|
|
||||||||
1University of California, San Diego, and 2PhiloMetron Inc., San Diego, California
Submitted 31 August 2006 ; accepted in final form 16 November 2006
| ABSTRACT |
|---|
|
|
|---|
sudden death; heart failure; dextrocardia; mechanoelectric feedback
Thus it is unknown whether TWA reflects the mechanisms of volume overload, which may be ameliorated, or intrinsic myocardial disease, which is likely to be more permanent. A stretch-related contribution to TWA is plausible, since acute volume load alters ventricular repolarization and conduction and induces premature beats (4, 24). A direct contribution of myocardial disease to TWA is also plausible, since CHF patients at low SCA risk are more likely to have normal TWA, although it is unclear whether they also have less volume overload (2, 9). Volume load has a complex interaction with myocardial disease and may alter arrhythmic risk differentially, whether regional or global (19, 24). Thus studies in normal hearts may clarify the contribution of volume load to TWA, independent of this interaction.
We hypothesized that acute volume overload would elevate TWA magnitude (Valt) in normal hearts. We tested this hypothesis in dogs during controlled, volume infusion, measuring TWA from atrial pacing at baseline, peak volume infusion, and after diuresis, calibrated against infused volume, pulmonary capillary wedge (PCWP), and pulmonary and systemic arterial pressures.
| MATERIALS AND METHODS |
|---|
|
|
|---|
Canine preparation. Animals were preanesthetized with xylazine (2 mg/kg im) and ketamine (20 mg/kg im) and then anesthetized with isoflurane (0.61.5%). They were intubated and then ventilated with 100% O2 at 16 breaths/min, with tidal volume of 10 ml/kg and a 1:2 inspiratory-to-expiratory ratio. After routine surgical preparation, we performed cutdowns to cannulate the left femoral and external jugular veins and a branch of the left femoral artery. Under digital fluoroscopic guidance (OEC Medical Systems, model 9600), a Swan-Ganz catheter was advanced transvenously to the pulmonary artery, and pressures were electronically transduced (Hewlett-Packard model 78353B, Sunnyvale, CA). A 6-F quadrapolar catheter was advanced to the right atrial appendage for pacing from an external pacemaker (Medtronic 5336, Minneapolis, MN). Catheter placement is shown in Fig. 1A (dog 2). Dog 5 was found to have dextrocardia during fluoroscopy (Fig. 1B).
|
Volume infusions and hemodynamic measurements. After an acclimatization period of anesthesia lasting 4560 min, we maintained constant anesthesia while infusing isotonic fluid (0.9% saline with 20 meq/l KCl) at 500 ml/h.
The endpoint of infusion was PCWP >20 mmHg for >30 min or acute pulmonary congestion, whichever came first. Pilot observations suggested that 80100 ml/kg net infusate (23 liters) were needed after adjusting for urinary and estimated insensible losses. Every 15 min we measured heart rate and femoral arterial pressure (Criticare Systems, model 1100, Waukesha, WI), pulmonary arterial pressure and PCWP, oxygen saturation, temperature, and infused volume. Mean pressures were recorded directly. At the infusion end point, we paused for 15 min. We then commenced diuresis with 0.5 mg/kg iv furosemide and recorded measurements every 15 min until urine output slowed to <40 ml/h. At the end of the study, the dog was resuscitated, if possible.
Measurement of TWA. We applied clinical Hi-Res electrodes (Cambridge Heart, Bedford, MA) to the canine torso in positions analogous to those used for clinical TWA testing. Dog 5 had dextrocardia, and so we transposed right and left electrode positions (Fig. 1B). Meticulous, careful skin abrasion was used to achieve good electrode contact (1). TWA was measured using HeartWave (Cambridge Heart) in right atrial pacing at baseline, at the infusion midpoint, at peak infusion, and in diuresis. Prior studies report TWA in dogs at heart rates of 140 beats/min (17). Therefore, we measured TWA by accelerating heart rate from baseline (typically 90110 beats/min) to 160 beats/min in 10 beats/min steps every 90 s, followed by symmetrical deceleration to baseline (14).
Quantification and classification of TWA. We described TWA by its magnitude (Valt). When TWA was detectable, Valt was measured to the nearest 0.5 µV for the 1 min of peak TWA, using the method of least squares (25). We also recorded onset heart rate (i.e., threshold) for the contiguous period of TWA, including this peak.
Measurement of heart rate variability. To assess variations in autonomic activity during the protocol, we analyzed R-wave-to-R-wave intervals in sinus rhythm for two consecutive 5-min intervals at baseline and peak infusion. R-wave-to-R-wave intervals adjacent to an ectopic beat were excluded from analysis. Using standard methods (23), we computed power spectral density after applying Hanning Window in the frequency range of 00.5 Hz and then computed area integrals of total power (<0.40 Hz), low-frequency power (LF; 0.040.15 Hz), and high-frequency power (HF; 0.150.40 Hz), and the LF-to-HF ratio, at each volume state. We ensured constant anesthesia to minimize the effects of anesthetic fluctuations on heart rate variability (HRV) (7).
Statistical analysis.
Continuous data are presented as means (SD) and were compared using the two-tailed t-test. The relationship between Valt and net volume infusion was assessed using linear regression. The
2 test was applied to contingency tables of TWA vs. baseline or volume overload. A probability level of 5% was considered statistically significant.
| RESULTS |
|---|
|
|
|---|
|
As shown in Fig. 2A, infusion provided 2,784 ml (SD 924) [102 ml/kg (SD 30)] over 315 min (SD 50) to peak. This increased PCWP significantly from 9.7 mmHg (SD 3.1) at baseline to 21.3 mmHg (SD 1.5) (P < 0.0001; Table 1, Fig. 2B). Pulmonary mean arterial pressure rose in parallel from 12.8 mmHg (SD 2.6) at baseline to 24.8 mmHg (SD 3.5) at peak infusion (P = 0.004; Table 1). After peak infusion, in dogs the diuresis end point was reached at 438 min (SD 44). At this stage, subjects had a residual net volume overload of 2,090 ml (SD 837) [76 ml/kg (SD 27); Fig. 2A].
|
Response of TWA to volume overload. TWA rose and fell with volume load, PCWP, and pulmonary arterial pressure, as shown in Table 2. Valt was higher at peak infusion than baseline [3.4 (SD 1.6) vs. 0.5 µV (SD 0.35); P = 0.011] and occurred at a lower onset heart rate [128 (SD 6) vs. 151 beats/min (SD 12); P = 0.008]. Postdiuresis, Valt fell [1.2 µV (SD 1.6)] and the TWA onset heart rate rose [142 beats/min (SD 19)] to levels similar to baseline (P > 0.3 for each). These results occurred in each dog, with Valt being 4- to 10-fold higher at peak infusion than baseline (Table 2).
|
|
|
|
| DISCUSSION |
|---|
|
|
|---|
Mechanisms of elevated TWA during volume overload. Volume overload in this study significantly elevated Valt in all dogs, by 4- to 10-fold, and significantly reduced its onset heart rate threshold (Table 2). TWA reflects exaggerated dispersion of repolarization that may reflect abnormal rate response of repolarization or conduction (13). Studies of mechanoelectric feedback have shown that acute volume load may shorten or lengthen ventricular action potential duration and slow conduction (21, 24, 26).
The elevation of TWA with volume load in this study may reflect ventricular stretch that exaggerates physiological (24) or pathological (19) nonuniformities in repolarization or conduction and may be proarrhythmic. Notably, pulmonary and systemic arterial pressures rose with volume infusion in our study. Although blood pressure fell in dogs with decompensated pulmonary congestion (an infusion end point), this is typical for patients with decompensated CHF who continue to show evidence for ventricular stretch via hemodynamic measurements and elevated levels of brain natriuretic peptide (12). Further studies using invasive hemodynamic measurements should define the effects of stretch on the rate response and spatial dispersion of conduction velocity and repolarization in the presence and absence of regional disease.
Elevations in TWA in our study could also reflect elevated sympathetic nervous activity. Studies in dogs and humans show that
-blockers suppress TWA, suggesting an etiological role for the sympathetic nervous system (3, 10, 11, 20), although, conversely, isoproterenol may not consistently elevate TWA during constant pacing (8). Compared with baseline in our study, dogs at peak infusion exhibited sinus tachycardia, reduced total spectral power of HRV that, over 10 min, is analogous to SD of normal R-wave-to-R-wave interval and primarily reflects VLF power and nonsignificant trends for an elevated ratio of LF to HF frequency power. These results are consistent with an elevated sympathetic and/or markedly decreased parasympathetic control of heart rate (23). Studies could further support the contribution-increased sympathetic activity by documenting vasoconstriction at peak infusion or by testing whether
-blockers attenuate the effects of volume load on TWA, or whether anticholinergic agents augment them.
Other possible explanations for elevated TWA during volume load include hypothermia, which we avoided in all but dog 1, and electrolyte alterations, although electrolytes were repleted during infusion. Theoretically, variations in anesthesia could alter TWA, although we maintained constant anesthesia. Moreover, TWA tracked infusion volume (from baseline to peak to diuresis) rather than anesthestic time (which varied for each dog; Fig. 2). Finally, TWA has previously been studied in anesthetized dogs (17, 18).
Volume load, mechanoelectric feedback, and arrhythmias. There has been considerable recent interest in the relevance of mechanoelectric feedback to clinical arrhythmias (12). In patients with CHF, indexes of ventricular stretch, such as elevated brain natriuretic peptide, are associated with elevated TWA (22) and improve the predictive value of left ventricular dysfunction for SCA (12). In normal rabbit ventricles, acute volume load causes extra systoles from afterdepolarizations and alters action potential duration and conduction velocity that may facilitate reentry (4, 21, 24, 26).
Our results allow us to define a relationship between volume load and TWA under "idealized" conditions, independent of interactions with regional disease. This is relevant to emerging studies of TWA in individuals without identifiable myocardial disease (5) and may help to dissect the effects of volume overload from intrinsic myocardial disease in the genesis of TWA.
In the presence of myocardial disease, TWA may become exaggerated with less dramatic volume infusion, since volume load interacts unfavorably with regional disease to facilitate arrhythmias (19). Clinically, larger Valt has recently been linked to more extensive myocardial damage (9), to regional scar (16), and to the effects of premature beats (15) and adverse outcome (9, 25). However, studies are still needed to understand how volume overload, which may be ameliorated, and structural disease, which is more likely to be permanent, each contribute to TWA and arrhythmic risk. Such studies will enable better risk stratification for SCA in the face of dynamic changes in clinical status.
Technical issues in TWA analysis. Measuring TWA in dogs is well established, from the earliest reports in the contemporary literature to recent studies of ischemia-related arrhythmias (17, 18). TWA in dogs is similar to that in humans, but has a higher onset heart rate (1, 17).
Limitations. The major limitation of this study is its small sample size. Although results were consistent for all animals, this study needs to be extended to larger populations. Second, our study hypothesis was confined to normal heart limits, and ongoing studies will examine this interaction in the presence of myocardial disease. Third, we did not assess stretch using echocardiography or other tools, due to technical limitations. Fourth, anesthesia could influence both TWA and HRV, yet we maintained this at a constant level, and this study could not have been realistically performed without anesthesia. Fifth, we did not study the direct link between volume load and arrhythmias, and future studies in animals with structural disease could use arrhythmia induction or ambulatory ECG monitoring to further define that relationship.
In conclusion, acute volume loading elevates Valt in the absence of structural heart disease. Although dramatic, this effect was marked only at peak infusion and thus may operate clinically only in patients with marked volume overload. This canine study strengthens the value of abnormal TWA in most CHF patients, suggesting that it reflects proarrhythmic substrates rather than loading conditions per se. Studies are now required to define the interaction of volume load with regional myocardial disease and arrhythmic tendency.
| GRANTS |
|---|
|
|
|---|
| DISCLOSURES |
|---|
|
|
|---|
| ACKNOWLEDGMENTS |
|---|
|
|
|---|
| FOOTNOTES |
|---|
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
| REFERENCES |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
S. M. Narayan, M. R. Franz, G. Lalani, J. Kim, and A. Sastry T-Wave Alternans, Restitution of Human Action Potential Duration, and Outcome J. Am. Coll. Cardiol., December 18, 2007; 50(25): 2385 - 2392. [Abstract] [Full Text] [PDF] |
||||
![]() |
J. E. Madias Effect of acute volume overload on the magnitude of T-wave alternans J Appl Physiol, August 1, 2007; 103(2): 728 - 728. [Full Text] [PDF] |
||||
![]() |
S. M. Narayan, D. D. Drinan, R. P. Lackey, and C. F. Edman Reply to Madias J Appl Physiol, August 1, 2007; 103(2): 729 - 729. [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Visit Other APS Journals Online |