Relative tachycardia in ambulant children with borderline hypertension

Relative tachycardia in ambulant borderline hypertension

children

with

Borderline hypertensives who demonstrate tachycardia have a tendency toward the development of essential hypertension. However, the documentation of tachycardia in previous studies has been generally based on brief periods of observation. In the present study, we measured heart rates through a 24-hour period in 16 ambulatory mildly hypertensive subjects (ages 5 to 23 years). When compared with normal matched controls, significantly (p < 0.05) higher heart rates were observed during the waking periods (QQ + Q vs 90 f 11) and sleep periods (72 F 12 vs 62 f 7). Similar observations were also made for 24 hours (90 +- 6 vs 79 + 8). in addition, hypertensives also demonstrated thickened (during diastole) left ventricular posterior wall (0.96 + 0.17 vs 0.85 + 0.13 cm; p < 0.05) and interventricular septum (0.96 ? 0.17 vs 0.84 -t 0.19 cm; p < 0.05). It is suggested that tachycardia may be an early manifestation of borderline hypertension in children. (AM HEART J 1986;112:1257.)

Sudhir K. Mehta, M.D., R. C. Bahler, M.D., R. Hanson, R.N., J. Timothy Walsh, M.D., Ph.D., and L. Rakita, M.D. Cleveland, Ohio

There is increasing evidence that essential hypertension may have its origin in childhood. Young adults and adolescents with early and borderline hypertension have been shown to develop certain anatomic and functional cardiac abnormalities.l-4 Yet the long-term significance of borderline liypertension in childhood is unclear, since children do not have a strong tendency to maintain their blood pressure percentile rank with increasing age.5 Subjects with both tachycardia and transient elevations of blood pressure are prone to develop essential hypertension.6. ’ However, in clinical practice, it is difficult to use heart rate to predict whether an individual will ultimately develop hypertension, as tachycardia for short periods is frequently attributed to nervousness, anxiety, and variable response to different environmental stimuli and settings. This prospective study was undertaken to determine whether young children and adolescents with borderline hypertension, who were allowed to carry on their routine activities (1) manifest abnormal From the Department of Pediatrics and Medicine, Cleveland Metropolitan General Hospital and the Division of Pediatric Cardiology, Rainbow Babies and Children’s Hospital, Case Western Reserve University School of Medicine. Supported by a Grant from Ohio Affiliate, No. HANEO Received accepted

for publication Apr. 25, 1986.

Reprint Hospital.

requests: Sudhir 3395 Scranton

Dec.

the American 4245R. 23, 1985;

K. Mehta, M.D., Road, Cleveland,

Heart revision

Association, received

North March

Cleveland Metropolitan OH 44109.

East

18, 1986; General

heart rates during 24 hours, during waking and sleeping periods; and (2) demonstrate cardiac abnormalities that can be detected by echocardiography. METHODS Subjects. Sixteen patients between the ages of 5 and 23 years with borderline hypertension were prospectively evaluated at our pediatric hypertension clinic. None of the subjects had secondary causes of hypertension. Borderline hypertension was defined as a sitting systolic and/or diastolic blood pressure, measured on three or more occ+ions at intervals of at least 4 weeks apart, to be slightly more than the ninetieth percentile for age. Subjects with diastolic blood pressures of 100 mm Hg or more were considered to have established hypertension and therefore were excluded from the study. Fourteen subjects had diastolic pressures of 95 mm Hg or less.The remaining two patients, 16 and 23 years old, had diastolic pressures of 97 and 96 mm Hg, respectively. The 16 patients were matched for age, sex, race, height, and weight with 16 normotensive subjects (Table I). These individuals were selected from one of the neighboring schools and from the family members and friends of the

investigators. One patient with hypertension and one control subject had a past history of rheumatic fever and received regular penicillin prophylaxis. Neither patient had any evidence of residual cardiac involvement. All hypertensive patients and eight of the normotensive subjects had a positive family history of hypertension (NPH) in first-degree relatives. Blood pressure was measured in the right arm, which was positioned at heart level. During each visit, three to four readings were taken approximately 5 minutes apart 1257

1258

Mehta

et al.

American

I. Clinical data of patients with hypertension, normal controls, normal controls with positive family history of hypertension (NPH), and normal controls with negative family history of hypertension (NNH) Table

II. Comparisonof heart rate data from hypertensive patients and normal controls -Table

Average heart rate Cbpm) Awake

Normal controls Hypertensives (n = 16)

Age (yr) Sex Male

Female

15.6 + 4 9

NPH (n = 8) 15.5 * 5

NNH (n = 8) 15.8 + 2.5

5 3

4 4

4 1 64 + 7 143 rt_ 37

6 2 0 63 k 7.5 138 r 44

5 2 1 66 t 5 146 rt_ 32

138 * 13 90 k 6

120 * 14 71 & 10

115 _t 15 71 + 8

7

Race

White Black Hispanic Height (in) Weight (lbs) BP (mm Hg) Systolic

Diastolic

11

December 1966 Heart Journal

Asleep

Hypertensivepatients (n = 16) 99 ? 9* 72 * Controlsubjects (n = 16) 90 k 11 62 I

24-hour

12t 7

90 + 8$ 79 ?I 8

*p < 0.05.

tp < 0.01. tp < 0.001.

Ill. Comparison of heart rates during the echocardiographic examination in hypertensives, normal controls, normal children with a positive family history of hypertension (NPH), and normal children with a negative family history of hypertension (NNH) Table

Average heart rate (bpmi

Hypertensivepatients (n = 16) and the average of the last two or three readings was utilized. Diastolic pressurewas taken at phase IV of the Korotkoff sounds.All blood pressuremeasurementswere taken by one individual by meansof a mercury manometer, in accord with other recommendedprecautions.8 Study protocol. All medications were discontinued 2 weeks prior to and during the study. Twenty-four-hour Holter ECGs were recorded on an Oxford Medilog recorder (Oxford Medical Inc., Clearwater, Fla.). Most studies were begun at approximately noon and continued overnight to noon the next day. On the sameday, an M-mode echocardiographic study was performed. All participants were askedto continue their usual activities and to mark their sleepinghours in a diary. Heart rate analysis. From the 24-hour ECGs, the heart rate data were analyzed with the help of a microprocessor (North-Star Horizon, Berkeley, Calif.) and a specially designedinterface circuit. At replay of the patient’s ECG (60 times real time), the QRS spike of the ECG was converted to a pulse by means of an analog threshold circuit. With the aid of a 60 Hz recorded clock, an assembly language program counted the number of RS spikes in each minute and transferred this number to memory asthe heart rate for that minute. The accuracy of the time clock system was repeatedly checked by a test tape of known duration. The heart rate, as measuredby the computer, did not demonstrate any variations from the known heart rate of the test tape. Subsequent programs were used to average the heart rates for every 5-minute interval within the 24-hour recording period. The average heart rate for each 5-minute interval was usedfor computing average rates for the full 24 hours, at least 6 hours of sleep (nighttime) and 12 hours of waking (daytime). The programspermitted “patching” of the two segmentsof waking (separatedby sleep)to obtain at least 12 hours of waking heart rate. Baaed on the information from the diary, 1 to 2 hours of eachtransition period from

Control NPH NNH

subjects

*p < 0.05 compared

(n = 16) (n = 8) (n = 8)

77 73 80 66

1 k + i

8 14 14 11,

to NPH.

awake to sleepstates and from sleepto awake states were excluded from the daytime and/or nighttime analysis. This approach was used to avoid the uncertainty of the exact waking and sleeping periods and to exclude heart rate data from the transition phases.Neither patients nor control subjects had ventricular or atria1 ectopic rhythms during the study period. Echocardiograms. The echocardiogramfrom one of the obesehypertensive subjectswas unsatisfactory. His echocardiogram and that of the closestmatched control were excluded from the analysis. An M-mode echocardiogram (Hoffrel201 system, South Norwalk, Conn.) wasrecorded in all 32 subjects by meansof a 2.25 MHz transducer in older children and a 3.5 MHz transducer in younger subjects. Echocardiographic measurementswere taken in the supineleft lateral position according to the recommendations of the American Society of Echocardiography.g Measurements were made by two independent observers who had no prior knowledgeof the diagnosis.The correlation of repeated measurementsby the sameobserver was r = 0.99 and the correlation between two observers was r = 0.98. The maximum difference on any single measurement was0.2 cm. Final values were taken asthe averageof the measurementby the two different observers.Volume indices were derived by the cube method.‘OLeft ventricular masswas estimated by the method of Troy et al.” and also by a modified method that includes the thickness of the interventricular septum.I2Left ventricular wali thicknessrelative to ventricular size was expressedasthe ratio of left ventricular radius to wall thickness.‘*Left ventricular wall stresswascalculated by the method of Hermann et a1.13

Volume

112

Number

6

I

L

HYPERTENSWES

NORMOTENSIVES NEGATIVE HISTORY (NNH)

POSIT!VE HISTORY (NPH)

. . :

: , . . +

. I

8

ii 2-l ”

A.

; . :

NEGATIVE

POSITIVE

HISTORY (NNH)

HISTORY (NPH)

HYPERTENSIVES

i L

I I

I

I NORMOTENSIVES

@

:

NORMOTENSiVES

I HYPERTENSIVES

NEGATIVE HISTORY

POSITIVE HISTORY

(NNH)

(NPH)

1, A through C. Average heart rate (beats/minute) during the periods of waking (A), sleeping (E), and during 24 hours (C) for each normotensive and hypertensive subject. The horizontal line represents the meanheart rate for the group. There were no significant differences in heart rate amongnormotensive subjects with or without a positive family history of hypertension. Significant differences were present betweenhypertensive subjectsas comparedto normotensiveswith a negative family history (NNH) for all categoriesof heart rate (p < 0.05). A significant difference (p < 0.05) was alsopresent when the average heart rate for 24 hours was compared between hwertensives and normotensiveswith a positive family -_ history of hypertension (NPH). Fig.

Data analysis. Linear regressionanalysis was used to evaluate the relationships between heart rates, echocardiographic data, and clinical parameters such as height, weight, and age. Differences between groups with respect to a given variable were evaluated by an anaIysis of variance. When significant differences (p < 0.05) were found, multiple comparisons between groups were performed with Tukey’s method as modified by Steele and Torrie for unequal sample size. RESULtS Heart rate. Hypertensive patients demonstrated significantly higher mean heart rates than normotensive controls during the periods of waking, sleeping, and total 24 hours (Table II). Data from one of the hypertensive subjects, a nighttime worker, was excluded from the separate analyses of sleep and waking heart rates, but total 24hour analysis was included. The best separation of the heart rate data between patients and controls occurred only when heart rates were analyzed for 24 hours and if the patients were compared to normal subjects who had

a negative family history of hypertension (Fig. 1). When heart rates during sleep were considered as basal, the average increment of 40% in heart rate during awake periods in the hypertensive children was not significantly different from the 43 5%increase observed in normal controls. The correlations were poor between heart rate (24 hours, sleep, and awake periods) and the age, height, and weight in either hypertensive patients or normal controls. Heart rate was unrelated to blood pressure in the control group. However, within the hypertensive group, heart rate was inversely related to the average diastolic blood pressure during sleep (r = -0.70, p < 0.01) and during 24 hours (r = -0.64, p < 0.01). There was no significant difference in the heart rates between hypertensive and control subjects during echocardiographic examination (Table III). However, among the normal controls, subjects with positive family history of hypertension (NPH) had significantly higher heart rates than subjects with negative family history of hypertension (NNH).

1260

Mehta

et al.

American

IV. Echocardiographic data in hypertensive children and controls Table

Echocardiographic parameters

Left ventricular posterior wall thickness, diastole (cm) Interventricular septum/left ventricular posterior wall ratio Left ventricular end-diastolic dimension (cm) Left ventricular shortening fraction (% ) Left atrium (cm) Left atrium/aorta Interventricular septal thickness, diastole (cm) Left ventricular mass (gm/m’) Stroke volume (cm’) Left ventricular wall stress (dynes I 10’) Left ventricular radius/wall thickness ratio

Hypertensive patients

0.85 k 0.13

1.03 + 0.1

0.99 +- 0.2

4.5 _+ 0.7

4.6 _+ 0.5

36 -+ 5

36 t 6

3.15 * 0.4 1.26 _t 0.24* 0.98 k 0.17*

0.84 + 0.19

89.5 k 27 73 -I- 33

79.5 +- 15 77 * 22

188 If: 44

174 +- 30

2.3 +- 0.38*

2.8 +- 0.24

3 2 0.5

1.10 + 0.1

Auerage

Age W Height (in) Weight (lbs) Systolic blood pressure (mm JW Diastolic blood pressure (mm Hg) LVPWD (cm) IVSD (cm) Left ventricular mass (dm’) Left atrium (cm) LVEDV (cm’)

Heart rates observed during the echocardiographic examination and the heart rate for the total 24 hours were significantly related (r = 0.56, p < 0.001). A similar correlation (r = 0.71, p < 0.05) was also observed among controls with a positive family history of hypertension. Echocardiography. Thickness of the left ventricular posterior wall and the interventricular septum was greater in the hypertensive patients, whether measured separately or expressed in relation to the left ventricular radius (Table IV). The somewhat greater left ventricular mass in the hypertensive subjects was not statistically significant whether determined by the method of Troy et al.” or by the modified method.12 Although left atrial dimensions were similar in the two groups, a significantly higher left atrium/aorta ratio was observed in the hypertensives. All the other measured parameters including stroke volumes did not differ significantly between the groups (Table IV). in hypertensive

children.

Since the mean 24-hour heart rate for the hypertensive subjects was 90 bpm, we compared the data from those with heart rates above 90 bpm to those with rates less than 90 bpm (Table V). There were no significant differences between the two groups in age, height, weight, systolic and diastolic pressures (Table V). However, the group with the lower average heart rates had a significantly greater left ventricular mass (corrected for body surface area). Age

24-hour

heart

rate

Heart rate <9O/min (n = 8)

Heart rate >90/min (n = 7)

175 66? 149 t 141 t

3 5 26 13

14+ 63* 139 +137 t

5 9 49 12

92 rt

3

87+

8

1.02 t 0.18 1.0 -t 0.15 102.5 i 31*

0.88 It 0.13 0.92 t 0.18 74.5 +- 11

3.35 k 0.2 115 rt 49

2.9 iY 0.3 80 k 30

IVSD = interventricular septal thickness, diastole; LVEDV ular end-diastolic ness, diastole. *p < 0.05.

*p < 0.05.

Heart rate and Echo findings

Table V. Comparisonof echocardiographicdata in hypertensive children with average24-hour heart rates above or below 90 bpm

Controls

0.96 k 0.17’

December 1966 lieart Journal

volume; LVPWD

= left ventricular

= left ventricposterior wall thick-

was not a significant covariant in this analysis. Left ventricular mass index remained significant. A trend toward a larger left atrial dimension was also noted in subjects with a lower average heart rate 0, = 0.07) (Table V). The echocardiographic data from the hypertensive children with average 24-hour heart rates greater than 90 bpm was similar to that of the normal controls (Fig. 2). The patient group with lower 24-hour heart rates had echocardiographic evidence of early hypertrophy. DISCUSSION

Subjects with borderline hypertension generally have increased cardiac output with either normal or decreased peripheral vascular resistance.14-l6 Increased cardiac output may result from an increase in heart rate and/or an increase in stroke volume. However, high heart rates with a normal stroke volume16~17 and normal cardiac outpuP have been reported more commonly. Hemodynamic studies in adolescents with idiopathic hypertension have also documented high heart rates with variable cardiac output.ls From a recent review,2o it is apparent that tachycardia in early and borderline hypertension is a frequent finding and is probably secondary to an increase in sympathetic tone and a decrease in parasympathetic tone.21 It is possible that tachycardia may precede or accompany the onset of hypertension. Levy et al. in

volume Number

112 6

Heart rate in hypertrnsiw

*

Significantly different with

from

(p = <

-05)

the

group

r-d&en

126t

Hf?
Fig. 2. Left ventricular posterior wall thickness during diastole (L VPWD) and left ventricular CL,V) mass are similar for the normal @II,) subjects and hypertensive subjects with average heart rates above 90 bpm. Bars represent standard error of the mean. A0 = aorta; LA = left atrium: R = radius; TH = thickness.

194P reported that normal adults with transient tachycardia develop hypertension more often than subjects with normal heart rates. Subjects with both tachycardia and transient hypertension had a strong tendency to eventually develop hypertension.7 A recent study22 also suggested that adolescents with borderline hypertension who have a positive family history of hypertension, elevated heart rate, and a greater cardiovascular response to mental stress have a greater risk for developing essential hypertension. However, in all of these studies documentation of tachycardia was based only on relatively brief duration of observation. In clinical practice, documentation of tachycardia may be difficult due to normal variability in the heart rates. Even when observed, it is often attributed to nervousness or anxiety of the patient and thus loses it predictive value. Our observation of a relative tachycardia during the periods of sleeping, waking and over 24 hours indicates that the increased heart rate in hypertensive subjects is more than just a transient episode that can be attributed to short-term anxiety or nervousness. It may be argued that higher heart

rates during waking periods in these patients may be related to their physical activities, the influence of athletic habits, or their response to environmental stress. However, tachycardia was also observed during the relatively basal condition of sleep, indicating that the factor(s) leading to tachycardia is (are) continually present. Furthermore, since the increments in heart rate from sleeping to awake periods were similar in hypertensive and normal subjects, it is unlikely that the aforementioned extrinsic factors play a major role in the genesis of the tachycardia. Additionally, if the period of echocardiographic examination is considered as relatively stressful, no significant difference was noted in the heart rates between hypertensive and control subjects during the procedure. The difference in the 24-hour heart rate between controls and hypertensive subjects was most striking when the hypertensive subjects were compared to normotensive controls with a negative family history of hypertension. The tendency toward higher heart rates in several of our normotensive control subjects with a positive family history of hypertension is intriguing and warrants follow-up to determine

December

1262

Mehta

et al.

whether tachycardia may be a risk factor for the development of essential hypertension in later life. It is quite possible that, with more numbers in each group and giving due consideration to age, 24-hour heart rate measurements could become a useful clinical parameter to predict the development of hypertension. Although the heart rates during the 24-hour period and during the echocardiographic examination were significantly correlated, the intersubject variability in heart rates during the procedure focuses on the limited diagnostic potential of only brief periods of heart rate monitoring. Enlargement of the left atrium has been observed in adults3 as well as in children4 with blood pressures in the upper quintile. This finding may be related to decreased ventricular compliance, but probably more so to the abnormalities of ventricular relaxation. Although no attempt was made to measure the cardiac output directly, estimated stroke volume, as determined by echocardiography, was similar in both groups. If borderline hypertensives manifest increased cardiac output, its major contribution seems to be through increased heart rate. Recent studies have demonstrated increased left ventricular wall’~ 2,l2 and septal thickness” I2 in children with hypertension, and our findings are similar. The asymmetric septal hypertrophy observed by Safar et. al.24 and by Niederle et. a1.25 was not observed in the present study. Our observations indicate that the echocardiographic signs of left ventricular hypertrophy are encountered predominantely in subjects with 24hour heart rates similar to normal subjects. On the contrary, hypertensive subjects with 24-hour heart rates above 90/min had echocardiographic parameters similar to the normal controls. It is possible that one of the first manifestations of hypertension is the presence of tachycardia. With the progression of hypertension, heart rate decreases and left ventricular hypertrophy appears. The inverse relation between heart rate and diastolic pressure in our hypertensive group is consistent with this concept. Alternatively, mild hypertension with tachycardia could be a different entity, with a pathophysiology that is distinct from another process that results in the concurrent development of left ventricular hypertrophy and hypertension without an intervening phase of relative tachycardia. Our study provides the first 24-hour quantitative data confirming the presence of a relative tachycardia in some children and adolescents with borderline hypertension, and demonstrates that this relative tachycardia is not associated with left ventricular hypertrophy. We have also demonstrated that quan-

Amerlcsn

Heart

1986 Journal

titative 24-hour evaluation of heart rate in children and adolescents is now possible and will permit a more precise longitudinal evaluation of the prognostic significance of a relative tachycardia in borderline hypertension. The expert assistance of Doris Lotz, Molly Palcko, Lyn Hanak. William Lamont, and Annette Tolles is greatfully acknowledged. Dr. A.D. Goldberg and Dr. Thomas A. Riemenschneider deserve our special thanks for their continuing intellectual support and invaluable guidance during this study. We also thank Marlene Leppelmeier and Crystal Galbreath for typing the manuscript.

REFERENCES

Culpepper WS III, Sodt PC, Messerli FH, et al: Cardiac status in juvenile borderline hypertension. Ann Intern Med 1983;98:1. 2. Laird WP, Fixler DE: Left ventricular hypertrophy in adolescents with elevated blood pressure: Assessment by chest roentgenography, electrocardiography and echocardiography. Pediatrics 1981;67:255. 3. Savage DD, Drayer JM, Henry WL, et al: Echocardiographic assessment of cardiac anatomy and function in hypertensive subjects. Circulation 1979;59:623. 4. Schieken RM, Clarke WR, Lauer RM: Left ventricular hypertrophy in children with blood pressures in the upper quintile of the distribution. The Muscatine Study. Hypertension 1981;3:669. 5. Clarke WR, Schrott HG, Leaverton PE, et al: Tracking of blood lipids and blood pressures in school children: The Muscatine Studv. Circulation 1978:58:626. 6. Levy RL, White”PD, Stroud WD, et al: Transient tachycardia. JAMA 1945,129:585. 7. Paffenbarger RS Jr, Thorne MC, Wing AL: Chronic disease in former college students. VIII. Characteristics in youth predisposing to hypertension in later years. Am J Epidemiol i968;88:25. 8. Blumenthal S, Epps RP, Heavenrich R, et al: Report of the task force on blood uressure control in children. Pediatrics 1977;59(suppl):797. 9. Sahn DJ, Demaire A, Kisslo J, et al: Recommendations regarding quantitation in M-mode echocardiography: Results of a survey of echocardiographic measurements. Circulation 1978;58:1072. 10. Pombo JF, Troy BL, Russell RO: Left ventricular volumes and ejection fraction by echocardiography. Circulation 1971;43:480. 11. Troy BL, Pombo J, Rackley CE: Measurement of left ventricular wall thickness and mass by echocardiography. Circulation 1972;45:602. 12. Zahka KG, Neil1 CA, Kidd L, et al: Cardiac involvement in adolescent hypertension. Hypertension 1981;3:664. 13. Hermann HJ, Singh R, Dammann JF: Evaluation of myocardial contractility in men. AM HEART J 1969;77:755. 14. Eich RH, Peters RJ, Cuddy RP, et al: The hemodynamics in labile hypertension. AM HEART J 1962;63:188. 15. Julius S, Pascual AV, Sannerstedt R, et al: Relationship between cardiac output and peripheral resistance in borderline hypertension. Circulation 1971;43:382. 16. Lund-Johansen P: Haemodynamics in early essential hypertension. Acta Med Stand 1967;482(suppl):l. 17. Julius S, Conway J: Hemodynamic studies in patients with borderline blood uressure elevation. Circulation 1968; 1.

38:282.

Foud FM, Tarazi RC, Dustan HP, et al: Hemodynamics of essential hvpertension in adolescents (abstr). Circulation 1976;(suppi54):II-24. 19. Dustan HP, Taraxi RC: Hemodynamic abnormalities of 18.

Volume Number

112 6

adolescent hypertension. In New Juvenile hypertension. New York: 20. Lund-Johansen P: Haemodynamics sion. Clin Sci 1980;59:3435. 21. Julius S, Pascual AV, London R: inhibition in the hyperkinetic type sion. Circulation 1971;44:413. 22. Falkner B. Kushner Ii, Onesti characteristics in adolescents who tension. Hypertension 1981:3:521.

MI, Levine LS, editors. Raven Press, 1977:181. in essential hypertenRole of parasympathetic of borderline hypertenG, et al: Cardiovascular develop essential hyper-

23. Tarazi RC: The progression from hyprttrophv to heart failure. Hosp Pratt 1983;18:101. 24. Safar ME, Lehner JP, Vincent MI, et al: Echocardiographiz dimensions in borderline and sustained hvpertension. Am .I Cardiol 1979;44:930. 25. Niederle P, Widimsky J, Jandova R. et al: Echocardiographic assessment of the left ventricle in juvenile hypertension. In! .I Cardiol 1982;2:91.

Exercise in children before and after coarctectomy: Hemodynamic, echocardiographic, and biochemical assessment After repair of coarctation of the aorta, some patients with normal blood pressure at rest have an exaggerated hypertensive response to activity. Blood pressure response to exercise was studied in 15 children, aged 5 to 15 years, prior to and at periods up to 6 months following coarctectomy. Preoperatively, 11 of 15 children had systolic hypertension at rest and 12 of 15 after exercise. After surgery, only one child had mild systolic hypertension at rest, whereas exercise-induced hypertension persisted in 33% of patients (all older than 10 years). Exercise plasma renin activity was efevated preoperltively but normalized following surgery. No significant difference was seen in resting and exercise plasma catecholamine levels measured before and after surgery. Over the follow-up period of 6 months, echocardiographic evidence of left ventricular hypertrophy regressed in the younger patients but not in the older patients with exercise-induced hypertension. Exercise testing defines a subgroup of patients with exercise-induced hypertension evident soon after surgery. Structural upper segment arterial vessel wall changes in the older patient may explain these observations. (AM HEART J 1966;112:1263.)

Andrew N. Pelech, M.D., Wahono Kartodihardjo, M. Phys. Ed., Judith A. Balfe, B.A., John W. Balfe, M.D., Peter M. Olley, M.D., and Frans H. H. Leenen, M.D. Toronto, Ontario, Canada

Patients who have undergone corrective surgery for coarctation of the aorta continue to have an increased incidence of premature cardiovascular disease and early death,1,2 possibly related to persistent hypertension. Blood pressure at rest remains abnorFrom the Divisions of Cardiology and Nephrology, Department of Pediatrics, Hospital for Sick Children and Hypertension Unit, Department of Medicine. Toronto Western Hospital. Supported by an operating grant from the Hospital for Sick Children Foundation. Toronto, Ontario, and by the Canadian Heart Foundation. Received for publication Sept. 4, 1985; revision received Feb. 24, 1986; accepted April 1, 1986. Reprint requests: Frans H. H. Leenen, M.D., Ph.D., Hypertension Unit, Division of Clinical Pharmacology, Toronto Western Hospital, 399 Bathurst St.. Toronto. Ontario M5T ZSR, Canada.

mal in 10% to 30% of patients following coarctectomy.3-7 During exercise, exaggerated elevations of upper limb pressure may occur in 30 % to 65 % of postcoarctectomy patients.s-‘O Exercise blood pressure reflects the dynamic nature of the cardiovascular response to activity and, particularly in the young patient, may be more representative of the average daily blood pressure. Moreover, blood pressure during exercise correlates better with the degree of left ventricular hypertrophy in young mildly hypertensive adults than does resting blood pressure.” Postoperative hypertension has been attributed to: (1) recoarctation or residual coarctation, (2) 1263