Cocaethylene: The Third Drug Your Liver Makes

Drinking on cocaine is not just two drugs at once. Your liver builds a third one out of them. Cocaine and ethanol in the body together react to form cocaethylene, which happens with no other common drug combination. Cocaethylene is more cardiotoxic than cocaine, clears more slowly, and accumulates every time you redose. Cocaine and alcohol are the most common two-drug pairing in emergency department presentations, and this is why.

What your liver does with both

Normally an enzyme called carboxylesterase 1, hCE1, breaks cocaine down by cleaving a bond and putting a water molecule in its place. The product, benzoylecgonine, is pharmacologically inert.

Put ethanol in the liver at the same time and hCE1 uses that instead of water, swapping cocaine’s methyl group for an ethyl group off the alcohol. The result, cocaethylene, is structurally almost identical to cocaine with a different tail, and the tail changes how the body handles it. This reaction only happens when both substances are in the liver together, and hCE1 is the only enzyme that runs it.

Harris and colleagues quantified the yield in a controlled human study published in Drug and Alcohol Dependence in 2003: roughly 17 percent of cocaine converts to cocaethylene when alcohol is present. Not a trace metabolite. A sixth of the dose becoming a different drug.

Why it is worse than cocaine

Cocaethylene blocks the same dopamine and norepinephrine transporters cocaine does, and produces the same stimulation and euphoria. Four things make it more dangerous.

It blocks cardiac ion channels harder. A 2024 systematic review of 42 studies found cocaethylene is a threefold more potent blocker of hERG potassium channels than cocaine, and a more potent sodium channel blocker as well. Both channels are load-bearing for normal rhythm, and blocking them prolongs the QTc interval, which sets up fatal arrhythmias including torsades de pointes and ventricular fibrillation. Emergency department data match: cocaine-alcohol patients show QTc dispersion of 58 to 82 ms against 24 to 43 ms in cocaine-only patients.

Sudden death risk rises sharply. Andrews’ 1997 review of cocaethylene toxicity put the increase in sudden death risk at 18 to 25 fold when cocaethylene was present, compared with cocaine without alcohol. Evidence tier: forensic case series, not a trial. The signal is consistent across independent datasets.

Emergency departments see it directly. Shastry’s prospective cohort at two urban EDs, in Academic Emergency Medicine in 2023, covered 199 patients: 150 cocaine-only, 49 cocaethylene-positive. Cardiac arrest occurred in 6.1 percent of the cocaethylene group against 0.7 percent of the cocaine-only group (adjusted OR 12.61, 95% CI 1.10 to 144.18, p=0.048). That confidence interval is enormous, which is what happens when a cohort this size produces very few arrests, so treat the point estimate loosely and the direction seriously. The cocaethylene patients also ran higher lactate, meaning worse tissue oxygen deprivation even short of arrest.

It hangs around. Perez-Reyes measured both directly in humans: cocaine’s elimination half-life is about 1.1 hours, cocaethylene’s about 1.7. That gap compounds. Every extra line taken while drinking adds cocaethylene to a pool draining more slowly than the cocaine is, so a night of redosing produces more total exposure than the half-life arithmetic suggests.

Neither drug tells you to stop

This combination is so common because it feels good and it feels controlled, and both of those are the problem.

Cocaine counteracts alcohol’s sedation. It cuts perceived intoxication, restores reaction time and coordination at moderate blood alcohol levels, and removes the drowsiness that would otherwise register as “that’s enough,” so people drink substantially more than they meant to. It runs the other way as well. Alcohol takes the edge off cocaine’s anxiety and cardiovascular discomfort, which makes it easier to keep going, and Harris’ subjects rated the combination as more intoxicating and more pleasurable than either substance alone.

A night that starts as a couple of drinks and a couple of lines escalates because neither drug is issuing its usual stop signal, and every round adds to a toxic load that outlasts both.

The liver, over years

Each drug is hepatotoxic alone, and together the burden is greater than the sum. Tamargo measured cocaethylene in the blood of 649 participants by mass spectrometry, publishing in Drug and Alcohol Dependence in 2022. People with detectable cocaethylene had 3.17 times the odds of liver fibrosis compared with non-users (95% CI 1.61 to 6.23, p=0.0008), after adjusting for HIV status, hepatitis C and other covariates, and the risk from the combination exceeded what either drug contributed individually.

Fibrosis precedes cirrhosis, liver failure and hepatocellular carcinoma. It accrues over years rather than over a night, which makes it easy to discount and worth knowing anyway.

Seizures, in rodents only

One rodent study found 90 percent status epilepticus rates in cocaethylene-treated mice on repeated exposure, higher than with cocaine, and relative toxicity rankings put cocaethylene above cocaine for acute seizure risk. All of it is rodent data at doses well above human recreational exposure, and no controlled human study has compared seizure thresholds directly. Real enough to be careful about, not established human pharmacology, and we are not going to present it as though it were.

Fentanyl is the other problem

Cardiac toxicity is one danger. Contamination is a separate one, and by now just as pressing. Wagner and colleagues, pooling community drug checking data from 25 US states in 2023, found fentanyl in roughly 14.8 percent of powder cocaine samples, and nearly half of US overdose deaths that year involved both opioids and stimulants.

Fentanyl in cocaine is odorless, tasteless and invisible. A strip is the only way to find it: dissolve a small amount in water and dip. One line is POSITIVE, fentanyl detected. Two lines is NEGATIVE. That reading is the reverse of what almost everyone assumes, and it is worth checking twice before you act on it. The method, including the dilution that stops a false positive on a stimulant, is in our fentanyl test strip guide.

If a strip comes back positive, use far less than usual, keep naloxone within reach, and do not use alone. A strip tells you a sample contained fentanyl somewhere, not how much or how evenly, so a negative is reassurance rather than a guarantee.

The threefold channel blockade, the cardiac arrest gap, the tripled fibrosis odds: they are all describing one thing, which is a drug your body made that is worse than either of the two you took, lasts longer than both, and builds up while the two of them quietly switch off the signals that would have told you to stop.

For the broader risk profile see the cocaine guide, and the interaction checker for other combinations.

Sources

Andrews 1997, cocaethylene toxicity, Journal of Addictive Diseases PMID 9243342 | Perez-Reyes 1994, potency and pharmacokinetics of cocaethylene versus cocaine in humans PMID 7701044 | Harris 2003, pharmacology of cocaethylene in humans PMID 14636972 | Shastry 2023, cocaethylene cardiotoxicity in emergency department patients, Academic Emergency Medicine PMID 36000306 | Tamargo 2022, cocaethylene and liver fibrosis PMID 35033954 | Wagner 2023, fentanyl prevalence in community drug checking samples PMID 37826988 | 2024 systematic review of cocaethylene cardiotoxicity