Tuesday, 26 May 2015

Why does it take so long for antidepressants to work?

With my degree finally finished I'd like to go through a topic that has only been covered by one lecture in my whole degree, but one that has always interested me. Why does it take so long for antidepressants to take effect? They increase levels of a brain chemical (or a neurotransmitter as we call them in neuroscience) called serotonin within a very short period of time, but it can take up to 12 weeks before there is any improvement in symptoms of patients with depression. Well finally I have heard an idea about why that might be. 

But first, you need to know a little bit about depression and the drugs that we use to treat it. Clinical depression is a condition in which the patient suffers from incredibly low mood, loss of enthusiasm and they have a generally horrible time. I'm sure most of you have a fairly decent idea of what depression is like, maybe you know someone or are someone who has suffered from it, or is suffering from it now. 

The generally accepted idea behind depression, is that it is due to a lack of a certain neurotransmitters, serotonin. serotonin, amongst other things, plays a role in mood. Now this may not exactly be the case, and there are many other factors involved such as elevated stress responses and that possibly serotonin depletion is just a symptom, rather than a cause of depression. At least for now, increasing serotonin levels is our best way of treating depression, alongside psychiatric therapy. The vast majority of these  drugs are selective serotonin reuptake inhibitors (SSRIs), which basically increase the amount of serotonin that is able to generate feelings of good mood in the brain by blocking neurones from taking it back up, once it has been released. 

However, a big draw back of these drugs is that they take 8-12 weeks before they are effective. This seems strange, as they increase serotonin right away. So why doesn't it improve mood right away? If you take another drug that massively increased serotonin, Ecstasy, you see a huge increase in mood and,most importantly, they think everyone is their friend! SSRIs don't give such a strong increase in mood, and over time there are complex interactions between receptors being sensitised in neurones. Serotonin can only act when it is able to act on its receptor, but adaptive mechanisms in the brain cause changes in the sensitivity of the receptors. It's not too essential here but you can read more about it here and it's essentially a mechanism by which brain circuits can be tuned and these processes underly how learning and memory works, through the strengthening and weakening of connections within the brain. But don't worry if that doesn't really make sense because it doesn't to the top neuroscientists in the world, so you're fine! 

A generic diagram of a synapse using serotonin as a neurotransmitter, when the serotonin neurone is excited, it releases serotonin into the gap between neurones (called a synaptic cleft), it then acts on receptors on the outside of the second neurone  and causes excitation in the second neurone. 
What is more established, though, is the cognitive biases involved in depression. Those suffering from depression have a negative bias towards most aspects of life. For example; if you gave people a list of 20 words to memorise, 10 positive and 10 negative, "normal" and depressed participants would remember the same number of words, but the depressed participants would show a bias towards remembering the negative words. In another example, if you show people a picture showing a face with varying degrees of emotion (a little happy, to very happy or a little sad to very sad) those with depression are tuned to see sad and angry expressions more easily. This creates a situation where they are perceiving a sea of neutral faces  as a sea of angry and sad faces. This leads to thoughts that people are angry and upset at them all the time, leading to further negative thoughts and withdrawal from social situations. 

This bias has been found in patterns of brain activation when patients are put in imaging machines, using a technique called fMRI to look at brain activation. It's not the most accurate of techniques, but across wide samples sizes you can consistently see an increase in activation in the major fear centre of the brain, the amygdala, at rest ie not thinking about any tasks. 

Other brain areas can override this, but that causes extra strain on these systems which means they need to put in extra cognitive effort to reach the same performance as those with no depression, because the amygdala is constantly diverting away resources towards these false feelings of fear and danger. 

With this information, we were shown how antidepressants can affect these negative biases within just a week of treatment. So why the delay? Well, the drugs help to shift these negative biases towards the more normal perception across the population, so now they don't see quite so many angry or sad faces and they start to reintegrate into society. Once they start becoming more sociable and they don't have such negative biases, they gradually improve in mood and eventually the depression goes away. So the delay is due to the lag between the changes in neurochemistry and changing in biases having effects on the overall psychology of the patient. 

patients biases are shifted from more readily judging a face to be sad to more readily seeing faces as happy, this subtle shift reflects a shift in the overall outlook of the patient's life (or the filter that they view the world through is changed) .

What is really exciting about using facial expression recognition tests, is that they are a very good predictor of whether an antidepressant is working for that patient. This is really important, as antidepressants are only effective 50% of the time, so you may need to take a larger dose, or take another antidepressant. The problem here, is that if it takes 8 weeks to know whether it's worked or not, you then have to wait another 8 weeks until you know whether the next drug worked.

All this time, the patient is off work, so the economic cost is huge! These tests are able to predict whether the drug will work at the end of the course, with a pretty good accuracy, within just a week. So when the patient is first diagnosed, they do the face expression test and are then given the treatment. One week later, they do the test again and if the drug is working they'll have lost some of that negative bias.

This means that, after a week, the patient can be told that their chances of recovery are looking good and that they should keep on as they are, or that they need to up the dosage, or that they need to take a different drug. This could be really useful for helping in increasing the chances of people recovering from depression and integrating back into society faster, which will benefit them and society as a whole. 


Harmer et al 2002. Enhanced recognition of disgust in bipolar disorder. Biological Psychiatry
Harmer, 2012. Emotional Processing and antidepressant action, Current topics in behavioural neuroscience
Breen et al, 2000. Models of face recognition and delusion misidentification: a critical review. Cognitive Neuropsychology


Post a Comment