Abstract
Evolution and post-traumatic stress disorder
It is a pity that Cantor and Price, in their timely paper in the most recent edition of the Journal [1], did not pursue their evolutionary perspective on post-traumatic stress disorder (PTSD) more widely. The Stockholm syndrome sits comfortably in a biopsychosocial framework and the “inescapable” element in circumstances leading to PTSD is generally considered important. However, the further back you go in the evolutionary process the less relevant psychosocial issues become.
While defensive behaviours are certainly associated with PTSD, McEwen and Lasley suggest that the condition itself is the result of “allostatic load” (overstretched stress response systems) [2], implying a pathological, degenerative, process rather than a past adaptive mechanism.
Garcia et al. noted that “all successful living organisms have evolved elaborate mechanisms to select food and equally complex mechanisms to defend against becoming food, at least until they have passed on their genes to their offspring” [3]. Indeed, as Bradley reports, so stressful is the effort of achieving parenthood that the males of some species of salmon die after spawning and the male (but not the female) antechinus dies at the end of just one mating season [4]. While survival of the species may be at risk the behaviour is not defensive. The chemistry involved is closely similar to the chemistry in PTSD, including in particular abnormal levels of cortisol.
Garcia et al. have demonstrated primitive “disgust” in the garden snail and conditioned “fear” escape behaviour in sea slugs, and a functional division into visceral and somatic learning is present throughout the animal kingdom, defending the gut from poison and the skin from predatory attack. Their work has demonstrated different behavioural manifestations of disgust and fear in rats, cougars, coyotes and monkeys [3]. Both gastric and skin complaints are associated with stress in humans and the pathogenesis of gastric ulcers has been studied, according to Bradley, in a range of animals including rats, hares and rabbits, carnivores and primates [4]. Le Doux indicates that “the behavioural expression of fear conditioning and its neural basis appears very similar in all species that have been examined in detail,” including insects, fish, birds, reptiles, mammals and man [5].
Given the importance of survival PTSD must surely involve contributions from all three areas of the “triune brain” [1] with the new mammalian era contributing the most significant psychosocial component. Le Doux suggests that the immediate response to unexpected threat bypasses the cerebral cortex [5] and it makes sense that such a fundamentally necessary response to highly, life-threatening, environmental challenge should be entrenched in the oldest parts of the brain.
Finally, Salzet reports that all animals, including parasites, have the same proopiomelanocortin-derived peptides involved in communication between the immune and neuroendocrine systems; both of which are involved in response to a wide range of threats and challenges [6].
Thus PTSD research needs not only to focus on context but also to be aware of the evolution of stress responses and the consequences of those responses.
Evolution and post-traumatic stress disorder: reply
Chris Cantor, Department of Psychiatry, University of Queensland, Brisbane, Queensland, Australia and
Paddy Burges Watson wishes we had pursued our evolutionary perspective of post-traumatic stress disorder (PTSD) more widely. It has been done [1]! JP's correspondence regarding CC's book and JP's work on appeasement led us to take the book's appeasement section to a higher level.
We agree that the further back you go in the evolutionary process the less relevant psychosocial processes become. Appeasement's dependence on social interactions in complex PTSD represents PTSD's most recently evolved form. PTSD, a disorder of defence [1], although one of the most recently recognized psychiatric disorders, represents one of the oldest in its evolutionary origins. Evolutionary sources of major threat were predators, the environment and conspecifics (one's own species). Predation threat dominated the evolution of early defences and thereby laid the foundations for PTSD.
MacLean's triune brain concept [2] is immensely useful. While the palaeomammalian and neomammalian brain regions are more relevant to complex PTSD, the reptilian and palaeomammalian regions are more relevant to ordinary PTSD. However, cognitive behaviour therapy (CBT) requires patients to use their neomammalian brains to counter these older regions. Frequently patients when pushed to explain their disappointing exposure exercises respond, “I don't know, I just couldn't.” Simply put, their reptilian brains acting outside of awareness overrode their neomammalian brains. Usually they are also too afraid: palaeomammalian input. It beats us how anyone does CBT without reference to the triune brain.
Pathology versus adaptation is like debating sound versus light waves. Neither is right or wrong – the issue is relevance. Psychopathology appears more correct because of clinicians’ familiarity with patients’ time frames, as opposed to the usually thousands to millions of years relevant to evolution. Take the obvious pathology of a broken leg. View it from an evolutionary perspective and you will find that limb strength evolved as a result of selection pressures. Neanderthals were stocky, short-limbed early humans who coexisted with the taller, slightly built, first modern humans whose limbs were more fragile [3]. Stocky Neanderthals did well in the cold. However, a changing environment and complex context-dependent cost–benefit issues determined the demise of Neanderthals and to some extent today's incidence of broken legs. Not too useful for the individual currently in pain, but potentially relevant to research and scientific progress.
A veterinary text advocates using the DSM-IV for animal psychopathology. A reservation expressed was “… given the importance of evolutionary and ethologic theory in comparative biology, a comparative clinical psychopharmacology might immediately move toward classification systems that rely on these theories.” [4]. Psychiatric neuroscience needs to recognize this. Genes are all very well, but most are associated with function and we need to understand those functions. Assuming genetic pathology for the ultrahigh-prevalence conditions that dominate psychiatry is naïve. What are the social activators of these genes? Future psychiatry mostly requires informed psychosocial interventions, not gene therapy for all.
In the clinician's office evolutionary theory is immensely useful for psychoeducation; it delivers understandability. CC asks irritable individuals with PTSD how might an animal licking its wounds following a mauling by a predator respond to approaching others? If in doubt snap. Patients identify more closely with this than with neurophysiological explanations of anger.
Burges Watson also raised McEwen and Lasley's 2004 work [5] on “… ‘allostatic load’ (overstretched stress response systems) implying a pathological, degenerative, process rather than a past adaptive mechanism …” (BW's quote). More recently McEwen wrote on the same subject “… undergo a reversible remodelling of their dendrites in conditions such as hibernation and chronic stress. The role of this plasticity may be to protect against permanent damage. As a result the hippocampus undergoes a number of adaptive changes in response to acute and chronic stress.” [6]. What is currently adaptive is likely to have been so in evolutionary times.
As for Burges Watson's references to Le Doux's work [7] and his other points, all we can say is we agree [1] and thank him for his interest.
Catatonia and telephone use
The seminal descriptions of psychopathology antedated general availability of the telephone (particularly, the user-friendly mobile telephone). Over the last year we have observed four patients with stuporous catatonia (absence of movement and speech) who have, unexpectedly, been able to communicate with family members and friends (when they have called) on a mobile telephone.
A 35-year-old mother with a recent history of bipolar disorder had become progressively slower and finally immobile and incontinent of urine over a 4 day period prior to being brought to hospital. On admission she was mute. Her limbs had markedly increased tone (assessed by palpating her muscles) with resistance to passive movement and some cog-wheeling. There was “forced grasping” [1]; she grasped sheets and examiners and did not release her grasp on request. Careful examination revealed changing symptoms over a half-hour period. At one point there was gegenhalten [2]; she resisted movements with the same degree of pressure exerted by the examiner. At another point there was less resistance, and when placed in a position she would retain that posture despite being told that she could return to a resting position. One example of this was when her hip and knee were flexed so that her heel was 30 cm above the bed; this position was retained for 5 min although she was invited to return her leg to the bed. Computed tomography was normal and she made progress with electroconvulsive therapy.
The father was a senior, reliable health professional. He reported that in the hour before he brought the patient to hospital, her brother had telephoned from overseas, to wish her well. The patient took the telephone and conducted an apparently normal conversation, quite different to that which she had been conducting with the rest of her family, or subsequently conducted with her medical attendants. There was nothing to suggest an episode of catatonic excitement.
The other three patients mentioned here had been in the psychiatric intensive care ward and were observed by nursing staff when they had taken telephone calls from relatives and responded tolerably well. One of these was a middle-aged former teacher with a well-established pattern of stuporous catatonia (including occasional incontinence of urine and faeces), but uncertain diagnosis. The other two were young men, both of whom suffered schizophrenia and had been mute and immobile.
All four patients showed temporary but unsustained mild improvement on i.v. diazepam (20 mg over 5 min).
The inconsistency of being able to converse with close relatives on the telephone but not with other close relatives or medical or nursing staff can raise questions about the patient being selective or manipulative, and call the diagnosis into question. However, we are in no doubt about the diagnosis in any of these cases.
Hamilton states that “(n)egativism depends to some degree on the environment”, and continues, “fellow patients evoke the negativistic reaction much less easily than doctors and nurses” [1]. Catatonic stupor may result from “excessive cerebral excitatory processes” [3]. It may follow that face-to-face interactions increase pathological excitation, thereby increasing the difficulties of responding. The telephone (an inanimate object) may allow communication of support by others, without the arousing effect of their physical presence.
Ketamine for the treatment of depression: what about the addictive potential?
Recent studies suggested a high therapeutic value of ketamine in treating depression [1]. This was shown for i.v. application of a low analgesic dose of ketamine in patients with major depression. However, ketamine is also used as a so-called “psychedelic club drug” in many countries and is known to have a clinically relevant addictive potential. Ketamine is widely used for anaesthesia but reports are numerous regarding abuse and dependence [2],[3]. It has been reported that ketamine (regarding the addictive potential) has much in common with drugs such as cocaine and amphetamines as well as with opiates, alcohol or cannabis [4]. Furthermore, similar to other high-affinity N-methyl-D-aspartic acid (NMDA) receptor antagonists such as phencyclidine (PCP, “angel dust”), ketamine can produce psychosis-like symptoms and may induce an acute phase of schizophrenia (glutamate hypothesis of schizophrenia) [5].
In the recent study by Zarate et al. ketamine was applied once in patients with major depression, who were followed up for a period of 1 week [1]. In that study the authors describe a stable anti-depressive effect for at least 7 days (end of study period). In our clinical experience the effect declined rapidly in most depressive patients 1 week after ketamine application. When we applied ketamine for a second time in several patients the anti-depressive effect was reproducible in most cases but again did not last longer than after the first application (in contrast to Zarate et al., who used a racemic mixture of equal quantities of (+)-ketamine and (−)-ketamine, we used the (+)-enatiomer). This raises the question of whether repeated i.v. applications of ketamine may help to stabilize the anti-depressive effect for a longer time, similar to electroconvulsive maintenance therapy. If applying ketamine infusions either once or more often, clinicians need to consider the addictive potential and the risk of inducing psychosis-like symptoms. This probable low risk of dependence should not restrict the use of ketamine to treat major depression, which is an intriguing new pharmacological approach. However, caution should be used with regard to patients with comorbid or premorbid substance abuse, including abuse of alcohol, nicotine, and benzodiazepines. The risk of developing dependence may exist especially in patients with repeated ketamine applications. Further research should therefore focus on additional pharmacological strategies to stabilize the positive effects of ketamine infusions.
Error monitoring is associated with the clinical state of schizophrenia
Frith proposes that a number of the “first rank” symptoms of auditory hallucinations, thought insertion and thought withdrawal may share one of the cognitive processes, namely, a deficit in self-monitoring [1]. Error monitoring is the online detection of error commission and the subsequent regulation that covers the self-awareness and behaviour adaptation. Previous studies have focused on the correlation between error detection dysfunction indicated by smaller error-related negativity (ERNs) and schizophrenic patients’ symptoms such as severe hallucination [2] and severe psychomotor poverty [3]. Other research has focused on the relationship between error monitoring and cognitive abilities [4]. However, whether error detection and error regulation is impaired in patients with schizophrenia might depend on patients’ clinical state [5]. The present study aims to examine whether error regulation is defect in schizophrenia and to explore the relationships between error monitoring, executive inhibition and clinical symptoms
Thirty-four patients with schizophrenia (25 men and nine women) meeting the diagnostic criteria of DSM-IV [6] were recruited. The mean age and education level of the patients were 33.12 years (SD = 11.27) and 11.29 years (SD = 2.29), respectively. The mean duration of illness was 4.51 years (SD = 7.13). One (2.9%) had undifferentiated schizophrenia; two (5.9%) had disorganized schizophrenia; 23 (67.6%) had paranoid schizophrenia and eight (23.5%) had residual schizophrenia. The mean medication (chlorpromazine equivalent) was 25.92 mg day–1 (SD = 87.60). The clinical symptom assessments were performed by clinical psychiatrists using the Positive and Negative Syndrome Scale (PANSS) [7], medication side-effects were assessed by Simpson Angus Scale(mean = 3.21, SD = 2.43) [8], Barns Scale (mean = 0.90, SD = 1.50) [9] and Abnormal Involuntary Movement Scale (AIMS) (mean = 1.31, SD = 2.79) [10]. Executive inhibition was assessed by go–no go and stop signal paradigms. In the go–no go task, patients were required to respond when a red light (70%) appeared and to inhibit the response when a green light (30%) appeared. In the stop signal task, patients were required to stop the ongoing response when they saw a frame appear outside the signal.
In the stop signal task there was an inverse correlation between error regulation and executive inhibition (r = −0.398, p = 0.010). Partial correlation controlling for medication side-effects also indicated that post-error slowing (PES) was associated with negative symptom (r = −0.421, p < 0.05, n = 22). In the go–no go task, we compared the mean reaction time (RT) of correct response that immediately precedes a commission error and that immediately following the commission error with a covariate of ‘age’. The results indicated that the mean RT of correct response following the commission error was significantly longer than the mean RT of correct response preceding the commission error (F(1,21) = 22.24, p < 0.01, MSe = 3404.23; mean correct response preceding = 341.42 ms, SD = 11.33; mean correct response following = 445.15 ms, SD = 13.81). Partial correlations, controlling for medication side-effects, indicated that PES tended to be inversely correlated with positive symptoms (r = −0.498, p = 0.07, n = 12), general psychopathology (r = −0.503, p = 0.067, n = 12) and the total PANSS score (r = −0.506, p = 0.065, n = 12).
The fact that patients significantly slowed down their RT after they committed commission errors might suggest that error monitoring could remain intact in chronic schizophrenia with regard to behaviour. However, this result does not challenge previous studies but instead suggests that error monitoring is associated with the clinical state of schizophrenia. That is, the more severe the clinical symptoms, the smaller the PES. Smaller PES might suggest a poorer error regulation. The preliminary findings presented here indicate that relationships exist between error monitoring and clinical symptoms of patients with schizophrenia. However, further study should be carried out to examine this primary result and determine whether there are relationships between error monitoring and some specific clinical symptoms in schizophrenia.