Laryngology Brief, August 2026, Issue 3

Welcome to the third issue of Laryngology Brief,

Every month I scan the laryngology literature across my core journal set, this time 168 articles published between July 1 and August 1, 2026, and select the handful that I believe deserve a working clinician's attention. Each featured paper comes with a short summary of its design and numbers, followed by my own reading: why it matters at the bedside or in the office, where the data is solid, and where I would hold back. This issue leans toward questions where the headline and the actual analysis part company: whether delay really costs a child their voice in unilateral paralysis, whether earlier trafermin injection is better, what to offer the refractory neurogenic cough patient now that cryotherapy has prospective data alongside last issue's nerve block meta-analysis, how to prevent scar after cordectomy rather than treat it, and whether bowing index deserves to be reported at all. My aim is not to replace your own reading but to save you time and give you a starting argument. Comments and disagreements are welcome; they make the next issue better and increase my motivation to continue!

One apology before the papers. This issue is late, for two reasons. August is August, and the clinic did not slow down because everyone else was on holiday. The other reason is that I changed how I read this year: instead of working from abstracts I have been sitting down with the full papers, and it turns out the abstract and the body do not always tell the same story. It happened three times this month. One paper's headline says earlier treatment is better while its primary analysis says nothing of the kind. Another reports a measurement so sensitive to which video frame you froze that the artefact is larger than the difference between mild and moderate disease. A third publishes its own complication rate in full, which almost nobody does. None of that is visible from 250 words of abstract. Reading properly takes longer, I think it is the only way this newsletter is worth your time, and I will probably be late again. Two papers below I could only reach at abstract level, and I say so where that is the case.

⭐ Featured

Gobillot TA, Carroll LM, Zur KB. Outcomes of Recurrent Laryngeal Nerve Reinnervation for Pediatric Unilateral Vocal Fold Paralysis. The Laryngoscope. 2026. 10.1002/lary.70750 · Open access · Level of evidence 4

Eighteen years of Karen Zur's practice at the Children's Hospital of Philadelphia, 2007 to October 2025: 102 children reinnervated for unilateral vocal fold paralysis. The operation is a package, not a single manoeuvre. Suspension microlaryngoscopy to assess arytenoid mobility and look for posterior intubation deficiency or scar, then transoral laryngeal EMG to confirm chronic denervation, then possible carboxymethylcellulose gel injection, then anastomosis of an ansa cervicalis branch to the proximal recurrent laryngeal nerve.

The cohort is medically complex in a specific way. Eighty-nine percent were iatrogenic cardiac injuries, most often PDA ligation. Seventy percent were premature, with a median gestational age of 25 weeks among those; 43% were extremely preterm. Fifty-nine percent had been intubated for more than ten days, mean 2.7 months.

Seventy-one children had evaluable voice outcomes over a mean 31 months of follow-up, the longest reaching 17 years. Fifty-three (75%) were rated dramatically improved, 11 (15%) moderately, 7 (10%) minimally or not at all. In the dramatic group pVHI fell 15.9 points, GRBAS 3.0, CPPS rose 2.1 dB, and maximum phonation time gained 2.9 seconds; 22 of 46 reached normal CPPS for age and sex and another 9 came within 1 dB. Benefit accrued with time rather than appearing at once: mean GRBAS change was −1.4 at 6 to 12 months, −2.8 at 12 to 20 months, and −4.8 beyond 20 months. Aspiration was present in 25 children and resolved in 12 of the 15 who were followed up. Six had dyspnea or stridor suggesting synkinesis; all resolved.

Four factors predicted a less than dramatic result: female sex (OR 0.29 for males, p = 0.03), lower gestational age (OR 0.89 per week, p = 0.01), prolonged intubation (OR 5.19, p = 0.008), and posterior intubation deficiency found at surgery (OR 4.63, p = 0.04). Length of intubation itself was not significant. Neither was age at surgery.

Why it matters. Start with the term "posterior intubation deficiency". It is a structural defect in the posterior glottis left by the endotracheal tube. It is not a neural problem, and reinnervation does nothing about it. Reinnervation restores tone and bulk to a denervated thyroarytenoid; it cannot close a gap that the tube carved out of the tissue. So when a reinnervated child does not sound the way you promised the parents, the honest answer is often that you treated the nerve and left the architecture alone. Zur's group showed in 2021 that higher-grade posterior deficiency predicts worse perceptual outcomes after phonosurgery, and this paper extends that to reinnervation. It is the same lesson arriving from a second direction.

But the finding I keep returning to is a negative one, and it is buried in a table. Time from nerve injury to reinnervation had no association with outcome: odds ratio 1.04, p = 0.52. The mean interval in this cohort was eight years, with a range from 0.7 to 20.5 years. Children operated two decades after their PDA ligation did as well as children operated within a year.

I want to be careful about how far to push that, because a null in an underpowered subgroup analysis is not proof of equivalence, and there is no comparison arm. But it is consistent with what Smith and others have reported, and it changes a conversation I have regularly. When a family arrives with a fourteen-year-old whose voice has been breathy since infancy, I have always had ambivalent feelings about whether to reinnervate. This cohort says the reinnervation window may not close, at least not on the timescale we assume, and that the thing which actually limits the ceiling is not delay but what the tube did to the posterior glottis when the child was a neonate. That is a different clinical model. Under the old model you hurry. Under this one you examine.

There is a mechanistic thread the authors raise briefly and I would pull harder. Infants born before 28 weeks have reduced development of the macula flava. If the deficit in the worst-performing children is partly a lamina propria that never fully formed, then no amount of neural input fixes it, and the four risk factors are really one risk factor with four faces: this larynx was structurally compromised before the nerve was ever cut.

Now the conditions, and there are real ones. Thirty-one of 102 patients had no evaluable voice data, and 40% of the children with preoperative aspiration were lost to follow-up entirely, so the swallowing result rests on 15 patients. In a single-surgeon series that pattern usually flatters the outcome. Outcome grading was an "integrated clinical assessment" made by the same team that performed the surgery, not a blinded rating, and the perceptual and acoustic numbers are reported by that grouping, which is close to circular. The limitations section is three sentences long and does not mention any of this. Most importantly, there is no control arm. Paediatric paralysis compensates spontaneously in a real fraction of children over years, and a case series with a 90% improvement rate is not the same claim as a 90% attributable benefit. That is not this paper's job, but anyone citing it should say so out loud.

What I would take into clinic tomorrow: stop leading with timing, start leading with the posterior glottis. I will be more cautious in kids with history of being preterm or long intubation.

Neurolaryngology

Terasaki M, Hasegawa T, Iinuma R, et al. Effect of Treatment Timing on Voice Outcomes After Intracordal Trafermin Injection for Vocal Fold Paralysis Following Aortic Surgery. Journal of Voice. 2026.10.1016/j.jvoice.2026.07.014 Hasegawa T, Fujita R, Watanabe Y. Long-term and Longitudinal Outcomes of Intracordal Trafermin Injection in Vocal Fold Paralysis. Journal of Voice. 2026. 10.1016/j.jvoice.2026.07.008

Two papers from the Tokyo Voice Center and Gifu University, published together and best read together. Trafermin is recombinant basic fibroblast growth factor (Fiblast, Kaken), injected intracordally as a regenerative agent rather than a bulking one. The preclinical rationale is Goto's rat work: high-dose bFGF into the thyroarytenoid one day after nerve transection produced larger muscle cross-sectional area at 28 and 56 days, more satellite cells to day 14, and more neuromuscular junctions.

The timing study took 50 patients paralysed after aortic surgery (42 men, 8 women, mean age 65), mean interval from surgery to injection 340 ± 524 days, range 27 to 2600. The primary analysis was negative. ANCOVA adjusting for sex, age, dose, days to injection and baseline values found no significant association between days-to-injection and any outcome. The weak correlations quoted in the abstract (VHI R = −0.381, MPT R = −0.323) appear only after excluding the six cases beyond 500 days. What did reach significance was dose, for MPT and mean flow rate, and baseline severity, for everything except MPT.

The long-term study pooled 221 cases across all aetiologies, excluding repeat injections, prior injectables, and patients with no follow-up data. Averaged in 90-day bins, all four voice measures improved immediately and held for about two years. A generalised linear mixed model found no association between days since injection and VHI; the associations for MPT and pitch range were statistically significant but trivially small (MPT estimate 0.0027 seconds per day, about one second a year). Between-patient variance accounted for 62% to 78% of total variance depending on the parameter. And then the authors write, in print, that the longitudinal results "did not fully align with our clinical experience."

Why it matters. First the terminology, because trafermin gets discussed loosely. This is not augmentation. Nothing is added to fill a gap. The claim is that a growth factor changes the tissue, promoting hyaluronic acid in the superficial lamina propria, clearing abnormal collagen, and blocking muscle atrophy, and that the change outlasts the injected substance. That is a different category of intervention from hyaluronic acid and should not be benchmarked against it as though it were the same thing.

Now the detail that reframes both papers, and it is in the methods, not the abstract: trafermin was injected into the paralysed fold and the contralateral fold, in every patient. The authors are open about why, and their reasoning is defensible in clinical terms. Their goal was the best voice for the patient, not restoration of the paralysed side, and augmenting the healthy fold narrows the glottal gap. But it means the design cannot support the hypothesis it was built to test. If you treat both sides, you cannot attribute improvement to preventing atrophy in the denervated one. The authors half-acknowledge this and then note, correctly, that injecting only the paralysed fold "might have resulted in voice improvement more consistent with our hypothesis." That is an unusually honest sentence and it is also a description of a study that has not been done.

The second detail cuts the same way. Only eight patients were injected within 90 days, and the authors report that those eight had relatively poor improvement. So the headline "earlier is better" rests on a correlation that survives only after outlier exclusion, in a dataset where the earliest-treated patients did worst. Their own explanation is sobering and true everywhere, not just in Japan: aortic disease is often fatal, cardiovascular surgeons are saving lives, and there is no pathway that gets these patients to a laryngologist quickly. Sample size at the early end is small because the referral system does not exist.

What I take from the pair is not an answer about timing. It is that 62% to 78% of the variance in these outcomes sits between patients, not across time. By the way, it's not available outside of Japan, so we will not have a chance to try in our own patients. If I had, probably I would prefer to use it as an additional tool instead of solely as a treatment option.

Dhar SI, Allen E, Liu YL, et al. Laryngeal Cryotherapy for Neurogenic Cough: Safety, Feasibility, and Prospective Outcomes. The Laryngoscope. 2026. 10.1002/lary.70723 · Open access · Level of evidence 4

Thirty patients with refractory neurogenic chronic cough, 96.7% of whom had already failed cough suppression therapy, underwent laryngeal laser sensory testing followed immediately by awake selective laryngeal cryotherapy with a contact device, delivered in 30-second freeze-thaw cycles. Swallowing safety was checked with FEES before proceeding; anyone with aspiration would not have been treated. All thirty completed the awake procedure. Seven patients had a treatment-emergent adverse event, six grade 1 and one grade 2, one leading to study discontinuation; one event was grade 1 dysphagia which self-resolved. No serious adverse events. The laser power threshold to trigger a cough rose from 2.8 W to 7.5 W (p < 0.001). Cough severity index fell 8.25 points and urge-to-cough VAS fell 18.07 by six months, both significant at every timepoint.

Why it matters. What I like is not the effect size, it is the mechanism check. Most chronic cough trials measure a symptom scale and leave you guessing whether anything physiological moved. This group measured the sensory threshold directly, before and after, and it shifted in the predicted direction by a factor of nearly three. That converts "patients felt better" into "we blunted laryngeal hypersensitivity and patients felt better," which is a stronger claim and, more usefully, a falsifiable one.

For those of us who treat neurogenic cough the toolbox is thin: neuromodulators half of patients cannot tolerate, superior laryngeal nerve blocks that fade, behavioural therapy that works but needs a motivated patient and a therapist. A tolerable single office procedure with a six-month signal deserves attention. I would also flag the aside the authors make near the end, because it may outlast the cough indication: if contact CO2 cryotherapy is tolerable awake in a hypersensitive larynx, the same platform could serve office ablation of polyps, granulomas and perhaps stenosis without the inflammatory response laser produces. That is a bigger idea than cough. I will be at Fall Voice this September. Maybe I will catch a workshop on cryotherapy, but I need to check if it's the same device or not.

Now the conditions, and the authors state most of them themselves, which I respect. There is no control arm, and they say plainly that placebo response in refractory cough is powerful and that a placebo group is needed. The protocol evolved during the trial: target site, freeze time and cycle number all changed, so the thirty patients did not all receive the same intervention. Patients were limited to a single treatment for safety.

And there is one confounder that deserves more weight than it gets. Every patient received a modified endoscopically guided superior laryngeal nerve internal branch block as the local anaesthetic for the cryotherapy. In a study whose entire premise is sensory neurolysis, giving every subject a sensory nerve block is not a trivial co-intervention. The authors' defence is the right one and it is empirical rather than theoretical: most of these patients had received the same block before enrolment without sustained benefit. That helps. It does not fully separate the two, and the authors' own proposed fix, moving to general anaesthesia so the block can be dropped and the protocol blinded, is exactly right.

I will add one caution they do not: deliberately reducing laryngeal sensation is a strategy with an obvious downside in an older population. They screened with FEES and saw one transient grade 1 dysphagia, which is reassuring at this scale. Thirty patients cannot detect a rare aspiration signal, and the phase II should be powered with swallowing as a named safety endpoint, not an incidental one.

Phonosurgery and Laryngeal Framework Surgery

Chitose S, Umeno H, Ono T. Endoscopic Sealing With a PGA Sheet to Reduce Scarring After Vocal Fold Mucosal Resection. Laryngoscope Investigative Otolaryngology. 2026. 10.1002/lio2.70481 · Open access · Level of evidence 3

Read at abstract level only; I did not have the full text for this issue. Patients undergoing type I or II cordectomy for dysplasia or Tis/T1a glottic carcinoma had the mucosal defect either left to heal (38 patients) or covered with a polyglycolic acid sheet secured with fibrin glue (22 patients). The PGA group had lower postoperative GRBAS Grade scores and longer maximum phonation time, with videostroboscopic advantages in symmetry, glottal closure and mucosal wave at three months and sustained benefit in closure and mucosal wave later. Roughness dominated breathiness in the postoperative perceptual profile.

Why it matters. Scar is the problem we have not solved. Every laryngologist has a patient whose oncological result was perfect and whose voice never came back, and we have almost nothing to offer them afterwards. A paper about prevention rather than salvage has my attention before I read a word of it. The intervention is also unusually deployable: a PGA sheet and fibrin glue are on the shelf in most theatres, so this is not a study you file under "in ten years."

The conditions I can state from the abstract alone: not randomised, unequal groups, and in a series like this the sheet usually goes to the cases where the surgeon expects it to help. The benefit is concentrated in type I cordectomy and the authors say type II remains unsettled, which limits the claim precisely where scarring hurts most. And the perceptual profile is worth pausing on. Roughness dominated breathiness, meaning the sheet restores closure more readily than it restores pliability. That is not a failure, but it tells you what the thing is doing. It is a cover, not a lamina propria.

Which is why I want this read against the Notch1 work below: two interventions, two layers of the same problem.

Woo P. Laryngeal Electromyography-Guided Injection Laryngoplasty: A Prospective Series. Journal of Voice. 2026. 10.1016/j.jvoice.2026.06.038 · Level of evidence 4

Sixty-four patients, 118 EMG-guided vocal fold injections between 2020 and 2026, every one verified afterwards by video-fibrescope. The route was chosen deliberately for patients who are poor candidates for fibrescope-guided injection: aversion to the endoscope, no second operator available, concurrent botulinum toxin, absent anatomical landmarks after radiation or open surgery, or previous failure of the transcervical approach.

Twenty complications occurred across 118 folds (17%): inadequate injection in 8, partial lamina propria injection in 5, partial extrusion into the lumen in 6, and one requiring removal. Multiple complications on the same fold were common. Six patients (10%) were clinical failures. Fifty-eight of 64 (90%) got the result intended. The single serious event was a professional voice user with spasmodic dysphonia who requested simultaneous botulinum toxin and Restylane; the Restylane went into the lamina propria, the voice worsened, and it was removed in the office with cup forceps. Outcome showed no association with indication, age, sex, side or volume injected.

Why it matters. I want to praise this paper for something other than its success rate. Woo scoped every single injection afterwards and published what he saw, including a 17% technical complication rate in the hands of someone who has been doing this for forty years. Almost nobody reports that denominator.

The mechanistic explanation for the misses is the sharpest thing in the discussion. EMG tells you the needle is in muscle at the moment of the signal. Then the patient swallows or coughs and the needle moves, and you cannot reposition on the basis of an EMG signal the way you can under vision. That is the real trade-off, it is a property of the method.

Where I would push back is on framing. This is presented as a technique paper and I think its actual contribution is a verification protocol. The interesting question is not whether EMG guidance is good enough. Obviously it is not. I would prefer not to inject hyaluronic acid under EMG. Sorry, Peak. But I might use it in selected cases with PRP, where there is no harm in injecting into the lamina propria.

Laryngeal Rejuvenation

Miller CS, Kim BS, Al-Ghezi M, et al. Vocal Fold Opening Position Impacts Bowing Measures in Age-Related Vocal Atrophy. The Laryngoscope. 2026. 10.1002/lary.70684

Twenty-two patients over 60 with age-related vocal atrophy, with an unusually clean exclusion list: no prior thyroplasty or injection, no neck surgery or radiation, no paresis or paralysis, no lesion, no significant smoking history, no neurological disease. Mean age 72.5, mean VHI-10 only 12.4. Flexible videolaryngostroboscopy during an alternating voice-sniff task; bowing index measured on every non-phonatory frame by two blinded raters, a mean of 20.7 frames per patient.

Bowing index was higher during abduction (8.15 ± 3.29) than adduction (6.34 ± 2.21), mean difference 1.81 (95% CI 1.25 to 2.38, p < 0.0001), and the direction held within individual subjects. Across quintiles of opening, bowing peaked at 20% to 40% (8.79) and 40% to 60% (8.49) of maximal abduction and was lower at 1% to 20% (7.47) and 60% to 80% (7.31). Compared against two expert clinicians rating severity blind to the numbers, mean bowing index was 7.30 for mild and 8.90 for moderate atrophy, with 14.62 in the single severe case. Inter-rater reliability for the bowing measurement itself was 0.530, moderate at best.

Why it matters. Put two of those numbers side by side. The swing produced by when in the respiratory cycle you froze the frame is 1.81. The difference between mild and moderate atrophy is 1.60, and that difference is not statistically significant (95% CI −0.85 to 4.06, p = 0.25). The measurement artefact is larger than the clinical signal it is supposed to detect.

We had a study in 2020 in which we used the bowing index in supranuclear palsy patients as a tool to estimate vocal fold atrophy in this patient group. At that time, I also had similar questions about the bowing index that we are using, because according to the phase of the movement of the vocal fold, the bowing index might change.

The practical consequence is immediate and it implicates work many of us have done. If you measure bowing on whatever still frame looks clearest, or compare a preoperative image caught mid-breath against a postoperative image caught near full abduction, you can manufacture improvement or manufacture failure without anything happening to the patient. Add the 0.530 inter-rater reliability and the honest position is that bowing index in its current form is not fit to serve as an outcome measure at all.

Swallowing

Steele CM, Dharmarathna I, Benfield JK, et al. The Influence of Liquid Consistency and Food Texture on Swallowing: A Systematic Review Update. Dysphagia. 2026. 10.1007/s00455-026-10977-w

An update to the review that underpins much of IDDSI, synthesising 70 articles with samples of at least 10 participants reporting quantitative data for at least two bolus consistencies. Six headline findings:

Penetration and aspiration are highest for thin liquids (IDDSI 0) and for mixed consistencies containing a thin component, and reduce with incremental thickening beginning at slightly thick (level 1).
In healthy swallowing, pharyngeal residue rises from level 0 to level 2 when taken by sip, but does not rise further at levels 3 and 4, a pattern the authors attribute to the smaller volumes used when these are served by teaspoon. Results in patients are mixed.
Time to swallow initiation lengthens as consistency thickens, in both healthy and patient populations.
Laryngeal vestibule closure timing and duration, and pharyngeal area at maximum constriction, are stable across consistencies.
Duration and extent of UES opening are driven primarily by bolus volume, not consistency.
On high-resolution manometry, UES contractility and distension are more sensitive to consistency than pharyngeal contractility or bolus flow; pharyngeal contractile integral and hypopharyngeal intrabolus pressure vary minimally.

Why it matters. Texture and consistency modification is the most prescribed dysphagia intervention in the world and the evidence beneath it has always been thinner than the confidence with which we order it. This is the update to the review that built that evidence base.

Two findings should change practice, and both point the same way: less thickening than we use. The safety benefit begins at level 1, and the residue penalty accumulates from level 0 to level 2. So the reflexive jump to moderately or extremely thick is buying diminishing safety at a real efficiency cost, and the authors say directly that recommendations for thicker consistencies must be evaluated so they do not damage efficiency. The apparent plateau at levels 3 and 4 is a measurement artefact of volume, not a licence to thicken further.

Finding 5 is the one I would put on a wall. UES opening tracks volume, not consistency. We spend enormous clinical energy adjusting what the patient drinks and comparatively little on how much they take per swallow, and for the sphincter the second variable is the one that moves.

Finding 4 is quietly reassuring and worth knowing for the opposite reason: if laryngeal vestibule closure is stable across consistencies, then a patient who penetrates on level 2 as well as level 0 does not have a consistency problem, and thickening further is unlikely to be the answer.

The limitations are stated candidly and one is structural. The literature search ends 11 May 2024, so the review is roughly two years stale on publication, and the authors say a further update was not feasible given the volume of work. True meta-analysis was impossible because coverage of consistencies and volumes varied too much across studies, so effect sizes are not estimable. Studies not using IDDSI terminology were excluded, which removed most EMG, acoustic and accelerometry work. And only two of the 70 studies reported paediatric data, which is a striking gap in a field where texture modification is used heavily in children. The authors' proposal of an open-access living review database is the right response to a literature moving faster than five-yearly reviews.

🔬 From Bench to Clinic

Cai J, Thibeault SL, Lungova V. Effect of Notch1 on Vocal Fold Re-Epithelization After Acute Injury. The Laryngoscope. 2026. 10.1002/lary.70760

Conditional knockout mice (n = 5 per group) in which Notch1 was deleted specifically in Lrig1-expressing vocal fold epithelial stem cells, followed by naphthalene-induced acute epithelial injury and examination at 1, 3 and 7 days. Losing Notch1 disrupted repair, worst in homozygotes: P63-positive basal cells and Ki-67-positive proliferating cells expanded, peaking at day 3 and still elevated at a week; KRT14 spread abnormally toward the squamo-respiratory junction; KRT13 and the stress keratin KRT17 rose and stayed high; and IBA1-positive immune cells accumulated and remained elevated at a week, past the point at which murine vocal fold epithelial healing is normally complete.

Translational promise. The finding worth holding is not that Notch1 matters. It is that losing it produced moreproliferation and worse repair, alongside inflammation that would not switch off. That is the signature of a wound healing badly, and it reframes scar as a timing failure rather than an excess of activity. We tend to picture fibrosis as too much healing. This says it may be healing that never receives the signal to stop. If the epithelium holds the clock, then epithelial integrity in the first days after phonosurgery is not cosmetic, it is regulatory, and how roughly we handle mucosa during resection may matter more than what we lay on top afterwards.

▪️ In Brief

Semaglutide and the thyroplasty you thought was finished. A 74-year-old developed new dysphonia nine years after successful bilateral medialisation thyroplasty, coinciding with 25 pounds of weight loss on semaglutide; evaluation showed marked atrophy and glottic insufficiency, and revision thyroplasty was required. One case, but worth carrying as GLP-1 use spreads through our clinics. 10.1002/lary.70732
Nanofat injection laryngoplasty. Thirty-one patients with glottal insufficiency from atrophy, paresis or scarring, about 0.5 mL per side, median 12 months. VHI-10 improved and better closure tracked with better voice, but 29% showed unwanted absorption within the first year. 10.1002/lio2.70511
Deep brain stimulation for laryngeal dystonia. Systematic review with descriptive pooled analysis from the Sataloff group. Worth knowing the state of the evidence before a neurologist asks. 10.1016/j.jvoice.2026.07.033
Botulinum toxin into the posterior cricoarytenoid for aspiration in bilateral adductor paralysis. A case series proposing chemical weakening of the abductor as an aspiration strategy. Small, but a genuinely unusual use of an old drug. 10.1016/j.jvoice.2026.07.018
Volitional swallows are not saliva swallows. Pharyngeal high-resolution manometry in 50 healthy adults: cued volitional 2 cc thin-liquid swallows generated higher pressures than non-cued saliva swallows at the velopharynx, tongue base and post-closure UES, with longer UES relaxation. No sex effects. Almost everything we know about pharyngeal pressure comes from a task the patient does not perform all day. 10.1007/s00455-026-10988-7
How good is cepstral peak prominence, really? Systematic review and meta-analysis of smoothed CPP accuracy in identifying voice disorders. Open access. If you report CPP in clinic notes, this tells you what the number is worth. 10.1016/j.jvoice.2026.06.017
GRBAS between expert raters. Reproducibility of auditory-perceptual evaluation among experienced raters. Perceptual rating is still our reference standard, which makes its reliability everyone's problem, and this issue already contains one paper (Miller) where inter-rater reliability of an objective measure was only 0.530. 10.1016/j.jvoice.2026.06.024
Vocal fold atrophy in Alzheimer's disease and related dementias. An association study raising a question none of us ask in the dementia clinic. 10.1002/lary.70761
Where the air actually goes during jet ventilation. Electrical impedance tomography during low-frequency jet ventilation for subglottic stenosis surgery, six patients and four complete datasets. Jet ventilation preferentially filled apical lung and consistently under-ventilated basal regions, with apical-to-basal volume ratios more than 100% higher than with a laryngeal mask. A pilot, but if it holds it changes how we think about long microlaryngeal cases. 10.1002/lio2.70488

Next issue: September 2026. Reply with corrections, disagreements, or a paper I missed.

Necati Enver, MD

Istanbul Voice Center

https://istanbulvoicecenter.com

Youtube, Instagram

Doç. Dr. Necati Enver - Ses Yutma ve Nefes Hastalıkları

Laryngology Brief, Issue 3: when the abstract and the paper disagree