The 2019 Judy Dunn Conference focused on ‘Suicide and self-harm in young people’.
Keynote speaker Prof. Christian Schmahl, Professor of Experimental Psychology and Medical Director of the Department of Psychosomatic Medicine at the Central Institute of Mental Health in Mannheim on the ‘Neurobiology of self-harm in BPD’.
ACAMH members can now receive a CPD certificate for watching this recorded lecture. Simply email firstname.lastname@example.org with the day and time you watch it, so we can check the analytics, and we’ll email you your certificate.
Christian is a Professor of Experimental Psychopathology and Medical Director of the Department of Psychosomatic Medicine at the Central Institute of Mental Health in Mannheim, Germany. His research focus is on emotion regulation, self-injurious behavior and dissociation as well as the interaction of neurobiology and psychotherapy in Borderline Personality Disorder and Posttraumatic Stress Disorder. He received his M.D. in Germany at Mainz and Giessen Medical School and did his residency in Psychiatry at Freiburg Medical School and his residency in Psychosomatic Medicine at the CIMH. Since 2018 he is spokesperson of the Research Training Group “Impact of Adverse Childhood Experiences on Psychosocial and Somatic Conditions across the Lifespan”. He has published more than 200 articles and book chapters.
Dr. Dennis Ougrin – The second announcement is going to be shorter. It’s going to be about our first speaker who is… We have in German speaking world we have this thing where we say how many times you are a doctor and how many times you are a professor, so I think you are just the one doctor and one professor [inaudible 00:00:30] once [laughter], so this professor doctor, Christian Schmahl, he’s based in Mannheim, which for those who haven’t come across this before is like the equivalent of the Institute of Psychiatry but in Germany. So this is where they forge all of their scientists and researchers who are looking into mental health. And one thing I wanted to say to you about him, which is serious, is that he started a new centre for self-harm, personality disorders and trauma for adolescents. But interestingly, in line with what many of us are thinking about, that spans young people and also younger adults. So his research now scans into young people and adults too. And one fun thing about Christian is that he cycles to work every day for an hour, and apparently when he does this, he almost always beats all of his colleagues because he’s an [s.l. immense blagger 00:01:36] [laughter]. All right. So without further ado, as they say, allow me to give you, Professor, Doctor. Christian Schmahl.
Professor Christian Schmahl – Is it working? Can you hear me? So, thank you, first of all, Dennis, for this kind two part introduction for the fun part of the series, actually. And thank you all for coming. And it’s true that we started a centre for adolescents that self-harm, that have borderline personality disorder and trauma related disorders. That’s a unique thing in Germany because we tackle this transition phase, which I believe personally is very important. And I wanted to make a point here that the research we’re doing on self-harm that I will present on this also made to help improving therapy for these young adults and adolescents. And that’s, I think, important as a background for the research I’m presenting. We are interested in what is behind self-harm. Why do young people cut themselves or hurt themselves?
In the next 45 minutes I’ll give you some examples of the research we have done in the past couple of years. So the disorder we are mostly studying is borderline personality disorder, and why do we study that and why is it ideal to study self-harm? Of course, because NSSI… and first another announcement, in Central Europe and also in the US we mostly use the term non-suicidal self-injury, and NSSI is also now part of the DSM-5. And I think probably we’ll go into more detail in this part on the generality. NSSI or self-injury is a little bit… We found that it’s not as broad as self-harm. It is mostly related to the destruction of the body surface or the body tissue, which means mostly cutting, burning, head banging and so on. And that’s a frequent behaviour in BPDs. About 80 percent of patients with borderline personality disorder show [inaudible 00:04:05], one of the nine criteria before NSSI. And of course that’s also something for me also definitely go in more detail. It’s very frequent in other mentally ill young people and also the general population in adolescence.
But now I will focus on borderline personality disorder and in this disorder it’s very frequent. A lot of these patients, as you probably know from your clinical work, cut themselves frequently every day or once a week or so, and this makes for a lot of concern, both problems for the individual but also a lot of costs for the society, and BPD leads to a lot of hospitalisation, emergency room use and so on. This also I, Paul, will definitely talk about this more. What is interesting is that self-injury is very frequent in adolescents, as I mentioned, between 14 and 16 years and, big question, I personally think it’s very important is what happens with those young people that cut themselves or hurt themselves, and those that stop and those that don’t stop. And I think that some of them that don’t stop NSSI after adolescence end up as borderline personality disorder.
So this is the background to the neurobiological work I will show you today because we can use borderline personality disorders as a model disorder to study the mechanisms behind self-injury. And my talk will tackle three domains. Emotion dysregulation is at the heart of borderline personality disorder and also closely related to disturbed pain processing I will talk a lot about emotion dysregulation of pain processing. At the end I will have a brief section on social interaction, what is destructive social interaction, and how this is related to pain and emotion dysregulation in borderline people. So let’s start with NSSI in borderline personality disorder. We started the research in that field with asking the patients, what are your motives for self-injury? And by far the most frequent motives are to reduce stress or tension levels. You see that nearly 100 percent of borderline patients mentioned that as their primary motive for NSSI, and also related to reduction of unpleasant feelings, so emotion regulation is at the centre of NSSI motives.
There is another group that is related to dissociative symptoms. A lot of these patients show dissociation, personalisation, derealisation, and often NSSI in these young people is used to reduce these dissociation symptoms. So motive can be, for example, to regain the awareness of physical sensation. This is also a very frequent motive. Recently, we moved on to assess psychometrics with ecological momentary assessment, EMA, or ambulatory assessment. You probably are familiar with that. We can now use these handhelds to assess. The questionnaires that we used to have as a paper pencil version we now have on the smartphones, and the advantage of that is that we can assess the symptoms and also some biological data in everyday life. So these smartphones every hour, every other hour, it will ask for what the people have done, where they are, if they have hurt themselves, for example, so you can assess self-injury in real life. And of course, you can also use that tool for interventions.
Then this is for the EMI, ecological momentary intervention. That is also a relatively new field. And there are different assessments. For example, you can have random prompt, so this thing beeps every other hour or event based prompt, such as what we are doing, so once an individual has hurt her or himself they have to answer certain questions. I will show you what we did with that. But beforehand let me show you some older data that was before the smartphone era because we had these handheld PCs that were carried around over 48 hours, and you see here at first glance that the high level of tension or stress levels in borderline patients compared from those. This is a very significant marker of the borderline personality disorder, and what is even more interesting, if you look at the fluctuations, you can use these assessments every hour and look at the fluctuation from hour to hour.
For example, here, this NICE study found that the effectiveness can be shown here by changing levels from green, which is very good mood, to very bad mood, and see this high fluctuating states as compared to the healthy controls. The healthy controls are mostly in the green area and the borderline patients are fluctuating a lot. So this is a marker of affective instability. When you look at changes in emotions before and after NSSI there’s an interesting recent study that showed or that assessed negative and positive emotions before and after NSSI events in real life. And in the left side, you see the negative emotions, sadness, angriness, feelings of being hurt or frustrated, and on the right you see the positive emotions, content, proud, feeling relief and so on. And what is a bit scary, actually, is that all the negative emotions get better after self-injury, so it’s a reduction of negative emotions and the positive ones also get stronger, so they feel more happy, more content, more relief. So you see there’s a strong moment of reinforcement of NSSI that can be measured in everyday life.
So now we’re coming back to our EMA study that has been recently finished, so the data are not published yet. And what we did here is we studied young adolescents and adults between the age of 15 and 25 in everyday life. They’re carrying these smartphones that had the random prompt, so every second hour it beeped and we assessed emotions; we assessed dissociation with the [inaudible 00:10:50] called [inaudible 00:10:51], and we also assessed interpersonal events. So we asked them, what did you do before the last meeting? With whom were you gathered? What did you do? And then we had this so-called self initiated forms that were these forms that were carried out after NSSI. So the participants were instructed once they had cut or hurt themselves to answer questions and to mark that on their smartphones. And we assessed the same questions, emotions, dissociation, also interpersonal events. So we could study what happened in terms of emotions, but also in terms of what changed in their interpersonal life before and after self-injury. And another thing you can combine that with is to assess certain biological data. What we did we assessed saliva samples and looked for the opioid system. So we assessed endogenous opioids that are produced in the body that are related to pain before and after self-injury.
SO I will show you only a few of these data from this study. So here is the first data for the effect and the stress level. So here’s a negative effect and you see the NSSI condition and the control condition. And I forgot to mention that with this design you can compare problems after NSSI with problems before and after conditions where you had similarly high stress levels, but without NSSI. So you have the possibility to compare states of high stress with one followed by self-injury and one without self-injury, so we have a controlled condition. So you see that we have situation with high levels of negative affects and stress. And then you have the controlled condition where there is no NSSI event, then you have the same, similar situation of high stress followed by NSSI. You see that negative affect. And tension is lower when the NSSI has been carried out. So this confirms the older study that NSSI leads to a reduction of negative affect and tension levels.
Now coming to the biological data, which is, I think, the first one that has compared endogenous opioids in relation to real NSSI events in everyday life, how are these things related? And not just opioid system is a relatively complex system; there are different classes of opioids. Most importantly, endorphin studies are the better endorphins that are important for regulating of pain, better endorphins down-regulate pain, not only physical pain, but also social and emotional pain, and are related, of course, to positive reinforcement. So the question was, can the [inaudible 00:13:49] of better endorphins be related to NSSI? Because of course, the endorphin opioid system is related to pain, as I mentioned. So the question was always implicated to non-suicidal self-injury.
What did we do in this study? This was done all over Germany. And you see, we assessed individuals in different places and it was incredibly complex. So they had to have a little bit of saliva sample here that were collected and then sent all to Mannheim and analysed and [inaudible 00:14:27] in the lab that [inaudible 00:14:27] saliva opioid assessment. What we found is that there is a difference between what came before and after self-injury in terms of endorphin levels. So they actually increase after NSSI; before NSSI they’re relatively low; and after NSSI they get back to normal levels. So it seems that there is a depletion of endorphins before NSSI and through self-injury endorphin levels go back to normal levels. This might be another important factor to explain the mechanism and to explain the reinforcement of NSSI. That is a hypothesis that has been put forward. But we could demonstrate that in [inaudible 00:15:21] case in everyday life, that endorphin levels can be increased through self-injury.
And finally, I show you some of the findings in relation to interpersonal events. So we actually assessed negative and positive events before and after NSSI. Here you see the negative events we assessed. For example, someone criticised me. I was rejected. I was ignored. The positive events, just examples, someone supported me, showed me affection. And as you can see, there were relatively more negative events in general than positive events in everyday life here. But the interesting thing is what happened after self-injury. So we looked at the difference between before and after self-injury in that [inaudible 00:16:15] hypothesis actually our first hypothesis was that negative interpersonal events predict higher probability of NSSI. So if I had had a bad experience, a negative event, then the probability to hurt myself is higher, and this could be confirmed. That’s indeed the case.
Our second hypothesis was related to reinforcement. So that’s something negative social reinforcement, so meaning that the hypothesis was that NSSI reduces the probability of negative social events. So there has been hypothesised that when you hurt yourself negative social events become less important or you can avoid, for example, negative social events, negative social encounters and so on, but this could not be confirmed; this was not the case.
And the third hypothesis was related to positive reinforcement. So the question was, does NSSI increase the probability of positive social events that I’ve just mentioned? So I get affection, people take care of me, and that could also not be confirmed. So we could not confirm these two social reinforcement hypotheses, but we could confirm the hypothesis of a prediction, so that negative interpersonal events indeed predict higher probability of NSSI.
So now this was all not really in neurobiology so far, and now you probably wait for the brain in this now [laughter]. Are we coming to the brain? So first, before we talk about pain, we need to talk about emotion regulation, and just in a nutshell, what is going on in the brain of a borderline patient that is self-injurious? What is important here? A few facts. One is that, of course, the important brain region we are talking about here is the amygdala, here in red. The amygdala is the emotional centre of our brains. And for example, if we show these pictures, like these negative pictures from the International Affective Picture System, usually the amygdala is overactive in most people. And second, the amygdala is on a strong control from the prefrontal cortex. You can see here that also lateral prefrontal cortex has two pathways how it regulates the amygdala. One is through the cingulate cortex. That’s the dorsal pathway here and there’s a ventral pathway through the orbital front cortex. So amygdala activated by negative stimuli and is under strong, prefrontal control. And what we could show in a meta-analysis that has been recently published is that people with trauma related disorders, such as PTSD and borderline PD have overactive amygdala when you show these pictures. That’s different from major depression, where we have a rather hypoactive amygdala. So in these trauma and stress related disorders the amygdala is overactive and we assume that this regulating pathway from the prefrontal cortex is not functioning well. So that’s the background for that.
So remember that the amygdala is overactive because now we come to pain and here the picture is very different. And that’s interesting because we are interested in why people cut or hurt themselves. So why do you inflict pain on your own body, which is somehow not logical? So let me try to solve this puzzle a bit and I’ll show you in the next few minutes. So we started this research, as already mentioned, very long ago, nearly 20 years ago. We did studies on pain processing in borderline personality disorder. In the old days, we used the cold pressure test. It’s a very simple test. You take a bucket of ice-cold water, put your hand in there, and then you just simply ask the participant, how painful is that? And that you can see here. And you ask painfulness on a scale from zero to seven. And we here normal healthy women between 20 and 30 years old, and you see it’s pretty painful. So it goes up nearly to the highest pain fullness. Then we had borderline patients here. That’s in red, this line. And you’ll see how reduced painfulness is here. It only goes up to a level of two.
And the more interesting is the same patients who are studied here again… That’s the green line… the same patients were asked to come to the lab and to the cold pressure test when they were in high level of stress, when they were just about to cut themselves, and you see again another, this is still a significant reduction of painfulness under stress. The same could be confirmed in the second study, where we assess pain not with water, but with electric stimulation. So we used all kinds of painful torture methods [laughter], so one of them is electric pain, pain thresholds. You increase the level of current, and then you ask, where is the pain threshold? When do you feel pain? And you see the higher the arousal or the stress level, the higher the pain threshold. That’s exactly the same. So you can correlate individual arousal with painfulness.
Interestingly, the painfulness and cortisol system are closely related. And [inaudible 00:21:51] and colleagues did studies on the Trier social stress test so if you inflict social pain the Trier social stress test in social pain stimulation where you would sit in front of an audience, you have to perform, you have to make a presentation in front of an audience, which is a social stress, and you can see that normally the cortisol levels as a marker of social stress go up, but there’s a blunting of this social stress response of cortisol in an NSSI group, so nearly no reaction of cortisol. Interestingly here, and again, here you see how pain processing is disturbed in NSSI. The opposite is the case if you have inflicted pain. Then cortisol is rising more than in control. So NSSI subjects also they feel less pain, have a stronger rise of cortisol levels. So there’s, again, a disturbance of the pain stress axis, as you can see here. And this fact of reduced pain sensitivity that means something different with different methods have been recently confirmed in meta analysis, also by the Heidelberg group, Julian Koenig, nicely demonstrated that if you look at different studies all across the board, you will get our own borderline studies, there are other studies that study pain sensitivity in youth with NSSI, without any form of disorder, that all over the board it’s a reduction of pain sensitivity. So this is a… you can say it’s a fact now and has been confirmed in meta-analysis that pain sensitivity is reduced in those that hurt themselves.
So now coming back again to the brain, another of our pain methods is [inaudible 00:23:37] pain, which is ideal to study during fMRI. We can’t carry the ice bucket water in the fMRI scanner, so we need a more sophisticated method here. And what we did is this thermal induced pain. This is placed on the back of your hand and it’s heated up to temperatures between 40 and 48 degrees Celsius. And here again, this is not… Now it’s not new for you anymore. In borderline patients, here in red, you see the reduction of pain sensitivity. Let’s take the temperature, for example, 46 degrees, that we applied. In healthy women, it’s pretty painful; here you see 60 percent painfulness. In borderline patients because they are all frequent cutters, so all these borderline patients we studied here in this study have at least one NSSI event per week and they have a reduction of pain sensitivity. This is not new. But now the more important effect is if you remember the elevated amygdala activity that is normal in these patients, here with the pain stimulation, we see the opposite. When we afflict pain, normally in healthies it increases the amygdala because it’s an aversive stimulus. In the borderline patients it’s the opposite. It’s a decrease of amygdala activity, not only amygdala, but also other affective brain regions, such as the anterior cingulate cortex.
Interestingly, one of the control regions that controls pain, the dorsal 00:24:59] prefrontal cortex is overactive. So there’s an over-control of the amygdala, which leads to a dampening, and this is a hint that something is completely wrong here and that the amygdala is decreased to painful stimuli, and maybe also, again, reinforcing the mechanism [inaudible 00:25:18]. Then we went a little bit deeper into the system. We combined these two mechanisms. So we first tried to increase the amygdala activation by these pictures and then added the pain. So to mimic somehow this self-injury or the mechanism of self-injury. Of course, it’s not self-injury here because the pain is afflicted by the machine or by the investigator. I come to that later how you can disentangle that. But here it’s still simply pain. But what we can see is, again, the amygdala is overactive. That’s the borderline patients here in this circle. So in the case where we only show the pictures, there’s an overactive amygdala and again, [inaudible 00:26:02]. After adding pain here, the amygdala goes back to normal levels, so, again, demonstrating that pain can be used to dampening the overactive amygdala here.
The next step was to look at the impact of tissue injury. We tried to understand the mechanism of self-injury from different angles, and of course, the first and most important is pain. Second is tissue damage or tissue injury. So, in self-injury, as I mentioned, the definition is damage of the body surface, so usually cutting. So why people use cutting? So we use a pain model. It’s called incision. Incision means we do a little damage to the skin. That’s how it’s done; it’s done with a scalpel. It’s a 4 mm wide incision. It’s nothing compared to the real self-injury, which is often 10 centimetres or so. It’s very short, but it’s a little destruction of the skin. It’s comparable to a venipuncture in terms of the size and also of the painfulness. But interestingly, we combined it with a stress induction that’s a modular limiting stress task. We’ve seen that it’s forced arithmetics under time pressure. You are told that you are too slow, and it’s a very nasty task because the better you get, the more difficult the arithmetics get. And there’s also a social component in here, like in the Trier social stress task. You are told that you are here in the red area and that the other participants in the study managed to go to the green level, but you will never make it there [laughter]. You are always held in the red part. So it’s really stressful, and you can see it works here. We can really induce stress here. Borderline patients, healthy cohorts, they all get stressed. But the difference comes later when we add this incision after the stress induction. The borderline patients go down and the healthies go further up. Again there’s an opposite direction of the stress levels here. And we did that in the scanner. You can do that in the scanner. And again the amygdala here in the borderline patients… the reds are the borderline patients… goes down. So the amygdala activity is dampening after this incision while it is increased in healthy controls.
But we really wanted to understand, is it really the tissue injury or is it simply pain? So we needed the paradigm in direct comparison of these two paradigms and that’s what we did here. We compared the incision with the scalpel with another method of painfulness, pain induction, which is called Blade. It’s a pressure pain, but it doesn’t destroy the skin. It’s the same painfulness. These two instruments have the same painfulness, but the only difference is that this is destruction of the skin and this is not. And actually we thought that the incision is more potent in stress reduction because we thought that tissue injury is an important factor, but we could not demonstrate that. Interestingly, both incision and Blade match to a similar decrease of stress levels. You can see here that’s the Blade, the dotted line, and the solid line is the incision. It’s exactly the same reduction of [inaudible 00:29:42] of stress levels, so the stress reduction is the same. So this was contrary to our hypothesis: no influence of tissue injuries; it’s the simple effect of pain.
The last factor is likely the two… Last factor besides tissue injury is the role of seeing blood. So why is it so important to see your own blood flow? And the last one is again, as I already mentioned, the perspective. So is it important that you inflict the pain yourself, like in self-injury or is it also working if something else is doing it? So we did that by two design. We used our Blade and combined it with artificial blood. So it’s blood that is used in movies or in theatres. It has the same temperature, the same consistency and the same colour as real blood, but it’s artificial blood. And we combined that with a Blade and we had conditions of the pain infliction, either self-inflicted or other inflicted. You can put this instrument yourself on your skin. It can’t cause any harm. So we can compare the same painfulness, either self or other inflicted and with and without blood. So let me first show you the effects of seeing blood. There’s no difference in painfulness. That’s nice to see. But the important thing is here in terms of [inaudible 00:31:11] and we had the same in terms of subjective stress levels, there is an initial decrease that is stronger. This is a decrease of stress levels again, so going down means decrease, and the decrease is stronger when blood is added, although the painfulness is the same. There’s a significantly stronger decrease just by simply having this red colour running over your skin. And interestingly, it’s very short lasting. It’s only the first few minutes; after half an hour this thing is gone. So this is closely related to the clinical experience that the patients say, okay, once I’ve seen the blood flowing over my arm, I feel relieved, and this is an immediate and a very strong effect. Then finally, the effect of author, of self or other. And here the results are a bit mixed. So in terms of heart rate, directly after the stimulus, immediately after the stimulus, there’s a stronger decrease when it’s inflicted by others. Three point five minutes or a little bit later, it seems to turn around, so it’s not really clear yet, so we are not sure what to make of this dual effect here. So the role of doing it yourself or someone else is inflicting pain is a little bit unclear yet.
So let me sum all this kind of research, let me sum that up. So we could demonstrate now in a confirmed manner there is a clear finding of reduced pain sensitivity related to NSSI. Pain leads to a reduction of stress and also of amygdala activity; that’s also relatively clear. It seems to be the case that it’s not so important that the skin is destroyed in this mechanism. There is at least… We have two studies now with the effect of seeing blood that could be confirmed, but only by our group, so I would be happy to collaborate with anyone that is using similar methods. We need much more of this research. There are only a few groups worldwide that do that, and, of course, all in Vienna. And you’re doing pain research here, which is really important to increase the number of studies and also to confirm that not only by the same but by others. That’s really important. So that’s why it’s in brackets here because we only confirmed it ourselves. As I mentioned at the end the role of the perspective, so whether I inflict pain on myself, is that the important factor? We kind of say, yes. Okay, so that was the story of the mechanism of NSSI in relation to emotional regulation.
Now coming to social interaction, a little bit different topic. How is that related? We now know that social pain is related to physical pain. It feels similarly and it is related to the same brain regions. So social pain, like social exclusion, leads to an increase or activation of brain regions that is very similar to actual physical pain. And how do you induce social pain? It’s usually done by this paradigm called Cyberboy. You probably have seen or heard about that. It’s a paradigm where you are told that you are playing a ball game with two other people and you are told that this is your hand. For example, you can do the instruction that you are lying in an MRI scanner and you’re playing with two other people that are lying in different scanners in another country also. And you have two conditions. You have an including condition where you are allowed to play with them. And there is an excluding condition where only these two guys are playing with each other and you don’t get the ball. That’s the exclusion condition. And it’s very potent.
You can see here the exclusion condition leads to increase of feeling excluded very strongly. So it really does what it’s supposed to do. And what we look at is to borderline patients… or that also… They all had NSSI as well here… how do they behave in the inclusion exclusion condition? Not surprisingly, both groups, during the exclusion condition, they felt excluded. That’s not very surprising. The more important thing is that during the inclusion condition there’s a difference and also during the control condition… I forgot to mention the control condition, which is just playing in clockwise here, round and round, so it’s a control condition just to control for the movement of the ball… So in the control and the inclusion condition the borderline patients felt more excluded than the healthy controls, which is also related to clinical knowledge that they feel excluded even if they are included. That’s what a lot of patients tell you.
Now, coming again to the brain, what happens in the brain here? There are certain brain regions, for example, here, the mid cingulate cortex, which are overactive in the inclusion condition in borderline patients. You can see here the borderline and healthy controls. The borderline patients show an overactive during the inclusion condition again. This is the most important difference. And [inaudible 00:36:39] group quite nicely demonstrate that this is specific for the NSSI. We compared NSSI adolescent group with a depressed adolescent group and only the NSSI group showed this activation here, not exactly the same region, but in a more frontal region of the medial prefrontal cortex, but also of the lateral prefrontal cortex. So it somehow also seems to be related to NSSI here.
And finally, we combine these two things again. So combining the exclusion and inclusion and assess pain after this. It’s kind of a stress challenge similar to the other ones I showed you before. So we first did an inclusion, then exclusion, simulation and then we assessed pain and the brain activation of pain. And since the inclusion is the more potent in showing differences, I only show you here the data for the inclusion condition. So pain after inclusion leads again to amygdala deactivation, similar to the one I showed you before. Not after a controlled condition where there’s only a warm simulation. But after the inclusion, in all the patients, in contrast to the healthy control, again pain showed deactivation. And interestingly, this is related to something called rejection sensitivity. It’s a questionnaire that assesses rejection sensitivity and the higher the rejection sensitivity [inaudible 00:38:10] the solid line, the lower the amygdala activity. So the more you feel rejected in everyday life, the more the amygdala goes down after painful stimulation here.
So I showed you how or what is wrong in the brain, also in relation to NSSI mechanisms in youth and adolescents with NSSI and borderline personality disorder, which is maybe, in a sense, a pretty sad story, but in the end I will show you some things that may be a little bit more optimistic or gives rise to hope, which is related to what happens after remission of borderline personality disorder, which means after they had stopped cutting themselves. And we were in the lucky situation that we had a larger collaborative [inaudible 00:39:07] over the last seven or eight years, all studying mechanisms in borderline PD in Mannheim in Heidelberg in Germany. And one was we compared patients with borderline personalities with those that had remission of the disorder. Remission in this case means having only three out of the nine DSM criteria, and most importantly who have stopped cutting. So here’s a group of current BPD, ongoing NSSI, remitting BPD, meaning having stopped NSSI for at least one year, and 22 healthy controls. Again, you know this thermal thing here in pain and you can see nicely halfway normalisation of pain thresholds. Here’s the healthies, you see normal pain thresholds, so the good thing here is they feel pain better than the borderlines, and the remitting ones, after having stopped cutting, are halfway back to normal, which is, I think, a good sign.
Another thing is we looked at the changes of this mechanisms. So, again, this is the stress induction or the moderate injury stress induction. And here you can see the virtual NSSI, so here in the top line is the current borderline, so that you see after stress induction, urge for NSSI goes up. Here in the solid line is the remitting, so there is no difference to healthy cohorts anymore. So they have stopped cutting and if you induce stress the urge for NSSI is not increasing anymore, which is also good. And here, this is a bit more complicated. I explain what that means. This is the relation between painfulness and stress reduction. So that means it’s painfulness of the stimulation related to stress reduction. In current BPD, that’s this line here, the more painful the stimulus is, the more stress is reduced, so that is why patients with ongoing cutting behaviour always increase the amount of pain, for instance cutting is getting deeper and deeper because they need more and more pain. So this is from here. But this relation is not found any more in remitted. So there’s no relation between painfulness and stress reduction. So this second mechanism or this relation between painfulness and stress reduction is not found any more in remitted, and here this shows that there is no increase anymore, for NSSI after stress induction. So, again, a good sign.
And here finally, what I want to show you is that after psychotherapy, we find the same in even shorter time. What we do is we have a residential treatment, DBT treatment, three months of residential DBT, and what you can see here is painfulness before and after. Here is healthy controls; here is BPD; here going down means reduced pain sensitivity as compared with healthies, and here again you can see tendency to normalisation of painfulness, and what I like even more is the change of the amygdala mechanisms. You remember this decrease of amygdala in current BPD patients; that’s fond here again before therapy. So the amygdala is deactivated and after three months of DBT, you find the normal amygdala activation. So that means pain is perceived as being stressful or aversive again, which is a good sign in borderline because that is related to stopping the self-injury.
Great. So you see there’s some light on the horizon, so to speak. So this is not all hardwired, but very important is now, and that’s also part of the STAR consortium that Paul is heading, and which I’m lucky to be a part of as well, to look at the cause of NSSI and painfulness through the course of adolescence. We have the chance to study young adolescents and adults between the age of 14 and 20, 21, and to study their painfulness and also these mechanisms of social exclusion, social inclusion and emotion regulation over the course of several years and just see how that develops and what happens when they stop cutting. That’s the STAR paradigm. We study 300 subjects with NSSI, adolescents. They don’t need to fulfil any disorders. They just have to have regular NSSI, compare that with other controls. We study several genetic, endocrine and also psych-physiological markers. We do EMA in everyday life. We study heart rate, heart rate variability and compare them, and it’s also embedded here is a E-treatment, an online treatment of NSSI, so we can have the comparison of E-treatment with the psycho-education group. And in the subgroup of 60 participants with NSSI and 60 healthy controls we also studied these fMRI paradigms I showed you this pain sensitivity, social exclusion and emotion regulation in this subgroup.
All right. So I would like to come to the end and sum it up. So I wanted to show you that we now understand a little bit more about NSSI and how this is related to pain sensitivity. That’s clearly a proven fact now that NSSI is related to reduced pain sensitivity. And we have demonstrated that underlying neural mechanisms, mostly relate to the amygdala prefrontal axis and how this is disturbed in people that hurt themselves. So there is amygdala deactivation, which is maybe related to the reinforcement of self-injury and this can be changed again. So in remission in patients that remitted or that have been run through successful psychotherapy these mechanisms can go back to normal. And I believe and I hope you agree with me, that the better understanding of this correlates is helpful in two ways. It helps to improve psychotherapy. We can now, for example, tackle these amygdala mechanisms, for example, with brain stimulation or neurofeedback methods more precisely. That’s one thing. And the second thing is, which I think is as important as the first one, is that it helps to destigmatise this behaviour. You all know that NSSI behaviours also is very much stigmatised and was long related to attention-seeking. Of course, it has these aspects as well, but there is also a very strong biological component behind NSSI. And then that’s my personal experience that it helps the patients and also their relatives to better understand it and to treat it as a disturbance of the brain as well, besides all the psychological components it has.
All right, I want to thank my co-workers and the collaborators here. These are the people in the group. [Inaudible 00:46:48] who was running a lot of these studies on ecological momentary assessment and also on the opioid system that were actually done here in the lab of Alexander [inaudible 00:47:01] here in [inaudible 00:47:01], this lab where Lisa* analysed all these saliva samples, which was quite some work recently. And I want to thank you for your attention.