- Design
- systematic review with random-effects pooling of proportions; no comparator arm
- Population
- 12 studies, 69 implanted ears, in patients with hearing loss after radiation for nasopharyngeal carcinoma
- Primary outcome
- audiometric and speech perception improvement, and surgical or device complications
- Effect
- audiometric improvement 91.3% (95% CI 73.1-98.8); no postoperative complication 96.5% (88.3-99.5); mean 6.5 years deaf before implantation
Radiation for nasopharyngeal carcinoma damages hearing, and the assumption has been that an irradiated temporal bone makes cochlear implantation risky and a long-deaf cochlea makes it futile. A systematic review pooled 12 studies covering 69 implanted ears to test both halves of that.
Safety held up. Implantation was usually via standard mastoidectomy with posterior tympanotomy, with alternative approaches used selectively; 80.4% of cases had no intraoperative complication (95% CI 55.1-96.7) and 96.5% no postoperative complication (88.3-99.5). Efficacy held up too: audiometric improvement was seen in 91.3% (73.1-98.8), with gains in speech perception and patient-reported outcomes. The mean duration of profound hearing loss before implantation was 6.5 years.
That last number is the finding that should change something. It means these patients are not being referred — they are spending most of a decade deaf after successful cancer treatment before anyone arranges rehabilitation. The review's own conclusion is that this supports routine audiological surveillance after radiation so that referral happens in time.
The practical change is a scheduled audiogram in survivorship follow-up rather than a reactive one when the patient complains, because hearing loss after radiation is progressive and gradual and patients accommodate to it. That matters particularly in India, where nasopharyngeal carcinoma is concentrated in the north-eastern states and survivorship care is often delivered far from the treating centre — so the surveillance has to be written into the discharge plan to happen at all.
The evidence base is 69 ears from 12 small studies with wide intervals, so the pooled figures should be quoted as encouraging rather than precise. The referral behaviour is what changes; the operation remains an individual decision.
- Write scheduled audiological follow-up into the survivorship plan after radiation for nasopharyngeal carcinoma
- Do not wait for the patient to complain — the loss is gradual and progressive, and patients accommodate
- Refer for cochlear implant assessment rather than assuming an irradiated temporal bone excludes it
- Quote the complication and benefit figures as encouraging but imprecise; they rest on 69 ears
- Where survivorship care happens away from the treating centre, name who is responsible for the audiogram
Why it matters
A mean of 6.5 years deaf before implantation is a referral failure, not a clinical finding about the operation.
Don't overread it
Twelve uncontrolled small studies with wide confidence intervals cannot establish complication rates precisely, and published series favour good outcomes.
The statistics, in plain English
A pooled proportion of 80.4% with an interval from 55.1% to 96.7% is barely informative — the true rate could be just over half or almost all — and that width reflects 69 ears drawn from 12 small studies. The narrower interval on postoperative complications (88.3 to 99.5) is more useful. None of these studies had a comparison group, so improvement in 91.3% describes what happened after implantation rather than an effect measured against an alternative. Publication bias also runs one way here: centres publish the series that went well.
Read the rest in the app
You have read your two free briefings this month. The app carries all 27 specialties, every morning, free — and this finding is waiting in it.
Scan to keep reading on your phone. No account needed to start.
Tomorrow morning, before your first patient
One edition a day for ent & head and neck surgery, written by the desk, every claim tied to its paper. Six minutes.
Get the app — free