Welcome back to the journal review — this is our walkthrough of Dermatologic Surgery, November twenty twenty-six, and we've got four pieces worth your time this month: a retrospective comparison of laser-assisted drug delivery regimens for burn scars, a first-in-human trial of a novel synthetic wound matrix for post-Mohs defects, a SEER-based survival analysis in extramammary Paget disease, and a reconstructive conundrum covering a nasty multi-subunit periocular defect. Let's get into it. First up, an original article out of the MedStar Washington burn center looking at laser-assisted drug delivery — combination triamcinolone and five-fluorouracil versus triamcinolone monotherapy — for hypertrophic burn scars. This is a single-institution, prospectively collected but retrospectively analyzed chart review, fifty-five patients enrolled in an established laser scar revision program. The clinical gap here is real: intralesional triamcinolone is first-line for hypertrophic scarring, but it's poorly suited to large or irregular burn scars, and it carries its own adverse effect burden — atrophy, telangiectasias. Fractional ablative CO2 laser creates microchannels that let you deliver topical medication transdermally, so the question is whether adding topical five-FU to topical triamcinolone, delivered this way, buys you anything over triamcinolone alone, extrapolating from prior data suggesting intralesional five-FU plus triamcinolone beats triamcinolone injection alone. Methodologically, patients weren't randomized — allocation was essentially determined by date, since the institution's standard of care switched from triamcinolone monotherapy to the combination regimen in March of twenty twenty-one. That's a quasi-experimental, before-after design dictated by practice change rather than investigator choice, and it's worth flagging because it introduces an obvious confounder: anything else that changed in practice patterns around that time — laser settings, patient selection, concurrent therapies — rides along with the group assignment. The authors do show the two groups were statistically similar on the usual demographic and scar variables, which helps, but it's not randomization. Statistically, they used one-way ANOVA with multiple comparisons, splitting this into two comparisons: each post-laser session against the preoperative baseline, corrected with Dunnett's method, and the two treatment groups against each other at each session, corrected with Sidak's method. This two-pronged approach makes sense here because they're asking two different questions — did either treatment work at all, and did one treatment outperform the other — and each needs its own correction for multiple testing to avoid false positives from running comparisons at three separate treatment timepoints. What's genuinely instructive is that the authors ran a formal power analysis and found they'd have needed sample sizes ranging from around one hundred forty to over four hundred patients, depending on the outcome measure, to reliably detect a between-group difference. With only fifty-five patients, this study was never adequately powered to find a group difference even if one existed — that's an important lens for interpreting the null result that follows. On results: both groups improved significantly from baseline across the subjective scar scales and the objective durometer firmness measurements after three treatments. But there was no statistically significant difference between combination therapy and monotherapy at any timepoint. The authors do note that when you look at magnitude of improvement from baseline, the combination group trended numerically better — but this did not reach significance, and given the power analysis above, you should read that as hypothesis-generating noise rather than a signal. Adverse effects were low and equivalent in both groups — no telangiectasia, atrophy, or infection in either arm, and the occasional blistering or hypopigmentation tracked with laser settings rather than drug choice. Bottom line for your practice: this is not practice-changing on its own, but it's a reasonable data point suggesting you don't need to reflexively add topical five-FU to your laser-assisted triamcinolone protocol for burn hypertrophic scars — the combination doesn't clearly buy you more efficacy and doesn't meaningfully reduce adverse effects either, undercutting the rationale for routinely adding a drug with its own irritant profile. Given the underpowering, I'd call this supportive rather than definitive — it doesn't close the door on combination therapy, but it also doesn't give you a reason to open it if you're not already using it. Second article: a randomized controlled, multicenter, first-in-human safety study of a microporous annealed particle wound matrix — they call it MAP — for post-Mohs defects left to heal by secondary intention. This is genuinely novel technology: a fully synthetic, flowable suspension of hydrogel microspheres that you inject into the wound and then light-anneal in place for ten minutes to form a stable, porous, bioresorbable scaffold. The pitch is that it fills wound volume better than existing options, promotes a pro-regenerative type-2 immune response rather than the fibrotic type-1 response you get with many synthetic fillers or tissue substitutes, and does this with a single application rather than the serial dressing changes secondary intention usually demands. Design-wise, this was a two-to-one randomization — MAP versus a hydrocolloid dressing control — open-label, across three US dermatology centers, conducted under an FDA investigational device exemption. The primary endpoint was safety, specifically serious adverse device effects, which is exactly what you'd want for a first-in-human study of a new biomaterial — you're establishing that it doesn't do harm before you ever power a trial to prove superiority on efficacy. Everything on the efficacy side — pain, wound bed score, time to granulation, time to closure, scarring — was explicitly exploratory, and the authors are upfront that their p-values here are hypothesis-generating, not confirmatory. That's an important distinction to carry through the results: with only forty patients split twenty-six to fourteen, this trial was not powered to prove efficacy claims, only to look for safety signals and generate promising trends. On the numbers: no serious adverse device effects in either group. Nonserious adverse events trended lower with MAP — roughly twenty-seven percent of patients versus fifty percent with hydrocolloid — and serious adverse events occurred only in the control group, in two of fourteen patients, zero with MAP. Neither of these differences reached statistical significance, but directionally they favor safety of the new matrix. Granulation and time-to-closure were essentially identical between groups — around ten weeks for both — so the headline reassurance is that this product does not delay healing, which was a real theoretical concern given it's filling the wound bed with foreign material. The one standout finding, and the one I'd flag as clinically interesting, is wound bed quality: MAP-treated wounds reached a favorable wound bed score about two weeks earlier than control. And scarring at six months, scored by the Patient and Observer Scar Assessment Scale, was significantly better in the MAP group. Limitations are substantial and the authors essentially telegraph them through the exploratory-endpoint framing: small sample, unequal group sizes, open-label design with no blinding for patients or treating investigators — though they did use blinded reviewers for the wound bed score and photograph-based assessments, which is a reasonable bias-mitigation step within an otherwise unblinded trial. This is a first-in-human study, so by definition it's not meant to be definitive on efficacy. Practical takeaway: this is interesting, not yet practice-changing. The safety profile looks clean and reassuring, which is the necessary first step for any new biomaterial heading toward broader adoption in post-Mohs reconstruction. The scarring benefit is the result I'd watch for in a larger, blinded, adequately powered follow-up trial — if it holds up, a single-application product that improves cosmetic outcome without delaying healing would be a meaningful addition to the secondary-intention toolkit, particularly for larger or irregular defects where current fillers like Integra or collagen substitutes have volume limitations. For now, file it as promising early-phase data, not something to change tomorrow's dressing orders. Third, a short communication — a national retrospective cohort study using the SEER database comparing Mohs micrographic surgery against wide local excision for disease-specific survival in extramammary Paget disease. The background gap is clean: we already know Mohs gives better local control and lower recurrence than wide excision in extramammary Paget, but survival data comparing the two approaches has been sparse, and that's what this paper tries to fill. They pulled one hundred sixty-six patients from SEER's seventeen-registry database spanning two thousand to twenty twenty-two — ninety-nine treated with wide excision, sixty-seven with Mohs. The statistical backbone here is Kaplan-Meier survival estimation paired with Cox proportional hazards regression, and it's worth spending a moment on why. Kaplan-Meier lets you estimate the probability of surviving disease-free over time while properly handling patients who drop out of follow-up at different points — some followed two years, some followed fifteen — without throwing away their partial information. Cox regression then lets you ask whether surgical modality predicts survival while adjusting for other variables simultaneously, by modeling the instantaneous risk of death at any given moment as a function of covariates, rather than requiring you to specify an exact survival distribution. Because they detected non-proportional hazards for sex — meaning the relative risk associated with sex wasn't constant over time, violating a core Cox assumption — they stratified sex in the model rather than including it as a standard covariate, which is a sound fix that lets each sex have its own underlying hazard shape while still adjusting for its confounding effect. They also stratified age and stage rather than estimating individual hazard ratios for each category, because event counts within those subgroups were too sparse to estimate stable coefficients — a sensible pragmatic call given a cohort of this size. The results: five-year disease-specific survival was around ninety-eight percent with Mohs versus ninety percent with wide excision. In the unadjusted, univariate analysis, that difference did not reach statistical significance. But after adjusting for household income, year of diagnosis, and the stratified variables of age, sex, and stage, Mohs was associated with a significantly lower risk of disease-specific death — roughly an eighty to ninety percent relative reduction in hazard, and this adjusted result was statistically significant. That's a notable example of confounding or power masking an effect in unadjusted analysis that becomes apparent once you account for the structure of the data — worth remembering next time you see a flat univariate result that doesn't match the adjusted story. Limitations are the ones you'd expect from any SEER-based retrospective design: no granular staging detail, no margin or recurrence data captured in the registry, and an inherent asymmetry since Mohs coding didn't begin until two thousand three, meaning the Mohs cohort by definition skews to more recent years — reflected here in the mean diagnosis year being five years later for Mohs patients. Even with adjustment for year of diagnosis, residual secular trends in staging, imaging, or systemic therapy could still be baked in. This is also an inherently small cohort for a rare disease, and the wide confidence intervals around the hazard ratio reflect that. Practically: this adds real weight to the existing local-control argument for Mohs in extramammary Paget disease, now with a survival signal attached, though I'd stop short of calling this definitively practice-changing given the retrospective design and small numbers — it won't be news to anyone already favoring Mohs for this indication, but it gives you a stronger evidence-based talking point when counseling patients on surgical approach, particularly in anatomically sensitive sites like the vulva, scrotum, or perianal region where subclinical extension is common and margin control genuinely matters. Last up, a reconstructive conundrum — a case-based technique discussion rather than a study, so no methods or statistics here, just the thinking behind a tricky repair. A sixty-five-year-old woman with melanoma in situ of the right nasal sidewall ends up with a two-by-one-and-a-half centimeter Mohs defect spanning four cosmetic subunits simultaneously — nasal sidewall, medial canthus, inferior eyelid, and superomedial cheek — with the canthal involvement split roughly eighty-five percent inferior and fifteen percent superior to the canthal angle. The core design challenge the authors walk through is that the medial canthus is a tethered junction with its own concavity that's easy to flatten or convert into a convex mound if you don't actively reconstruct the negative space, and any periocular repair in this territory lives or dies on whether it avoids ectropion. Their solution is a combination repair: a glabellar bilobed transposition flap to resolve the nasal sidewall and medial canthus, paired with a crescentic cheek advancement flap for the superomedial cheek. The logic for the bilobed is elegant — the primary lobe is harvested from the upper eyelid and superior nasal sidewall, which gives you a better color and thickness match for repairing the inferior canthus and nasal sidewall than the glabellar skin would alone, while the secondary lobe comes from the glabella itself and gets aggressively thinned to match. The flap gets tacked to the periosteum at the medial canthal tendon specifically to recreate the boundary between superior and inferior canthus — that anchoring point is doing real reconstructive work, not just securing the flap. For the cheek component, the crescentic advancement is designed around horizontal tension vectors rather than vertical ones, specifically to avoid pulling down on the lower eyelid — and again, a periosteal tacking stitch at the nasofacial sulcus is the key maneuver, simultaneously supporting the flap, preventing ectropion, avoiding canthal webbing, and delineating the subunit boundaries. The authors are explicit about what they rejected and why: second intention and full-thickness grafts would blur the subunit junction lines and risk ectropion or poor cosmetic match; a contralateral forehead flap would be overkill in bulk and morbidity; a rotation flap for the cheek would need an incision extending to the temple with unwanted vertical tension. Outcome was good functionally and cosmetically, with only minor residual fullness at the medial canthus managed with a little triamcinolone at three months, and the expected minor blunting of the melolabial fold that comes with any cheek advancement. The teaching points to carry forward: multi-subunit periocular defects need you to think about both concavity restoration and tension vector direction simultaneously, the bilobed sourcing from the superior canthus is a nice trick for color-matching the inferior canthus specifically, and periosteal fixation is your main lever for distributing tension horizontally and protecting eyelid position in these combined repairs. That wraps our four articles this month — a burn scar study reinforcing that less may be just as good as more when it comes to combining topical agents in laser-assisted delivery, promising early safety and scarring data on a new synthetic wound matrix for post-Mohs secondary intention healing, a SEER cohort adding a survival argument to the existing case for Mohs in extramammary Paget disease, and a nicely reasoned combination flap approach for a genuinely difficult multi-subunit periocular defect. Thanks for listening, and I'll see you next month.