Exercise assessment of sinoatrial node function following the Mustard operation

Exercise following

assessment of sinoatrial the Mustard operation

node function

To screen for sinoatrial node dysfunction following the Mustard procedure for transposition of the great arteries, we studied the chronotropic response to graded maximal treadmill exercise in 29 patients at mean 6.7 years after operation. Although 93% of patients had normal resting heart rate (HR), 83% demonstrated significant depression of maximum HR and/or recovery HR after termination of exercise. These findings were similarly present among a subset of 13 patients with normal exercise tolerance. Resting and exercise-induced HR in 10 patients receiving chronic digoxin therapy were no different than in the 19 patients without medication. Sixteen patients with abnormal chronotropic responses to exercise had intracardiac electrophysiologic evaluation which confirmed sinoatrial node dysfunction in nine. Abnormal HR responses did not correlate with clinical symptoms, cardiac arrhythmias, or postoperative hemodynamics. Maximal exercise testing may be a sensitive noninvasive method to identify sinoatrial node dysfunction in postoperative children. (AM HEART J 103:351, 1982.)

Peter S. Hesslein, M.D., Howard P. Gutgesell, Dan G. McNamara, M.D. Houston, Texas

M.D., Paul C. Gillette,

Cardiac rhythm disturbances are common following the Mustard operation for transposition of the great arteries.‘-* The most frequent arrhythmias are due to the sick sinus syndrome, characterized by tachyand/or bradyarrhythmias. Because sinoatrial node dysfunction may be progressive, and since such patients are susceptible to symptoms or sudden death as a result of their arrhythmia, a sensitive method is needed to identify those children with sinus node disease. It was the purpose of this study to evaluate the treadmill exercise as a provocative test of sinoatrial node function in patients after Mustard repair. METHODS Patients studied. We analyzed the results of treadmill exercise performed by 33 patients who had previously undergone Mustard operation. These patients were not randomly selected, but comprised a group for which there

wasoften a clinical suspicionof sinusnode (SN) disease.A total of 40 tests were performed by these patients between

From the Lillie Frank Abercrombie Section of Cardiology, Department Pediatrics, Baylor College of Medicine; and Texas Children’s Hospital.

of

Supported in part by three grants from the National Institutes of HealthNational Research Service Award HL-07190, General Clinical Research Branch Grant RR-00188, and Research Career Development Award HL-09571 (Dr. Gillette). Received Reprint Children’s

for publication requests: Howard Hospital, 6621

0002~8703/82/030351

Sept.

17, 1981;

accepted

P. Gutgesell, M.D., Fannin St., Houston,

+ 07$00.70/O

@ 1982

Oct. 26, 1981.

Pediatric Cardiology, TX 77030.

The

C. V. Mosby

Texas

Co.

M.D., and

January 1,1974, and December31,198O.All 40 tests in the 33 patients were evaluated for heart rhythm response to

exercise; however, only 36’ tests (29 patients) could be included for analysis of heart rate response. This was becausethree tests were excluded for inadequate patient effort and a fourth was eliminated since the patient was receiving propranolol and had atrial flutter throughout the test. Ten of these 29 patients whoseexerciserates were analyzed were taking oral digoxin at the time of study. The age range at the time of stresstesting was 5 to 25 years, with median age 11 years, and the postoperative interval at the time of testing was 0.8 to 13.3 years (mean = 6.7 years). Exercise test procedure. In our laboratory each test followed the Bruce protocol,6 asmodified by Cumming et al.,l according to which treadmill speed and grade are increased every 3 minutes through a maximum of ten stages.A physician was present for every test, and each patient wasencouragedto continue until exhausted. The ECG was continuously monitored and recorded at lminute intervals. Blood pressurewasintermittently measured using a Doppler transducer and an appropriately sized cuff before, during, and for at least 5 minutes after exercise. Data recorded included duration of exercise, cardiac rhythm, and heart rates at rest, at maximum exercise,and at 2 minutes and 5 minutes into the recovery period. Evaluation of exercise data. The exercise values obtained from each patient were analyzed and individually comparedto age- and sex-related values for normal children in a clinic population, previously published by Cumming et al.’ The validity of using published normal values was confirmed by statistical concordancebetween 351

March,

352

Hesslein et al.

Amerksn

Table I. Reproducibility and second tests

22-20--

(‘CitegOt-?;

of results: First

Mean test

Heart

1982 Journal

values for first Second

test

IS-I&-

.

: BSANORMALIZED

14--

(YRS)

Exercise duration (min) Resting heart rate (bpm) Maximum heart rate (bpm) 2 min recovery heart rate (bpm) 5 min recovery heart rate (bpm )

11.7 Ik 2.0 71.7 F 9.5 174.3

i 22.8

102.3 zk 18.1 90.3 i

n = seven pairs of tests; values are for bpm = beats per minute; min = minutes.

15.3

mean

11.9

k 2.1

13.4 ri- 7.5 182.3

i

10.5

110.4 + 27.0 89.4 + 14.2

+ 1 standard

deviation;

RESULTS Fig.

1. Correlation of body size and chronologic age in

patients at the time of exercise evaluation. BSA-normalized age = the chronologic ageat which body surface area equaled the 50th percentile for normal children of the samesex. When each patient’s BSA-normalized age was plotted against true chronologic age, the values clustered around the identity line. The correlation coefficient was r = 0.92, p < 0.05 by Student’s t test of independent means.

those results and data derived from a group of normal children undergoing the sametest in our clinic. To avoid emphasisof those patients who were exercisetested more than once, their test results were averaged to yield one value in each category for each patient. Significant deviation from normal wasdefined asa variation of morethan 2 SD from the published mean value for age and sex. We also compared each patient’s treadmill findings with the individual’s own clinical and catheterization data to examine the relationships between treadmill response and intracardiac electrophysiology, hemodynamic status, medications, and evidence of cardiac arrhythmias. Statisticsl andysis. To further validate the experimental data and their comparability with the normal values published by Cumming et al.,7we subjectedour results to two statistical analyses.First, the reproducibility of our data was confirmed by comparing the results of the first and secondtest for those seven patients who were tested twice (Table I). There were no statistically significant differences between the two tests for exerciseduration or for resting, maximal, and recovery heart rates. Second, since any deviation from normal body build could affect the comparability of these patients with their normal age-mates,we assessedthe body build of our patients. Specifically, each patient was assigneda “BSA-normalized” age, representing that age for which body surface area equalled the 50th percentile value for normal children (Fig. 1). By this criterion we determined that in our total group of patients the true chronologic age and the “BSA-normalized” age were statiaticalIy comparable (r = 0.92, p < 0.05).

Chronotropic responws. Fig. 2 illustrates the achieved heart rate (HR) values for one patient, a g-year-old boy who ran for 13 minutes on the treadmill. This duration of exercise ranked him in the 60th percentile among boys his age. His HR responses were typical for what we observed among postoperative Mustard patients as a group. The resting HR of 80 beats per minute (bpm) was well within normal limits (84 ? 6 bpm). However, despite full effort his maximum HR of 160 was significantly below the norm for boys his age (200 rt 6 bpm), and his recovery HR at 2 minutes (98 bpm) and 5 minutes (85 bpm) postexercise approached resting values abnormally early (normal 122 + 11 bpm and 116 or 9 bpm, respectively). Within the entire group of 29 patients, a high proportion of abnormal exercise and recovery HR values was encountered (Fig. 3). Fig. 4 shows an example of one such patient’s ECG. Although most patients had a somewhat lower resting HR than normal (nine patients had junctional rhythm prior to exercise), only 2 of 29 (7%) resting HR values were more than 2 standard deviations below the expected mean. With maximal exercise, however, 22 of 29 patients (76 % ) had a significantly reduced HR. Two minutes into the recovery phase, the HR was significantly reduced in 13 of 29 patients (45%). Five minutes after exercise, the HR was abnormally low in 22 of the 29 patients (76%). By chi-square analysis, the proportion of significantly depressed HR during and after exercise was highly significant (p < 0.001). Although maximal effort was sought from each patient, some of these children demonstrated reduced tolerance for exercise. The exercise duration was at least 1 standard deviation below normal in 16 of the 29 children. The remaining 13 patients remained on the treadmill an average 12.7 minutes, well within the expected range. To eliminate the

Volume Number

103 3

Fig.

2. Chronotropic responseto treadmill exercise of

Post-Mustard

exercise HR: Relation

353

to SN dysfunction

3. Proportion of abnormal heart rate responsesat various stagesof maximal treadmill stress test. “Abnormal” is defined as a heart rate more than 2 standard deviations below the normal mean for age and sex. Solid bars bars are for all patients (n = 29) and cross-hatched represent the subgroup with normal exercise tolerance (n = 13). MHR = maximum heart rate; RHR = resting heart rate; R2 = 2-minute recovery heart rate; R5 = 5minute recovery heart rate. Fig.

patient S.B. (age9.4 years; BSA-normalized age8.8 years), 5.4 years after Mustard operation. Dashed line and shaded area represent the normal range (mean +- 2 standard deviations) for 8 to g-year-old boys.? See text for details. MHR = maximum heart rate; RHR = resting heart rate; R2 = 2-minute recovery heart rate; R.5 = 5minute recovery heart rate.

possibility of inadequate’effort modulating the HR responses, we analyzed the results for those 13 patients with clearly normal effort and exercise tolerance in the same manner as for the entire group (Fig. 3). Once again, the resting HR was generally reduced, but in only one case was it significantly so. By contrast, 8 of 13 of this group (62%) had significantly reduced maximal HR, and 9 of 13 (69 % ) had significantly reduced &minute recovery HR (p < 0.001). Therefore, even when considering only those patients with normal exercise capability, a high proportion of patients still had distinctly abnormal HR responses. Cardiac rhythm responses. Table II shows the effect of exercise on the basic cardiac rhythm for the entire group of 33 patients. Since these effects were identical on subsequent tests for those patients tested twice, only one response is reported for each patient. All patients who began the test in sinus rhythm maintained that rhythm throughout the test. Eight of nine patients who began in junctional rhythm promptly converted to sinus rhythm during exercise; seven of these reverted to junctional

Table II. Effects of exercise on cardiac rhythm (n = 33 patients) At

With

After

rest

exercise

exercise

Sinus rhythm Junctional rhythm Atria1 flutter

rhythm within 5 minutes after completion of exercise. One child had atrial flutter with 3:l conduction at rest. During exercise she remained in flutter but developed 1:l conduction to the ventricle at a rate of 240. Atrioventicular block returned 3 minutes after termination of exercise. Although resting rhythm was generally stable prior to exercise, paroxysmal sinus pauses occurred in the first minute after exercise in 5 of 40 tests (Fig. 5). Premature atrial or ventricular extrasystoles were not encountered prior

354

Hesslein

et al.

American

March, 1982 Heart Journal

Fig. 4. ECG tracings during exercisetesting in a 6-year-old boy 4 years after Mustard operation. Panel A,

Prior to exercisethe basiccardiac rhythm wasa sinusmechanismwith sinuspausesand junctional escape beats. The overall ventricular rate was 87 beats per minute. Panel B, At maximal exertion after 11 minutes, the patient exhibited normal exercisetolerance for age,sex, and weight, but his heart rate of 161 beatsper minute wassignificantly lower than expected (normal range t 2 SD = 187to 215bpm). Panel C. Five minutes into the recovery phasethere wassinusbradycardia with a junctional escaperhythm. Again, the heart rate of 60 beatsper minute wassignificantly below the normal response(82 to 150bpm) for this stage of exercise. Calibration: 10 divisions = 1 mV. Paper speed= 25 mm/set.

to exercise but occurred during or after exercise in 11 of 40 tests (Fig. 5). There were no paired extrasystoles or sustained tachyarrhythmias, except for the patient with incessant atria1 flutter. Digoxin effect. As Table III illustrates, the 10 patients receiving maintenance digoxin at the time of exercise testing demonstrated no statistically significant differences in mean resting HR, exercise HR, or recovery HR, compared to the 19 patients taking no medication. EPS findings and exercise-clinical relation. Sixteen of the 29 patients also underwent postoperative electrophysiologic catheterization study (EPS). All 16 of these patients had abnormal maximal and/or &minute recovery HR on the treadmill. On the basis of corrected SN recovery time and, in some cases sinoatrial conduction time, SN function was judged to be abnormal in only 9 of 16 cases. In reviewing other clinical data from our patients (Table IV), we were unable to find a relationship between HR response to exercise and resting cardiac rhythm, cardiac symptoms compatible with an arrhythmia,

objective evidence of cardiac arrhythmias, operative hemodynamics.

or post-

DISCUSSION Comparison of exercise of SN function. Cardiac

testing

vs EPS in evabath

arrhythmias are common after the Mustard operation and these most often take the form of the brady-tachyarrhythmia syndrome.2-4 Previous studies from our institution have demonstrated that the usual mechanism for these dysrhythmias is damage to the sinoatrial node with subsequent reduction in sinoatrial automaticity.5 Altered surgical techniques in recent years have reduced but have not eliminated postoperative SN dysfunction8 the consequences of which have included syncope and sudden death. These abnormalities are not limited to patients with transposition of the great arteries, but are sometimes encountered in children following other surgical procedures involving the atria, or may occur de nouo or as a result of drug therapy for ta&ycardia.9,‘” The usual invasive test for confirming reduced sinoatrial auto-

Volume Number

103 3

Post-Mustard

exercise

HR: Relation to SN dysfunction

355

Fig. 5. Continuous (arrow) lead II ECG tracing 30 secondsafter termination of exercisein a 12-year-old girl with normal exercise tolerance 6 years after Mustard repair. The rhythm was irregular due to intermittent sinuspauseswith atria1 and junctional escapebeats.This pattern in the immediate recovery phase,which occurred in 5 of 40 tests, is uncommonin normal children. A ventricular premature beat was present at the beginning of this tracing. Although no patient demonstrated premature complexesprior to exercise, 11 of 40 tests elicited premature atria1 or ventricular complexes during or after exercise. Calibration: 10 divisions = 1 mV. Paper speed= 25 mm/set.

Table

III. Effect of chronic digoxin on heart rate responseto treadmill exercise Testing circumstance

Digoxin (n = 10) No digoxin (n = 19)

RHR

MHR

R2

R5

75.5 + 10.3 73.2 + 12.7

172.2 k 16.0 173.9 + 21.4

100.3 + 20.2 103.6 k 26.1

84.6 iz 14.4 89.4 k 21.5

Values are for heart rate (mean + 1 standard deviation) recovery heart rate; R5 = 5 minute recovery heart rate.

in beats per minute.

maticity has been the intracardiac measurement of corrected sinus node recovery time (CSNRT). While CSNRT is a specific determination, previous studies have revealed false negative results in one half to two thirds of patients with sick sinus syndrome.“-l3 In view of the increasing number of children at risk for SN failure, a more sensitive noninvasive test would be helpful to identify those patients in whom further investigation is warranted. Consideration dren. Maximal

of dynamic

exercise

testing

in chil-

dynamic exercise provides a stimulus and challenge to sinoatrial node function which is both extreme and yielding of reproducible results within a narrow range in normal children.7*15 We have chosen to administer the Bruce protocol treadmill test because it requires lower average energy expenditure, while eliciting a slightly greater and more reproducible chronotropic response than other dynamic exercise modalities.14s ~3l7 We believe, however, that the chronotropic response to cycle ergom-

MHR

= maximum

heart rate; RHR = resting

heart rate; R2 = 2 minute

etry is sufficiently similar to what is observed with treadmill testing to allow the use of either as a means of SN evaluation. The disadvantages of all standardized forms of dynamic exercise testing in children are that the patient must be highly cooperative and old enough (generally at least 4 or 5 years of age) to perform. Since many open-heart procedures are now carried out in infancy or early childhood, there is often a several year hiatus before exercise assessment of sinoatrial node function can be accomplished. Potential diff@wlties in evaluating exercise responses in congenital heart disease. In comparing exercise

data from patients with severe congenital cardiac malformations to data from normal subjects, several pitfalls may exist. There may be differences between groups in motivation, although Cumming18 has demonstrated that when normal values are determined from a clinic population rather than a school population, this bias is minimized. Although we

March.

356

Hesslein et al.

American

IV. Comparison of treadmill findings with clinical and hemodynamic data Table

Category 1. Resting cardiac rhythm Normal (n = 20) Abnormal (n = 9) 2. Symptoms compatible with arrhythmias Present (n = 10) No symptoms (n = 19) 3. Clinical arrhythmias Present (n = 25) No arrhythmias (n = 4) 4. Hemodynamics Satisfactory result (n = 13) Suboptimal result (n = 4)

Patients with abnormal heart rate response 17120 719

8llO 16/19 22125 214 9113 P/4

believe the 29 patients included in this study exerted maximal effort, only 13 had exercise tolerance within 1 standard deviation of that expected (24 were within 2 standard deviations of normal). It is not unreasonable to expect reduced exercise tolerance in some of these children, but reduced motivation may also have played a part in some patients with subnormal duration of exercise. To eliminate this latter possibility, we examined the HR responses of those 13 children with clearly normal exercise tolerance. In doing so we may have introduced a reverse bias, i.e., evaluating a group of patients who were supernormally motivated to yield a normal response. It is therefore even more striking that this subset of 13 children demonstrated chronotropic depression comparable to. that observed in the entire study group of 29 patients. Consideration of BSA. Children with congenital heart disease may have a different body size and build than their normal age-mates. This is more likely to affect their exercise capacity than their HR response. The patients in this study closely approximated their age-mates in terms of body surface area, but in assessing the maximal exercise response of individual patients who may be much larger or smaller than normal, it is probably desirable to compare them to their size-mates as well as to their age- and sex-mates. Consideration of concurrent drug therapy. Heart rates may be altered by medications. We eliminated from consideration the one patient receiving propranolol. The 10 patients in this study receiving chronic oral digoxin (for inotropic support and/or as prophylaxis against rapid ventricular response to atria1 flutter) had virtually identical chronotropic responses to exercise compared to those patients on

Heart

1982 Journal

no medication. This is not surprising, since Kugler et all9 showed that ouabain had no effect on the CSNRT (hence automaticity) in a group of normal children. Consideration of the relation of reduced exercise HR to SN dysfunction. There is some hazard in equating a

reduced HR response to exercise with sinoatrial node dysfunction. The reason for this is that there may be a variety of causes for reduced chronotropic response to exercise. Moreover, there is no consensus on what shall represent the “gold standard” in the determination of significant SN disease. The treadmill test correctly identified all nine patients whose sinoatrial node dysfunction was “confirmed” by intracardiac electrophysiology. Seven other “false positives” may mean either that the treadmill test is a more sensitive evaluation of SN function than EPS, or that exercise testing is not very specific and reflects additional conditions such as abnormal adrenergic state or altered cardiac innervation. If stress testing is indeed more sensitive than other methods to detect SN disorders, implications of this increased sensitivity are unclear, since an abnormal chronotropic response did not correlate with clinical signs or symptoms of cardiac arrhythmias. Although a correlation between abnormal SN function and later arrhythmias has been made,20 we do not advocate that therapeutic decisions be based on the exercise test alone. It does appear, however, that the maximal exercise test may be useful in identifying those patients who require further investigation or closer scrutiny. Conclusions. From these studies we conclude that the maximal treadmill exercise test may be a useful provocative modality for the evaluation of SN function in children. We found that abnormal HR responses to exercise are common in children after the Mustard operation, as has been reported by other investigators.21 These abnormalities are encountered regardless of whether the patient exhibits normal or reduced exercise tolerance, and regardless of whether or not the patient is taking digoxin. Depressed HR responses to exercise were present in every catheterization-proven case of sinoatrial node dysfunction and were also found in seven patients whose intracardiac electrophysiology was interperted as within the normal range. This suggests that the treadmill test may be more sensitive than intracardiac study in uncovering abnormal SN function, although the clinical implications of this increased sensitivity are unclear. The treadmill-

elicited responses are obtained in what is probably a more physiologic state than catheterization data acquired in sedated patients. Moreover, the tread-

Volume 103 Number3

Post-Mustard

mill test is applicable to the outpatient setting. We therefore recommend maximal dynamic exercise testing as a screen for all patients at risk for SN dysfunction. PS, Gutgesell HP, Gillette PC: Exercise assessment of sinoatrial node function in children after Mustard’s operation (abstr). Pediatr Cardiol 1:322, 1980. 2. El-Said GM, Rosenberg HS, Mullins CE, Hallman GL, Cooley DA, McNamara DG: Dysrhythmias after Mustard’s operation for transposition of the great arteries. Am J Cardiol 1. Hesslein

3.

4.

5

6.

11.

12.

REFERENCES

30:526,

10.

13.

14. 15.

19’71.

Lewis AB, Lindesmith GG, Takahashi M, Stanton RE, Tucker BL, Stiles QR, Meyer BW: Cardiac rhythm following the Mustard procedure for transposition of the great vessels. J Thorac Cardiovasc Surg 73:919, 1977. Clarkson PM, Barratt-Boyes BG, Neutze JM: Late dysrhythmias and disturbances of conduction following Mustard operation for complete transposition of the great arteries. Circulation 53:519, 1976. Gillette PC, Kugler JD, Garson A Jr, Gutgesell HP, Duff DF, McNamara DG: Mechanisms of cardiac arrhythmias after the Mustard operation for transposition of the great arteries. Am J Cardiol 45:1225, 1980. Bruce RA, Kusumi F, Hosmer D: Maximal oxygen intake and nomographic assessment of functional aerobic impairment in cardiovascular disease. AM HEART J 85:546, 1973. Cumming GR, Everatt D, Hastman L: Bruce treadmill test in children. Normal values in a clinic population. Am J Cardiol 41:69, 1978. El-Said GM, Gillette PC, Cooley DA, Mullins CE, McNamara DG: Protection of the sinus node in Mustard’s operation. Circulation 53:788, 1976. Yabek SM, Jarmakani JM: Sinus node dysfunction in children, adolescents and young adults. Pediatrics 61:593, 1978.

16.

17.

18. 19.

20.

21.

exercise HR: Relation

to SN dysfunction

357

Kugler JD, Gillette PC, Mullins CE, McNamara DG: Sinoatrial conduction in children: An index of sinoatrial node function. Circulation 58:1266, 1979. Rosen KM, Loeb MS, Sinno MZ, Rahimtoola SH, Gunnar RM: Cardiac conduction in patients with symptomatic sinus node disease. Circulation 43:836, 1971. Narula OS. Samet P. Javier RP: Sienificance of the sinus node recovery time. Circulation 45:1iO, 1972. Gupta PK, Lichstein E, Chadda KD, Badui E: Appraisal of sinus nodal recovery time in patients with sick sinus syndrome. Am J Cardiol 34:265, 1974. Bruce RA: Methods of exercise testing: Step test, bicycle, treadmill, isometrics. Am J Cardiol 33:715, 1974. Rozanski JJ, Dimich I, Steinfeld L, Kupersmith J: Maximal exercise stress testing in evaluation of arrhythmias in children: Results and reproducibility. Am J Cardiol 43:951, 1979. Blackburn H, Winckler G, Vilandre J, Hodgsen J, Taylor HL: Exercise test: Comparison of the energy cost and heart rate response to five commonly used single-stage, non-steadystate, submaximal work procedures. In Brunner D, Jokl E, editors: Medicine and Sport. vol 4. Physical activity and aging. Basel, 1970, S. Karger, pp 28-36. Shephard RV, Allen C, Benade AJS, Davies CTM, di Prampero PE, Hedman R, Merriman JE, Myhre K, Simon R The maximum oxygen intake. An international reference standard of cardiopulmonary fitness. Bull WHO 38:757, 1969. Cumming GR: Maximal exercise capacity of children with heart defects. Am J Cardiol 42:613, 19’78. Kugler JD, Garson A Jr, Gillette PC: Electrophysiologic effect of digitalis on sinoatrial node function in children. Am J Cardiol 44:1344, 1979. El-Said GM, Gillette PC, Mullins CE, Nihill MR, McNamara DG: Significance of pacemaker recovery time after the Mustard operation for transposition of the great arteries. Am J Cardiol 38448, 1976. Issenberg HJ, Freed MD: Exercise capacity and heart rate response after Mustard operation (abstr). Pediatr Cardiol 1:322,

1980.