Unexpected Interaction Between /3-Adrenergic Blockade and Heart Rate Variability Before

Transcription

1 976 Unexpeted Interation Between /3-drenergi Blokade and Heart Rate Variability Before and fter Myoardial Infartion Longitudinal Study in Dogs at High and Low Risk for Sudden Death Philip B. damson, MD, MS; Ming H. Huang, MD; Emilio Vanoli, MD; Robert D. Foreman, PhD; Peter J. Shwartz, MD; Stephen S. Hull, Jr, PhD Downloaded from by on September 27, 218 Bakground Heart rate (HR) variability is a marker of toni ardia autonomi ativity and ontributes in assessing risk for sudden death after myoardial infartion. Reent linial observations have indiated that attenuation of HR variability, whih ours after myoardial infartion, may be transient. his study addresses the issue of whether autonomi ontrol of heart rate reovers at different rates after myoardial infartion in subjets at high and low risk for ventriular fibrillation (VF). Methods and Results hirty dogs, 22 with myoardial infartion and 8 sham-prepared animals, ompleted the study. Changes and reovery in ardia autonomi ativity after myoardial infartion were examined by measuring HR variability before and at defined intervals during the first 3 days after infartion. Eah HR variability measurement was made before and after p-blokade in dogs at high (n=1) and low (n=12) risk for VF. rrhythmia risk was determined on the basis of development of VF during exerise and transient myoardial ishemia 3 days after infartion. No sham-prepared animals developed VF. Preinfartion measurements of HR variability were not different between the groups before 3-blokade, but HR variability inreased muh more in response to,l-blokade in animals destined to be resistant ompared with suseptible animals (289±26 to 369±35 mse, C ardia autonomi balane, measured by heart rate (HR) variability,12 is altered early after myoardial infartion in humans3-7 and in experimental animals.8 Quantifiation of this alteration an predit mortality after myoardial infartion, espeially the risk for arrhythmi death.3-7 Clinial observations show a trend toward signifiant reovery of ardia autonomi ativity within the first 2 months after myo- Reeived Otober 12, 1993; revision aepted pril 9, From the Department of Internal Mediine, Cardiovasular Diseases Setion, University of Oklahoma, HSC, Oklahoma City (P.B.., E.V.); the Department of Physiology and Biophysis, College of Mediine, University of Oklahoma, HSC, Oklahoma City (P.B.., M.H.H., E.V., R.D.F., P.J.S., S.S.H.); the Instituto di Clinia Media Generale e erapia Media, Universita di Milano (Italy) (E.V.); and the Dipartimento di Mediina, Universita di Pavia (Italy) (P,J.S.). Correspondene to Stephen S. Hull, Jr, PhD, University of Oklahoma College of Mediine, Department of Physiology and Biophysis, BMSB, Room 653, PO Box 2691, Oklahoma City, OK merian Heart ssoiation, In. 27.7%, versus 27±36 to 283-±34 milliseonds, 4.8%, respetively, P<.1). Immediately after infartion, HR variability was signifiantly attenuated in all dogs, but in the resistant dogs it reovered to pre-myoardial infartion levels within 1 days. fter the infartion, P-blokade did not inrease HR variability in either group of animals. Postoperative inreases in HR variability from P-blokade were preserved in the sham group. Suseptible animals were haraterized by a persistent attenuation of HR variability throughout the 3 days. Conlusions he depression in HR variability produed by myoardial infartion has a learly different temporal reovery pattern between low- and high-risk animals. fter myoardial infartion, P-adrenergi blokade does not alter HR variability, thus preserving its preditive value. Before myoardial infartion, however, P-blokade inreases HR variability only in the animals destined to be at low risk for lethal arrhythmias after the infartion. he reovery pattern of HR variability after myoardial infartion may ontribute to the early reognition of individuals at high risk for sudden death. (Cirulation. 1994;9: ) Key Words heart rate myoardial infartion. death, sudden nervous system reeptors, adrenergi, beta ardial infartion.9 By 1 year, autonomi reovery does our,1"l but measurements are still attenuated ompared with noninfarted ontrol subjets.12 Previous studies that examined autonomi reovery after myoardial infartion reported large varianes12 in the measures of HR variability, indiating that the groups omprised subjets having a wide range of effetive vagal ontrol of heart rate, possibly at different levels of reovery or destined never to reover. Still, these studies have ontributed to generation of the onept that vagal ontrol of heart rate ultimately reovers after myoardial infartion. Reent observations indiate that depressed HR variability identifies high-risk patients even 2 years after myoardial infartion,'3 whih is potentially in onflit with the onept of time-dependent vagal reovery. herefore, the temporal reovery patterns of HR variability after myoardial infartion remain ontroversial and linially relevant. nother unanswered question of linial importane is how p-adrenergi blokade affets HR variability after myoardial infartion. Kleiger and olleagues4

2 damson et al HR Variability and P-Blokade Before and fter MI 977 Downloaded from by on September 27, 218 showed that mortality was redued by use of fl-blokers after myoardial infartion, but low HR variability (<5 milliseonds) in the presene of adrenergi blokade was still assoiated with twofold inrease in mortality ompared with patients on (-blokers, who had higher HR variability (>5 milliseonds). Cook et a114 found that HR variability showed a moderate inrease during 13-adrenergi blokade in young healthy subjets, but the relevane of these results to older patients after myoardial infartion is unknown. reent study in patients after myoardial infartion showed an inrease in HR variability after 3-adrenergi blokade, but the signifiane is obsure, sine the sample size of the population used in the study was not reported.15 Overall, linial studies are limited by their inability to provide information on a most ritial issue: namely, whether or not a hange in any given autonomi marker is assoiated with a hange in risk status. he present study was designed to test the hypothesis that reovery of ardia autonomi ativity after myoardial infartion has prognosti signifiane. Methods Overview hirty dogs, 22 with a small anterior wall myoardial infartion and 8 prepared with the surgial proedure but no infartion (sham), ompleted the study. Eah animal underwent HR variability analyses before and after,3-blokade several days before the myoardial infartion and periodially during the 3 days immediately after the infartion. On day 3, eah dog underwent an exerise test oupled with transient myoardial ishemia, and the outome of the test determined risk status for sudden death: animals experiening ventriular fibrillation during the transient myoardial ishemia were labeled suseptible (n=1), and those without ventriular fibrillation were labeled resistant (n=12). Values of HR variability were then orrelated with risk status to assess the preditive value of the test. Surgial Preparation he surgial proedure used to reate the myoardial infartion has been desribed in detail.16 Briefly, during anesthesia, the heart was exposed via a left fifth interostal spae thoraotomy, an arterial atheter was implanted in the aorti arh, and a Doppler flow probe (2 MHz) and distally plaed pneumati oluder were implanted around the irumflex branh of the left oronary artery. he left anterior desending oronary artery was ligated immediately proximal to the first major diagonal branh after a 3-minute subtotal stenosis. ll instrumentation leads were exteriorized on the dorsal nek area. he eight sham-prepared animals had the same instrumentation as outlined above, inluding dissetion of the left anterior desending oronary artery, but the artery was not ligated. Every dog reeived immediate postoperative short-term analgesis (pentazoine latate, 1 mg/kg IM) followed by a longer-ating analgesi (nalbuphine HCl,.5 mg/kg IM). Guidelines (National Institutes of Health, merian Physiologial Soiety, and merian Heart ssoiation) pertaining to the appropriate are and use of animals were stritly followed. limation to the Reording Environment Our previous experiene indiates that HR variability measurements are signifiantly altered after the dogs are alimated to the reording environment and personnel.8 herefore, eah animal underwent measurement protools several days before the atual reording period began. dditionally, if the HR was deemed to be too high ompared with baseline measurements or if the behavior was onsidered to be inappropriate for basal autonomi reordings, measurements were postponed and redone after further alimation. HR variability was not measured if any animal had signs of overt linial illness. Exerise and Ishemia est Dogs were evaluated for the risk of lethal arrhythmias by use of submaximal exerise stress tests on a motor-driven treadmill aording to a protool previously desribed.16 Speifially, dogs ran for approximately 12 to 15 minutes while the workload progressively inreased every 3 minutes (3 mph at % grade, 4 mph at % grade, 4 mph at 4% grade, 4 mph at 8% grade, and 4 mph at 12% grade). he irumflex oronary artery was oluded when heart rate reahed a target level of 21 to 22 beats per minute, whih represents approximately 7% of the maximal heart rate attainable in a dog. Our experiene with this protool showed that vagal ativity at this level of exerise is progressively but never ompletely withdrawn beause it an be further modulated by vagolyti interventions.17 Continuous measurement of transthorai ECG, beat-to-beat HR (from the ECG signal), phasi and mean arterial blood pressures, and phasi blood flow through the irumflex oronary artery were transribed simultaneously on a diret-writing osillograph (model 7D, Grass) while the dog initially stood on the treadmill and throughout the exerise stress test. When the target heart rate was reahed, the pneumati oluder surrounding the irumflex oronary artery was inflated for 2 minutes. he animal ontinued to exerise during the first minute, and exerise was stopped during the seond minute. Ventriular fibrillation ourred in 1 of 22 dogs (45%) (suseptible), and those animals were defibrillated immediately. he remaining 12 animals (55%) had no sustained ventriular arrhythmias during exerise and myoardial ishemia (resistant). he ourrene of ventriular fibrillation in this model is known to be highly reproduible.18 HR Variability HR variability was measured in eah animal, without and with 3-blokade (atenolol 1 mg/kg IV), before the myoardial infartion and on days 5, 9, 13, 17, 21, 25, and 29 after the infartion. he pre-myoardial infartion reording was made after at least 7 days of laboratory alimation. For all reordings, 25-minute ECG samples were obtained while the dogs were resting quietly, but not sleeping, on a padded and warmed table. he reording room was loked and kept quiet during data aquisition. Detailed methodology has been desribed.8 Briefly, a transthorai modified lead I ECG was amplified (Grass 7P4 H), filtered at a low frequeny (5 Hz) and high frequeny (5 Hz), and digitized at 4 Hz. ll digitally enoded files were analyzed with a ommerially available program (Corazonix Corp). Premature ventriular omplexes and their adjaent RR intervals were rejeted by the software, in addition to eletrial noise or other aberrant ECG signals. HR variability measurements were made only on visually onfirmed sinus rhythm reordings. he RR interval data from 25 minutes of ontinuous ECG reordings were plotted as a frequeny histogram distribution using 4-milliseond bins from whih the mean RR interval and its SD were derived. Sine the SD of the mean RR interval is influened by the prevailing HR, another measurement, the oeffiient of variane (quotient of the SD of the mean RR interval and the mean RR interval x 1), was alulated to normalize HR variability to HR. Infart Size Estimation When the experimental protools were ompleted, the animals were killed with an overdose of pentobarbital (1 mg/kg IV), and the hearts were immediately removed. Infart size was measured in 8 of the 1 suseptible and 11 of the 12

3 978 Cirulation Vol 9, No 2 ugust a. I ' 6 Q " e O'_H 4U..E to 6 I 4 6 S i v -, - 4.-C, O E _ 3- a Ut 1- C ~ ' e F _ I. B,, 4- j' E o E..., 2- C 3o a,- 1. C a v_ 1_- _ v B II,. Downloaded from by on September 27, 218 C. Q) i lj -5 t 47 I, _C 1. L o Resistant Suseptible Sham Days Relative to Myoardial Infartion FIG 1. Graphs showing heart rate (beats per minute, ), SD of the mean RR interval (milliseonds, B), and the oeffiient of variane (SD/mean RR intervalx 1, C) from 1 dogs suseptible to ventriular fibrillation (e), 12 dogs resistant (o) to ventriular fibrillation, and 8 sham-prepared animals (). Measurements were made 3 days before and periodially after an anterior wall myoardial infartion or surgery (sham). Signifiant differenes (P<.1) when suseptible dogs are ompared with both resistant and sham animals. tsignfiant differene (P<.5) when suseptible and resistant animals are ompared with the sham group. Values are mean±sem. resistant dogs with the tetrazolium enzymati staining tehnique.19 he left ventrile was isolated and ut through a line perpendiular to the long axis from the apex in 1-m slies and inubated at 38 C in a buffered 5% tetrazolium blue solution for 2 minutes. he infart zone was identified by only two investigators (P.B.., E.V.) to avoid exessive subjetive variation. Normal and infarted tissues were then arefully disseted, and the infart size was expressed as a perentage of total left ventrile and septal wall mass. Statistis Statistial signifiane of the data was determined by NOV suitable for repeated measures, and mean differenes were assessed with ukey's test. Data are presented in the test as the mean+sem unless otherwise noted. n a level of P<.5 was onsidered statistially signifiant, and hanges are signifiant unless otherwise noted. U a, CX a a, Q U3-5 9 O Before atenolol fter atenolol Days Relative to Myoardial Infortion FIG 2. Graphs showing heart rate (beats per minute, ), SD of the mean RR interval (milliseonds, B), and the oeffiient of variane (SD/mean RR intervalx 1, C) before (o) and after () 1 mg/kg intravenous atenolol in 12 onsious dogs shown to be resistant to ventriular fibrillation. Measurements were made 3 days before and periodially after an anterior wall myoardial infartion. Signifiant differene when the baseline value is ompared with the measurement after atenolol (P<.1). Values are mean±sem. Results Heart Rate HRs in all groups of dogs were equivalent before surgery (Fig 1). fter myoardial infartion, HRs were persistently higher in suseptible animals throughout the reording period (76+5 beats per minute preinfartion versus 89±5 beats per minute day 29 after infartion, P<.5). Compared with preinfartion onditions, resting HRs in resistant dogs were higher immediately after the infartion but were not signifiantly different by day 17 (72+4 beats per minute preinfartion, 82±4 beats per minute day 5 after infartion, 76+5 beats per minute day 17 after infartion). Sham-prepared dogs had a transient elevation in heart rate after surgery but returned to presurgial levels within 5 days (Fig 1). tenolol signifiantly redued HR before surgery in all dogs as shown in Figs 2 (resistant), 3 (suseptible), and 4 (sham). Early after myoardial infartion, adrenergi blokade had no influene on heart rate (Figs 2, 3, and 4), but its effet beame evident in all dogs from day 5 after surgery. SD of the Mean RR Interval SD of the mean RR interval measurements before the infartion were not different between groups (289±26 milliseonds, resistant; 27±36 milliseonds, suseptible; 285+±24 milliseonds, sham; Fig 1B). However, before surgery an important differene already

4 damson et al HR Variability and P-Blokade Before and fter MI 979 E I) a U) X s \_?K 4.. M. E Ua.) t U 6 lw 4 E 3,- ~_ B. o 4.,- U) 1 O E X -@---Q B ot 1- C Co 1 v 7 f--v ' 2- C) C 1 U) O 11..,, en > i X Downloaded from by on September 27, 218 U).L C :U O Before atenolol fter atenolol 4i. -5 s 1 1S Days Relotive to Myoardial Infartion FIG 3. Graphs showing heart rate (beats per minute, ), SD of the mean RR interval (milliseonds, B), and the oeffiient of variane (SD/mean RR interval x 1, C) before (o) and after () 1 mg/kg intravenous atenolol in 1 onsious dogs shown to be suseptible to ventriular fibrillation. Measurements were made 3 days before and periodially after an anterior wall myoardial infartion. Signffiant differene when the baseline value is ompared with the measurement after atenolol (P<.1). Values are mean+sem. existed that was unmasked by,-blokade. tenolol inreased the SD only in the sham and resistant dogs and not in the suseptible group, as shown in Fig 5 (27.7%, resistant; 29.8%, sham; 4.8%, suseptible; P<.1 omparing resistant and sham with suseptible). On day 5 after myoardial infartion, the SD of the RR intervals was markedly attenuated in both groups. he derease was more evident among suseptible dogs (from 27±36 to 113±2 milliseonds) than among resistant dogs (from 289±26 to 178±22 milliseonds), resulting in an HR variability value that was learly lower among suseptible dogs even though it fell just short of statistial signifiane (P=.6). hese measurements were redued in sham-prepared animals as a result of surgial intervention, although to a muh smaller extent than in infarted animals, and quikly reovered to ontrol levels (Fig 1B). t day 9 after myoardial infartion and thereafter, suseptible animals had SD of the RR intervals signifiantly lower than the resistant dogs (133±22 milliseonds suseptible versus 24±23 milliseonds resistant, day 9; P<.3) (Fig 1B). Resistant and sham animals had equivalent levels of SD for eah day after day 9 (Fig 1B). C 4' ': 3 : ~ 2. 1 CD! (N /- _, $ s 3 O Before atenolol fter atenolol Days Relotive to Myoardial Infartion FIG 4. Graphs showing heart rate (beats per minute, ), SD of the mean RR interval (milliseonds, B), and the oeffiient of variane (SDIRR intervalxlooo, C) before (o) and after (X) 1 mglkg intravenous atenolol in 8 onsious dogs (sham) who previously underwent the idential surgery as the resistant and suseptible dogs (see text) exept that no myoardial infartion was indued. Measurements were made 3 days before and periodially after surgery. Signfflant differene when the baseline value is ompared with the measurement after atenolol (P<.1). Values are mean+sem. he effets of atenolol on the SD of the mean RR interval after surgery are shown in Figs 2B, 3B, and 4B. Only on day 5 after myoardial infartion, but not thereafter, did,3-blokade slightly but signifiantly inrease the SD of the RR intervals in the resistant animals (from 178±22 to 215+±23 milliseonds; P=.2) (Fig 2B). 1-Blokade did not signifiantly alter the SD of the RR intervals in the suseptible animals before or at any point after myoardial infartion (Fig 3B). In ontrast to the infarted animals, sham-prepared dogs had signifiant hanges in SD of the mean RR interval after 13-blokade before and on eah day after surgery (Fig 4B). Coeffliient of Variane o ensure that the hanges in HR variability after myoardial infartion and 1-blokade were not simply due to hanges in HR, the oeffiient of variane was alulated to normalize HR variability to the prevailing HR (SD of the mean RR interval/mean RR intervalx 1). Baseline values for the oeffiient of variane from all dogs before and after surgery are shown in

5 Downloaded from by on September 27, Cirulation Vol 9, No 2 ugust 1994 CD CD 2 CD 1 Resistant Suseptible n=12 n=1 FIG 5. Bar graph showing perent hange in the SD of the mean RR interval in response to 1 mg/kg intravenous atenolol before myoardial infartion in 12 dogs later shown to be resistant (left) to ventriular fibrillation and 1 dogs who were later found to be suseptible (right) to ventriular fibrillation. Data shown are group means. P<.1 when resistant and suseptible values are ompared. Fig 1C. Before myoardial infartion, the groups were not different. Measurements for resistant and suseptible animals dereased early after myoardial infartion but were different between the two groups of dogs by 9 days (33±21 versus 194±29, respetively, P<.1). his signifiant differene was maintained for eah reording thereafter. Coeffiient of variane values in sham animals were not signifiantly hanged by the surgial intervention (3±27 presurgery versus 275±5 at day 5 postsurgery, P=.42). Coeffiient of variane measurements inreased signifiantly in response to fi-adrenergi blokade before surgery in the resistant dogs (from 331±21 to 368±24; P<.5) (Fig 2C) and sham dogs (from 3±27 to 349±49; P<.5) (Fig 4C). In ontrast to these groups, the preinfartion oeffiient of variane from the suseptible animals had a trend toward derease after fl-blokade, but the hange did not reah statistial signifiane (from 327±33 to 32±24, P=.2) (Fig 3C). fter surgial intervention, the oeffiient of variane inreased signifiantly after fl-blokade only in the sham animals (Figs 2C, 3C, and 4C). Infart Size Estimation Hearts from 8 of the 1 suseptible and 11 of the 12 resistant animals were available for infart size estimation. he myoardial infartion averaged 13.4±.9% (range, 9.5% to 16.4%) of the left ventriular mass in the suseptible animals and 13.6±1.6% (range, 9.% to 27.2%) in the resistant dogs (P=NS). Disussion he present study provides a longitudinal analysis on the evolution of HR variability after a myoardial infartion with two features not readily available in linial investigations. One is the omparison of HR variability after myoardial infartion with values before the infartion, whih allows a orret assessment of the extent of reovery; the other is the omparison between animals at low and at high risk for sudden ardia death, whih allows a refined assessment of the prognosti value of HR variability. hree main findings emerged, and they all have signifiant linial impliations. he first is that myoardial infartion produes a depression in HR variability that shows a strikingly different temporal evolution between low- and high-risk animals. mong resistant dogs, the depression is quite transient, sine HR variability returns to pre-myoardial infartion values within 1 days. In ontrast, HR variability in suseptible dogs remains signifiantly depressed at 1 month after the infartion. he seond finding is that after myoardial infartion,,8-adrenergi blokade, highly effetive in preventing ventriular fibrillation in this animal model,2,21 does not alter HR variability. Speifially, the expeted inrease in HR variability after 8-adrenergi blokade is absent in all dogs after myoardial infartion. hus, treatment with f-blokers would not be a onfounding fator in the analysis of HR variability. he third finding is the unexpeted and intriguing observation that before myoardial infartion, 13-blokade inreases HR variability only in the animals destined to be at low risk for lethal arrhythmias after the infartion. his raises the possibility that individuals at high risk for arrhythmi death after a myoardial infartion an be identified very early, even before the infartion. HR Variability fter Myoardial Infartion Previous linial studies found that HR variability is diminished within 2 weeks after myoardial infartion3-7,9,1,12 and signifiantly reovers within 6 months to 1 year.9'1"2 In ontrast, ardia sympatheti ativity inreases after aute myoardial infartion and dereases over time.9 Parallel to these observations is the derease in baroreflex sensitivity that takes plae shortly after myoardial infartion and that reovers between 1 and 3 months.'1 he design of the present study has provided information about several issues that are not easily obtained in linial studies. fter myoardial infartion, both low- and high-risk animals underwent repeated assessments of HR variability. he objetive and nonontroversial ourrene of ventriular fibrillation was used to identify the high- and low-risk animals. Reovery patterns after the infartion and the influene of fl-blokade on HR variability were examined separately by internal ontrol analysis. Most importantly, 3 days after the myoardial infartion, all dogs were exposed to the same situation of risk, namely, a brief episode of aute myoardial ishemia in the presene of a healed myoardial infartion while sympatheti ativity was elevated physiologially by exerise. his step provides a fundamental differene from linial studies. Indeed, depending on the extent of their oronary artery disease, many patients may never experiene a new ishemi event. hey therefore esape the true assessment of their arrhythmi risk. he partiular onditions of the exerise and ishemia test apply speifially to arrhythmias, whih are mehanistially related to myoardial ishemia oupled with elevated sympatheti ativity. he appliability of findings generated from this anine model of autonomially mediated ventriular arrhythmias during myoardial ishemia to linial sudden ardia death has reently been reviewed.22 he present data indiated that HR variability reovered in the resistant animals within 1 days of the infartion and at that time was similar to that of the noninfarted sham-operated animals. Depression in HR

6 damson et al HR Variability and P-Blokade Before and fter MI 981 Downloaded from by on September 27, 218 variability was not due to the stress of the thoraotomy, sine the sham-operated animals had insignifiant hanges in HR variability after the surgery. dditionally, the reovery patterns observed were not due to differenes in the infart size, whih, onsistent with previously published data from our laboratory,82 were not different between the resistant and suseptible animals. hus, the different reovery harateristis in ardia autonomi ativity between the two groups of animals did not depend on differenes in the extent of left ventriular damage from the oronary artery ligation. HR Variability and p3-bloker herapy he widespread use of,b-adrenergi bloking therapy in post-myoardial infartion patients has generated onerns about the validity of risk stratifiation by HR variability; ie, therapy may modify the risk marker. Kleiger et a14 performed a subanalysis in patients treated with,3-blokers and reported that 24-hour HR variability of <5 milliseonds was still assoiated with inreased mortality ompared with patients treated with /3-blokers but with HR variability of >5 milliseonds. However, their study was not designed to determine whether /-blokers influene HR variability. Cook et al'4 addressed this issue in normal young healthy adults and showed that HR variability inreases after /-adrenergi blokade with atenolol. hese results are in agreement with our findings in the sham-prepared dogs and in the resistant animals before the myoardial infartion. However, the 3-bloker-mediated inrease in HR variability was no longer present after myoardial infartion. dditionally, after myoardial infartion, HR variability measured before and after /3-blokade aurately predited risk status in this population. hus, /3-adrenergi blokade does not impair the apability of HR variability to disriminate high- and low-risk subjets after myoardial infartion. utonomi tivity Before Myoardial Infartion We have previously reported that baseline measurements of HR variability before myoardial infartion were not able to disriminate between resistant and suseptible dogs.8 his is at variane with the fat that baroreflex sensitivity is already lower before the infartion in the suseptible ompared with the resistant animals.16 Both these findings were supported by a study in whih "toni" and "reflex" ardia vagal efferent ativity (assessed by the single-fiber reording tehnique) was measured in anesthetized ats before oronary artery olusion.23 he animals destined to develop ventriular fibrillation during myoardial ishemia showed a marked impairment in the apability to inrease vagal efferent ativity in response to a blood pressure rise, ie, the baroreeptor reflex, but were not different in terms of "toni" ativity from the animals destined to survive. he present study adds a signifiant piee to this puzzle by showing that, before myoardial infartion, the response of HR variability to P-blokade is already different between the dogs destined to beome resistant and suseptible after the infartion. aken together, these data strongly support the onept that there is great individual diversity in the autonomi make-up of normal individuals and that this diversity already ontains important information pertaining to arrhythmi risk in ase of future ishemi events.16,22 However, the mehanism underlying this differene remains a matter of speulation. he previously reported inreases in HR variability after /3-blokade have been interpreted as being largely seondary to entral modulation of ardia reflexes,14,24 partly on the basis of reports that atenolol dereases toni and reflex renal sympatheti nerve ativity.25'26 his explanation, however, has to ontend with the fat that atenolol does not ross the blood-brain barrier.27 he present data suggest that the proposed parasympathomimeti effet of atenolol either was lost or was muh less effetive in suseptible animals, thus produing the different responses to /3-blokade that we had observed before myoardial infartion. his might reflet an individual predisposition of either geneti or environmental origin. tentative explanation for the differential effet of atenolol on HR variability before the infartion is possible. /3-Blokade redues the sensitivity of ardia sympatheti afferents28 and therefore an be expeted to dampen transient inreases in sympatheti afferent ativity. s a onsequene, this reveals differenes in ardia vagal efferent ativity23'29 resulting in higher HR variability in the resistant animals. Redued baroreflex sensitivity among suseptible dogs before myoardial infartion'6 has already indiated that these animals are less able to augment ardia vagal ativity. Similarly, one might expet that HR variability in dogs destined to beome suseptible is also less likely to respond to /3-blokade. he ourrene of a myoardial infartion might further depress vagal efferent ativity to at on the premorbid differenes in autonomi ativity found between high- and low-risk individuals.'6 Vagal efferent depression probably ours through a sympathovagal reflex23'29 resulting from the myoardial infartion. Inreased sympatheti afferent traffi seondary to the mehanial onsequenes of the neroti sar22 would produe inhibitory effets at the level of the entral nervous system. he present results learly indiate that further studies on the sympathovagal interation in the entral nervous system are needed to better understand the relation between the autonomi nervous system and the propensity toward sudden ardia death. Independent of the orret underlying mehanism, the finding that /3-blokade produes different responses in HR variability aording to the future risk of lethal arrhythmias during an ishemi episode opens exiting perspetives for early risk stratifiation among still healthy individuals. Highly relevant in this respet is the reent finding that exerise training improves all autonomi markers of vagal ativity and strikingly redues the risk for ventriular fibrillation not only in infarted animals3 but also in animals with normal hearts.31 hese data suggest that the ombination of baroreflex sensitivity assessment and the response of HR variability to P-blokade may provide early identifiation of individuals potentially at risk for sudden death. hese individuals, one identified, may be proteted by a nonpharmaologial intervention suh as exerise training.

7 Downloaded from by on September 27, Cirulation Vol 9, No 2 ugust 1994 knowledgments his study was supported by National Institutes of Health grant HL , the Presbyterian Health Foundation, and the merian Heart ssoiation. Referenes 1. Ekberg DL. Human sinus arrhythmia as an index of vagal ardia outflow. J ppl Physiol. 1983;54: Katona PG, Jih F. Respiratory sinus arrhythmia: noninvasive measure of parasympatheti ardia ontrol. Jppl Physiol. 1975; 39: Malik M, Farrell G, Camm J. Ciradian rhythm of heart rate variability after aute myoardial infartion and its influene on the prognosti value of heart rate variability. m J Cardiol. 199; 66: Kleiger RE, Miller JP, Bigger J Jr, Moss J, and the Multienter Post-Infartion Researh Group. Dereased heart rate variability and its assoiation with inreased mortality after aute myoardial infartion. m J Cardiol. 1987;59: Casolo GC, Stroder P, Signorini C, Calzolari F, Zuhini M, Balli E, Sulla, Lazzerini S. Heart rate variability during the aute phase of myoardial infartion. Cirulation. 1992;85: Farrell G, Bashir Y, Cripps, Malik M, Polonieki J, Bennett ED, Ward DE, Camm J. Risk stratifiation for arrhythmi events in postinfartion patients based on heart rate variability, ambulatory eletroardiographi variables and the signal-averaged eletroardiogram. Jm Coll Cardiol. 1991;18: Pedretti R, Etro MD, Laporta, Braga SS, Caru B. Predition of late arrhythmi events after aute myoardial infartion from ombined use of noninvasive prognosti variables and induibility of sustained monomorphi ventriular tahyardia. m J Cardiol. 1993;71: Hull SS Jr, Evans R, Vanoli E, damson PB, Stramba-Badiale M, lbert DE, Foreman RD, Shwartz PJ. Heart rate variability before and after myoardial infartion in onsious dogs at high and low risk of sudden death. Jm Coll Cardiol. 199;16: Lombardi F, Sandrone G, Pernpruner S, Sala R, Garimoldi M, Ceruti S, Baselli G, Pagani M, Malliani. Heart rate variability as an index of sympathovagal interation after aute myoardial infartion. m J Cardiol. 1987;6: Mazzuero G, Lanfranhi P, Colombo R, Giannuzzi P, Giordano, and the EMI Study Group. Long-term adaptation of 24-h heart rate variability after myoardial infartion. Chest. 1992;11: 34S-38S. 11. Shwartz PJ, Zaza, Pala M, Loati E, Beria G, Zanhetti. Baroreflex sensitivity and its evolution during the first year after myoardial infartion. Jm Coll Cardiol. 1988;12: Bigger J, Fleiss JL, Rolnitzky LM, Steinman RC, Shneider WJ. ime ourse of heart period variability after myoardial infartion. Jm Coll Cardiol. 1991;18: Bigger J, Fleiss JL, Rolnitzky LM, Steinman RC. Frequeny domain measures of heart period variability and death in hroni oronary heart disease (1 year after infartion). Cirulation. 1992; 86(suppl I):I-66. bstrat. 14. Cook JR, Bigger J, Kleiger RE, Fleiss JL, Steinman RC, Rolnitzky LM. Effet of atenolol and diltiazem on heart period variability in normal persons. Jm Coll Cardiol. 1991;17: Molgaard H, Mikley H, Pless P, Bjerregaard P, Moller M. Effets of metoprolol on heart rate variability in survivors of aute myoardial infartion. m J Cardiol. 1993;71: Shwartz PJ, Vanoli E, Stramba-Badiale M, De Ferrari GM, Billman GE, Foreman RD. utonomi mehanisms and sudden death: new insights from analysis of baroreeptor reflexes in onsious dogs with and without a myoardial infartion. Cirulation. 1988;78: De Ferrari GM, Vanoli E, Stramba-Badiale M, Hull SS Jr, Foreman RD, Shwartz PJ. Vagal reflexes and survival during aute myoardial ishemia in onsious dogs with a healed myoardial infartion. m J Physiol. 1991;261:H63-H Vanoli E, De Ferrari GM, Stramba-Badiale M, Hull SS Jr, Foreman RD, Shwartz PJ. Vagal stimulation and prevention of sudden death in onsious dogs with a healed myoardial infartion. Cir Res. 1991;68: Fishbein MC, Meerbaum S, Rit J, et al. Early phase aute myoardial infart size quantifiation: validation of the triphenyl tetrazolium hloride tissue enzyme staining tehnique. m Heart J. 1981;11: De Ferrari GM, Salvati P, Grossoni M, Ukmar G, Vaga L, Patrono C, Shwartz PJ. Pharmaologi modulation of the autonomi nervous system in the prevention of sudden ardia death: a study with propranolol, methaholine and oxotremorine in onsious dogs with a healed myoardial infartion. Jm Coll Cardiol. 1993; 21: Vanoli E, Hull SS Jr, Foreman RD, Ferrari, Shwartz PJ. lpha, adrenergi blokade and sudden ardia death. J Cardiovas Eletrophysiol. 1994;5: Shwartz PJ, La Rovere M, Vanoli E. utonomi nervous system and sudden ardia death: experimental basis and linial observations for post-myoardial infartion risk stratifiation. Cirulation. 1992;85(suppl I):I-77-I Cerati D, Shwartz PJ. Single ardia vagal fiber ativity, aute myoardial ishemia, and risk for sudden death. Cir Res. 1991;69: Coker R, Koziell, Oliver C, Smith SE. Does the sympatheti nervous system influene sinus arrhythmia in man? Evidene from ombined autonomi blokade. J Physiol. 1984;356: Sott E. he effets of atenolol on spontaneous and reflex ativity of the sympatheti nerves in the anaesthetized at. BrJ Pharmaol. 1981;73: Sott E. he effet of atenolol on the spontaneous and reflex ativity of the sympatheti nerves in the at: influene of ardiopulmonary reeptors. Br J Pharmaol. 1983;78: Cruikshank JM, Neil-Dwyer G, Cameron MM, Mnish J. Betaadrenoeptor-bloking agents and the blood-brain barrier. Clin Si. 198;59:453s-455s. 28. Lombardi F, Casalone C, Malfatto G, Gnehi-Rusone, Casati R, Malliani. Effets of propranolol on the impulse ativity of ardiovasular sympatheti afferent fibers. Hypertension. 1986;8: Shwartz PJ, Pagani M, Lombardi F, Malliani, Brown M. ardioardia sympathovagal reflex in the at. Cir Res. 1973;32: Billman GE, Shwartz PJ, Stone HL. he effets of daily exerise on suseptibility to sudden ardia death. Cirulation. 1984;69: Hull SS Jr, Vanoli E, damson PB, Verrier RL, Foreman RD, Shwartz PJ. Exerise training onfers antiipatory protetion from sudden death during aute myoardial ishemia. Cirulation. 1994;89: