 of hospitalization for enteral feedings with a liquid diet (Jevity 1.5 CAL 8.0 fl oz; Abbott Laboratories), administered as a CRI from days 7 to 18. Initial medical therapy at admission included maropitant (Cerenia; Zoetis, Kalamazoo, Michigan, USA), 1 mg/kg BW, SC, q24h, which was continued for 21 d, pantoprazole (Protonix IV; Pfizer, New York, New York, USA), 1 mg/kg BW, IV, q12h, continued for 21 d, and ondansetron (Ondansetron injection; Heritage Pharmaceuticals, Eatontown, New Jersey, USA), 0.5 mg/kg BW, IV, q12h, continued for 15 d, for management of nausea associated with uremia. Ampicillin (Ampicillin; Sandoz GmbH, Princeton, New Jersey, USA), 22 mg/kg BW, IV, q12h, was started on day 2 and continued for 12 d. On the third day of hospitalization, enalapril (Enalapril maleate; Wockhardt, Parsippany, New Jersey, USA), 0.25 mg/kg BW, PO, q24h, continued for 10 d, and amlodipine (Amlodipine besylate; Ascend Laboratories, Parsippany, New Jersey, USA), 0.18 mg/kg BW, PO, q24h, continued for 10 d, were started when the dog became hypertensive. The dog received metoclopramide (Metoclopramide injection; Teva Pharmaceuticals, Irvine, California, USA), 0.042 mg/kg BW per hour, IV, CRI, from days 10 to 22 to improve gastric motility with increased trickle feedings through his nasogastric tube. The patient received vitamin B complex injections (Vitamin B complex 150; Neogen, Lexington, Kentucky, USA), 1.5 mL, SC, twice on day 15 and day 22. On day 18, the dog had an episode of hematemesis and subsequent coughing. Ampicillin/sulbactam (Unasyn; Auromedics Pharma, East Windsor, New Jersey, USA), 40 mg/kg BW, IV, q8h, and sucralfate (Sucralfate tablets; Teva Pharmaceuticals), 1 g as a slurry, PO, q8h were initiated due to concerns for aspiration pneumonia and gastrointestinal ulceration, respectively (both continued until day 21). Thoracic radiographs were consistent with fluid overload, and a single dose of furosemide (Salix; Merck, Summit, New Jersey, USA), 1 mg/kg BW, IV, was given. Clopidogrel (Clopidogrel tablets; Apotex, Weston, Florida, USA), 2 mg/kg BW, PO, q24h, from days 19 to 24, was started due to concern for shower pulmonary thromboemboli from the hemodialysis catheter, although no thrombus was visible on echocardiogram. Aluminum hydroxide (Aluminum hydroxide gel; Rugby, Livonia, Michigan, USA), 50 mg/kg BW per day, PO, divided with meals, was administered beginning at day 26, at which time the dog’s phosphorus was 2.5 mmol/L (RR: 0.9 to 1.8 mmol/L), and was continued after discharge. Aural enrofloxacin [Enrofloxacin 1% otic suspension — compounded in-house using Enroflox (enrofloxacin injection for dogs 2.27%; Norbrook, Lenexa, Kansas, USA)] and 0.9% sodium chloride Injection (Hospira, Lake Forest, Illinois, USA), 1 mg/kg BW, q24h, and miconazole [Conzole (miconazole nitrate) 1% solution; VetOne MWI, Boise, Idaho, USA], 1 mg/kg BW, q24h, were applied to the ear canal from day 15 until approximately 1 wk after discharge, when culture from the left ear on day 33 revealed no growth.

The dog was hospitalized in our facility for a total of 33 d and was discharged with a creatinine of 274.1 μmol/L (RR: 44.2 to 59.1 μmol/L), BUN 10 mmol/L (RR: 2.5 to 9.6 mmol/L). He was eating and drinking well, receiving additional water through his esophageal tube, and his only oral medication was aluminum hydroxide (Rugby), 40 mg/kg BW per day. He was sent home on aural enrofloxacin, aural miconazole, and ear wash [Keto + TRIS EDTA (ketoconazole 0.1%, Tris-EDTA, phytosphingosine 0.01%), SoGeval, Oldsmar, Florida, USA]. Nine days after discharge from our hospital, the esophageal tube was removed and the dog was continued on subcutaneous fluids twice weekly. Seven months after discharge, the dog was still deaf according to owner observations at home, had a stable creatinine between 194.5 and 221.1 μmol/L (International Renal Interest Society stage 3) for 3 mo, and was on no oral medications. No further substaging of his kidney disease was available at the time of follow-up.

Discussion

This report documents the first published case of combined cochleotoxicity and acute kidney injury following parenteral gentamicin administration in a dog. The initial management of AKI with high dose of furosemide may have contributed to a lesser extent to the renal and cochlear effects observed in this patient. The toxicity of aminoglycosides is due at least in part to oxidative stress of affected cells. Reactive oxygen species activate caspase and caspase-independent pathways leading to cellular necrosis and apoptosis (5). Co-administration of antioxidants, including D-methionine and silymarin, has been recommended to minimize these effects (12,13). Binding of the aminoglycoside to iron may potentiate cellular degeneration in the cochlea, and iron chelating drugs have been suggested to reduce ototoxic effects (14).

Gentamicin cochleotoxicity is multifactorial but is primarily attributed to destruction of outer hair cells in the cochlea by reactive oxygen species. This produces the bilateral, progressive sensorineural deafness that was seen in this case (14). The drug enters the inner ear rapidly after parenteral administration and is concentrated in the perilymph and endolymph (15). Demyelination and degeneration of spiral ganglion neurons also contribute to cochleotoxicity (16). The similarity between mammalian mitochondrial RNA and the prokaryotic ribosomal 30S subunit targeted by aminoglycosides plays a role in the ototoxicity identified in humans (1). Mitochondrial mutations are a risk factor for aminoglycoside ototoxicity in humans, but no specific mutation has been identified in domestic animals (2,5).

The nephrotoxicity of gentamicin is due to uptake and accumulation of the drug into proximal tubular epithelial cells, and to a lesser extent distal and collecting ducts. Disruption of lysosomal phospholipid metabolism leads to leakage of proteolytic enzymes into the cytosol (17). Cytosolic gentamicin acts on mitochondria to activate apoptotic pathways, impair ATP production, and produce radical oxygen species, leading to proximal tubular cell death (8,18). No significant relationship has been found between aminoglycoside nephrotoxicity and ototoxicity (5). This dog experienced anuric kidney injury and was initially managed with crystalloids and then loop diuretic therapy. The use of diuretics in AKI is controversial, as the literature shows no significant benefit in outcome and potential inappropriate delay of referral for renal replacement therapy in humans (10). However, in veterinary medicine, diuretic therapy is often attempted in anuric AKI cases, as hemodialysis is not widely available (9). This dog remained anuric despite adequate volume resuscitation and diuretic therapy, and was volume overloaded at the time of referral, leaving hemodialysis as the last effective treatment option. Prognosis in this case without renal replacement therapy would have been poor.

Both nephrotoxic and ototoxic effects of aminoglycosides are potentiated by furosemide (19,20). Aminoglycosides increase cell membrane permeability in the inner ear, allowing furosemide to enter the cells at higher concentrations (19). High-dose furosemide alone can occasionally cause permanent deafness, especially when it is given in conjunction with other ototoxic drugs (5). Furthermore, concurrent administration of furosemide with gentamicin results in earlier and more marked increases in serum creatinine and more rapidly progressive renal tubular necrosis. Clinical uremia is more severe in dogs treated with concurrent gentamicin and furosemide (20). The half-life values for gentamicin in the organ of corti (13 h) and kidney (57 h) are significantly longer after a bolus dose in rats than in plasma (43 min, and it takes approximately 5 d for gentamicin to reach undetectable levels in the inner ear) (15). This dog likely had residual gentamicin concentrations in his inner ear and renal cortex at the time of initial furosemide administration, which may have contributed to his deafness and the severity of his azotemia.

There are many published doses of gentamicin, but the dose administered in this case was significantly higher and of longer duration than those routinely prescribed. Current recommendations include sensible patient selection, daily monitoring of renal values, and maintenance of hydration through the course of aminoglycoside therapy. Average aminoglycoside courses are 5 to 10 d, as the risk of nephrotoxicity increases with longer duration courses (3). Other antibiotic choices should be strongly considered for patients predisposed to renal disease (or with pre-existing renal disease), patients receiving other nephrotoxic or ototoxic drugs (such as loop diuretics), and service animals whose working capacity would be limited by deafness.

To the authors’ knowledge, this is the first reported case of combined cochleotoxicity and acute kidney injury following parenteral gentamicin administration in a dog. This case highlights the importance of exercising caution in dosing of parenteral aminoglycosides and the necessity of patient monitoring throughout the duration of treatment. CVJ