By late 2023, a respiratory illness that nobody could name had made dogs sick across sixteen states, and some of them had died. The symptoms were maddeningly ordinary right up until they were not: coughing, sneezing, discharge from the eyes or nose, lethargy, and then labored breathing and gums going blue or purple.
Into that vacuum came a word that spread faster than the illness. Chloramphenicol. A powerful antibiotic that had, in a handful of desperate cases, pulled critically ill dogs back from the edge. The internet promptly filed it under “cure.”
It is not a cure. And the reason it is not a cure is the most useful thing in this entire story.
The drug is aimed at the wrong kind of enemy
Chloramphenicol kills bacteria. Specifically, it kills bacteria that happen to be susceptible to it, which is a narrower claim than it sounds.
The mystery illness, based on everything vets were seeing, looked viral. And viruses do not care about antibiotics. Handing an antibiotic to a dog with a viral disease does not help that dog, and it does help drive antibiotic resistance, which quietly makes the drug less useful for the next dog who genuinely needs it, the one with a raging bacterial pneumonia and no other options left.
That is the reasoning veterinarians kept trying to get across while the cure story was going around, and it did not travel nearly as well.
“When a drug gets a reputation as the thing that saved somebody’s dog, every owner in the waiting room wants it, and I understand exactly why,” says Dr. Lena Park. “But this one is not a stronger version of the antibiotic your dog got last year. It is a different tier of drug, with a different risk profile, and reaching for it out of fear costs something.”
Where the cure story came from
There was a real dog behind it. In an interview with Good Morning America, one owner described watching her dog, Ike, spend day after day in the hospital with pneumonia after catching the illness, and being told he probably would not make it. Somebody floated chloramphenicol. The first response was that this was an extremely strong drug and close to an end-of-the-road option. An internal medicine specialist overruled that. Seven days later the dog went home healthy.
That is a genuinely good outcome and a genuinely bad basis for a treatment protocol. Ike was critically ill with pneumonia, meaning a bacterial infection had taken hold, whatever started the whole thing. The drug was doing the job it exists to do, at the exact point in the illness where a last-resort option is what is left. It was not treating the mystery, it was treating what the mystery had opened the door to.
Why “last resort” is a literal description
Chloramphenicol is used in humans and animals alike. On the human side it shows up most often as an ointment for conjunctivitis, and it can also be taken orally or by injection for serious infections including meningitis, typhoid fever, and cholera. In dogs it comes in tablet, capsule, or injectable form, and is prescribed for infections in bone, in wounds, and in the lungs.
But the reason your vet’s face changes when you say the word has to do with the risks, and they run in two directions.
For the dog: vomiting, diarrhea, and loss of appetite are common enough. The margin is thinner still if your dog is anemic, or has failing kidneys or a compromised liver, and thinner again if she is pregnant.
For you: this is the part almost nobody knows. Anyone giving this drug to an animal should be wearing gloves. The reason is aplastic anemia, a shutdown of the bone marrow that can kill a person, and the drug is capable of triggering it in the human holding the pill. Pregnancy and liver disease raise that stakes further. You are handling a drug that carries a warning for you, not just your dog.
That combination is why the standard veterinary position is narrow: reach for it only when a culture has come back and the results say the shortlist of drugs that will still work has gotten very short indeed. Not when a dog has a cough. Not because the internet said so.
What actually protected dogs
The unglamorous answer, then and now.
Vaccination. Dogs vaccinated against the known respiratory diseases tended to have milder cases, according to reporting from the Oregon Veterinary Medical Association. That is not a claim that a vaccine prevents an unidentified pathogen. It is a claim that a dog whose immune system is not simultaneously fighting kennel cough, parainfluenza, or canine influenza has a better afternoon than one who is.
Distance. During an active outbreak, dogs should not be gathering in large groups, and should not be sharing water sources. If you see a coughing dog, take your dog the other way.
Boarding decisions, especially over the holidays. Crowded kennels at peak season are exactly the wrong environment. A pet sitter at home is worth considering, because stress itself suppresses immune function, and a stressed dog in a full boarding facility is stacking the odds badly. If you have to board, use a facility that enforces vaccine requirements and strict hygiene, and ask them about it directly rather than assuming.
None of that makes headlines. All of it works better than a drug you should not be asking for.
What’s new since this was written
The mystery has largely dissolved, which is a quieter ending than the one people expected.
Investigations that followed the 2023 outbreak did not turn up a single novel pathogen sweeping the country. What they found instead was canine infectious respiratory disease complex, the familiar cluster of viruses and bacteria that has always circulated in dogs, behaving badly in clusters and defeating routine diagnostic panels often enough to look like something new. Oregon, where the alarm started, had logged more than two hundred reports from veterinarians by August 2023, and common respiratory testing kept coming back negative, which is precisely how a familiar problem acquires a mysterious reputation.
One organism did get more interesting. A 2025 case series in the Journal of Veterinary Emergency and Critical Care looked at thirteen dogs with pneumonia from the outbreak window and found Mycoplasma cynos as the sole pathogen in eight of them, and alongside Bordetella bronchiseptica in five more. That does not make M. cynos the cause of the whole event, and researchers have been careful not to say so. It does suggest that in the sickest dogs, the ones who ended up in oxygen cages, a bacterial player was often part of the picture.
Which, notably, is the one scenario where a drug like chloramphenicol was ever the right call in the first place.
References
- Mycoplasma cynos-Associated Canine Infectious Respiratory Disease Complex Pneumonia in 13 Dogs (Journal of Veterinary Emergency and Critical Care, 2025)
- Was the canine respiratory illness in 2023 really a ‘mystery’? (Veterinary Practice News)
- Chloramphenicol (VCA Animal Hospitals)
- Predominance of Canine Parainfluenza Virus and Mycoplasma in Canine Infectious Respiratory Disease Complex in Dogs
