Welcome back to the journal review. This is your walkthrough of the September 2026 issue of Dermatologic Surgery, and we've got four articles on deck today — a cost-modeling piece pitting superficial radiotherapy against Mohs, a big database study on modifiable risk factors for surgical complications, a prospective look at referral documentation quality, and a technique paper on a novel helical rim closure. Let's get into it.

First up is an original article, essentially a health-economics modeling study, comparing image-guided superficial radiotherapy — IGSRT — against Mohs surgery for squamous cell carcinoma, using CPT and Medicare rates. The background here is worth sitting with because you've all seen the marketing: IGSRT vendors advertise lower recurrence, no scar, less morbidity, and — critically — lower cost than Mohs. The authors wanted to actually pressure-test that cost claim using real Medicare fee schedule numbers rather than anecdote.

Methodologically, this isn't a clinical trial — it's a modeled cost analysis. The authors picked three representative SCC scenarios — an in-situ lesion, a T1, and a T2 — and priced them out separately for head-and-neck versus trunk-and-extremity locations, using actual CPT codes at Minnesota's Medicare Administrative Contractor rates. For Mohs, they anchored the number of stages to a previously published mean of 1.4 stages to clear, so the Mohs cost included that fractional extra stage plus the reimbursement for whatever repair was used — closure, flap, or graft. For IGSRT, they built the cost from the actual billing components: treatment planning, ultrasound guidance, simulation, hyperthermia, and the delivery code itself, drawing on a recent real-world series of over three thousand lesions to know what dosing patterns and lesion sizes are typical in practice. This is a straightforward accounting exercise dressed as research — no inferential statistics, just multiplying real fee-schedule numbers by realistic utilization scenarios, so there's no statistical technique to teach here, but the modeling logic itself is the pedagogical point: cost claims in the literature should always be checked against the actual billing architecture, not stated as a given.

The results are pretty stark. Mohs with repair ran about one thousand to sixteen hundred dollars depending on repair complexity, fairly consistent whether the lesion was on the head and neck or the trunk and extremities. IGSRT, by contrast, showed an enormous range — from about fourteen hundred dollars up to over ten thousand dollars for a ten-session hypofractionated course, and from about two thousand up to over twenty-five thousand dollars for the guideline-recommended twenty-five session hyperfractionated course. The swing depends almost entirely on how aggressively ancillary codes — the ultrasound guidance, hyperthermia, and simulation codes — are billed per session versus just once. And here's the number that should stick with you: the actual radiation delivery code itself accounts for less than four percent of the total session cost. Nearly all the cost is in these adjunct codes. Bottom line, the authors calculate SRT can run anywhere from about ten percent more expensive up to nearly twenty-three times more expensive than Mohs, and even in the cheapest billing scenario, hyperfractionated IGSRT is still at least one and a half times pricier than Mohs. Time-wise, hypofractionated treatment takes similar total chair time to Mohs but spread across ten separate visits, and hyperfractionated dosing doubles total time and stretches it across twenty-five visits — which the authors reasonably flag as a real burden, especially factoring in travel and caregiver time that isn't even captured in the cost model.

Limitations are inherent to the design — this is a theoretical model built on published billing patterns and one real-world dosing series, not an actual claims analysis of matched patients, so it can't capture regional fee variation, bundling practices, or facility-fee settings outside Minnesota. It also doesn't touch efficacy or recurrence data head-to-head — it's purely a cost argument. But the practical takeaway is genuinely useful for you: the next time a patient arrives having been told SRT is the cheaper, scarless option, you now have a citable, code-level rebuttal on cost — this is practice-relevant ammunition for patient counseling, not just an academic curiosity, even though it doesn't change your surgical indications one bit.

Second article, a retrospective cohort study using the TriNetX network, asking which modifiable and non-modifiable patient factors predict acute complications after excisional dermatologic surgery — Mohs, and both benign and malignant excisions. The gap they identify is real: prior work has looked at individual risk factors like anticoagulation or immunosuppression in isolation, but nobody had done a broad multi-factor sweep specifically in dermatologic surgery while trying to control for the fact that comorbid patients tend to stack risk factors together.

This is a huge cohort — roughly nine hundred thirty thousand patients — pulled from a federated international electronic health record database, though the vast majority were US-based. Because this is retrospective registry data rather than a trial, the authors used propensity score matching to deal with confounding, and this is worth actually teaching. Propensity matching takes all your covariates — here, the Charlson comorbidity diagnoses plus age, sex, race, and ethnicity — and collapses them into a single predicted probability of belonging to the exposed group, essentially a summary score of how "sick" or "typical" a patient looks. Then it pairs each exposed patient — say, someone with obesity — to an unexposed control with a similar propensity score, one-to-one, discarding anyone who can't be matched closely. The logic is that if the two groups end up statistically indistinguishable on all those baseline characteristics, any difference in complication rates that remains is more plausibly attributable to the exposure itself rather than to the fact that, for instance, smokers also tend to have more peripheral vascular disease. It's a nice alternative to just throwing everything into one giant multivariable regression, particularly here because the analytic platform itself didn't support multivariable modeling — so matching was really the only tool available to isolate each variable's effect one at a time. The authors confirmed adequate balance after matching using standardized mean differences, which is the standard check that the matching actually worked.

On to results: the overall complication rate across nearly a million procedures was low, under two percent, with infection being the most common single complication at just over one percent, followed by dehiscence, then hematoma or seroma. Nicotine use, obesity, anticoagulant use, and antiplatelet use all predicted every complication category studied. Escalating BMI tracked with escalating infection and dehiscence risk in a dose-dependent way, and severe obesity carried a risk profile comparable to active smoking. Vitamin D deficiency and oral corticosteroid use — particularly at higher prednisone-equivalent doses — predicted infection and dehiscence specifically. Interestingly, oral immunosuppressants as a class — methotrexate, azathioprine, mycophenolate, the calcineurin inhibitors, the mTOR inhibitors — showed no significant association with complications, which is a genuinely useful negative finding given how often we reflexively worry about these agents perioperatively. And critically, glycemic and blood pressure control mattered more than the diagnosis itself: well-controlled diabetics and hypertensives looked essentially like patients without the disease, aside from a persistent modest bump in infection risk even in well-controlled diabetics, whereas poorly controlled disease in either category raised odds across the board.

Limitations are the usual registry caveats — this is ICD-10 code-based outcome ascertainment, so complications are only as good as what got coded, there's no adjudication of complication severity, and residual confounding is always possible even after matching, since propensity scores can only balance what you measured, not unmeasured factors. The complication window was fixed at thirty days as a surgical convention, which is reasonable but means anything presenting later gets missed. The practical takeaway here is genuinely practice-relevant for your preoperative counseling and optimization conversations: this reinforces that smoking cessation, weight management, glycemic and blood pressure optimization, and vitamin D correction aren't just generic wellness advice — they're the specific modifiable levers tied to your acute complication risk, and the reassurance about oral immunosuppressants not raising risk is something you can use directly when timing surgery around a patient's immunosuppressive regimen.

Third article, a prospective observational cohort study out of a single private Mohs practice, looking at the quality of information in surgical referrals and what predicts a referring provider getting the anatomical site wrong. This sits squarely in the patient-safety literature around wrong-site surgery, which remains a real risk in our specialty given the time lag between biopsy and surgical treatment.

The design is simple and pragmatic: a thousand consecutive referrals were tracked, and for each one the surgeon recorded whether the documented biopsy site matched what was clinically evident at the time of surgery, along with what documentation elements were present — clinical photographs, photo quality, anatomical diagrams, and triangulation or distance measurements from landmarks. Statistically, this is chi-square testing of independence across categorical variables, which is exactly the right tool here because both the outcome — correct or incorrect description — and every predictor — provider type, presence of a photo, anatomical location — are categorical rather than continuous. A chi-square test essentially asks whether the observed distribution of correct-versus-incorrect across subgroups deviates from what you'd expect if there were no relationship at all, and the standardized residuals the authors report afterward tell you which specific subgroups are driving that deviation, so you can say not just "location matters" but specifically "nasal lesions are the problem."

Overall correctness was high, about ninety-three percent, so this isn't reporting an epidemic of wrong-site errors, but the factors driving the misses are informative. Anatomical location mattered significantly — nasal lesions were the standout problem area, significantly more likely to be misdescribed than any other site, while extremities, face, and neck trended toward better-than-expected accuracy. Documentation elements mattered a lot: referrals containing a clinical photograph were significantly more likely to be correct, and so were those containing an anatomical diagram — with diagrams showing the stronger effect of the two. Interestingly, photo quality itself didn't matter once a photo was present at all — even a mediocre photo beat no photo. And triangulation, meaning distance measurements from fixed landmarks, was rarely used, in fewer than five percent of referrals, but when it was used, site description was correct one hundred percent of the time — a small numbers effect, worth noting as a trend rather than a proven significant finding given how few cases used it. Tumor type, patient sex, age, lesion size, and precision of description all failed to reach statistical significance, though smaller lesions and older patients trended toward worse accuracy.

The limitations are honestly stated: single-center, private-practice, single-surgeon design limits generalizability, and there's no downstream outcome data linking these documentation gaps to actual wrong-site events or medicolegal consequences — this measures a process metric, not a hard patient-safety endpoint. Practically, though, this is directly actionable for your own practice and for how you train referring offices: nasal lesions deserve an extra beat of scrutiny before you incise, and pushing your referral network toward routinely including even a basic photo and a simple diagram — not necessarily a perfect photo — meaningfully reduces your site-identification risk. This is a low-cost, high-yield administrative fix, not just an interesting academic observation.

Last article, a technique piece — a prospective case series introducing what the authors call the "Sculpo" repair, a linear closure with cartilage sculpting for helical rim defects up to two and a half square centimeters. The clinical problem is familiar: helical rim reconstruction is traditionally size-staged, with straightforward linear closure reserved for defects under about one and a half square centimeters, and anything larger pushed toward flaps — helical advancement, postauricular interpolation — that bring their own morbidity, staged procedures, and hematoma risk, particularly relevant in your anticoagulated patients.

The technique itself is the heart of this paper. After Mohs clearance, rather than widely undermining skin to achieve tension-free closure, the surgeon incises the underlying cartilage and removes a narrow strip — about two millimeters wide — centered over the defect and tapering into the Burow's triangles, then trims and sculpts the cut cartilage edges with scissors until the skin edges reapproximate without needing skin undermining at all. The design intentionally reshapes the helical arc, and the resulting closure line typically ends up four to five times the width of the original defect. The rationale, as the authors state it, is that removing a small amount of structural cartilage lets you convert what would be a high-tension linear closure into a low-tension one while preserving the postauricular sulcus completely intact — avoiding the hematoma risk that comes with undermining that space, and sidestepping the second wound and staged healing time that flaps or grafts require.

This was a prospective single-arm cohort of fifteen patients who completed follow-up, evaluated at suture removal for early complications and at a minimum of three months for scar outcomes, rated by two blinded Mohs surgeons using a standardized scar scale plus patient-reported outcomes on the Patient Observer Scar Assessment Scale. There's no comparator arm and no inferential statistics here — this is a descriptive case series, so there's no statistical technique to unpack, just outcome reporting.

The results were favorable across the board. No infections, no hematomas, and just one dehiscence, in a superior helical rim defect. On blinded scar grading, the vast majority of ratings showed minimal scar spread, minimal erythema, essentially no visible suture tracking, and ninety percent of ratings were judged cosmetically desirable overall. Patient-reported satisfaction scores were strongly favorable as well on the POSAS instrument.

The authors are appropriately modest about limitations: small sample, no control group against the standard flap alternatives, and most follow-up capped at three months, which is short given that scar maturation continues well beyond that window. This is a promising technique article, not a practice-changing one yet — it hasn't been compared head-to-head against advancement or interpolation flaps, so you can't say from this data alone that outcomes are equivalent or superior, only that this single-arm series looks good. That said, for your own case selection, particularly in anticoagulated patients or those who strongly prefer a single-stage procedure, this gives you a reasonable, low-morbidity option to consider for helical defects in that one-and-a-half-to-two-and-a-half-square-centimeter range where you might otherwise default straight to a flap — worth trying and tracking your own outcomes, but not yet evidence strong enough to displace flap reconstruction as your default for that size range.

That wraps up this month's four articles — a sobering cost reality-check on superficial radiotherapy, a large database confirmation of which modifiable risk factors actually matter before you operate, a practical referral-documentation fix you can implement immediately, and a promising new option for helical rim closure worth keeping on your radar. Thanks for listening, and we'll see you next month.