Diazoxide has been an oral prescription drug since the 1970s, and one of its frequent side effects is excess body hair. People have put it on the scalp before, in animals and in small human reports, with no controlled trial published. Poor solubility is a likely reason it went no further.
The short answer: Diazoxide opens ATP-sensitive potassium (K-ATP) channels like minoxidil sulfate does, without needing SULT1A1, the follicle enzyme that activates minoxidil. Excess body hair is among the most frequently reported side effects of oral diazoxide in children treated for hyperinsulinism. In a 16-month study of five treated stumptailed macaques on 5% topical diazoxide, three regrew hair and two kept theirs, with no statistical test (Uno 1990). We found no published randomized trial of topical diazoxide for hair loss, so its efficacy in people is unknown.
Anagen's Topical Diazoxide is a compounded 3% scalp solution, prepared on prescription after a licensed clinician's review. It is not FDA approved, no diazoxide product is approved for topical use or for hair loss, and no controlled trial has tested it.
How does topical diazoxide work for hair growth?
Opening ATP-sensitive potassium channels is the leading proposed mechanism for how this class of drug grows hair. The support is organ culture, where the blocker tolbutamide abolished the effect, and nobody has measured any of it in a human scalp. The circumstantial case: minoxidil, pinacidil and diazoxide come from unrelated chemical families, were all developed for blood pressure, and have all been reported to cause excess hair growth in people (Buhl 1992).
The channel pairs a pore-forming subunit (Kir6.1 or Kir6.2) with a regulatory sulfonylurea receptor subunit (SUR1, SUR2A or SUR2B), and the SUR subunit decides which drugs can open it. Diazoxide's approved use is the same action in a different organ: it opens the channel on pancreatic beta cells, which stops insulin release and raises blood sugar (Trube 1986).
Human follicles carry two versions. Shorter and colleagues found SUR1 with Kir6.2 in the epithelial matrix at the base of the follicle, SUR2B with Kir6.1 in the dermal papilla and sheath, and no SUR2A detected in the follicle. The same paper reported that minoxidil stimulates only the SUR2 version (Shorter 2008).
Diazoxide acts on both, with different potency and efficacy at each. In cloned human channels read by thallium flux (Dyhring 2023):
- pancreatic Kir6.2/SUR1: EC50 6.2 micromolar, 97% of the response to the reference opener NN414
- vascular Kir6.1/SUR2B: EC50 9.7 micromolar, 38% of the response to the reference opener P1075
- cardiac Kir6.2/SUR2A: essentially nothing up to 100 micromolar
The two EC50 values overlap within their reported spread, so the difference lies in how fully diazoxide opens each channel. An earlier study using a different method put its SUR1 potency at 31 micromolar (Dabrowski et al. 2003). So diazoxide may engage the matrix channel minoxidil has not been shown to act on. Its strongest target is also the pancreatic channel, which is why absorption into the blood matters: serum diazoxide was detectable in all four people in our pilot, below.
Is diazoxide better than minoxidil for hair loss?
Unknown: we found no human trial comparing the two. They differ in how they are activated and in which channels they open.
Minoxidil is a prodrug. The molecule you apply to your scalp does not open the channel. A sulfotransferase in the follicle has to convert it to minoxidil sulfate first. Buhl and colleagues at Upjohn, minoxidil's maker, showed that the sulfate is 14 times more potent in cultured follicles, and that blocking sulfation blocks minoxidil's effect while leaving the sulfate's intact (Buhl 1990). The enzyme most often named is SULT1A1, and scalp activity varies between people. The developers of a minoxidil response assay put the share of patients who regrow hair on 5% topical minoxidil at about 40% (Goren 2015). The same group reported low follicular sulfotransferase activity in poor responders (Goren et al. 2014), in a small retrospective study we found no independent validation of. Minoxidil sulfate versus minoxidil covers that conversion step.
Diazoxide skips it, since the parent molecule opens the cloned channel itself. Which invites a hypothesis: someone who converts little minoxidil might still respond to a direct opener. We found no test of it in any person.
On potency, minoxidil came out ahead in the one experiment that tested both side by side, isolated red deer follicles in organ culture. Minoxidil increased growth at all four concentrations tested, 0.1 to 100 micromolar (p<0.001), while diazoxide reached significance only at 10 micromolar (p<0.01) and did nothing significant at the other three. Tolbutamide abolished the effect of both, and the authors read minoxidil's stronger response as pointing to a SUR2-based channel (Davies et al. 2005).
One caveat on that ranking. We found no activation profile of minoxidil sulfate on cloned subunits comparable to diazoxide's, beyond a 1998 binding assay at cloned SUR2B (Schwanstecher et al. 1998). So the two cannot be ranked at the channel itself.
Does diazoxide cause hair growth in people?
Oral diazoxide, taken for a different condition, causes excess body hair in a share of patients. That is the strongest human evidence for diazoxide and hair. The scalp evidence is thinner: one case report, from 1975.
Body hair was described first in children treated for low blood sugar (Koblenzer and Baker 1968), then in adults treated for high blood pressure (Burton et al. 1975). The case report, in The Lancet, describes a 42-year-old man on 800 mg a day whose scalp grew "a pronounced growth of hair on his head, where he had previously been bald" (Morris and Montuschi 1975). His baldness returned to its pretreatment state after he stopped. One uncontrolled patient, whose hair followed the drug on and off.
The reported rate depends on who was asked. Post-marketing surveillance in Japan followed 384 children on diazoxide for up to seven years and recorded hypertrichosis, through physician reporting, in 8.6% (Fukutomi et al. 2018). A survey of the Hyperinsulinism Global Registry asked families directly and found increased body hair in 89% of 121 current users, their most commonly reported side effect (Pasquini et al. 2025). A systematic review pooled the two cohorts that reported it at all and arrived at 45%, with heterogeneity so extreme that its own confidence interval ran from below zero to above 100% (Chen et al. 2021). Read that as "common" and nothing more.
Two things keep these rates off a scalp. The doses are systemic, titrated to control blood sugar, a median of about 6.7 mg/kg a day in children (Kizu et al. 2017), not a few milligrams rubbed into a scalp. And the effect reported is generalized body hair, with no measured scalp density endpoint. The Proglycem prescribing information calls "hirsutism of the lanugo type, mainly on the forehead, back and limbs" a frequent adverse reaction. The same label lists "loss of scalp hair".
Has topical diazoxide been tested on humans before?
Yes, more than once, and we found no controlled trial. The clearest record is a 1988 review by Henry Roenigk of Northwestern, reporting a trial of a 3% diazoxide solution. Among 60 patients treated at the time of the report, "about 25% of the patients have had some visible terminal hair growth, and the remainder of the patients have had some indeterminate hair growth; only one patient did not have any response," with no side effects or laboratory abnormalities noted (Roenigk 1988). That is weak evidence: an interim summary six months into a trial whose description includes no control group, no hair counts and no statistics, with no follow-up publication we can find. It does not describe the vehicle, so it cannot be assumed to match Anagen's formulation.
An earlier attempt found no growth. Upjohn's minoxidil patent describes applying 1% and 2% diazoxide in ethyl alcohol to human deltoid skin, alongside minoxidil solutions. It reports that "the Diazoxide tests were a failure in the sense that no increased hair growth was noted" (Chidsey and Kahn 1986). Deltoid skin is not scalp, the vehicle was plain alcohol, and the patent does not say how many people were treated. Weak evidence against diazoxide, from the company that made minoxidil.
The most relevant animal study we found ran 16 months in stumptailed macaques, the model used to validate minoxidil and finasteride. Seven adult animals: five on 5% topical diazoxide, two on vehicle alone, once daily five days a week. All five treated animals maintained or thickened their frontal hair, three showed progressive regrowth from month five or six, and the hair thinned again three months after treatment stopped. Body weight, blood pressure, heart rate, serum androgens, hematology and intravenous glucose tolerance were all unremarkable (Uno 1990). A companion report from the same lab says topical minoxidil and topical diazoxide both "induced significant hair regrowth in the bald scalps of macaques", without ranking them (Uno and Kurata 1993).
Two caveats. Five treated animals, photographs graded by eye, no statistical test. And the vehicle was 50% isopropyl alcohol with benzyl alcohol and two surfactants at pH 10.0 to 10.5. That holds a poorly soluble molecule in solution for a primate experiment, and would be hard to wear daily.
Solubility is a likely reason the drug went no further. Of seven potassium channel openers Upjohn tested in cultured mouse follicles, diazoxide was the only one that failed to increase hair shaft protein. The authors noted it "was poorly soluble and thus was tested only at low doses" (Buhl 1992).
What did our own blood and glucose pilot measure?
Our pilot measured drug exposure and blood sugar. It did not measure hair growth. Four adults, three Anagen team members and one volunteer, applied 1 mL of 3% a day, which is 30 mg of drug. Each gave one blood sample after four or five daily doses. Serum diazoxide by LC/MS came back between 12.9 and 28.2 ng/mL. This was an informal self-experiment among adults who chose to take part, not a registered or protocol-reviewed study, with no control group, no randomization and no blinding.
For comparison, children treated with oral diazoxide for low blood sugar run steady-state serum levels in the tens of micrograms per millilitre (Kizu et al. 2017). They averaged 44.6 on twice-daily dosing and ranged up to 120.7 in that cohort. Our highest level, a single sample, sits more than a thousandfold below the lowest of those steady-state averages. That compares our levels with oral treatment levels only, because no no-effect concentration for diazoxide has been established in adults.
Three of the four wore Dexcom Stelo continuous glucose monitors from four days before the first application. Their average glucose rose by between 2.1 and 6.2 mg/dL, a mean increase of 3.9 mg/dL. The 95% confidence interval ran from 1.4 below baseline to 9.1 above (p = 0.09), and widens to roughly 5 below and 13 above once the sensors' early-wear error is allowed for. More than 99% of readings stayed between 70 and 180 mg/dL in all three. All three averages rose a little, the direction oral diazoxide would push glucose. But at the blood levels measured the drug is an unlikely cause of a rise this size, and three unblinded people for a week cannot rule out a small effect. Method and charts are on our topical diazoxide science page.
What have the first patients reported?
Three people have completed the day-30 check-in, between 12 and 14 November 2026. All three reported no visible change in their hair yet. The check-in asks patients to pick from a list of specific symptoms, and all three answered that they had none. Extra shedding is not one of the options on that list, and one person reported mild extra shedding outside it. Three answers cannot show how common any effect is, and Anagen gives patients a reward for each completed check-in whatever they answer.
Nothing in those reports isolates diazoxide. All three were also taking oral minoxidil and a dutasteride, so any change in either direction has at least three plausible causes. Early shedding on minoxidil is the subject of minoxidil shedding: when it starts and stops. Anagen's other compounded products include MINX, a smooth-release oral minoxidil, and Precision Dutasteride, and neither is FDA approved.
What we still do not know
We found no published or registered controlled trial of topical diazoxide for androgenetic alopecia (PubMed and ClinicalTrials.gov, searched 15 November 2026). What it has is a mechanism argument, an animal study, human reports that describe no control group, and our own exposure measurements. Until a trial with a vehicle control and a hair count endpoint is published, topical diazoxide for hair loss has no controlled human evidence behind it.
The dose is unknown too. No one has measured how much diazoxide reaches a human follicle from any topical vehicle, or what concentration a follicle needs. So the gap between a working dose and one that touches the pancreas cannot be calculated. Long-term scalp safety is also unknown: the macaque study reported no local tolerability assessment across its 16 months. The non-responder idea remains a hypothesis. And the evidence that the SUR1 channel drives hair growth rests on that single 2005 deer-follicle experiment, which we found no repeat of in human follicles. What we still do not know about minoxidil applies here too, and the diazoxide list is longer.
Human genetics supports the target, as far as it goes. Gain-of-function mutations in ABCC9, the gene for SUR2, cause Cantu syndrome, whose features include congenital hypertrichosis. The authors noted the resemblance between the syndrome and minoxidil's side effects (Harakalova et al. 2012). That mutation sits in the SUR2 pair minoxidil already opens, so it supports the class effect and says nothing about SUR1, diazoxide's distinctive target. Cantu syndrome and hair growth covers the condition itself.
Frequently asked questions
Does diazoxide work for people who do not respond to minoxidil?
Unknown. Minoxidil needs the follicle enzyme SULT1A1 to become active, and in cloned-channel assays diazoxide opens potassium channels without that step, so the idea is that poor converters might still respond. We found no study of diazoxide in minoxidil non-responders, and no test on offer explains any one person's response.
How is topical diazoxide different from oral diazoxide?
Oral diazoxide is FDA approved for low blood sugar caused by excess insulin, and children taking it run serum levels in the tens of micrograms per millilitre. A topical aims for far lower exposure: in our four-person pilot, serum ran between 12.9 and 28.2 ng/mL. No topical diazoxide product is FDA approved.
Is topical diazoxide safe for the scalp?
Not established. We found no controlled trial measuring scalp tolerability for any topical diazoxide, and the 1990 macaque study reported no local tolerability assessment across its 16 months. The oral label carries warnings including sodium and fluid retention. Any use is a prescribing decision for a clinician who knows your history.
Does topical diazoxide raise blood sugar?
Oral diazoxide raises blood sugar by design, because it opens the same channel on pancreatic beta cells. In our pilot, three people on 1 mL of 3% a day saw average continuous-glucose readings rise between 2 and 6 mg/dL over about a week. That group mean has a confidence interval that includes no change at all.
Can you use diazoxide and minoxidil together?
We found no study of the combination in any species, so there is no evidence either way. The authors who mapped the two follicular channels suggested SUR1 drugs could work alongside minoxidil acting through SUR2B, on different parts of the hair bulb (Shorter 2008). Both are vasodilators that can cause fluid retention, so this is a prescriber's decision.
References
- Trube G, Rorsman P, Ohno-Shosaku T. Opposite effects of tolbutamide and diazoxide on the ATP-dependent K+ channel in mouse pancreatic beta-cells. Pflugers Arch. 1986;407(5):493-499. DOI 10.1007/BF00657506. PMID 2431383Cited in How does topical diazoxide work for hair growth?
- Shorter K, Farjo NP, Picksley SM, Randall VA. Human hair follicles contain two forms of ATP-sensitive potassium channels, only one of which is sensitive to minoxidil. FASEB J. 2008;22(6):1725-1736. DOI 10.1096/fj.07-099424. PMID 18258787Cited in How does topical diazoxide work for hair growth?
- Dyhring T, Jansen-Olesen I, Christophersen P, Olesen J. Pharmacological profiling of KATP channel modulators: an outlook for new treatment opportunities for migraine. Pharmaceuticals (Basel). 2023;16(2):225. DOI 10.3390/ph16020225. PMID 37259373Cited in How does topical diazoxide work for hair growth?
- Dabrowski M, Larsen T, Ashcroft FM, Bondo Hansen J, Wahl P. Potent and selective activation of the pancreatic beta-cell type K(ATP) channel by two novel diazoxide analogues. Diabetologia. 2003;46(10):1375-1382. DOI 10.1007/s00125-003-1198-1. PMID 12961066Cited in How does topical diazoxide work for hair growth?
- Schwanstecher M, Sieverding C, Dorschner H, et al. Potassium channel openers require ATP to bind to and act through sulfonylurea receptors. EMBO J. 1998;17(19):5529-5535. DOI 10.1093/emboj/17.19.5529. PMID 9755153Cited in Is diazoxide better than minoxidil for hair loss?
- Buhl AE, Waldon DJ, Baker CA, Johnson GA. Minoxidil sulfate is the active metabolite that stimulates hair follicles. J Invest Dermatol. 1990;95(5):553-557. DOI 10.1111/1523-1747.ep12504905. PMID 2230218Cited in Is diazoxide better than minoxidil for hair loss?
- Buhl AE, Waldon DJ, Conrad SJ, et al. Potassium channel conductance: a mechanism affecting hair growth both in vitro and in vivo. J Invest Dermatol. 1992;98(3):315-319. DOI 10.1111/1523-1747.ep12499788. PMID 1545141Cited in How does topical diazoxide work for hair growth? · Has topical diazoxide been tested on humans before?
- Goren A, Shapiro J, Roberts J, et al. Clinical utility and validity of minoxidil response testing in androgenetic alopecia. Dermatol Ther. 2015;28(1):13-16. DOI 10.1111/dth.12164. PMID 25112173Cited in Is diazoxide better than minoxidil for hair loss?
- Goren A, Castano JA, McCoy J, Bermudez F, Lotti T. Novel enzymatic assay predicts minoxidil response in the treatment of androgenetic alopecia. Dermatol Ther. 2014;27(3):171-173. DOI 10.1111/dth.12111. PMID 24283387Cited in Is diazoxide better than minoxidil for hair loss?
- Davies GC, Thornton MJ, Jenner TJ, Chen YJ, Hansen JB, Carr RD, Randall VA. Novel and established potassium channel openers stimulate hair growth in vitro: implications for their modes of action in hair follicles. J Invest Dermatol. 2005;124(4):686-694. DOI 10.1111/j.0022-202X.2005.23643.x. PMID 15816824Cited in Is diazoxide better than minoxidil for hair loss?
- Koblenzer PJ, Baker L. Hypertrichosis lanuginosa associated with diazoxide therapy in prepubertal children: a clinicopathologic study. Ann N Y Acad Sci. 1968;150(2):373-382. DOI 10.1111/j.1749-6632.1968.tb19062.x. PMID 5243880Cited in Does diazoxide cause hair growth in people?
- Burton JL, Schutt WH, Caldwell IW. Hypertrichosis due to diazoxide. Br J Dermatol. 1975;93(6):707-711. DOI 10.1111/j.1365-2133.1975.tb05123.x. PMID 1220817Cited in Does diazoxide cause hair growth in people?
- Morris AI, Montuschi E. Growth of head hair during treatment with diazoxide. Lancet. 1975;305(7921):1384. DOI 10.1016/s0140-6736(75)92300-4. PMID 48973Cited in Does diazoxide cause hair growth in people?
- Fukutomi M, Shimodera M, Maeda Y, Iwakura M, Hara M. Safety and effectiveness, including intelligence prognosis, of diazoxide in pediatric patients with hyperinsulinemic hypoglycemia: special survey in Japan (long-term, all-case survey). Clin Pediatr Endocrinol. 2018;27(3):131-143. DOI 10.1297/cpe.27.131. PMID 30083030Cited in Does diazoxide cause hair growth in people?
- Pasquini TLS, Rohli KE, Almeida FJ, et al. Real-world experience with the use of diazoxide among people living with congenital hyperinsulinism and their caregivers. Front Endocrinol (Lausanne). 2025;16:1628125. DOI 10.3389/fendo.2025.1628125. PMID 40917359Cited in Does diazoxide cause hair growth in people?
- Chen X, Feng L, Yao H, Yang L, Qin Y. Efficacy and safety of diazoxide for treating hyperinsulinemic hypoglycemia: a systematic review and meta-analysis. PLoS One. 2021;16(2):e0246463. DOI 10.1371/journal.pone.0246463. PMID 33571197Cited in Does diazoxide cause hair growth in people?
- Kizu R, Nishimura K, Sato R, et al. Population pharmacokinetics of diazoxide in children with hyperinsulinemic hypoglycemia. Horm Res Paediatr. 2017;88(5):316-323. DOI 10.1159/000478696. PMID 28715810Cited in Does diazoxide cause hair growth in people? · What did our own blood and glucose pilot measure?
- Roenigk HH Jr. New topical agents for hair growth. Clin Dermatol. 1988;6(4):119-121. DOI 10.1016/0738-081x(88)90075-2. PMID 3214817Cited in Has topical diazoxide been tested on humans before?
- Chidsey CA 3rd, Kahn G. Methods and solutions for treating male pattern alopecia. US Patent 4,596,812, granted 24 June 1986. https://patents.google.com/patent/US4596812Cited in Has topical diazoxide been tested on humans before?
- Uno H, Kemnitz JW, Cappas A, Adachi K, Sakuma A, Kamoda H. The effects of topical diazoxide on hair follicular growth and physiology of the stumptailed macaque. J Dermatol Sci. 1990;1(3):183-194. DOI 10.1016/0923-1811(90)90130-6. PMID 2085505Cited in Introduction · Has topical diazoxide been tested on humans before?
- Uno H, Kurata S. Chemical agents and peptides affect hair growth. J Invest Dermatol. 1993;101(1 Suppl):143S-147S. DOI 10.1111/1523-1747.ep12363275. PMID 8326148Cited in Has topical diazoxide been tested on humans before?
- Harakalova M, van Harssel JJT, Terhal PA, et al. Dominant missense mutations in ABCC9 cause Cantu syndrome. Nat Genet. 2012;44(7):793-796. DOI 10.1038/ng.2324. PMID 22610116Cited in What we still do not know
- Teva Branded Pharmaceutical Products. PROGLYCEM (diazoxide) oral suspension prescribing information. DailyMed, National Library of Medicine. https://dailymed.nlm.nih.gov/dailymed/drugInfo.cfm?setid=b16c7832-2fd9-49af-b923-1dc0d91fd6e2
- US Food and Drug Administration. Drugs@FDA: PROGLYCEM, NDA 017453, original approval 28 May 1976. https://www.accessdata.fda.gov/scripts/cder/daf/index.cfm?event=overview.process&ApplNo=017453
- Anagen. Internal observation: serum diazoxide by LC/MS after four or five daily doses of Topical Diazoxide, n = 4, November 2026; continuous glucose monitoring before and during Topical Diazoxide, n = 3, October to November 2026. https://anagen.xyz/treatment-science/diazoxide
This article is for educational purposes only and is not medical advice. Topical minoxidil is FDA approved for pattern hair loss, as is oral finasteride 1 mg in men; oral minoxidil, dutasteride, topical finasteride and spironolactone are approved drugs used off-label for hair loss in the United States. Diazoxide is FDA approved only as an oral drug for hypoglycemia caused by excess insulin, and no diazoxide product is approved for topical use or for hair loss. Anagen's Topical Diazoxide, MINX and Precision Dutasteride are compounded prescription preparations, are not FDA approved, and have not been tested in an efficacy trial that we found; nothing here is a claim that any of them treats, cures or prevents any condition. Any treatment decision belongs with a licensed clinician who knows your history.


