Welcome back to the journal review. This is Dermatologic Surgery, April 2026, and we've got four pieces to get through — two retrospective analyses touching on reconstruction, a small pilot study out of hair transplantation, and a prospective pain-control study you'll want to think about for your injectable and laser days, if not your excisional ones.

Let's start with the first original article, a retrospective cohort study out of Cooper University Health Care looking at early takedown of staged interpolation flaps — postauricular, melolabial, and paramedian forehead.

You know the conventional wisdom here: leave the pedicle in for around three weeks to guarantee neovascularization off the recipient bed before you divide it. The gap this group is addressing is that the evidence supporting earlier takedown — under three weeks — is reasonably established for melolabial and paramedian forehead flaps from prior single-center work, but the postauricular flap has essentially no data behind early division. So they set out to fill that specific hole.

Methodologically, this is a straightforward single-center retrospective chart pull spanning 2015 through 2024, roughly 141 patients total across the three flap types — 75 postauricular, 38 melolabial, 28 paramedian forehead — split into early takedown, defined as under 21 days, versus conventional, 21 days or more. A retrospective design here makes sense for the obvious reason: you're not going to randomize pedicle duration in a way that's ethically comfortable or practically feasible when the conventional interval is so entrenched, so mining existing takedown timing variation across your own practice is the pragmatic way to generate a comparison. They used chi-square for categorical variables and t-tests for continuous ones, which is appropriate for a straightforward two-group comparison like this.

The results are reassuringly boring, which is exactly what you want here. About two-thirds of the whole cohort actually underwent early takedown already, averaging around 14 to 16 days for postauricular and melolabial flaps and around 15 and a half days for paramedian forehead flaps, compared to roughly 22 to 23 days in the conventional groups. Complication rates — infection, dehiscence, necrosis — were low across the board and did not differ significantly between early and conventional groups for any of the three flap types. The postauricular group, the one with the least prior data, showed infection, dehiscence, and necrosis each in roughly one in ten early-takedown patients versus single-digit percentages in the conventional group, but none of those differences reached significance. Melolabial flaps had a somewhat higher infection rate in the early group — about 22% versus zero in conventional — which sounds concerning until you register that with only 23 and 15 patients per arm, that gap simply doesn't clear statistical significance, and the paramedian forehead group had zero complications in the early arm at all. Flap and defect sizes were comparable between groups throughout, so size wasn't confounding the comparison.

The authors' takeaway is that this extends the accelerated-takedown literature to the postauricular flap specifically, which had been the one flap type without dedicated safety data. Worth flagging honestly: this is retrospective, single-center, and the subgroups — particularly paramedian forehead at 14 and 14 — are underpowered to detect anything but a fairly large difference in complication rates. The demographic imbalances between arms, especially in the melolabial cohort where the early group was substantially older and had far higher rates of anticoagulant use and male sex than the conventional group, aren't adjusted for statistically, so residual confounding is possible even though the topline complication numbers look clean.

Practically, I'd call this one modestly practice-affirming rather than practice-changing on its own — it doesn't overturn anything, but it adds postauricular flaps to the growing list of interpolation flaps where you likely don't need to make your patient live with an external pedicle for a full three weeks. If you've already been taking these down around two weeks based on clinical judgment, this gives you a citable rationale; if you've been rigidly holding at three weeks, this is reasonable evidence to start flexing that, particularly given the quality-of-life burden of the pedicle that the authors and prior literature both emphasize.

Next is a pilot study — genuinely small, this one's a short communication rather than a full powered trial — looking at 24-hour delayed implantation in follicular unit extraction hair transplantation.

The background tension is a known controversy in hair restoration: immediate implantation after extraction is the traditional standard, but delayed implantation would give real workflow flexibility for staged or high-volume sessions. Prior literature is a mixed bag — some showing minimal viability loss over several hours in chilled saline, one series showing a drop to around 79% survival at 24 hours, another showing a sharp fall to 40% survival at room temperature but improved to about 76% with 4-degree storage, and another suggesting specialized cold solutions can extend viability well past 24 hours. So the question of optimal storage temperature and solution remained open, and that's the specific gap this pilot addresses.

The design is about as minimal as you can run and still call it a study: three male patients, split-scalp design, fifty grafts harvested per patient, twenty-five implanted immediately on one side and twenty-five stored at 4 degrees Celsius in HypoThermosol for 24 hours before implantation on the other side. The rationale for this split-body, intrapatient design is obvious and the authors would likely say the same — it controls for every patient-level variable, since each subject serves as his own control, which matters enormously when your total sample is only three people. Follow-up was global photography and trichoscopic analysis at six and nine months.

The results: regrowth was clearly present on both sides in all three patients, and at nine months follicular unit density was essentially equivalent — about 19 per square centimeter on the same-day side versus 18 and a third on the delayed side, a difference the authors themselves note is not statistically significant, though they're upfront that with an n of three, the confidence interval around that comparison is wide enough that this significance testing is basically illustrative rather than meaningful. Hair width, hairs per follicular unit, and terminal-to-vellus ratios tracked comparably between sides too, with the modest declines in hair caliber attributed to the deliberate use of single-hair grafts along the frontal hairline rather than to any storage effect.

The authors propose plausible mechanisms — hypothermic slowing of metabolism reducing oxidative stress and apoptosis, and the HypoThermosol solution itself mitigating cold-induced cellular stress — borrowing conceptually from solid organ preservation science. They're honest that the sample size sharply limits generalizability, and they call explicitly for larger randomized comparisons across storage solutions and durations.

Practically: this is interesting, not actionable. It's hypothesis-generating for surgeons doing high-volume FUE who might want workflow flexibility to stage extraction and implantation across days, but with three patients and no formal statistical power, you shouldn't be changing storage protocols off this alone. File it as a proof-of-concept worth watching for a properly powered follow-up.

The third piece is a benchmark analysis using the TriNetX federated research network, looking at national repair-type selection after Mohs surgery.

The gap here is refreshingly simple to state: despite reconstruction being central to Mohs outcomes, nobody had published real-world national data on what repair types are actually being used, broken down by anatomic site. The authors pulled every CPT-coded Mohs case from 2006 to 2024 — north of 600,000 cases — and bucketed the subsequent same-day repair into reconstructive classes using CPT groupings, then stratified by anatomic subunit using ICD-10 codes. Using a federated claims database like this is really the only way to get a number this large; the tradeoff, which the authors acknowledge, is that you lose surgeon-level and even patient-level clinical detail and you're trusting CPT coding accuracy across a hundred contributing institutions.

Across all sites combined, complex linear repairs were the single most common category at about a third of cases, followed closely by no-same-day-repair — meaning second intention, delayed grafting, or referral off-service — and local flaps, each at around one in five cases. Intermediate repairs also sat around one in five. Skin grafts were used in about 8%, and both interpolation flaps and island pedicle flaps were rare, under 1% each. Breaking it down by site tells the clinically expected story: local flaps dominated on the nose at nearly 30%, no-same-day-repair was highest on the eyelid at 55% and the lip at 43%, reflecting reliance on second intention or specialist referral in those areas, and complex repairs were heavily favored on the trunk, extremities, scalp, and neck — over 40% on scalp and neck alone — where you'd expect more straightforward linear closure of larger, less cosmetically constrained defects.

The temporal trend analysis is where this gets genuinely useful rather than just descriptive. Island pedicle flap use fell sharply — from under 1% to about one-tenth of a percent — right around 2013, which lines up precisely with a CPT clarification requiring an axial artery for that code, meaning this is very likely a coding artifact rather than a true practice shift. Complex repair usage dropped from roughly 37% down to about 28% after a 2020 CPT revision that redefined what qualifies as complex closure, with a corresponding rise in intermediate repair coding — again, this reads as reclassification rather than surgeons actually changing what they physically do to a wound. Local flaps, skin grafts, and no-same-day-repair stayed relatively flat with only a slight downward drift.

Limitations are stated plainly by the authors: TriNetX is deidentified, so there's no way to stratify by surgeon experience, region, or training background the way the Improving Wisely campaign did for stage numbers. CPT miscoding is possible, especially for rarer repair types, and cases where a Mohs defect was reconstructed by a different specialty on the same day still get captured in the dataset — though they note nondermatologists perform under 1% of Mohs cases nationally, so this is a small contamination risk.

Practically, this is a genuinely useful benchmarking tool rather than a practice-changer in the traditional sense — it won't tell you to alter your technique, but it gives you a national reference point to sanity-check your own repair distribution against, in the same spirit as the Improving Wisely stage-utilization reports. If your personal practice looks wildly out of step with these national percentages for a given site — say, you're doing complex linear closure on the nose far more than local flaps — that's worth a moment of self-audit, even if there's no right answer being asserted here.

Last up, a prospective split-lesion study out of India evaluating a wearable vibrating kinetic anesthesia device — brand name Vybe — for procedural pain during intralesional steroid injections and platelet-rich plasma therapy.

The rationale draws on gate control theory: vibrating large-diameter A-fibers is thought to dampen nociceptive transmission, and while this has been explored before in cosmetic dermatology, the authors note prior studies were limited by small samples, healthy-volunteer populations, or narrow procedure types. Their gap is real-world generalizability across a large, diverse dermatologic population.

The design is a split-lesion, intrapatient-controlled study — 629 patients, each undergoing the same procedure on two comparable sites, one with the vibrating device and one without, rated by a blinded observer using the Visual Analog Scale, the Numeric Rating Scale, and a modified verbal descriptor scale. The intrapatient split design is the right choice for the same reason as the hair transplant pilot — it cancels out interpatient variability in pain threshold, which is otherwise a huge confound in pain research, and here it's paired with genuine statistical power given the sample size.

The results were consistent and large in magnitude. Mean VAS pain scores dropped from about six and a half down to just under four with the device, and NRS scores showed essentially the same drop — differences that were highly statistically significant with large effect sizes, Cohen's d over one, which is well beyond the threshold where you'd worry this is statistically significant but clinically trivial. About three-quarters of patients cleared the two-point minimal clinically important difference threshold on these scales, composite success — meaning a two-point improvement on both scales plus a stated preference for the device — was achieved in just over 80% of procedures, and patient preference for future use of the device was overwhelming, at roughly 95%. Verbal pain descriptors shifted meaningfully from "stabbing" and "agonizing" at baseline to "dull" and "prickling" with the device. No adverse events were recorded in any of the 629 procedures. Subgroup analysis showed PRP patients benefited more than those getting intralesional steroid injections, and patients with alopecia areata were the weakest responders relative to androgenetic alopecia or keloid patients — the authors speculate this may reflect differences in anticipatory anxiety or inflammatory nociceptive thresholds in alopecia areata, though that's interpretive on their part. Procedure type, baseline pain score, and age under 40 all independently predicted a better response on regression analysis.

I'd flag one thing myself, not stated as a limitation in the provided discussion: the procedures studied here — intralesional injections and PRP — are needle-based but not excisional, so direct extrapolation to Mohs defect anesthesia, punch biopsies, or larger local anesthetic infiltration for flap reconstruction is my own inference, not something the authors tested or claimed. The mechanism and the device generalize plausibly to your injection of local anesthetic itself, which is really the more relevant application for a Mohs practice — vibration at the injection site rather than during the procedure that follows.

Practically, this one leans toward practice-changing, at least for anyone doing high volumes of injectable procedures — intralesional steroids, PRP, filler, even the local anesthetic infiltration step before excision — given the size of the effect, the safety profile, and the low barrier to adoption. It's a reasonable device to pilot in your own injection workflow if patient-reported comfort and needle anxiety are a friction point in your practice.

That wraps this month's four. To sum up the throughlines: two flap and repair-pattern papers that mostly reassure and benchmark rather than mandate change, a hair transplant pilot that's a hypothesis worth watching rather than acting on, and a pain-control device study with genuinely actionable, large-effect data for your injection-heavy days. Thanks for listening — see you next month.