This Week in Otolaryngology — Aug 6, 2026

Welcome to This Week in Otolaryngology. This week we're covering 10 notable papers spanning head and neck oncology and value-based care, sinonasal and airway safety, and otology and sleep medicine outcomes. Let's dive in.

We'll start with head and neck oncology, where two papers ask how far we should extend an intervention, and a third resets the guideline framework. In JAMA Otolaryngology-Head and Neck Surgery, a cost-effectiveness analysis of perioperative pembrolizumab modelled the KEYNOTE-689 population using a partitioned survival model with Medicare and Healthcare Cost and Utilization Project derived costs, and the answer depends entirely on PD-L1 expression [4]. For patients with a combined positive score above 10, adding pembrolizumab bought roughly 1.35 additional quality-adjusted life years at an incremental cost of about 182,000 dollars, giving an incremental cost-effectiveness ratio near 135,000 dollars per quality-adjusted life year, with a 57 percent probability of being cost-effective at a 150,000 dollar willingness-to-pay threshold. For patients with a combined positive score between 1 and 10, the picture collapses: about eight-hundredths of a quality-adjusted life year gained for roughly 167,000 dollars, an incremental cost-effectiveness ratio above two million dollars, and only an 11 percent probability of cost-effectiveness. The practical message for tumour boards is that biomarker stratification, not simply trial eligibility, should drive who gets perioperative immunotherapy. A similar de-escalation logic runs through a retrospective series of 145 hypopharyngeal squamous cell carcinoma patients in Otolaryngology-Head and Neck Surgery, where occult contralateral nodal metastasis occurred in about 13 percent overall [8]. In clinically node-negative patients with strictly unilateral tumours not crossing the midline, the contralateral rate was around 9 percent, and no contralateral disease was found in the small subgroup with midline involvement and node-negative necks, while macroscopic extranodal extension in the ipsilateral neck emerged as a significant driver of contralateral spread. When occult disease was present, it clustered in levels two and three. The authors argue for bilateral dissection in central tumours, and for lateral lesions, contralateral levels two to three only when the ipsilateral neck is node-positive, the tumour is T3 or T4, or it crosses the midline — with intraoperative extranodal extension as a trigger to convert. Rounding out this theme, a commentary in the same journal walks surgeons through the 2025 American Thyroid Association guidelines for differentiated thyroid cancer, which now cover cancer only and no longer address thyroid nodule management [1]. That scope change matters for how you counsel patients with nodules, and the authors are frank that these guidelines will only change practice if they are actively disseminated and debated.

Next, a cluster of papers on sinonasal disease, safety, and the diagnosis you cannot afford to miss. In The Laryngoscope, a longitudinal cohort of 664 adults undergoing endoscopic sinus surgery with five-year follow-up found that obesity tracked with worse outcomes at every stage [2]. Obese patients started with higher symptom burden on the 22-item Sinonasal Outcome Test — a difference of roughly seven points, which is below the minimal clinically important difference on its own — but their symptoms then worsened at a faster rate over time, close to a point a year relative to normal-weight patients, so the cumulative gap exceeded the minimal clinically important difference by year three. Elevated body mass index roughly doubled the risk of revision surgery, and in polyp patients the association was considerably stronger. Notably, there was no significant association with subsequent biologic use. This is observational, so causality is unproven, but it supports frank preoperative counselling and closer long-term surveillance in obese patients with polyps. Also in The Laryngoscope, the largest single-institution series of cavernous sinus thrombosis, 33 patients, is a reminder of how this presents before it becomes obvious [10]. A third of patients were initially discharged or transferred with something else — migraine, muscle strain, or uncomplicated sinusitis. Neck pain was strongly associated with delayed diagnosis, present in over half the delayed group versus about one in eleven of the rest, and delayed patients were substantially younger, a median of 40 versus 64 years. Sphenoid opacification was seen in nearly four in five, ninety percent underwent sinus surgery for source control and cultures, and 94 percent were anticoagulated with major haemorrhage in 12 percent. More than half of eyes had some visual impairment. The takeaway is concrete: sphenoid sinusitis plus headache, cranial neuropathy, orbital findings, or neck pain, especially in a younger patient, warrants dedicated imaging for cavernous sinus thrombosis. On the safety side, JAMA Otolaryngology-Head and Neck Surgery reports a prospective quality improvement study of a human factors programme in paediatric laryngotracheal endoscopy, comparing 44 procedures before and 36 after implementation of a structured checklist, identification caps, and aviation-style crew resource management simulation [5]. Verbal use of the surgical checklist rose from about 9 percent to 89 percent, preoperative surgeon-anaesthetist discussions from 71 percent to 100 percent, and postoperative debriefings from 57 percent to 92 percent, with fewer task interruptions and higher non-technical skills scores. Importantly, clinical indicators including oxygen desaturation and atropine use did not change — this is behavioural change, not yet demonstrated outcome change, and the authors say so.

Our third theme is measurement and prognostication in otology, sleep, and salivary surgery — papers that all challenge how we define success. Otolaryngology-Head and Neck Surgery reports a multicentre ossiculoplasty registry analysis of 499 ears that all met current success criteria, with a mean postoperative pure-tone average air-bone gap under 10 decibels [6]. Yet the air-bone gap at 4 kilohertz averaged over 18 decibels, and more than half of these supposedly successful ears had a 4-kilohertz gap of 20 decibels or more, with about a quarter at 25 decibels or more. Air-conduction gains shrank steadily as frequency rose. In short, our reporting standards can certify a good result while a substantial high-frequency conductive deficit persists, which is precisely the range patients need for speech clarity in noise. Complementing that measurement theme, The Laryngoscope reports a cohort of adult cochlear implant recipients showing that improvement is front-loaded: gains between one and three months ran roughly thirty-five-fold faster than between six and twelve months [7]. Cluster analysis identified three performance phenotypes already distinguishable at one month, and preoperative word scores did not differ across them. A one-month consonant-nucleus-consonant score of 44 percent or higher predicted functional success at twelve months with an area under the curve of 0.82 — a large gain in predictive value over preoperative testing. Device use was a stable trait that did not itself predict improvement, but sustained or increasing engagement through six months was associated with better twelve-month scores. That gives you an actionable one-month checkpoint for counselling and for triaging intensive rehabilitation. In sleep medicine, a prospective multicentre study of 77 obstructive sleep apnoea patients across three hospitals used drug-induced sleep endoscopy with a titratable mandibular positioner to select candidates for mandibular advancement therapy [3]. Among the 66 who completed treatment, the apnoea-hypopnoea index fell from about 32 to 7, minimum oxygen saturation rose, and time below 90 percent saturation dropped from about 11 to 4 percent, with high positive predictive value across moderate and severe disease. There was no randomised comparator, so this speaks to selection accuracy rather than superiority. Finally, a retrospective comparison of 212 parotidectomies in Otolaryngology-Head and Neck Surgery found robot-assisted surgery took roughly 40 to 55 minutes longer but produced less blood loss, shorter incisions, better patient-reported outcomes, and — versus endoscopic surgery — substantially lower odds of long-term facial nerve palsy [9]. With only 28 robotic cases from a single surgeon, this needs prospective multicentre confirmation before it changes referral patterns.

If you only have time for one paper this week, make it the cost-effectiveness analysis of perioperative pembrolizumab in JAMA Otolaryngology-Head and Neck Surgery [4]. It takes a positive phase three trial and tells you which of your patients the benefit is actually concentrated in, which is the conversation multidisciplinary teams are having right now.

Here are the key takeaways from this week in Otolaryngology. First, perioperative pembrolizumab for locally advanced head and neck squamous cell carcinoma appears economically defensible only when the PD-L1 combined positive score exceeds 10; below that, the cost per quality-adjusted life year is prohibitive. Second, contralateral neck dissection in hypopharyngeal cancer can be selective — reserve it for central tumours, node-positive or advanced-stage lateral lesions, midline crossing, or intraoperative macroscopic extranodal extension. Third, obesity predicts worse baseline symptoms, faster symptom deterioration, and roughly double the revision risk after endoscopic sinus surgery, so counsel and monitor accordingly. Fourth, a one-month post-activation word score of 44 percent or higher is a practical prognostic checkpoint after adult cochlear implantation, and standard pure-tone reporting after ossiculoplasty hides clinically meaningful 4-kilohertz deficits in more than half of so-called successes. And fifth, in a younger patient with sphenoid sinusitis, headache, and neck pain, think cavernous sinus thrombosis before you discharge.

That's your roundup for This Week in Otolaryngology. The full transcript and references are available on the episode page in your AudioScholar library. This is an AI-curated summary — for clinical decisions, always consult primary sources and current guidelines. See you next week.

If this weekly briefing is useful, follow the show in your podcast app so new episodes arrive automatically. And if you have a paper you have been meaning to read, upload the PDF, or paste any link, at audioscholar dot C C. We will turn it into audio like this one, in any of thirty-one languages.

This is an automated summary generated by artificial intelligence, which can make mistakes. Always review the original source materials.