Electrocardiographic Manifestations in Patients with Thyrotoxic Periodic Paralysis

Electrocardiographic Manifestations in Patients with Thyrotoxic Periodic Paralysis YU-JUEI HSU, MD; YUH-FENG LIN, MD; TOM CHAU, MD; JUN-TING LIOU, MD; SHI-WEN KUO, MD; SHIH-HUA LIN, MD

ABSTRACT: Background: Thyrotoxic periodic paralysis (TPP) commonly precedes the overt symptoms and signs of hyperthyroidism and may be misdiagnosed as other causes of paralysis (non-TPP). Because the cardiovascular system is very sensitive to elevation of thyroid hormone, we hypothesize that electrocardiographic manifestations may aid in early diagnosis of TPP. Methods: We retrospectively identified 54 patients who presented to the emergency department (ED) with hypokalemic paralysis during a 3.5-year period. Thirty-one patients had TPP and 23 patients had non-TPP, including sporadic periodic paralysis, distal renal tubular acidosis, diuretic use, licorice intoxication, primary hyperaldosteronism, and Bartter-like syndrome. Electrocardiograms during attacks were analyzed for rate, rhythm, conduction, PR interval, QRS voltage, ST segment, QT interval, U waves, and T waves. Results: There were no signifi-

cant differences in age, sex distribution, and plasma K⫹ concentration between the TPP and non-TPP groups. Plasma phosphate was significantly lower in TPP than non-TPP. Heart rate, PR interval, and QRS voltage were significantly higher in TPP than non-TPP. Forty-five percent of TPP patients had first-degree atrioventricular block compared with 13% in the non-TPP group. There were no significant differences in QT shortening, ST depression, U wave appearance, or T wave flattening between the 2 groups. Conclusion: Relatively rapid heart rate, high QRS voltage, and first-degree AV block are important clues suggesting TPP in patients who present with hypokalemia and paralysis. KEY INDEXING TERMS: Electrocardiography; Hyperthyroidism; Hypokalemia; Paralysis. [Am J Med Sci 2003;326(3): 128–132.]

H

paralysis may be the first symptom of hyperthyroidism, preceding the familiar classic signs. These compounding factors make early diagnosis of TPP very difficult.1,6 –9 The definitive therapy for TPP is treatment of the underlying hyperthyroidism. Although K⫹ replacement therapy is generally suggested to hasten recovery, excessive K⫹ supplementation may lead to rebound hyperkalemia in patients with TPP.10 Therefore, failure to diagnose TPP early may lead to serious management errors. Because thyroid function tests are usually not available in the ED, it is useful to identify a set of clinical clues that differentiate between TPP and non-TPP. The cardiovascular system is very sensitive to thyroid hormone, and cardiovascular manifestations are prominent presenting features of hyperthyroidism, including rate and rhythm disturbances, especially for those TPP patients who do not have overt clinical symptoms of hyperthyroidism, such as weight loss, heat intolerance, palpitations, increased appetite, excitability, and diaphoresis.11–13 Therefore, we hypothesize that useful differentiating factors between TPP and non-TPP may be found in the electrocardiogram (ECG), a simple and frequently-performed test in the Emergency Department (ED).

ypokalemic paralysis (HP) is a potentially reversible metabolic disorder characterized by severe muscle weakness and severe hypokalemia.1–3 It can be simply divided into 2 groups: hypokalemic periodic paralysis (HPP), caused by sharp shifts of potassium (K⫹) into cells1–5 and non-HPP, caused by excessive body K⫹ deficit.1–3 Among the causes of HPP, thyrotoxic periodic paralysis (TPP) is the most common in Asians,1– 4,6 whereas familial periodic paralysis is most common in Western countries.2,3,5 In patients with non-HPP, excessive vomiting, diuretic use, primary hyperaldosteronism, licorice intoxication, Bartter syndrome/Gitelman syndrome, toluene abuse, and distal renal tubular acidosis are common causes.1 The neuromuscular presentations of TPP and non-TPP are indistinguishable. In addition, periodic

From the Divisions of Nephrology (Y-JH, Y-FL, TC, S-HL), Cardiology (J-TL), and Endocrinology and Metabolism (S-WK), Department of Medicine, Tri-Service General Hospital, National Defense Medical Center, Taipei, Taiwan. Submitted January 21, 2003; accepted June 6, 2003. Correspondence: Shih-Hua Lin, M.D., Division of Nephrology, Department of Medicine, Tri-Service General Hospital, Number 325, Section 2, Cheng-Kung Road, Neihu 114, Taipei, Taiwan (E-mail: [email protected])

128

September 2003 Volume 326 Number 3

Hsu et al

Table 1. Patient Characteristics in the TPP and non-TPP Groups

Patients with TPP Patients with non-TPP Sporadic periodic paralysis (SPP) Primary aldosteronism Licorice intoxication Renal tubular acidosis Diuretics Bartter or Gitelman syndrome

Number

Age (years)

M:F

31 23 12 1 1 4 2 3

27 ⫾ 5.7 29 ⫾ 8.4 27 ⫾ 2.4 39 49 35 ⫾ 8.3 50 ⫾ 2.1 21 ⫾ 2.5

31:0 19:4 10:2 1:0 1:0 2:2 2:0 3:0

M, male; F, female.

Methods Patients. Retrospectively reviewing charts between July 1999 and December 2002, we identified 54 patients with HP who were all first seen in the ED of Tri-Service General Hospital, an urban academic medical center in Taipei, Taiwan. HP was defined as severe muscle weakness with an inability to ambulate and plasma K⫹ less than 3.0 mmol/L at presentation. HPP was diagnosed in patients with hypokalemia and paralysis caused by a sharp shift of K⫹ into cells and low K⫹ excretion. TPP, a hyperthyroidism-induced HPP, was confirmed by thyroid function test. Sporadic periodic paralysis (SPP), a unique subgroup of HPP, was diagnosed in patients with a family history of neither paralysis nor hyperthyroidism. Measurement. Blood biochemistry, including acid-base and electrolytes, was determined in the ED. Arterial blood gas was measured by an ABL 510 (Radiometer, Copenhagen). Blood and urine biochemical values and electrolytes were determined with the use of automated methods (AU 5000 chemistry analyzer; Olympus, Tokyo, Japan). Thyroid function tests were determined by radioimmune assays. A standard 12-lead ECG was also obtained in the ED. The ECG during attack was analyzed for rate, rhythm, conduction pattern, duration, and amplitude of the P wave in lead II, P-R interval, voltage of the QRS complex, amplitude of R in lead II, maximum amplitude of R, maximum amplitude of T, and corrected Q-T interval (QTc). The QRS voltage was defined as the sum of the S wave in V1 and the R wave in V5 or V6. Statistical Analyses. All results are expressed as mean ⫾ SD. The Student’s unpaired t test was used to compare the differences between TPP and non-TPP. The prevalence of certain ECG parameters between TPP and non-TPP group were analyzed using the ␹2 test. The sensitivity and specificity of significant ECG parameters were calculated, and simultaneous testing was used when multiple significant parameters were combined. Differences were considered significant at P values less than 0.05.

Results Patient Characteristics in TPP and non-TPP Groups. There were 31 patients who had TPP and 23 patients with non-TPP, including SPP (n ⫽ 12), distal renal tubular acidosis (n ⫽ 4), diuretic use (n ⫽ 2), licorice intoxication (n ⫽ 1), primary aldosteronism (n ⫽ 1) and Bartter-like syndrome (n ⫽ 3) (Table 1). The male-to-female ratio was 31:0 in TPP and 19:4 in non-TPP. Age ranged from 19 to 49 years with a mean age of 27 in TPP and 19 to 52 years with mean age of 29 in non-TPP. Forty-five percent of patients with TPP had clinical manifestations of thyrotoxicosis, including weight loss, heat intolerTHE AMERICAN JOURNAL OF THE MEDICAL SCIENCES

Table 2. Biochemical Studies in Patients with TPP and Non-TPP

Plasma pH Na⫹ (mmol/L) K⫹ (mmol/L) Cl– (mmol/L) HCO3– (mmol/L) Phosphate (mg/dL) Calcium (mg/dL) Magnesium (mg/dL) Urea nitrogen (mg/dL) Creatinine (mg/dL) Thyroid function test Triiodothyronine (nmol/L) Thyroxine (nmol/L) Thyroid-stimulating hormone (mU/L)

TPP

Non-TPP

7.40 ⫾ 0.06 140 ⫾ 2.8 2.0 ⫾ 0.6 104 ⫾ 2.2 24 ⫾ 1.7 2.2 ⫾ 1.1* 9.5 ⫾ 0.6 1.9 ⫾ 0.6 15 ⫾ 5.6 1.0 ⫾ 0.2

7.41 ⫾ 0.05 142 ⫾ 2.4 2.1 ⫾ 0.5 102 ⫾ 2.9 25 ⫾ 1.9 3.2 ⫾ 1.0 9.6 ⫾ 0.5 2.0 ⫾ 0.5 16 ⫾ 4.8 1.0 ⫾ 0.2

5.2 ⫾ 2.8* 184 ⫾ 42* ⬍0.03*

1.8 ⫾ 1.0 93 ⫾ 16 1.2 ⫾ 0.5

* P ⬍ 0.05 in TPP versus non-TPP.

ance, palpitations, increased appetite, excitability, and diaphoresis during attacks. Biochemical Results in TPP and non-TPP Groups. As shown in Table 2, triiodothyronine and thyroxine concentrations were 5.2 ⫾ 2.8 nmol/L (reference range, 1.1–2.9 nmol/L) and 184 ⫾ 42 nmol/L (reference range 64 –154 nmol/L), respectively, and thyroid-stimulating hormone level was ⬍ 0.03 mU/L (reference range 0.35–5.0 mU/L) in TPP. Thyroid function tests (triiodothyronine, 1.8 ⫾ 1.0 nmol/L; thyroxine, 93 ⫾ 16 nmol/L; thyroid-stimulating hormone, 1.2 ⫾ 0.5 mU/L) were normal in non-TPP. There were no significant differences in age, sex distribution, and plasma Na⫹, Cl⫺, HCO3⫺, pH, calcium, magnesium, urea nitrogen, creatinine, or K⫹ concentration (2.0 ⫾ 0.6 versus 2.1 ⫾ 0.5 mmol/L) between the TPP and non-TPP groups. However, plasma phosphate concentration (2.2 ⫾ 1.1 versus 3.2 ⫾ 1.0 mg/dL, P ⬍ 0.05) was significantly lower in TPP than non-TPP. ECG Manifestations in TPP and non-TPP Groups. The comparison of ECG measurements is shown in Table 3. Heart rate (96 ⫾ 4 versus 72 ⫾ 3 Table 3. ECG Parameters in Patients with TPP and Non-TPP Variable

TPP

Non-TPP

Rate (beats/min) P wave duration (ms) Height of P (mma) P-R interval (ms) Maximum height of R (mma) Maximum height of T (mma) QRS voltage (mma) QTc (ms)

96 ⫾ 4* 80 ⫾ 4 1.8 ⫾ 0.2 236 ⫾ 11* 22 ⫾ 1.9* 3.4 ⫾ 0.6 39 ⫾ 2* 437 ⫾ 20

72 ⫾ 3 90 ⫾ 9 1.7 ⫾ 0.2 191 ⫾ 8 17 ⫾ 1.1 3.4 ⫾ 0.6 28 ⫾ 2 444 ⫾ 12

* P ⬍ 0.05 when TPP versus non-TPP. a 10 mm ⫽ 1 mV.

129

Electrocardiographic Findings in Thyrotoxic Periodic Paralysis

Table 4. Prevalence of ECG Features in Patients with TPP and non-TPP

Parameters Sinus tachycardia (⬎100 beats/min) Sinus bradycardia (⬍60 beats/min) First-degree AV block (⬎200 ms) Mobitz I second-degree AV block AV block (Total) Increased QTc (⬎430 ms) Increased QRS voltage (V1 ⫹ V5 or V6 ⬎35 mm)† T wave flattening “U” wave ST depression

TPP (%)

Non-TPP (%)

14 (45%)* 1 (3%) 14 (45%)* 3 (10%) 17 (55%)* 17 (55%) 23 (74%)*

0 1 (4%) 3 (13%) 0 3 (13%) 9 (39%) 6 (26%)

21 (68%) 11 (35%) 12 (38%)

16 (70%) 6 (26%) 5 (22%)

* P ⬍0.05 when TPP versus non-TPP. † 10 mm ⫽ 1 mV

beats/min, P ⬍ 0.001), PR interval (236 ⫾ 11 versus 191 ⫾ 8 ms, P ⬍ 0.05), maximum height of R (22 ⫾ 1.9 versus 17 ⫾ 1.1 mm, P ⬍ 0.05), and QRS voltage (39 ⫾ 2 versus 28 ⫾ 2 mm, P ⬍ 0.001) were significantly higher in TPP than non-TPP. Table 4 shows the frequency of ECG parameters and their statistical significance. Elevated QRS voltage was the most common ECG sign (74%) in patients with TPP. Forty-five percent (14 of 31) of TPP patients had first degree atrioventricular (AV) block compared with 13% (3 of 23) in the non-TPP group (P ⬍ 0.05). There were no significant differences in QT shortening, ST depression, U waves, or T flattening between the 2 groups. The sensitivity and specificity of significant ECG parameters as well as their definitions are shown in Table 5. Elevated QRS voltage had the highest sensitivity (74%) with specificity of 87% when used alone. Expanding the criteria to high QRS voltage or sinus tachycardia resulted in sensitivity/specificity of 87%/74%; for sinus tachycardia or AV block, 74%/87%. AV block combined with high QRS voltage had a sensitivity of 90% and a specificity of 65%. The sensitivity and specificity of combining all 3 parameters were 97% and 65%, respectively. Figure 1A shows a “typical” TPP ECG featuring increased QRS voltage mimicking left venTable 5. Sensitivity and Specificity of Significant ECG Parameters in TPP

Parameters Sinus tachycardia AV block (total) High QRS voltage Sinus tachycardia/AV block Sinus tachycardia/High QRS voltage AV block/High QRS voltage Sinus tachycardia/AV block/High QRS voltage

130

Sensitivity (%)

Specificity (%)

45 45 74 74 87 90 97

100 87 87 87 74 65 65

tricular hypertrophy. Another “typical” ECG pattern with sinus tachycardia and first-degree AV block is shown in Figure 1B. Discussion TPP is a hyperthyroidism-related electrolyte and muscle disorder manifesting as recurrent episodes of hypokalemia and muscle weakness.1–3 Although the incidence of the disorder is relatively higher among Asians,1– 4,6 it has been reported in many other racial groups.2,3,5 TPP has been reported to occur in 1.9 to 8.8% of Japanese14,15 and 1.9% of Chinese patients suffering from hyperthyroidism.6 Among nonAsian populations, the incidence rate has ranged from 0.15 to 0.2% in the United States.9 Typically, the first episode occurs between the ages of 20 and 40, and there is a male predominance.1,4 –7,10 Similar to previous findings, all 31 patients with TPP in this current series were male with a mean age of 27 years. The clinical symptoms of hyperthyroidism are usually subtle before the onset of paralysis. It has been reported that many patients with TPP do not have any symptoms related to hyperthyroidism during an attack.1,6 –9 In this study, 55% (17/31) of patients were clinically silent at presentation. Therefore, TPP may be misdiagnosed as non-TPP unless special laboratory tests for thyroid function are available in the ED. The cardiovascular system is very sensitive to increased levels of thyroid hormone, and we believe that electrocardiographic changes in patients with TPP provide helpful diagnostic clues.11–13 Although the electrocardiographic features of hypokalemia are well known, the ECGs of TPP patients are somewhat different. To our knowledge, few studies describe electrocardiographic manifestations in TPP.6,16 In a study of 30 patients with TPP, the electrocardiographic changes were mainly attributed to thyrotoxicosis and hypokalemia. In addition to the typical changes of hypokalemia (U waves, T wave flattening, ST depression, and QT prolongation), sinus arrest occurred in 2 patients and seconddegree AV block developed in 3 patients.16 The other study by McFadzean et al6 found prolongation of the P-R interval in 27%, and among 25 Chinese TPP patients, 3 patients developed right bundle-branch block. Both of these studies demonstrated conduction disturbances in patients with TPP. In this study, the incidence of conduction abnormalities is much higher than in the previous studies. Fourteen of 31 (45%) patients with TPP had first-degree AV block, 3 patients (10%) had second-degree AV block, and 1 patient had right bundle-branch block. The mechanism of conduction disturbances in TPP remains unclear. In this study, plasma phosphate concentrations were significantly lower in TPP.17–19 Although hypophosphatemia has been reported to cause ventricular tachycardia and impaired cardiac September 2003 Volume 326 Number 3

Hsu et al

Figure 1. (A) 12-lead ECG showing “typical” features of TPP with increased QRS voltage mimicking left ventricular hypertrophy. (B) 12-lead ECG showing “typical” features of TPP with sinus tachycardia and first-degree AV block.

contractility, no conduction abnormalities have been observed,20 diminishing the role of hypophosphatemia in the conduction disturbances. Perhaps, inflammation of the conduction system as a result of thyrotoxicosis is involved in this defect.11–13 High QRS voltage is a common ECG finding in patients with hyperthyroidism.11–13,21 In this study, THE AMERICAN JOURNAL OF THE MEDICAL SCIENCES

23 patients (74%) had increased QRS voltage, considered a diagnostic clue in patients with TPP. This finding may be the result of thyrotoxicosis itself or the effect of thyrotoxicosis-related left ventricular hypertrophy. Furthermore, the sensitivity of high QRS voltage increased significantly when used in combination with other ECG parameters, such as 131

Electrocardiographic Findings in Thyrotoxic Periodic Paralysis

sinus tachycardia and AV block. This unique characteristic has not been emphasized in previous studies of TPP. Thus, this simple test may be used to guide management decisions in cases of clinical suspicion. The definitive cure for TPP is achievement of euthyroid state by medical, surgical, or radioactive iodine therapy. Although potassium chloride (KCl) supplementation is generally practiced to hasten strength recovery during severe attacks, excessive KCl administration may lead to rebound hyperkalemia in patients with TPP.10 Differentiation between TPP and non-TPP may prevent the overly aggressive treatment of an apparent K⫹ deficiency, leading to potentially life-threatening rebound hyperkalemia on recovery. Manoukain et al10 suggested that KCl replacement should not exceed 90 mmol per 24 hours. Because hyperadrenergic activity is implicated in the pathogenesis of TPP, nonselective ␤-adrenergic receptor blockers may be an alternative therapy.1,22 Some reports suggest that TPP can be rapidly terminated with intravenous or oral propranolol.23–26 Acetazolamide treatment has been reported to be effective in patients with familial periodic paralysis or SPP, but it may aggravate clinical symptoms in patients with TPP.27,28 With the increasing number of immigrants from Asia, TPP will only become more commonplace in the EDs of Western countries. Distinctive ECG features, including rapid heart rate, high QRS voltage, and first-degree AV block are clinical clues favoring the diagnosis of TPP when patients present to the ED with HP. Early diagnosis and prompt treatment of TPP will prevent misdiagnosis and improper management of this curable disorder. References 1. Lin SH, Lin YF, Halperin ML. Hypokalemia and paralysis. QJM 2001;94:133–9. 2. Stedwell RE, Allen KM, Binder LS. Hypokalemic paralyses: a review of the etiologies, pathophysiology, presentation and therapy. Am J Emerg Med 1992;10:143– 6. 3. Ahlawat SK, Sachdev A. Hypokalemic paralysis: Postgrad Med 1999;75:193–7. 4. Ko GTC, Chow CC, Yeung VTF, et al. Thyrotoxic periodic paralysis in a Chinese population. QJM 1996;89:463– 8. 5. Ober KP. Thyrotoxic periodic paralysis in the United States: report of 7 cases and review of the literature. Medicine 1992; 71:109 –20. 6. McFadzean AJS, Yeung R. Periodic paralysis complicating thyrotoxicosis in Chinese. Br Med J 1967;1:451–5.

132

7. Okinaka S, Shizume K, Iino S, et al. The association of periodic paralysis and hyperthyroidism in Japan. J Clin Endocrinol Metab 1957;17:1454 –9. 8. Engel AG. Thyroid function and periodic paralysis. Am J Med 1961;30:327–33. 9. Kelley DE, Gharib H, Kennedy FP, et al. Thyrotoxic periodic paralysis. Report of 10 cases and review of electromyographic findings. Arch Intern Med 1989;149:2597– 600. 10. Manoukain MA, Foote JA, Crapo LM. Clinical and metabolic features of thyrotoxic periodic paralysis in 24 episodes. Arch Intern Med 1999;159:601– 6. 11. Sandler G. The effect of thyrotoxicosis on the electrocardiogram. Br Heart J 1959;21:111– 6. 12. Gordan G, Soley MH, Chamberlain FL. Electrocardiographic features associated with hyperthyroidism. Arch Intern Med 1944;73:148 –53. 13. Hoffman I, Lowrey RD. The electrocardiogram in thyrotoxicosis. Am J Cardiol 1960;6:893–904. 14. Okihiro MM, Nordyke RA. Hypokalemic periodic paralysis: experimental precipitation with sodium liothyronine. JAMA 1966;198:949 –51. 15. Satoyoshi E, Murakami K, Kowa H, et al. Periodic paralysis in hyperthyroidism. Neurology 1963;13:746 –52. 16. Ee B, Cheah JS. Electrocardiographic changes in thyrotoxic periodic paralysis. J Electrocardiol 1979;12:263–77. 17. Guthrie GP, Curtis JJ, Beilman KM. Hypophosphatemia in thyrotoxic periodic paralysis. Arch Intern Med 1978;138: 1284 –5. 18. Nora NA, Berns AS. Hypokalemic, hypophosphatemic thyrotoxic periodic paralysis. Am J Kidney Dis 1989;3:247–9. 19. Norris KC, Levine B, Ganesan K. Thyrotoxic periodic paralysis associated with hypokalemia and hypophosphatemia. Am J Kidney Dis 1996;28:270 –3. 20. Clark WR. Ventricular tachycardia associated with hypophosphatemia. Nutr Int 1985;11:102– 6. 21. Slovis C, Jenkins R. ABC of clinical electrocardiography: conditions not primarily affecting the heart. Br Med J 2002; 324:1320 –3. 22. Yeung RT, Tse TF. Thyrotoxic periodic paralysis: effect of propranolol. Am J Med 1974;57:584 –90. 23. Shayne P, Hart A. Thyrotoxic periodic paralysis terminated with intravenous propranolol. Ann Emerg Med 1994;24:736 – 40. 24. Birkhahn RH, Gaeta TJ, Melniker L. Thyrotoxic periodic paralysis and intravenous propranolol in the emergency setting. J Emerg Med 2000;18:199 –202. 25. Lin SH, Lin YF. Propranolol rapidly terminates the hypokalemia, hypophosphatemia and paralysis in thyrotoxic periodic paralysis. Am J Kidney Dis 2001;37:620 –3. 26. Haung TY, Lin SH. Thyrotoxic hypokalemic periodic paralysis reversed by propranolol without rebounding hyperkalemia. Ann Emerg Med 2001;37:415– 6. 27. Resnick JS, Engel WK, Griggs RC, et al. Acetazolamide prophylaxis in hypokalemic periodic paralysis. N Engl J Med 1968;278:582– 6. 28. Griggs RC, Engel WK, Resnick JS. Acetazolamide treatment of hypokalemic periodic paralysis. Prevention of attacks and improvement of persistent weakness. Ann Intern Med 1970;73:39 – 48.

September 2003 Volume 326 Number 3