Tomorrow marks the 11th anniversary of the terrorist attacks on the World Trade Center, which killed nearly 3,000 people and traumatized hundreds of thousands of others. One out of four witnesses to that awful scene — fires, blood, flying glass and metal and stone and people — developed post-traumatic stress disorder (PTSD), characterized by fearful memories that just won’t recede.
It’s no wonder that many people with PTSD — about 14 percent — try to self-medicate with alcohol. Booze helps us forget, right? The idea lurks in idioms — I might get “trashed” or “wasted,” or “drown my sorrows” — and in Proverbs 31:7, and in the gospel of Dave Matthews: Excuse me please / one more drink / Could you make it strong / ‘Cause I don’t need to think… One drink to remember / Then another to forget…
So I was fascinated by a new study showing the opposite. Alcohol, it seems, helps cement painful memories into neural circuits.
In the study, a team led by Andrew Holmes of the National Institute on Alcohol Abuse and Alcoholism first made murine alcoholics, exposing mice to 16 hours of vaporized ethanol followed by 8 hours of withdrawal for four days in a row. They did that whole thing four more times, with 80 hours of withdrawal between each cycle. Each dose of alcohol was equivalent to double the legal driving limit in humans.
After that, the researchers taught alcoholic and non-alcoholic mice to be afraid of a sound by repeatedly pairing it with an electric shock to the feet. This is a common technique. Once the animals learn the association (sound = pain), they will freeze to the sound alone, Pavlov style. But this fear response doesn’t last forever. After a few dozen trials of a sound with no shock, mice eventually forget the bad memory and don’t freeze when they hear the sound. For the alcoholic mice, though, this so-called ‘fear extinction’ happened much more slowly than it did in controls.
The brains of the alcoholic mice were different, too. In part of the prefrontal cortex, neurons in alcoholic mice lacked a key protein of the NMDA receptor. Chemical signaling at this receptor underpins the learning process, but without the protein, the receptor can’t work properly. Fittingly, the researchers also found that the alcoholic mice showed less signaling at NMDA receptors than controls.
The whole thing makes sense, the researchers say, because alcohol is a known blocker of NMDA receptors. So the mechanism is something like this: Lots of alcohol –> Blocks function of NMDA receptors –> over time –> the receptors break down significantly –> the alcoholic mice can’t un-learn a fear memory.
The study is unusual because the alcohol exposure happened before the scary event, rather than after. That certainly has real-world relevance, as excessive drinkers have an increased risk of trauma, such as a car accident or life-threatening fight. But Holmes told me he’d predict the same extinction patterns if the mice were first traumatized and then inebriated, because presumably alcohol would be wreaking the same havoc at those NMDA receptors. So that’s pretty bad news for the thousands of people who have both PTSD and alcoholism.
On the upside, there’s pot.
In the past year, Holmes and others have shown that cannabinoids — a family of chemicals that includes THC, the active ingredient in pot — can curb stress responses in rodents. For example, a study in January showed that giving a THC-like compound to rats within one day of severe stress reduced their anxious behaviors. In June, Holmes’s team showed in mice that boosting anandamide, a cannabinoid produced naturally by the body, helps the animals extinguish fear memories.
And in July, scientists demonstrated this effect in humans for the first time. Researchers recruited 29 healthy adults and studied them over three days. On the first day, participants sat in front of a computer screen and learned a fear association: blue squares and yellow squares were paired with an unpleasant “white noise burst” on their headphones. On day two, half of the participants received a low dose of dronabinol, a form of synthetic THC that’s sometimes prescribed to stop chemo-related nausea, and the other half a placebo. Two hours later, they went through an extinction trial, in which blue squares were no longer paired with that hideous noise. Finally, on day three, participants came back and saw a series of blue and yellow squares. During all sessions, the researchers measured the electrical conductance of participants’ skin — a direct measure of sweat and indirect measure of stress.
For participants who got a placebo, the fear extinction procedure didn’t take: On day 3 they showed no difference in their stress response to blue and yellow squares, despite having seen the previous day that blue squares can appear without a noise. In contrast, participants who had received dronabinol showed a lower stress response to blue squares than to yellow ones. Bottom line: synthetic pot helped them un-learn a fear memory.
One caveat here is that while low doses of pot can lessen anxiety, high doses may bring on paranoia. Still, the drug has fairly minor side effects (less than or comparable to those of antidepressants). In states that have legalized medical marijuana, doctors are already prescribing it for PTSD. In New Mexico, apparently, one-quarter of medical marijuana patients are being treated for PTSD.
Anecdotal reports say that it helps. But what we really need is a rigorous clinical trial of pot’s effects on a large group of people with PTSD. Pulling that off will be politically tricky because the bodies that fund this kind of research, namely the Department of Veterans Affairs and the National Institutes of Health, are part of the same federal government that is fighting a futile war on drugs. But I think it’s an important debate to have — especially on this 9/11 anniversary, when we’re all reminded of the people who are still living, and re-living, the nightmare.
Read Maia Szalavitz’s excellent explanation of the rat THC study here.