Hannah Zulueta Bloodwork · Attribution
Nobody Ever Blames the CBD
The liver value climbs, the herbs get pulled, and the chew nobody mentioned keeps getting given.
The panel comes back with one value flagged and the conversation goes the same way every time.
The vet asks what changed. Sometimes it lands on the food, because she started cooking for him in March and that is the newest thing in the room and the least familiar. More often, in my practice, it lands on the herbs.
What nobody asks about is the soft chew. He has had one every night since last summer, it came from the pet store in a resealable pouch, and she did not think to mention it because it is a treat.
I do not use CBD. I have bought and tried ten different brands over the years, tinctures and chews both, which is how I know what is in the bottles and what is not. I am not telling anyone to stop. I am asking you to know what it does, so that when a number moves you can tell what moved it.
The one thing to know
CBD raises alkaline phosphatase in dogs.
This is not controversial, not fringe, and not rare. It shows up across the published studies, and it is the same enzyme that gets a fresh diet questioned and a herbal formula pulled.
The finding
What the studies actually found
Elevated ALP is the most consistent finding in canine CBD research. It turns up in the pharmacokinetic work, the osteoarthritis trials, the epilepsy studies, and the long-term feeding trials. It usually appears without any other liver value moving, and it usually comes back down after the CBD stops.
The standard explanation is that CBD upregulates the liver’s cytochrome P450 machinery, and the ALP rise reflects busier metabolism rather than damage.
That explanation is probably incomplete, and the reason why is the most interesting thing in this literature.
The part almost nobody reports
Four enzymes, one printed number
A six-month feeding study in healthy dogs measured something the other studies did not. Alongside total ALP, the researchers measured bone-specific ALP.
Both went up together. The correlation ran above 0.9, meaning the dogs with the largest total ALP rise were the same dogs with the largest bone ALP rise. The authors concluded that at least part of the elevation was bone-derived.
This is where dog physiology becomes the whole story. ALP is not one enzyme. Dogs carry four versions of it.
Dogs are the only species carrying that fourth steroid-driven form. So a single flagged ALP on a routine panel does not tell you which of the four went up. It is a number with four possible addresses, and the standard panel does not print the address.
The liver value everyone panics about may be partly coming from somewhere that is not the liver.
The other number
ALT is the one to watch
ALT is the more honest marker for actual hepatocellular injury, and it has stayed mostly quiet in dog studies at conventional doses.
Mostly. A mobility study using a higher dose than the usual one to two milligrams per kilogram twice daily saw mild ALT elevations in five dogs. An epilepsy study at higher dosing reported the same. The pattern that emerges is dose-dependent.
The human data sharpens it. At 1500 milligrams daily for around three and a half weeks, 31 percent of people had ALT rises of five times the upper reference limit, which meets the international consensus criteria for drug-induced liver damage.
Nobody is giving a dog the equivalent of that on purpose. Which brings us to the problem underneath all of this.
Dose
The dose in the study is not the dose in your house
Owners give tinctures. Owners give chews. Plenty of owners give both, a dropper in the morning and a chew at night, and count it as one thing because it is all just CBD.
The published dog research runs on measured oil, given at a stated number of milligrams per kilogram of body weight. That is where every dose figure you have read came from: the one to two milligrams per kilogram twice daily, the four milligrams per kilogram once daily. Those numbers describe a controlled amount going into a weighed dog.
Almost nothing about that survives the trip to a kitchen counter.
The chew does not scale. A fixed number of milligrams per piece goes to a Boston Terrier and to a Shepherd alike.
Two is the obvious next move. When one does not seem to be working, the dose doubles without anyone experiencing that as a dosing decision.
The tincture can be measured, in theory. If the dropper is read carefully and the bottle contains what it says.
Both together add up. Neither package is written to be added to anything.
So when a study reports that four milligrams per kilogram was well tolerated over six months, that finding is about a dose. It is not automatically about your dog’s dose, and the gap between the two is not knowable from the packaging.
None of that makes CBD dangerous. It makes it uncounted, which is a different problem and the one this page is about.
The bottle
Your dog’s dose is not the dose on the label
I have tried ten brands. The reason I stopped trying is that I could not tell what I was giving.
Eighty commercial CBD products went through mass spectrometry in one analysis. Thirty-seven were off their own label claim by more than ten percent. Products claiming somewhere between 7.5 and 60 milligrams per milliliter actually contained anywhere from 2.9 to 61.3.
Read a study that says four milligrams per kilogram is well tolerated and you still cannot act on it, because the bottle in your hand is not a reliable measuring device. A dog who seems to react badly to CBD may be reacting to a dose several times what the label promised, or to the lead, or to the THC.
The panel
Why the whole picture gets harder to read
CBD moves cytochrome P450 activity, and P450 is how the liver clears most other drugs.
Take a dog with skin disease. He is on prednisone, which drives up the corticosteroid-induced ALP that only dogs have. He is also getting a nightly CBD chew, which drives up ALP through a different route. His ALP is high, and there are now two documented explanations sitting in front of anyone reading that panel.
If the same dog also switched to home-cooked food in the spring, or started a herbal formula, that is the one that gets named. It is the newest variable and the one the reader has the least experience interpreting.
I wrote a separate page on what each itch drug is doing inside the body, including the steroid question.
The herbs
About the herbs
The herbs get pulled more often than the food does, and usually faster. A number moves, somebody gets nervous about the unfamiliar thing, and a formula written for a specific dog gets stopped by someone who never read the ingredient list.
I am not going to defend the herbs with a blanket. Some Chinese herbs do injure livers, and pretending otherwise would be the same lazy thinking in the opposite direction.
The list is short and it has names. Out of more than seven thousand herbs in the Chinese materia medica, the reference literature implicates a handful.
In one systematic review, He Shou Wu accounted for 65 of 114 cases by itself.
Here is the line that should end most of these conversations, and it is not mine. The LiverTox database notes that the specific responsible product is often unclear, and the injury simply gets ascribed to Chinese herbs without anyone naming which one.
That is the whole problem, written by the people who catalogue herbal liver injury for a living.
The principle
Dose is the thing, and it always was
Nothing on that list is poisonous in the way the word gets used. Amount determines harm, and this is not a special rule for herbs. It is how every substance a dog eats works.
Copper is required in the diet and is why I hold liver under a ceiling in every recipe I write.
Zinc is required and has an upper limit.
Vitamin D is required and is the fastest way I know to hurt a small dog with good intentions, which is why it stays out of supplement stacks for dogs under twenty pounds.
Iodine and kelp belong in the same conversation, and I keep them out of my recipes for the same reason.
None of those are dangerous ingredients. They are ingredients with ceilings, and the person formulating is responsible for knowing where the ceiling sits.
Herbs work the same way. An herb prescribed at an appropriate dose for an appropriate duration is a different object from the same herb taken at an unknown dose for a year because it was on a shelf. The reported injury cases skew toward long exposure, unclear products, and self-directed use.
The useful question is never whether an herb can hurt a liver. It is what dose, for how long, from where, and prescribed by whom. Cutting a formula because a vet got nervous about a category is not risk management. It is the removal of the one variable somebody actually thought about.
Sourcing
Where the herb came from
This is opinion and I will say so.
Chinese herbs must come from a reputable source.
Not a marketplace listing, not a bulk bag with no origin, not whatever was cheapest. From a company that tests, that identifies the species correctly, and that discloses where the material was grown and how it was processed.
The reason sits in the CBD data further up this page. When a product category is unregulated, roughly half the labels stop describing the contents. Heavy metals turn up. Species get substituted. That is true of the CBD aisle and it is true of the herb supply chain, and the herb supply chain has the longer history of both problems.
He Shou Wu is the sharpest example. Preparation changes it. The prepared root is considered substantially safer than the raw root, and the processing is not a formality. A formula written by someone trained specifies which form is intended. A bag bought on price does not tell you which one you received, and the distinction disappears entirely when a whole tradition gets discussed as one undifferentiated thing.
Buy from companies that stand behind their material. Then the dose you were prescribed is the dose you gave, and a liver value that moves means something.
Two more reasons
Why the herbs are usually the wrong suspect
The pattern does not match
In that same systematic review, 81 percent of confirmed cases presented as hepatocellular injury. An isolated ALP with a normal ALT does not look like herbal injury. It looks like a number of other things, and one of them is CBD.
The timeline is tight
Median latency in those cases was about thirty days from starting the herb. If a formula started in January and the number moved in August, the formula is a poor suspect.
And most of those cases resolved once the herb stopped. If an herb ever is the cause, stopping it fixes it. That is an argument for finding the right variable, not for pulling everything at once.
What to do
How to actually tell
Position
Where I sit
I am not against CBD. Owners use it, some of them see real benefit for pain and for anxiety, and that is theirs to decide with their vet.
What I want is for it to be counted. It belongs on the medication list, with a start date and a dose, the same as anything else the dog swallows every day. A substance with a documented effect on liver enzymes should not be the only thing in the room that never gets asked about.
Where this ends up
She stopped the herbs
Not the food. The food was never raised. What worried her vet was the formula, and she trusted him, and she stopped it. Four months of a prescription written for her specific dog, ended on a suspicion that was never attached to an ingredient.
The medication continued. The one whose label says not to break the capsule, to wear gloves when you hand it to your dog, and to wash your hands afterward. Nobody proposed pausing that one. The soft chews continued too, every night, and were not mentioned in a single appointment.
I am not questioning anyone’s judgment. I am asking for one standard. If an unfamiliar substance has to prove itself before it stays in the plan, that rule belongs on everything the dog swallows, including the things that arrived in a pharmacy bag and the things that came in a pouch from the pet store aisle.
His ALP is still elevated. Only one variable in that dog’s life was ever removed, and it was the one nobody could name a mechanism for.
She was told to wash her hands after handling one of the things she gives him every day. It was not the herbs.
References
Bradley, S., Young, S., Bakke, A. M., Holcombe, L., Waller, D., Hunt, A., Alexander, K., Watson, P., Logan, D. W., Mendl, M., & Zulch, H. (2022). Long-term daily feeding of cannabidiol is well-tolerated by healthy dogs. Frontiers in Veterinary Science, 9, 977457. https://doi.org/10.3389/fvets.2022.977457
Corsato Alvarenga, I., Gustafson, D., Banks, K., Wilson, K., & McGrath, S. (2024). Tolerability of long-term cannabidiol supplementation to healthy adult dogs. Journal of Veterinary Internal Medicine, 38(1), 326–335. https://doi.org/10.1111/jvim.16949
Gidal, B. E., Vandrey, R., Wallin, C., Callan, S., Sutton, A., Saurer, T. B., & Triemstra, J. L. (2024). Product labeling accuracy and contamination analysis of commercially available cannabidiol product samples. Frontiers in Pharmacology, 15, 1335441. https://doi.org/10.3389/fphar.2024.1335441
Johnson, E., Kilgore, M., & Babalonis, S. (2022). Label accuracy of unregulated cannabidiol (CBD) products: Measured concentration vs. label claim. Journal of Cannabis Research, 4, 28. https://doi.org/10.1186/s42238-022-00140-1
National Institute of Diabetes and Digestive and Kidney Diseases. (2022). Chinese and other Asian herbal medicines. In LiverTox: Clinical and research information on drug-induced liver injury. https://www.ncbi.nlm.nih.gov/books/NBK548589/
Zhang, P., Ye, Y., Yang, X., & Jiao, Y. (2016). Systematic review on Chinese herbal medicine induced liver injury. Evidence-Based Complementary and Alternative Medicine, 2016, 3560812. https://doi.org/10.1155/2016/3560812
Read next
Hannah Zulueta obtained her Certificate in Canine Nutrition from CASI Institute. She is also studying for her Doctorate in Acupuncture, Traditional Chinese Medicine, and Herbalism from the esteemed Pacific College of Health and Medicine. If you’re as keenly interested in health for yourself, visit her website for human health: Satori Acupuncture and Herbs.
She resides in San Diego with her three dogs, Maggie, Orbit, and Mr. Higgins.
She is available for one on one consultations. She develops evidence-based recipes that meet both NRC and AAFCO nutritional standards while integrating the therapeutic food principles of Traditional Chinese Veterinary Medicine into her formulations.
Additionally, you can find her sharing free content on Instagram.
