Introduction and objectives Updated WHO guidance recommends at least 8 months of aminoglycoside (AG) for MDR-TB but provides no definitions or monitoring strategies for otoxocity. Most UK centres perform 2–4 weekly audiograms; we perform weekly audiograms. We retrospectively investigated whether this strategy pre-emptively identifies ototoxicity before significant hearing loss (HL) is evident.
Methods Patients we treated with amikacin for MDR-TB from 2002–2015, with at ≥4 weekly pure tone audiograms, were included. Audiograms, treatment duration, symptoms, creatinine clearance and outcome were obtained from clinical records. All patients received amikacin at 15 mg/kg per day and had weekly amikacin levels and renal function. Definition of HL was defined as per the ASHA as >20 dB loss of pure tone threshold from baseline at one frequency or >10 dB at two adjacent frequencies.
Results 31 MDR-TB patients fulfilled selection criteria; 15 female, median age 36 years (IQR: 24–43) and median weight 61.5 kg (IQR: 52–65.) 22/31 (70.9%) patients had their first audiogram within 10 days; median 5.5 (IQR:4–7.) The median duration of amikacin treatment was 79.5 days (IQR: 61.75–94) and median total dose of 70.8 g (IQR:44.4–97.75.) 4/31 (12.9%) had moderate-severe baseline hearing loss (HL). A total of 17 (54.8%) patients met the ASHA definition of HL: 7 at 4 kHz, 10 at 6 kHz and 17 at 8 kHZ. The median time to meeting ASHA definition of HL among these patients was 59 days (IQR: 41–84.75). 16/31 (51.6%) patients stopped amikacin earlier than planned and 1 continued; 2 (6.5%) due to symptoms of deafness, 2 (6.5%) due to tinnitus and 12 (38.7%) due to asymptomatic high frequency HL on audiograms. Creatinine clearance and trough amikacin levels remained within range for all patients. Regarding outcomes, 17 (54.8%) completed TB treatment, 5 (16.1%) remain on treatment, 4 (12.9%) transferred, 3 (9.7%) were lost to follow up and 1 (3.2%) died.
Conclusions AGs are important in the treatment strategy of MDR-TB but this must be balanced with the long-term side effects. The ototoxicity of AG is unrelated to elevated drug levels or contributing factors and is a common adverse effect. Weekly audiograms led to earlier detection of pre-symptomatic amikacin ototoxicity and cessation in 38.7% of patients.