Assessment of Cardiac Reserve in Patients with Hyperthyroidism

Assessment of Cardiac Reserve in Patients with Hyperthyroidism· Rex B. Shafer, M.D.; and Jesus A. Bianco, M.D.

RadionucHde moltigBted cardiac Imlglng was performed

to assess functional cardJac reserve in patients with hy-

perthyroidism. In 11 normal individuals, submaxlmal supine exercise increased heart rate (IIR) by 60 percent, systoUc blood pressure (BP) by 54 percent, and the left ventricular ejection fraction (LVEF) by 13 percent. AD of these increments were statisticaDy significant

I t is well established that thyroid hormones in-

duce positive inotropic and chronotropic effects on the heart in hyperthyroidism.' Positive inotropic effects of thyroxine (T4) have been demonstrated in papillary muscle, left atrial and right ventricular muscles, and hearts of conscious dogs.U In hyperthyroid patients, studies employing systolic time intervals have shown that thyrotoxicosis results in shortening of the left ventricular ( LV) isovolumic contraction and ejection times,s shortening of the LV preejection period,'/ and lowering of the LV preejection period/ejection time ratio," Similarly, echocardiographic investigations have confirmed presence of enhanced LV contractile state in patients with Graves' disease.s,D These investigations have also reported normal cardiac size in hyperthyroid patients. 8 To date, little has been reported on the LV functional reserve in patients with hyperthyroidism. Availability of exercise radionuclide ventriculographic studies has allowed us to assess the exercise response in patients with clinical and chemical manifestations of thyrotoxicosis. METHODS

Control Subfects The control group consisted of 11 men whose age range was 24-45 years (mean 37). None had clinical or anglographic evidence of ischemic heart disease, or of other cardiac disease. Six patients had undergone coronary angiographic examination and were found to have normal coronary °From the Nuclear Medicine Service, Veterans Administration Medical Center and University of Minnesota, Minneapolis, Minnesota Manuscript received September 10; revision accepted October29 Reprint reqtle8t8: Dr. Shafer, VA HOBf)ItDl Fori SnelUng, St. Paul, Minnesota 55111

CHEST, 78: 2, AUGUST, 1980

studied. Their mean ± standard error of the mean (SEM) serum thyroxine level was 17.7 ± 0.5 PC percent. There was a negative correlation between senun thyroDne levels and LVEF, both at rest and during exercise, III the patients with hyperthyroidism. Our data suaest that hyperthyroidism Is a.sodated with impaired functloaal

cardiac reserve.

arteries. The remaining five individuals were healthy volunteers who were asymptomatic, had normal ECGs and no clinical or family history of cardiac disease. All had normal thyroid function.

Patients with Thr/rotoxicosis Eight male patients, ages 25-56 years (mean 43), had diffuse toxic goiter (Graves' disease) with clinical criteria of

hyperthyroidism, te, weight loss, muscle fatigue, heat intolerance, tachycardia, tremor and nervousness. Symptoms had been present from three months to one year with a mean of four months. Laboratory criteria for hyperthyroidism included elevated total T., total biiodothyronine (T8)' T 8 resin sponge uptake (Ta RU), and 24-hour 181iodineuptake. Total senun T. of the group determined by radioimmunoassay was 17.7 ± 0.5 pg percent (mean ± SEM). The normal values for T. in our laboratory are 4-10 p.g percent No patient had clinical evidence of coronary disease, and all the patients had normal rest and exercise ECG and chest x-ray film findings. Two patients were receiving oral propranolol at the time of study (patient 12, 20 mg per day and patient 16, 160 mg per day). Exercise Radionuclide Ventriculogram

The multigated cardiac study was performed using methodology previously described. to Briefly, after red blood cell labeling with 99mtechnetium,11 the patient was placed in the modified 45 0 LAO projection, utilizing a S7-PMT large field-of-view Anger scintillation camera. Data collection was conducted under R-wave synchronization. Data were organized in the histogram mode employing 28 frames. Framing rate, at rest and during exercise, was pre-set by the minicomputer (Unicam, Medical Data Systems), in accordance with the prevailing heart rate (range 20-40 msec/data frame), The LV edge was defined by the minicomputer in each frame using threshold and second derivative criteria. The LV ejection fraction (LVEF) was computed as end-diastolic counts minus end-systolic counts/end-diastolic counts. Background activity was determined in the end-systolic frame from an area adjacent to the lateral wall of the LV; this background was subtracted from each matrix element in the region of interest of each. frame in the cardiac cycle. Each data

CARDIAC RESERVE IN PATIENTS WITH HYPERTHYROIDISM 268

Table I-Data in Normallndi,,",,",,- (n = 11)

Resting Study

Exercise Study

A

A

Patient

HR beats/min

BP mmHg

DP

(X 1()I)

1

75

120

2

56

3

EF

(%)

HR beats/min

BP mmHg

(XI0t )

(%)

90

60

105

180

189

74

140

78

65

123

190

234

73

80

130

104

60

120

190

228

75

4:

65

125

81

66

125

200

263

74

5

60

100

60

58

100

180

180

68

6

67

160

107

64

127

210

266

69

7

80

130

104

77

125

200

250

82

8

82

140

115

64

115

220

253

68

9

55

110

61

66

115

170

196

72

10

60

120

72

67

110

190

209

83

11

82

120

98

75

128

220

282

82

73±2

127±2

88±2

66±1

117±1

196±2

232±4

75±1

0.29

0.58

0.87

0.66

<0001

<0.001

<0.001

<0.001

Mean±SEM

r (correlation with resting value) p (versus resting value)

DP

EF

HR-heart rate; BP-=systolic blood pressure; DP=double product; EF=ejection fraction. acquisition period, at rest and under exercise, spanned a two to three minute interval as previously descnbed.w Frames

supine. Thereafter, both normal subjects and patients performed supine exercise on a bicycle ergometer, beginning at a workload of 100 kpm and increasing by 50 kpm increments until submaximal load was reached (approximately 300 kpm) . LV ejection fraction and wall motion were determined at a preselected end-point, under a steady workload. A tightly fitted cloth harness prevented excessive patient motion. The end-point of exercise was defined as at least a doubling of the double product (double product = heart rate X systolic blood pressure/100). LVEF, heart rate and systolic blood pressure were again determined during supine exercise

were presented in cinematic display after each acquisition. The validity, variability, and reproducibility of radionuclide LVEF, as calculated in our laboratory, were previously discussed. 10

E%perimenttJl Protocol The following protocol was followed in the normal subjects and in the patients with hyperthyroidism. LVEF, heart rate and systolic blood pressure were obtained with the patient

Table 2--Data in lIypenlayroid Patient. (n = B)

Exercise Study

Resting Study A

Patient

Serum T.

1«%

EF

----'"

EF

HR beats/min

DP (XI0t )

(%)

BP

mmHg

beats/min

mmHg

DP (XI0t )

(%)

105

125

131

84

150

175

263

68

84

150

126

65

110

220

242

66

BP

HR

12

16

13

22.6

14

21

100

120

120

68

125

150

188

70

15

12

93

140

130

73

137

165

226

72

16

14.8

107

150

160

75

151

230

347

88

17

19.9

65

130

84

61

129

190

245

52

18

21.5

85

120

102

52

132

190

241

52

19

14.1

92

110

101

82

149

190

283

70

17.7±0.5

91±2

131±2

119±3

70±1

135±2

189±3

254±6

67±2

0.41

0.60

0.35

0.66

<0.001

<0.001

<0.001

NS

Mean±SEM

r (correlation with resting value) p (versus resting value)

HR-heart rate; BP=-systoIro blood pressure; DP==double product; EF-ejection fraction; NS=not significant.

270

SHAFER,B~CO

CHEST, 78: 2, AUGUST, 1980

NORMALS

N=ll

300

• • • •• •

250

• •

cb .... x

t5

200

I

• • •

::>

~

no.

w

~

m

§

-s .... x

g ~ w

~

m

150

8 c 0 0

I 100

0

2

150

P<0.001

co

100

0

0 0 0 0

P
8

0 0

0 0 0

cP

50

~

~

REST EXERCISE

50

under a steady workload. The means cited in this paper refer to LVEF units, and to their fractional percentage changes. All values are expressed as mean ± standard error of the mean (SE·M). Statistical significance was ascertained by means of Student's paired t-test. Linear correlations were computed where appropriate. Linear fitting was done by the least-squares method. In all cases an informed written consent was procured prior to the exercise study.

...L ....L REST EXERCISE

FiCUBE 1. Effect of submaximal exercise on the double product (heart rate X systolic blood pressure/lOO) in 11 normal subjects and in eight patients with hyperthyroidism. Open circles = rest values; solid circles = exercise values; means ± SEM indicated.

correlation between the two variables (r = 0.65, l' <0.01). Analysis of LV wall motion at rest and during exercise in both the control group and the hyperthyroid group did not disclose noticeable differences. No wall motion abnormalities developed with exercise in either group. DISCUSSION

REsULTS

Data obtained in the control individuals and in the patients with hyperthyroidism are listed in Tables 1 and 2. In the control group, the increase in double product under exercise was 164 percent (Fig 1). This increment was accompanied by a fractional rise of 13 percent in the LVEF, as compared to rest (1' <0.(01), (Fig2). In the group of patients with hyperthyroidism, there was a 113 percent increase in double product under exercise (Fig 1). However, the mean L VEF did not significantly change during supine exercise (Fig 2). Figure 3 displays the relationship between serum T 4 levels and LVEF in the patients with hyperthyroidism. The graph shows values obtained both at rest and during exercise. There was a negative CHEST, 78: 2, AUGUST, 1980

The data from this study clearly suggest that functional LV reserve is impaired in patients with hyperthyroidism. While in normal individuals we found the expected increment of L VEF during exercise, there was no significant change in the L VEF with exercise in the group with Graves' disease. The lesser increase in double product in the hyperthyroid patients (2.1 times in hyperthyroid patients compared to 2.6 times in the normal subjects) was most likely due to the higher resting heart rates and systolic blood pressure seen in the hyperthyroid patients. The impairment of functional reserve is similar to ~at found in patients with coronary disease, aortic valvular disease, and other cardiomyopathies. Hyperthyroidism, as well as these clinical entities, must be considered when interpreting the exercise radionuclide ventriculogram.

CARDIAC RESERVE IN PATIENTS WITH HYPERTHYROIDISM 271

90

NORMALS N=11

-

85

--~

~a:

u,

~

fa ~ ~

N=8

•

o

o

80

" "-"

0

75

• •

0

...•••

70

J

0 00

t

0

00

P<0.001 60

-

85

• ••

80

65

90

HYPERTHYROID

00

Z 0

u
75

LL

70

a:

~

fa-, W

~

0

o

o

65

60

o

o

• •• • • P NS

I

0

55

FIctmB i. EJfect of submazfmal exercise on the left ventricular ejection fraction in 11 normal subjects and eight patients with hyperthyroidism. Open circles rest values; solid circles = exercise values; means ± SEM indicated; NS = nonsigniHcant.

55 o

=

..L.....L

50

••

-L REST

....L 50 REST EXERCISE EXERCISE showed a significant fall in LVEF from 61 percent to The negative correlation between serum T 4 level

and LVEF in the hyperthyroid patients, at rest and under exercise, further supports the presence of impaired LV function in thyrotoxic patients. We are not aware that this relationship between serum T4 and LVEF has previously been documented. Two of the eight hyperthyroid patients were receiving oral propranolol (patient 12, 20 mg per day and patient 16, 160 mg per day). While patient 16 r· -0.65 (p <0.01)

Y• 33.4 + X( -0.23) EXERCISE REST

25

e o

oe

12

m

"

M ~ " LV EJECTION FRACTION (I) FIc11BB 3. Relationship between serum throxine (T.) level and left ventricular ejection fraction in eight patients with hyperthyroidism. Correlation coefficient and signiHcance of it indicated. Regression equation (Ieast-squares method) is ~

also included. Diamond symbol depicts means ± SEM; open circles

= rest .values; solid circles = exercise values.

272 SHJfER,BIANCO

52 percent with exercise, this fall should not be attributable to propranolol. Marshall et al12 have shown that when oral propranolol was given to subjects in doses of 165 + 13 mg per day, no measurable change in resting LVEF resulted More recently, they have reported that the rise of L VEF during exercise is not blunted by oral propranolol given in doses up to 190 mg per day. Cardiac disease secondary to hyperthyroidism is a concept amply popularized by Samuel Levine, and accepted by many investigators. 1 Biventrieular cardiac enlargement may result from excess of thyroid hormones. Cardiac failure, however, is unusual today because of earlier diagnosis and therapy. Additional demands or limitations placed on the LV, as for example hypertensive vascular disease or coronary artery disease, may nevertheless provoke cardiac decompensation in patients with hyperthyroidism. Thyroid disease as the precipitating cause of heart failure may go unrecognized for a long time before hyperthyroidism is diagnosed.P Our experimental data suggest that the depression of LV functional reserve in a given patient with hyperthyroidism will importantly depend on the severity of the hypermetabolic state. Our investigation did not address the question of the mechanisms whereby altered thyroid state influ-

CHEST, 78: 2, AUGUST, 1980

ences myocardial function. Neither 'biochemical or histologic information has revealed the primary myocardial defect in patients with hyperthyroidism, Appropriate management of this condition reverses whatever mechanisms are responsible for the cardiomyopathy of hyperthyroidism. While this study casts little light as to the underlying primary pathogenetic defects in hyperthyroidism, certain conclusions may be drawn from the data. The LV functional reserve is impaired in patients with hyperthyroidism. Early diagnosis may expedite therapy and preclude appearance of heart failure. The degree of derangement of cardiac reserve in these patients probably determines their vulnerability for cardiac decompensation. One would expect that protracted hyperthyroidism with high levels of serum T. would be associated with greater propensity for development of cardiomyopathy. Patients with nonendocrine cardiac disorders may likely show cardiac decompensation with the onset of hyperthyroidism. Timely management of the latter situation should prevent worsening of the former condition.. In summary, this report establishes a decline in LV functional reserve in patients with hyperthyroidism, which appears to vary in intensity in relation to the serum level of thyroid hormones.

REFERENCES 1 Symons C. Thyroid heart disease. Br Heart J 1979; 41:257-262 2 Buccino RA, Spann JF Jr, Pool PE, et ale Influence of the

CHEST, 78: 2, AUGUST, 1980

thyroid state on the intrinsic contractile properties and energy stores of the myocardium. J Clin Invest 1961; 46:1669-1~

3 Murayama M, Goodkind MJ. Effect of thyroid hol'lllODe on the frequency-force relationship of atrial myocardium from the guinea pig. Cire Res 1968; 23:743-751

4 Zaimis E, Papadald L, Ash ASF, et al, Cardiovascular effects of thyroxine. Cardiovasc Res 1969; 3:118-132 5 Taylor RR, Coven JW, Ross J Ir, et ale Influence of the thyroid state on left ventricular tension velocity relations in the intact, sedated dog. J Clin Invest 1969; 48:775784 6 Amidi M, Leon DF, DeGroot WI, et aI. EJfect of the thyroid state on myocardial contractility and ventricular ejection rate in man. Circulation 1968; 38:229-239 7 Parisi AF, Hamilton BP, Thomas CN, et al. The short cardiac pre-ejection period, an index of thyrotoxicosis. Circulation 1974; 49:900-904 8 Lewis BS, Ehrenfeld EN, Lewis N, et aI. Echocardiographic LV function in thyrotoxicosis. Am Heart J 1979; 97:460-468 9 Otsuka K, Yanaga T, Kudo S, et ale Echocardiogram before and after subtotal thyroidectomy for hyperthyroidism. Jap Heart J 1977; 18:17-23 10 Bianco JA, Makey DC, Laskey WK, et ale Radionuclide left ventricular dV/ dt for the assessment of cardiac function in patients with coronary disease. J Nucl Med 1979; 20:1-6 11 Pavel DC, Zimmer AM, Patterson VN. In vivo labeling of red blood cells .with 99 1DTe : A new approach to blood pool visualization. J Nuel Med 1977; 18:305-308 12 Marshall RC, Berger HJ, Reduto CA, ef ale Effect of oral propranolol therapy upon left venbicular performance in coronary artery disease: Noninvasive assessment by sequential radionuclide angiography ( abs ). Circulation 1977; 55 and 56" (Supplm):nI-32 13 Somerville W, Levine SA. Angina pectoris and thyrotoxicosis. Br Heart J 1950; 12:245-257

CARDIAC RESERVE IN PAnENTS WITH HYPERTHYROIDISM 273