It is with interest that we read the paper by Buiten et al. (1) in which the authors, using the implantable cardioverter defibrillator (ICD) remote monitoring function in patients enrolled in the ICD-2 trial, showed that hemodialysis (HD) is a trigger for atrial fibrillation (AF) episodes. The study of Buiten confirms some observations we made in the recent past. Few years ago we studied, by high frequency electrocardiography, the effects of the HD session on P-wave duration (Pwd), an expression of intra -atrial conduction velocity, demonstrating that the procedure induced a prolongation of Pwd closely related to the reduction of the plasma potassium concentration. When the pre-and post-dialysis values of electrolytes were put in a computational model that simulates the atrial action potential, the atrial potential upstroke was slowed down from the beginning towards the end of the HD session. Also we observed a reduction of the effective refractory period (2). It is well known that both phenomena are part of the electrical remodeling at the base of AF onset. These data were also confirmed by a multiscale computer model (3). Buiten et al. show that a lower concentration of potassium in the dialysis bath (which usually is associated to a high shift of the electrolytes during the dialysis session) is associated with a higher probability of occurrence of AF episodes intra and peri-HD session. This observation is concordant with our data. Recently, we reported a case of a patient in whom HD session constantly induced episodes of paroxysmal AF. The episodes were preceded by a reduction of the plasma potassium levels and an increase of supraventricular ectopic beats and of the vagal component of heart rate variability as evaluated by spectral analysis. The computational model, feeded with the data of each session, consistently showed an increase in atrial potential upstroke duration and a reduction in the effective refractory period. Both phenomena were enhanced when the IK,ACh current, sensitive to acetylcholine, was added to the model, suggesting that the vagal rather than adrenergic stimulation plays an important role in the genesis of intradialytic AF episodes (4). We think that it would be very interesting to test this hypothesis using data obtained from the remote monitoring of ICD-2 trial patients.

References 1. Buiten MS, de Bie MK, Rotmans BA et al. The dialysis procedure as a trigger for atrial fibrillation: new insights in the development of atrial fibrillation in dialysis patients. Heart 2014; 100: 685-690 2. Severi S, Pogliani D, Fantini G et al. Alterations of atrial electrophysiology induced by electrolyte variations: combined computational and P-wave analysis. Europace 2010; 12: 842-849 3. Krueger MW, Severi S, Genovesi S et al. Alterations of atrial electrophysiology related to hemodialysis session: insights from a multiscale computer model. J Electrocardiol 2011; 44: 176-183 4. Vincenti A, Passini E, Fabbrini P et al. Recurrent intradialytic paroxysmal atrial fibrillation: hypotheses on onset mechanisms based on clinical data and computational analysis. Europace 2014; 16:396-404

Conflict of Interest:

None declared

The primary finding reported in this study [1] is a correlation coefficient of -0.042 between the inflammatory marker of interleukin IL-6 and the percentage flow-mediated dilation index (FMD%). The difference between this correlation coefficient and zero is statistically significant (P=0.02) for the large sample of participants that was studied (n=3404).

Here we have a clear illustration of why it is important to distinguish between clinical and statistical significance in research. The coefficient of determination, calculated from the square of the reported correlation coefficient, is 0.0018. This means that less than 0.2% of the variability in FMD% is explained by the variability in IL-6.

Weiner et al.[1] reported minimum correlations to achieve statistical significance for the various cohort sample sizes. Nevertheless, the minimum clinically important correlation coefficient was not reported, nor was a robust clinical anchor rationalised to interpret the findings properly. The appraisal of clinical significance is, nevertheless, straightforward in this study. The 95% confidence interval for a correlation of -0.042 with a sample size of 3404 is approximately -0.08 to -0.01. Even if the smallest worthwhile effect is as low as 0.1 ("small" by all definitions), then the true population correlation between IL-6 and FMD% is almost certainly trivial and not clinically important.

We were initially interested in this study because of the size- scaling problems with the FMD% index, which were described recently, using examples from the same MESA dataset [2]. These scaling issues are also clearly evident in the study by Weiner et al.[1]. For example, the apparently higher FMD% reported for women is merely because baseline diameter (Dbase) is smaller in women, and because FMD% does not allometrically scale for Dbase properly - illustrated, consistently in the literature, by the negative correlation between FMD% and Dbase [2, 3, 4]. This latter correlation, which is mediated by statistical rather physiological mechanisms, tends to be between -0.8 and -0.4, and is, therefore, almost certainly clinically significant {2, 3, 4].

References

1. Weiner SD, Ahmed HN, Jin Z, et al. Heart Published Online First: 8th April, 2014. doi:10.1136/heartjnl-2013-304893.

2. Atkinson G, Batterham AM. The percentage flow-mediated dilation index: A large-sample investigation of its appropriateness, potential for bias and causal nexus in vascular medicine. Vascular Medicine 2013;18: 354 -365.

3. Atkinson G, Batterham AM. (2013b) Allometric scaling of diameter change in the original flow-mediated dilation protocol. Atherosclerosis 2013; 226:425-427.

4. Atkinson G, Batterham AM, Thijssen DHJ, Green DJ. A new approach to improve the specificity of flow-mediated dilation for indicating endothelial function in cardiovascular research. J Hypertens 2013;31:287- 291.

Conflict of Interest:

None declared

We have read Ghadri et al' s article published in the issue of April 7, 2014 (1), which provide a very useful review to help us better understand Takotsubo cardiomyopathy (TTC), a sometimes puzzling disorder. In the "TTC TYPES" section, the authors state that TTC often affect different left ventricular (LV) regions in the recurrent episodes. The similar findings have been reported repetitively by other studies (2, 3) and found in our own TTC patients. A consistent theme appears to emerge is that recurrent TTC tends to change its original LV contractile pattern, to avoid significant hemodynamic compromise induced by the previous episode. This brings up an interesting question: does the heart have a mind of its own, to use memory imprint to protect itself?

TTC has several unique clinical features: 1) in spite of sometimes grave presentation, it usually improves without residual cardiac impairment, 2) although most patients with TTC appear to have poorer systolic function than those with acute myocardial infarction (AMI), their LV filling pressures are paradoxically lower, 3) There is significant discrepant hemodynamic features between LV and Right ventricle (RV) (4-6). The prevalence of RV involvement in TTC is largely underestimated, particularly when RV apex has not been imaged sufficiently in the echocardiographic examinations. In our TTC patients, regardless of various LV regional involvement, RV contractile pattern in is very constant (4-6): the basilar and middle segments of RV are hyperkinetic, but the apical segment is hypokinetic (reverse McConnell's sign, 4). The hyperkinesis of RV likely represents an adaptive physiologic response to the low LV filling pressure in TTC, to maintain cardiac output. Tethering of RV apex to an akinetic LV apex accounts for reduced RV apical motion. This characteristic RV contractile feature helps clinically differentiate TTC and AMI (6). More importantly, this significantly different, but also reciprocally dependent functional status between LV and RV provides us a hint to explore the unique pathophysiology underlying TTC.

Recent research shows that TTC is likely mediated by a switching of epinephrine signaling transduction through the pleiotropic ?2-adrenergic receptor from canonical stimulatory G-protein-activated cardiostimulant to inhibitory G-protein-activated cardiodepressant pathways (7). Clinical evidence also suggests that TTC is likely evolved as a self-protective strategy to limit catecholamine-induced myocardial injury and reduce hemodynamic compromise. During the acute phase of TTC, an active LV dilation will lower filling pressure, not only protecting the vulnerable myocardium within LV, but also preventing an abrupt increase in pulmonary capillary wedge pressure and pulmonary edema. Meanwhile, the less vulnerable myocardium, including RV and basilar LV walls, are activated. These myocardial portions likely form a temporarily alternative functional conduit that detours blood flow towards left ventricular outflow tract (LVOT), to help maintain cardiac output, until the excessive catecholamine surge subsides. This well-programmed interventricular discrepancy/dependency helps maintain benign clinical courses during the most acute TTC episodes. Heart failure/cardiogenic shock will not occur unless some structural (e.g. basal septal hypotrophy) or functional (e.g. dynamic LVOT obstruction) abnormalities significantly interrupt this otherwise harmony machinery. In recurrent TTC episodes, the heart can learn and adjust its contractile pattern to avoid affecting this key "rescue" system, and avoid hemodynamic compromise happening again. Therefore, maintaining appropriate LV filling pressure and avoiding dynamic LVOT obstruction should be the main therapeutic strategies during the acute phase of TTC (6). Unnecessary cardiac catheterization and inappropriate pressor (such as dobutamine) treatment will potentially disrupt the heart's self protective mechanism, and results in adverse outcome.

Reference 1. Ghadri JR, Ruschitzka F, L?scher TF, et al. Takotsubo cardiomyopathy: still much more to learn. Heart. 2014 Apr 7. doi: 10.1136/heartjnl-2013- 304691

2. Xu B, Williams PD, Brown M, et al. Takotsubo cardiomyopathy: does recurrence tend to occur in a previously unaffected ventricular wall region? Circulation. 2014;129: e339-340 3. Kaushik M, Alla VM, Madan R, et al. Recurrent stress cardiomyopathy with variable regional involvement: insights into etiopathogenetic mechanisms. Circulation. 2011;124:e556-e557. 4. Liu K, Carhart R. "Reverse McConnell's Sign?": A Unique Right Ventricular Feature of Takotsubo Cardiomyopathy. Am J Cardiol 2013;11:1232 -1235. 5. Boppana S, Bhatta L, Liu K. Reversible Tricuspid Regurgitation in Acute and Chronic Cardiomyopathies: Medical Management vs. Surgical Intervention? J Am Coll Cardiol Img 2013;6:920-921. 6. Liu K Krone RJ. What truly causes the adverse outcome of Tako-Tsubo cardiomyopathy? Lessons learned from echocardiography. J Am Coll Cardiol Img 2014; (in press). 7. Paur H, Wright PT, Sikkel MB, et al. High levels of circulating epinephrine trigger apical cardiodepression in a ?2-adrenergic receptor/Gi -dependent manner: a new model of Takotsubo cardiomyopathy. Circulation 2012; 126:697-706.

Conflict of Interest:

None declared

I read with great interest the "challenge" posed by the electrocardiographic recording in issue 7 of Heart for 2014. I also noted with interest that the authors suggest that Monckeberg provided the first account of such dual nodes in 1993 (their reference #1). Since Monckeberg was long since dead by that date, this must be some mistake! I think the appropriate date is 1913. It is certainly the case that Monckeberg was the first to describe this rare anatomic arrangement. Monckeberg's heart, however, was characterised by the presence of discordant atrioventricular connections, and double outlet right ventricle. Since then, dual atrioventricular nodes have been demonstrated in another heart with double outlet right ventricle, but in the setting of concordant atrioventricular connections (Br Heart J 1973; 35: 566-569). To the best of my knowledge, all other hearts subsequently published with dual atrioventricular nodes have either had isomerism of the atrial appendages (Smith et al, Cardiol Young 2006; 16: 437-454), or else discordant atrioventricular connections, as in the index case described by Monckeberg. Again, to the best of my knowledge, hearts have yet to be described with dual nodes located as shown in the cartoon provided by Paech and colleagues. It is possible that their patient, with hypoplastic left heart syndrome, could have had a so- called "atriofascicular tract", since that is the closest arrangement of which I am aware that fits their cartoon.The second node in these situations, however, is found anterolaterally within the right atrioventricular junction (Becker et al, Circulation 1978; 57: 870-879), rather than inferiorly as pictured in their cartoon. I wonder, therefore, what additional information they have to prove that their patient had dual atrioventricular nodes? On the basis of my own experience with assessing autopsied hearts, I would be very surprised to find the arrangement that they have depicted in a patient with hypoplastic left heart syndrome.

Conflict of Interest:

None declared