Letters to the Editor
Is High-Dose Catecholamine Administration in Small Animals an Appropriate Model for Takotsubo Syndrome? – Reply –
2015 Volume 79 Issue 4 Pages 898-899
Details
2015 Volume 79 Issue 4 Pages 898-899
We thank Drs Angelini and Tobis for their comments on our invited review entitled ‘Pathophysiology of takotsubo syndrome: temporal phases of cardiovascular responses to extreme stress’.1 They correctly point out that the aim of our review was to synthesise current thinking regarding the pathophysiologic mechanisms of takotsubo syndrome (TTS) from the perspective of animal models developed by ourselves and other groups working in the area and to integrate clinical observations.
We respond to each of their questions in order.
(1a) We agree that currently we do not have an animal model that exactly recapitulates clinical TTS in humans. Preclinical models allow the exploration of pathological mechanisms in a reductionist approach that is possible in the laboratory. These studies offer insights that individually or collectively will shed light on the pathology of the clinical condition. It is our belief, on the basis of many reports and reviews, consistently describing the experience of emotional or physical stress prior to the onset of symptoms, that a high level of sympathetic activity remains the best candidate for a common initiating factor in TTS. It is for this reason that we and many other research groups have applied catecholamine dosing as an experimental model for the disorder. It would be very useful for a study to investigate the correlation of plasma levels of catecholamines in TTS and non-TTS patients. No large, appropriately powered study has delivered this (Wittstein’s seminal paper demonstrated high serum catecholamine levels in a small TTS cohort2). We acknowledge that this has not been reproduced by other studies.
(1b) Regarding the reproducibility of TTS, we would suggest that the large number of reported cases of iatrogenic TTS following ‘epi-pen’ use or dobutamine stress echo (DSE), as well as other sympathomimetics, is at least some evidence of the ability of these agents to induce TTS. Although we agree that in such cases, the patient’s endogenous sympathetic activity also may be high prior to epinephrine administration (suggesting some synergy perhaps). Clinicians, somewhat understandably, are likely to be cautious about precipitating recurrent TTS in previous sufferers, given the potentially high risk of complications in the acute phase. Recurrence of TTS has been reported in 10% of cases;3 the prevalence of TTS may be higher in previous sufferers, making TTS a risk factor for subsequent TTS. It seems that these reported recurrences were usually following a further stressful event, meaning that TTS symptoms may re-emerge following another sympathetic “surge”.
(1c) In our review we detail that severe hypertension is rarely reported in patients. In our opinion this is because there are so few cases occurring within a setting where the immediate blood pressure (sub-5 min) response can be measured and recorded. This would need to take place within minutes of the precipitant, as the half-life of catecholamines in plasma is very short. We discuss the Gothenburg criteria, which allow pheochromocytoma to be present in the diagnosis of TTS. We believe that the Mayo Clinic criteria were essential in initially defining a complex diagnosis of exclusion and were a landmark in helping to identify cases. However, they are now too prescriptive, given that there is a greater awareness and understanding of TTS and the clinical variables.
(1d) We have suggested that TTS occurs within a narrow pharmacologic window between the complete lack of symptoms and acute arrhythmic death. Our reports detail a small number of dosing methodologies that are able to recapitulate contractile abnormalities in a phenotype consistent with TTS. The mortality rates are indeed higher than those reported in humans. We are able to report the exact number of deaths caused by dosing (these usually occur quite acutely). In the human population, in-hospital deaths following diagnosis of TTS therefore may be smaller in number. There is a selection bias because TTS sufferers presenting at hospital are by definition survivors vs. those who succumb to sudden cardiac death at the time of shock/stress. Fundamentally, mouse and rat responses may be milder than the responses of primates, meaning a more nonspecific range of pharmacologic activation must be entered before symptoms are produced. The study by Izumi et al in nonhuman primates used, comparably, much lower concentrations of epinephrine to precipitate TTS-like symptoms.4 The physiology of primates may be more sensitive to sympathetic toxicity or endogenous catecholamine levels may be higher.
(1e) We agree that small animal echocardiography is a challenging technique. Numerous steps have been taken by workers in this field to overcome some of the problems posed by rodent physiology. We used M-mode echo images from the apical and basal segments, which are paired to baseline recordings. Our M-Mode echocardiography protocol operates at a frame rate of 1,000 frames/s, which gives 100 s of images for each cardiac cycle in a rat and comfortably affords sufficient temporal resolution to study physiology at rapid heart rates. Shao et al were able to perform higher resolution B-Mode scans from which they were better able to visualize hypokinetic segments.5
(2) Finally, we do not dismiss the role of vasospasm in TTS. Reductionist and physiological studies are needed to dissect pathologies with a vascular or a direct myocardial basis in the whole heart. We have described studies in which perfusion defects have not been found.6 Regarding acetylcholine, it may be the case that future studies should assess whether any parasympathetic ‘re-bound’ is involved in sustaining animals for long enough to allow TTS to develop. Using atropine increases the mortality of animals injected with epinephrine to an unacceptable level.7 Given the high levels of oxidative stress in the vasculature following a sympathetic storm, it is perhaps not surprising that infusing vasospastic drugs demonstrates enhanced vasoreactivity. In this case, confirmatory studies are required to assess cause vs. association. It is conceivable that vasospasm is present in some cases of TTS and is part of the pathology. The fact that there are studies without vasospasm confirms that it is not absolutely necessary for TTS precipitation (examples of these cases include DSE and the use of isoprenaline in small animal models). We certainly encourage further studies (both clinical and laboratory) to explore the role of vasospastic lesions in TTS.
TTS remains something of a conundrum, but we hope that clinical and benchtop studies will continue to dissect the mechanisms underlying this increasingly recognized acute cardiac syndrome.
(Released online March 5, 2015)
