Welcome to this month's journal review. We're covering the September twenty twenty-six issue of the Journal of the American Academy of Dermatology, and there are four pieces on the docket today — a letter to the editor taking a hard look at a recent reconstructive surgery study, two "JAAD Game Changer" retrospectives on practice-shifting papers, and a brief report on the legislative landscape around firefighter skin cancer coverage. Let's get into it.

First up is a letter to the editor, responding to a retrospective cohort study by Wang and colleagues on cartilage grafting during nasolabial interpolation flap reconstruction of alar defects — so this isn't new primary data, it's a critical commentary, and it's worth understanding both what the original study found and why these correspondents think we should pump the brakes before changing practice. The underlying study looked at close to two hundred eighty consecutive nasolabial interpolation flaps done over a sixteen-year span, and found that auricular free cartilage grafts were used in roughly four in ten cases. Grafting was associated with nearly a three-fold increase in the odds of what they called "compromised breathing," and a little over a three-fold increase in the odds of scar revision. On its face, that reads as a warning against routine structural grafting.

The letter writers raise several concerns that are worth sitting with. First, the breathing endpoint itself: it was defined pragmatically, as patient-reported difficulty severe enough to trigger a revision, a referral, or more than one dedicated office visit. That's a threshold built on practice patterns and documentation habits as much as on actual physiology, and it likely misses mild-to-moderate obstruction that patients live with but never flag. The suggestion is to bring in validated instruments — things like the NOSE scale or SCHNOS — alongside objective measures like peak nasal inspiratory flow, so future work captures the full spectrum of outcome rather than just the events that generated a chart note.

The second point is the big one, and it's a genuine teaching moment: confounding by indication. This is the situation where the reason a treatment was chosen is itself tied to the outcome, independent of whether the treatment works. Here, cartilage grafting was more common in alar-only defects and in tumors with deeper invasion into fat or muscle — exactly the scenarios where native structural support has already been compromised and dynamic collapse is more likely. So when we see grafted patients doing worse on a breathing endpoint, we can't tell whether the graft caused the problem, or whether the graft was placed precisely because these were the patients already headed toward a worse anatomic outcome. Multivariable adjustment helps, but it can only adjust for what was measured — it can't account for baseline valve competence or lining quality that nobody assessed preoperatively. The letter proposes propensity-score methods as a partial fix. In plain terms, a propensity score estimates each patient's predicted probability of receiving the graft based on their observed characteristics, and then compares outcomes only between patients with similar predicted probabilities — essentially trying to reconstruct the kind of balanced comparison a randomized trial would give you, using observational data. It's a better tool than a simple multivariable model for isolating the graft's own contribution, though it still can't fix confounding from things that were never recorded in the first place.

Third, the letter flags surgeon and era effects — graft use tracked with less-experienced surgeons and declined over the study period, which in a multi-surgeon cohort raises the possibility that flap thinning technique, inset tension, and evolving scar management are doing some of the work that's being attributed to the graft. The fix they propose is a mixed-effects model with a random effect for surgeon, which accounts for the fact that patients operated on by the same surgeon aren't statistically independent observations — ignoring that clustering can make an association look more significant than it really is.

Finally, they note that "scar revision" may often reflect aesthetic preference rather than structural failure, since most revisions in the original study were for appearance alone — so that endpoint may be capturing a low threshold for optimization rather than a true complication.

The bottom line for practice: this is a thoughtful critique, not a rebuttal with new data, so it doesn't overturn anything — but it does mean you shouldn't treat the original odds ratios as a clean causal signal against grafting. The genuinely useful takeaway is the reminder that reflexive, routine cartilage grafting in every nasolabial flap probably isn't warranted, but the decision should still rest on your own assessment of structural need — deep invasion, alar-only defects with loss of native support, and dynamic collapse — rather than on this dataset alone. Not yet practice-changing in either direction; more of a call for better-designed prospective work with standardized preoperative valve assessment.

Next, a shorter piece — a JAAD Game Changer commentary looking back at a multisite population-based cohort study on hydrochlorothiazide and skin cancer risk, originally published by Azoulay and colleagues through the Canadian Network for Observational Drug Effect Studies. The commentary's message is simple and clinically pointed: hydrochlorothiazide, one of the most commonly prescribed antihypertensives out there, is associated with about a one-and-a-half-fold increase in keratinocyte carcinoma risk with longer duration and higher cumulative dose. The practical takeaway the commentary draws out is one we should already be building into our habits — take a real medication history at every visit, and when you see hydrochlorothiazide on the list, that's a natural cue to reinforce sun protection and sun avoidance counseling. This isn't a reason to ask patients to stop their blood pressure medication, but it is a reason to flag it as a modest, dose-dependent risk factor worth discussing.

Third is another Game Changer piece, this one revisiting a multicenter randomized controlled trial comparing surgical excision, topical five percent five-fluorouracil, and methylaminolevulinate photodynamic therapy for Bowen's disease — squamous cell carcinoma in situ. The original trial used a noninferiority design, which is worth pausing on because it's a different question than the superiority trials most of us are used to reading. A noninferiority trial isn't asking whether the new treatment works better than the standard — it's asking whether the new treatment is not unacceptably worse, using a prespecified margin of acceptable difference. That framework makes sense here precisely because Bowen's disease is slow-growing with low mortality risk, and the patients who get it skew older with more comorbidities — so a less invasive option that's "good enough" can be a genuine win even without beating surgery head-to-head. The trial found that topical five-fluorouracil met that noninferiority bar against surgical excision, while the photodynamic therapy data were not conclusive one way or the other.

The commentary frames this through a dermatoethics lens — autonomy and beneficence obligate us to actually lay out these evidence-based alternatives for patients, particularly older patients with competing comorbidities, rather than defaulting to surgery as the only option discussed. The practical takeaway: for an appropriately selected older or medically complex patient with Bowen's disease, topical five-fluorouracil is a reasonable, evidence-supported alternative to excision that's worth explicitly offering as part of shared decision-making. The photodynamic therapy arm is more "interesting but not yet actionable" — the data there just weren't conclusive enough to draw a firm equivalence claim.

Last is a brief report, and this one's a straightforward descriptive, cross-sectional analysis rather than a study with inferential statistics — so there's no modeling technique to unpack here, but it's a genuinely useful piece of practice intelligence. The background: firefighters carry meaningfully elevated skin cancer risk from ultraviolet exposure, combustion byproducts, and fire-suppression materials, with a skin-cancer mortality rate close to sixty percent higher than the general population. Whether that translates into workers' compensation coverage, though, depends entirely on state law, and the authors set out to map that variability systematically.

Methodologically, they did what's essentially a structured legislative content analysis — searching a legislative database called LegiScan for bills combining "firefighter" with cancer-related terms, screening down from over twenty-eight hundred identified bills to five hundred twenty-one that got full-text review, and extracting variables like cancer subtype coverage, whether paid versus volunteer firefighters were included, years-of-service requirements, benefit scope, and whether the presumption of work-relatedness was rebuttable or not. They double-coded everything and resolved disagreements by consensus — a sensible quality-control step for a study built entirely on interpreting legal text, where two reviewers can genuinely disagree on how a statute should be classified.

What they found: forty-four states currently offer some form of presumptive coverage for melanoma and or nonmelanoma skin cancer in firefighters, and most of those presumptions are rebuttable, meaning an employer or insurer can still contest the claim with contrary evidence. Wisconsin was the pioneer, back in 1998. About seven in ten of these forty-four states explicitly provide disability and death benefits, and a similar proportion explicitly cover medical care. Just over half explicitly extend coverage to volunteer firefighters, with a five-year median minimum service requirement — and that matters, because even though full-time career firefighters make up only about a third of the national workforce, and part-time firefighters roughly one in eight, the majority of the firefighting workforce in this country is actually volunteer. So the states restricting coverage to paid firefighters are, numerically, leaving out most of the workforce. Looking forward, a dozen states introduced new bills between 2025 and 2026, three-quarters of which still used a rebuttable presumption structure, and there was real inconsistency in which cancers were named — melanoma was explicit in four pending bills, nonmelanoma skin cancer alone in two others.

The authors are appropriately modest about limitations: this is statutory text analysis, not claims data, so they can describe what coverage is supposed to look like on paper without knowing actual approval rates or how often rebuttable presumptions are successfully challenged in practice. Still, statutory eligibility is the necessary first gate before any claim gets adjudicated, so mapping it has real value.

The practical takeaway here isn't a change to clinical management, but it is a change worth making in documentation habits. When you're evaluating a firefighter-patient with a new melanoma or keratinocyte carcinoma, being explicit and thorough in documenting the diagnosis alongside a clear occupational exposure history can materially affect whether that patient successfully accesses workers' compensation benefits — and given how much interstate variability exists in whether coverage is even on the books, it's worth knowing whether your state has a presumptive statute at all, whether it's rebuttable, and whether it extends to volunteer status. There's also an explicit invitation here for dermatologists to engage in advocacy, given how much these statutes are still actively evolving.

That wraps up this month's four pieces — a caution against over-reading confounded observational data in flap reconstruction, two retrospective looks at how a drug-risk cohort study and a Bowen's disease noninferiority trial changed real-world counseling and treatment offering, and a reminder that our documentation carries legal weight well beyond the chart. Thanks for listening, and we'll see you next month.