Naloxone Still Saves Lives. It Can't Touch the Drug That's Replacing Xylazine.

Bill Kinkle spent his third night in a Philadelphia hospital bed recording videos for his children. A doctor had just told him his labs gave him roughly a one-in-ten chance of walking out. He is a nurse. He is a paramedic. He has been in recovery for years, and he co-hosts a show for health professionals who are too. None of that mattered to what was happening in his bloodstream. He had used what he thought was his normal supply, contaminated — unknown to him — with a veterinary sedative called medetomidine. He wrote about what came next for Filter: hallucinations he could hear and feel, memory that wouldn’t hold, a body that wouldn’t stand. He survived. Not everyone does, and the drug that put him there is not going away.

The overdose-reversal drug that a generation of harm reduction work was built around does not work on the thing that is now flooding the fentanyl supply.

Naloxone reverses opioids. Medetomidine is not an opioid. It’s an alpha-2 adrenergic agonist — the same drug class veterinarians use to sedate large animals for surgery — and it slows a person’s heart rate, drops their blood pressure, and knocks them into a sedation naloxone cannot touch. You can give someone naloxone all day and it will do nothing for the medetomidine sitting alongside the fentanyl in their system. It will still restore their breathing if fentanyl is present, which is why the CDC and the White House Office of National Drug Control Policy still tell people to carry it and use it. But the sedation, the crashing vitals, the withdrawal that can look nothing like anything a paramedic has been trained for — none of that responds to the one tool most people have in their pocket.

The number went from 247 to 8,233, and nobody outside a lab noticed

Here is the part that should have been the headline in April and mostly wasn’t. In a joint health advisory, the CDC and ONDCP reported that lab detections of medetomidine in the national forensic system rose from 247 in 2023 to 2,616 in 2024 — a 950% jump — and then to 8,233 in 2025, another 215% on top of that. Ninety-eight percent of medetomidine-positive drug samples also contained fentanyl. The advisory’s own sentinel-site data, drawn from twenty testing locations nationwide between July and December of last year, found medetomidine in roughly 35% of opioid-positive samples at ten of those sites. The Northeast accounts for 52% of national detections, the Midwest 31%. This is not a fringe contaminant showing up in isolated seizures. It is becoming a baseline ingredient in a supply that was already, by any honest measure, unmanageable.

The CDC’s own language is a study in the gap between clinical caution and street reality. Its recommendation to providers is to “use heart rate to help differentiate between sedation from medetomidine toxicity” and ordinary opioid effects — a genuinely useful diagnostic tool, and also a sentence that assumes a clinician has the luxury of watching a monitor. A person alone in a bathroom does not have a heart rate strip to read. What they have is whatever naloxone they were told, correctly, to carry — and a growing chance that carrying it is necessary but no longer sufficient.

Rize tracks medetomidine under novel and emerging substances rather than folding it into opioids, precisely because it doesn’t behave like the drug it travels alongside — and its predecessor made this same trip first. Xylazine — the “tranq” that turned up in nearly every tested sample of Philadelphia dope by early 2023 and left users with the wounds that became its public signature — didn’t dominate the supply because it made anyone’s high better. It dominated because it was cheap, it stretched fentanyl further, and it did not, at first, draw attention from regulators who were still fighting the last war. Once the DEA moved to schedule xylazine, dealers didn’t need a memo to find the next thing. Kinkle calls this the Iron Law of Prohibition in his own account: crack down on one substance in an unregulated supply chain and the market doesn’t shrink, it substitutes — and the substitute is very often more dangerous than what it replaced, because nobody selected it for safety. Medetomidine is ten to twenty times more potent than xylazine. It is what filled the space xylazine’s own crackdown opened up.

Once the DEA moved to schedule xylazine, dealers didn’t need a memo to find the next thing.

Philadelphia got here first, and “here” is now the floor, not the ceiling

Philadelphia’s Medical Examiner’s Office started testing specifically for medetomidine in May 2024. By the end of that year, forty-six people who died there had it in their system alongside fentanyl. Clinicians at Thomas Jefferson University — Kory London, an emergency medicine faculty member, and Karen Alexander, on the nursing faculty — describe a drug that is now roughly twice as prevalent as xylazine in the city’s checked samples, without xylazine’s telltale wounds, which means the visual cue clinicians had trained themselves to look for is gone. One Philadelphia hospital has stood up a dedicated intensive-care unit specifically for medetomidine withdrawal, because the withdrawal syndrome — tachycardia, dangerously high blood pressure, drenching sweats, vomiting that doesn’t respond to the standard anti-nausea medications, and in the worst cases a swelling of the brain called posterior reversible encephalopathy syndrome — doesn’t behave like anything on the standard opioid-withdrawal protocol. A CDC MMWR field report from Pittsburgh documented the same pattern: symptoms starting within four to six hours of last use, far faster than typical opioid withdrawal, progressing to severe illness within six to twelve hours, in patients who showed little response to the GABA and opioid-agonist medications providers reached for first because those are the medications that work for everything else.

This is the twenty-one-year lesson harm reduction keeps relearning on a shorter and shorter clock. In 1988, Congress banned the use of federal funds for the syringe exchange programs that were, at the time, one of the only interventions slowing the spread of HIV among people who inject drugs. The ban held for twenty-one years, through an epidemic that public health researchers would later estimate cost thousands of preventable infections, because the political instinct was that funding a syringe looked like endorsing the drug use rather than treating the person in front of you. Medetomidine is the same instinct playing out at supply-chain speed instead of legislative speed: the response to one danger creates a new one, and the people paying for the gap between the two are the people who were never asked.

If you are not stopping tonight, here is what is actually true right now

If you are the person this article is actually for — not the provider reading it for protocol, but someone who knows they are using tonight regardless of what this says — some of what’s true is genuinely still useful to you. Naloxone still works on the fentanyl. Carry it, and carry more than one dose, because you may need to give several doses to get someone breathing even though the sedation itself won’t lift. Fentanyl test strips can pick up what they’re built to pick up, but they were not designed for medetomidine and can return a false negative on it — the CDC’s own guidance says as much — so a clean strip result is information, not a guarantee. Using alone raises the odds that whatever happens, happens with nobody there to call for help; using around someone else, even just within earshot, is the single intervention that shows up across every version of this crisis, xylazine and medetomidine both. None of this is a reason to use less carefully. It’s the opposite. The reversal drug in your pocket is still the right thing to carry. It’s just no longer the whole answer, and the people who wrote the health advisory know that, even if the advisory can’t say it in those words.

For the providers and case managers reading this for the practical version: the CDC’s clinical guidance to watch heart rate as a differentiator, and to consult a poison control center (1-800-222-1222) or toxicologist when sedation persists past an adequate naloxone dose, is worth putting directly into your team’s overdose-response training this week — not as a footnote, as the update. If your intake screening still treats “opioid withdrawal” as a single clinical picture, this is the month to add the question about heart rate and blood pressure trajectory, because a patient who doesn’t respond to your standard protocol may not be failing the protocol; the protocol may be built for a drug that isn’t the one in front of you anymore.

Kinkle’s hospitalization started, in part, because a clinic nurse refused to dispense his methadone dose for arriving seven minutes late — a decision made to teach him a lesson about punctuality that instead taught the rest of the field something about what punitive dosing policy actually costs. That story is its own indictment. This one is about what happened after he walked out the clinic door already in withdrawal and used whatever he could get: a drug he’d never heard of, in a supply nobody had warned him about yet, that very nearly killed him anyway. He’s alive to write about it because of a combination of luck, timing, and a body that held on three more days than the odds said it would. The advisory that might have prepared him for it existed by the time he needed it. It just hadn’t reached him — and four months after the government published it, there is no evidence it has finished reaching anyone else it needs to.

It just hadn’t reached him — and four months after the government published it, there is no evidence it has finished reaching anyone else it needs to.