Gentamicin Ototoxicity: a 23-Year Selected Case Series of 103 Patients
Total Page:16
File Type:pdf, Size:1020Kb
Research Gentamicin ototoxicity: a 23-year selected case series of 103 patients Rebekah M Ahmed entamicin is an important bac- MB BS, FRACP, Abstract Neurology Fellow1 tericidal antibiotic with two Objective: To review patients with severe bilateral vestibular loss associated serious potential adverse Imelda P Hannigan G with gentamicin treatment in hospital. RN, Neuro-otology Nurse1 effects: nephrotoxicity and ototoxicity. Design and setting: A retrospective case series of presentations to a balance Clinicians are well aware that rising Hamish G MacDougall disorders clinic between 1988 and 2010. PhD, serum creatinine levels in patients Vestibular Scientist2 treated with gentamicin could indicate Main outcome measures: Relationship between vestibulotoxicity and gentamicin dose or dosing profile; indications for prescribing gentamicin. Raymond C Chan nephrotoxicity. However, many do MB BS, FRACP, FRCPA, not know that, contrary to textbooks Results: 103 patients (age, 18–84 years; mean, 64 years) presented with Infectious Diseases imbalance, oscillopsia or both, but none had vertigo. Only three noted some Physician1 and antibiotic guidelines, gentamicin hearing impairment after having gentamicin, but audiometric thresholds for all ototoxicity causes impairment of G Michael Halmagyi 1,2 patients were consistent with their age. In all patients, the following tests gave MD, FRACP, vestibular, not auditory, function. positive results: a bilateral clinical head-impulse test, a vertical head-shaking Neurologist1 Vestibulotoxicity is frequently over- test for vertical oscillopsia, and a foam Romberg test. In 21 patients, imbalance 3-7 looked in patients having gentamicin, occurred during gentamicin treatment (ignored or dismissed by prescribers in 1 Department of Neurology, so that severe, irreversible, bilateral 20) and in 66 after treatment; the remaining 16 could not recall when symptoms Royal Prince Alfred Hospital, Sydney, NSW. vestibular loss can occur, causing were first noticed, except that it was after gentamicin treatment in hospital. 2 Vestibular Research permanent imbalance, which is partic- Total gentamicin dose range was 2–318 mg/kg (mean, 52 mg/kg), daily dose Laboratory, School of range was 1.5–5.6 mg/kg (mean, 3.5 mg/kg), and duration was 1–80 days Psychology, University of ularly debilitating in elderly people. Sydney, Sydney, NSW. We reviewed patients presenting to (mean, 17 days). Six patients had only a single dose; 26 had five or fewer doses. Serum gentamicin levels, measured in 82 patients, were in the recommended [email protected] a balance disorders clinic between range in 59. Time to diagnosis ranged from 4 days to 15 years. Nephrotoxicity 1988 and 2010 with severe, selective, developed in 43 patients. Gentamicin dosage complied with contemporary or MJA 2012; 196: 701–704 bilateral vestibular loss, who had been current Australian antibiotic guidelines in under half the patients. treated in hospital with gentamicin doi: 10.5694/mja11.10850 Conclusions: Gentamicin ototoxicity is vestibular, not cochlear, producing between 1975 and 2010. permanent loss of balance, but not of hearing. Gentamicin can be vestibulotoxic in any dose, in any regimen, at any serum level. Methods indications (clinical and microbiologi- vestibular testing of lateral semicircular Patients cal) for using gentamicin; (v) when canal function, or both (Box 1; available 12 Over the 23-year period, 552 patients imbalance was first noted by the at mja.com.au). were diagnosed with severe, symmet- patient; and (vi) when bilateral vestib- rical, selective (ie, normal hearing for ular loss was first clinically recognised. Auditory testing age), bilateral vestibular loss. Of these, Ethics approval was not required for Air-conduction, pure-tone threshold, 263 patients had gentamicin vestibulo- this retrospective review. clinical audiometric graphs (0.25– toxicity (GVT). In the remainder, the 8kHz), from hearing measurement in Diagnosis vestibular loss was caused by cisplati- each ear when GVT was diagnosed, num ototoxicity (9), meningitis (6), Vestibular loss were available in 73 of the 103 patients. The Medical Journal of Australia ISSN: 0025- Frequencies above 8kHz were not hereditary factors (29), or bilateral Bilateral vestibular loss was diagnosed 729X 18 June 2012 196 11 701-704 tested. In patients with air-conduction sequential vestibular neuritis (61); in if all the following gave positive results: ©The Medical Journal of Australia 2012 8 thresholds above normal for age and 184 the condition was idiopathic. (i) bi-directional, horizontal and verti- www.mja.com.au in those with middle-ear disease, Research Patients with GVT who had severe cal head impulse test (Videos 1–4);9 (ii) bone-conduction thresholds were also bilateral vestibular loss after having vertical oscillopsia with loss of at least measured. Audiologists’ descriptive gentamicin in hospital were inter- three lines of visual acuity on a Snellen reports of the audiometric results were viewed and examined, and hospital chart during vertical head shaking available for the remaining 30 patients. records for the “gentamicin admis- (Video 5);10 and (iii) negative results of sion” were obtained. Our study a Romberg test on a firm surface and Indications for gentamicin included 103 patients with normal positive results on a foam surface renal function at the start of gen- (Video 6) (Videos 1–6; available at We reviewed patients’ clinical diagnoses tamicin treatment (serum creatinine mja.com.au).11 to determine (i) whether gentamicin level, <121 mol/L [eGFR, > 59 mL/ therapy complied with contemporary min/1.73 m2]) and for whom the follow- Vestibular testing (at the time of admission to hospital) or ing data were available: (i) total dose To confirm the diagnosis and quantify with current (2010) Australian antibiotic and dosage regimen of gentamicin; (ii) vestibular loss, 95 patients with GVT guidelines;13 and (ii) whether the treat- Editorial p 665 renal function before and during treat- (the remaining eight were too frail to ment was empirical, but appropriate, or Online first 03/04/12 ment; (iii) serum gentamicin level; (iv) test) had either caloric or rotational based on results of cultures. MJA 196 (11) · 18 June 2012 701 Research Results and both ears, was not different from It was not possible from the treat- accepted age-group means (Box 2; ment chart information to determine The 103 patients who fulfilled our crite- available at mja.com.au).15,16 Audio- the exact method of intravenous ria comprised 47 men, 56 women; metry in the three patients who com- administration used: slow infusion mean age, 64 years (range, 18–84 plained of hearing loss also showed (recommended) or bolus injection. years). Forty-eight presented with both thresholds consistent with age. The imbalance and oscillopsia, 39 with pure-tone thresholds in the 30 descrip- Serum gentamicin levels imbalance only, and four with oscillop- tive reports were reported as normal or Serum gentamicin levels (trough or sia only; and in 12 patients we were showing only high-frequency hearing peak or both) were retrieved in 82 unable to determine their main pre- loss, consistent with age and noise patients; in 23, a trough or a peak level senting symptom. In all patients, the exposure. We were unable to retrieve was above the recommended range. following tests gave positive results: a audiometry data from before gen- Peak levels were higher in those who 9 bilateral clinical head impulse test, a tamicin treatment for any of the developed nephrotoxicity than in those vertical head-shaking test for vertical patients. who did not (Box 4). oscillopsia,10 and a foam Romberg test 11 (Videos 1–6; available at mja.com.au). Nephrotoxicity Indications and results of culture Thirty-eight patients had recurrent falls Adherence to contemporary or cur- or required a walking aid, and 44 During gentamicin treatment 43 of 103 14 patients developed nephrotoxicity, rent antibiotic guidelines for gen- required vestibular rehabilitation. 13 defined as a serum creatinine level tamicin use is shown by patient Twenty-one patients first noted diagnosis in Box 5 and by clinician symptoms in hospital during gen- > 120 mol/L (range, 121–841 mol/L) 2 subspecialty in Box 6. tamicin treatment (gentamicin treat- and eGFR <60 mL/min/1.73 m (range, 2 Culture results were available in 73 of ment was stopped in only one patient); 59–< 6 mL/min/1.73 m ) on two sequential daily measurements. 103 patients; 44 showed no growth. An 29 patients had completed treatment organism was isolated in 29 cultures, but were still in hospital; 37 experi- Dosage and administration sensitive to gentamicin in 11; in another enced symptoms after discharge; and 11 gentamicin was not indicated for 16 could not remember when they first The gentamicin dose ranged from 160 treatment of the organism isolated (eg, noticed symptoms, except that it was to 320 mg/day, equivalent to 1.5 to methicillin-sensitive Staphylococcus after gentamicin treatment in hospital. 5.6 mg/kg/day (mean, 3.5 mg/kg/day). aureus); and in seven gentamicin sensi- The total gentamicin dose ranged from tivity was not tested or reported. Time to diagnosis 160 to 16 520 mg, equivalent to 2– The delay to diagnosis of bilateral ves- 318 mg/kg (mean, 3639 mg or 52 mg/ tibular loss ranged from 4 days (the kg). The mean total dose in the 43 Discussion only patient in whom GVT was diag- patients who developed nephrotoxicity nosed during treatment) to 15 years. was 4363 mg, in contrast to 3240 mg in Loss of balance, not of hearing GVT was diagnosed less than 12 the 60 who did not. The gentamicin All these patients developed symptoms months after treatment in 69 patients, doses our patients with GVT received and signs of bilateral vestibular impair- and more than 12 months after treat- in hospital, between 1975 and 2010, ment after treatment with gentamicin, ment in 34 patients. indicate that there was no change in but only three noted any hearing the total dose or daily dose of gen- Absence of cochleotoxicity impairment.