- Email: [email protected]
Electrocardiographic changes simulating acute myocardial infarction caused by hyperkalemia: Report of a patient with normal coronary arteriograms
Electrocardiographic changes simulating acute myocardial infarction caused by hyperkalemia: Report of a patient with coronary arteriograms
Kamal K. Chawla, M.D., F.A.C.C., Jairo Cruz, M.D.** Neil E. Kramer, M.D., F.A.C.C.*** William D. Towne, M.D., F.A.C.C., Chicago, Ill.
F.R.C.P.
(C)*
F.A.C.P.,
F.C.C.P***”
Electrophysiological effects of moderate to severe hyperkalemia and its effect on the electrocardiogram have been well described by various authors.1-s Shortening of QT interval, tenting of T waves, lengthening of PR interval, loss of P wave causing “sinoventricular rhythm,” and finally broadening of QRS and “sine wave” are some of the more commonly described abnormalities.’ ST segment changes resembling a “current of injury” have been rarely reported.‘. 5 However, in none of the previously reported cases was coronary artery disease ruled out with any certainty. We observed a patient with diabetic ketoacidosis and hyperkalemia who showed marked ST segment elevations resembling acute anterior wall myocardial infarction. Treatment of hyperkalemia resulted in prompt return of the ECG towards normal. Subsequent work-up including exercise testing and selective coronary arteriography ruled out any significant coronary artery From the Division Chicago, Ill.
of
Adult
Cardiology,
Received
for publication
May
23, 1977.
Accepted
for publication
July
12, 1977.
Reprint requests: Kamal K. Chawla, ogy, 1825 W. Harrison St., Chicago,
Cook
M.D., Division Ill. 60612.
County
of Adult
Hospital,
Cardiol-
*Attending Physician, Division of Adult Cardiology and Electrocardiography and Human Performance Laboratory, County Hospital, Chicago, Ill.
Director, Cook
**Fellow, Chicago,
Hospital,
Division
of
Adult
Cardiology,
Cook
County
Ill.
***Attending Noninvasive ****Chairman, Chicago, Ill.
Physician, Laboratory, Division
Division of Adult Cardiology and Director, Cook County Hospital, Chicago, Ill. of Adult
0002-8703/78/0595-0837$00.40/O
Cardiology,
Cook
County
normal
Hospital,
0 1978 The C. V. Mosby Co.
disease, suggesting that the ECG changes were probably caused by hyperkalemia. Case report G. C., a 48-year-old black male, was admitted to Cook County Hospital because of diabetic ketoacidosis. The patient was a known diabetic for the last 12 years, controlled with 45 units of NPH insulin. Two weeks prior to admission he had, developed mild fever and a non-productive cough. He had also noted fatigue, weakness, and dyspnea on moderate exertion. He also complained of nausea and had vomited a few times. The patient denied any history of chest pain. In the past he had been treated for pulmonary tuberculosis. There was no history of hypertension. The patient had smoked 20 to 40 cigarettes a day for about 30 years. On examination the patient was a moderately built although somewhat undernourished black male in mild respiratory distress. He was drowsy but well oriented. The heart rate was 120 beats/minute, blood pressure 150180 mm. Hg, temperature 99” F. and respiratory rate 28/minute. The skin was dry and neck veins were flat. CVP measured 1.0 cm. of H,O. The lungs showed slightly decreased breath sounds in right upper zone. Examination of the heart was entirely normal. Specifically, no S,, S,, or friction rub was heard by any examiner. Initial laboratory data showed marked hyperkalemia, acidosis, hyperglycemia and ketonemia compatible with the picture of severe diabetic ketoacidosis (see Table I). The ECG on admission showed marked ST segment elevation in Leads V,, V,, and avL suggestive of acute anterior wall injury. There were also tall peaked T waves in the Leads V, through V, compatible with hyperkalemia (see Fig. 1, upper-panel). With the presumptive diagnosis of acute myocardial infarction the patient was transferred to the ICCU where close monitoring and standard treatment ‘of acute diabetic ketoacidosis consisting of regular insulin, sodium bicarbonate, and fluid replacement was instituted. Serial ECGs and enzymes determinations were obtained. Over the next 24 to 48 hours the patient showed marked improvement in his general condition as well as in the biochemical data. His blood sugar came down
American Heart Journal
637
Chawla
et al.
I
II
Ill
avl
11.8.76.
1~19 A.M.
avR Serum
11.8.76.
7~20A.M.
Swum K*4.3
11.10.76.
7~07 A.M.
K+4.0 Serum r/ _bt --f--y
avF
“1
“2
“3
K* 8.3 meq/L
maq/L
h
Fig. 1. Serial electrocardiograms. The top panel represents a twelve-lead ECG on admission showing marked ST segment elevations in Leads V,, V,, and avL. Middlepanel shows same twelve leads six hours later when the serum potassium has come down to 4.3 mEq./L. Notice the marked improvement in the ST segment elevation and T wave abnormality. Lower panel recorded two days later shows normal ECG when serum K+ is 4.0 mEq./L. from 840 to 452 mg per cent. The serum K’ came down from 8.3 to 4.0 mEq./L. and the pH increased from 7.09 to 7.44 (see Table I). The electrocardiogram taken six hours after admission when the serum K+ had came down to 4.3 mEq./L. showed marked decrease in the ST segment elevation. However, slight J point and early ST segment elevation and some T wave tenting persisted (Fig. 1, middle panel). Subsequent ECGs, two and three days later, were completely within normal limits (Fig. 1, lower panel). Slight QRS prolongation, which had been noted on the admission ECG, was absent in subsequent tracings. The frontal plane QRS axis remained unchanged throughout the patient’s course. The cardiac enzymes remained grossly within normal limits. The patient made an uneventful recovery and was placed back on NPH insulin. Ten days later a treadmill exercise test was performed using the Bruce protocol. The patient reached the predicted level of 6 to 7 mets and a maximal heart rate of 150 beats/minute (87 per cent of maximum predicted for the age). The ECG did not show abnormal ST segment changes or arrhythmia and the patient did not complain of chest pain. The test was considered negative for ischemic response. Cardiac catheterization was performed two weeks after the admission and selective coronary arteriograms were obtained in multiple views. Both the left and right coronary arteries and their major branches were visualized and appeared completely free of any obstructive lesions (Fig. 2). Left
636
ventricular function of 69 per cent.
was also normal
with
an ejection
fraction
Discussion
Increased plasma potassium may occur spontaneously or iatrogenically. Some of the more frequent clinical causes include renal failure, untreated diabetic ketoacidosis, adrenocortical insufficiency, and over-enthusiastic therapeutic use of potassium salts or drugs with high potassium content. An increase in the concentration of extracellular potassium lowers the resting membrane potential, leading to a reduction of the upstroke velocity, amplitude, and duration of the action potential. This effect has been recorded in atrial, ventricular, and Purkinje fibers and may be manifested on the ECG by changes in the amplitude and duration of the P wave and QRS complex.“, 6 The mechanism of T wave and ST segment changes is not well understood and has not been well correlated with action potential abnormalities.
May, 1978, Vol. 95, No. 5
EC’G changes
Fig. 2. Coronary arteriograms. right coronary artery in the left
Table
I. Laboratory
A, Left coronary anterior
oblique
artery in the left anterior oblique (60 degree) (40 degree) projection, showing all branches
Serum glucose (rngm.%)
Time
s. bicarb. (mEq./L.)
A B
11/B/76 11/B/76
1:20 7:10
A.M. A.M.
8.3 4.3
7.09 7.35
840 348
6.8 14.1
C
11/10/76
790
A.M.
4.0
7.44
452
25.8
The hyperkalemia of ketoacidosis differs from that of other causes in that there is a shift of potassium out of the cell, resulting in an altered ratio of intracellular to extracelluar potassium.‘. R This may contribute in part to the severity of the ST segment elevations in the setting of diabetic ketoacisosis. It is also possible that the associated severe acidosis and marked hyperglycemia present in our patient may have added to the effect of hyperkalemia in producing the observed ECG changes. Electrocardiographic abnormalities resembling myocardial infarction in the presence of severe hyperkalemia have been reported by several Q waves as observers in the past. 1-SPathological well as ST-T changes resembling acute injury pattern have been described separately. The appearance of pathological Q waves has generally been attributed to altered initial QRS forces secondary to intraventricular conduction abnormalities resulting from hyperkalemia.3 Levine and associate9 reported four cases of acute renal failure with hyperkalemia and ST-T patterns suggesting myocardial infarction. In all
Heart
projection and B, widely patent.
data
Date
American
caused by hyperkalemia
Journal
ECG Injury pattern and tented T waves Slight ST elevation and tented T waves Normal
of these patients lowering of the serum potassium level by hemodialysis was associated with return of the electrocardiogram towards normal. However, in two of the four patients, fibrinous pericarditis was noted to be present at autopsy and this may have had an effect on the ECG of these two patients. In the third patient a left bundle branch block pattern was present which, of course, renders the ECG diagnosis of acute myocardial infarction difficult. The fourth patient was not autopsied so that neither coronary artery disease nor pericarditis, both of which can be associated with renal failure, were ruled out. Gelzayd and Holzmar? reported a case of diabetic ketoacidosis with hyperkalemia presenting with ECG abnormalities similar to that of our patient. However, neither coronary arteriography nor exercise testing was reported as performed in their patient. Lichstein and colleagues’ reported two patients with hyperkalemia induced fascicular block, one of whom had marked ST segment elevation in Leads V, and V, similar to those of our patient. The resemblance of the ST-T changes to those of
639
Chawla
et al.
acute anteroseptal myocardial infarction is evident from their Fig. 1 although the authors do not call attention to this point. The pattern is described as showing right bundle branch block and left anterior hemiblock. However, these conduction abnormalities, unlike left bundle branch block, do not by themselves explain the observed ST-T changes. The findings in our patient reemphasize the previously reported observation that hyperkalemia may produce marked ST-T changes resembling those of acute myocardial infarction. To the best of our knowledge, there is no previous report in the literature of a patient with the hyperkalemia-induced ECG changes of acute myocardial infarction in whom coronary artery disease was judged unlikely on the basis of negative exercise electrocardiography or ruled out by normal selective coronary arteriograms. We would conclude that hyperkalemia should be considered in the differential diagnosis of ST-T changes resembling acute myocardial infarction, especially when these occur in those clinical settings where hyperkalemia is frequently observed. Summary
A patient is described with severe diabetic ketoacidosis and hyperkalemia who presented with an ECG resembling an acute anterior wall myocardial infarction, Treatment of hyperkalemia resulted in prompt return of the ECG towards normal. Subsequent work-up including exercise testing and selective coronary arteriography ruled out any significant coronary artery
640
disease suggesting that the ECG changes were probably caused by hyperkalemia. While similar changes have rarely been described in the past, this would appear to be the first such case in whom coronary artery diseasewas ruled out by a negative exercise testing and coronary arteriography . We thank assistance.
Miss
Suzanne
Adams
for her expert
secretarial
REFERENCES 1.
2.
3.
4.
5.
6.
7.
8.
Ettinger, P. O., Regan, T. J., and Oldewurtel, H. A.: Hyperkalemia, cardiac conduction, and the electrocardiogram: A review, AM. HEART J. 88:360, 1974. Gelzayd, E. A. and Holzman, D.: Electrocardiographic changes of hyperkalemia simulating acute myocardial infarction: Report of a case, Dis. Chest 51:211, 1967. Arnsdorf, M. F.: Electrocardiogram in hyperkalemia: Electrocardiographic pattern of anteroseptal myocardial infarction mimicked by hyperkalemia-induced disturbance of impulse conduction, Arch. Intern. Med. 136:1161, 1976. Lichstein, E., Gupta, P. K., and Grumwald, A. A.: Hyperkalemic fascicular block: Electrocardiogram of the month, Chest 70:290, 1976. Levine, H. D., Wanzer, S. H., and Merrill, J. P.: Dialyzable currents of injury in potassium intoxication resembling acute myocardial infarction or pericarditis, Circulation 13:29, 1956. Fisch, C., Knoebel, S. B., Feigenbaum, H., and Greenspan, K.: Potassium and the monophasic action potential: Electrocardiogram, conduction and arrhythmias, Progr. Cardiovasc. Dis. 8:387, 1966. Cohen, H. C., Rosen, K. M., and Pick, A.: Disorders of impulse conduction and impulse formation caused by hyperkalemia in man: Case report, AM. HEART J. 89:501, 1975. Nadler, C. S., Bellet, S., and Lanning, M.: Influence of the serum potassium and other electrolytes on the electrocardiogram in diabetic acidosis, Am. J. Med. 5:838, 1948.
May, 1978, Vol. 95, No. 5
Kamal K. Chawla, M.D., F.A.C.C., Jairo Cruz, M.D.** Neil E. Kramer, M.D., F.A.C.C.*** William D. Towne, M.D., F.A.C.C., Chicago, Ill.
F.R.C.P.
(C)*
F.A.C.P.,
F.C.C.P***”
Electrophysiological effects of moderate to severe hyperkalemia and its effect on the electrocardiogram have been well described by various authors.1-s Shortening of QT interval, tenting of T waves, lengthening of PR interval, loss of P wave causing “sinoventricular rhythm,” and finally broadening of QRS and “sine wave” are some of the more commonly described abnormalities.’ ST segment changes resembling a “current of injury” have been rarely reported.‘. 5 However, in none of the previously reported cases was coronary artery disease ruled out with any certainty. We observed a patient with diabetic ketoacidosis and hyperkalemia who showed marked ST segment elevations resembling acute anterior wall myocardial infarction. Treatment of hyperkalemia resulted in prompt return of the ECG towards normal. Subsequent work-up including exercise testing and selective coronary arteriography ruled out any significant coronary artery From the Division Chicago, Ill.
of
Adult
Cardiology,
Received
for publication
May
23, 1977.
Accepted
for publication
July
12, 1977.
Reprint requests: Kamal K. Chawla, ogy, 1825 W. Harrison St., Chicago,
Cook
M.D., Division Ill. 60612.
County
of Adult
Hospital,
Cardiol-
*Attending Physician, Division of Adult Cardiology and Electrocardiography and Human Performance Laboratory, County Hospital, Chicago, Ill.
Director, Cook
**Fellow, Chicago,
Hospital,
Division
of
Adult
Cardiology,
Cook
County
Ill.
***Attending Noninvasive ****Chairman, Chicago, Ill.
Physician, Laboratory, Division
Division of Adult Cardiology and Director, Cook County Hospital, Chicago, Ill. of Adult
0002-8703/78/0595-0837$00.40/O
Cardiology,
Cook
County
normal
Hospital,
0 1978 The C. V. Mosby Co.
disease, suggesting that the ECG changes were probably caused by hyperkalemia. Case report G. C., a 48-year-old black male, was admitted to Cook County Hospital because of diabetic ketoacidosis. The patient was a known diabetic for the last 12 years, controlled with 45 units of NPH insulin. Two weeks prior to admission he had, developed mild fever and a non-productive cough. He had also noted fatigue, weakness, and dyspnea on moderate exertion. He also complained of nausea and had vomited a few times. The patient denied any history of chest pain. In the past he had been treated for pulmonary tuberculosis. There was no history of hypertension. The patient had smoked 20 to 40 cigarettes a day for about 30 years. On examination the patient was a moderately built although somewhat undernourished black male in mild respiratory distress. He was drowsy but well oriented. The heart rate was 120 beats/minute, blood pressure 150180 mm. Hg, temperature 99” F. and respiratory rate 28/minute. The skin was dry and neck veins were flat. CVP measured 1.0 cm. of H,O. The lungs showed slightly decreased breath sounds in right upper zone. Examination of the heart was entirely normal. Specifically, no S,, S,, or friction rub was heard by any examiner. Initial laboratory data showed marked hyperkalemia, acidosis, hyperglycemia and ketonemia compatible with the picture of severe diabetic ketoacidosis (see Table I). The ECG on admission showed marked ST segment elevation in Leads V,, V,, and avL suggestive of acute anterior wall injury. There were also tall peaked T waves in the Leads V, through V, compatible with hyperkalemia (see Fig. 1, upper-panel). With the presumptive diagnosis of acute myocardial infarction the patient was transferred to the ICCU where close monitoring and standard treatment ‘of acute diabetic ketoacidosis consisting of regular insulin, sodium bicarbonate, and fluid replacement was instituted. Serial ECGs and enzymes determinations were obtained. Over the next 24 to 48 hours the patient showed marked improvement in his general condition as well as in the biochemical data. His blood sugar came down
American Heart Journal
637
Chawla
et al.
I
II
Ill
avl
11.8.76.
1~19 A.M.
avR Serum
11.8.76.
7~20A.M.
Swum K*4.3
11.10.76.
7~07 A.M.
K+4.0 Serum r/ _bt --f--y
avF
“1
“2
“3
K* 8.3 meq/L
maq/L
h
Fig. 1. Serial electrocardiograms. The top panel represents a twelve-lead ECG on admission showing marked ST segment elevations in Leads V,, V,, and avL. Middlepanel shows same twelve leads six hours later when the serum potassium has come down to 4.3 mEq./L. Notice the marked improvement in the ST segment elevation and T wave abnormality. Lower panel recorded two days later shows normal ECG when serum K+ is 4.0 mEq./L. from 840 to 452 mg per cent. The serum K’ came down from 8.3 to 4.0 mEq./L. and the pH increased from 7.09 to 7.44 (see Table I). The electrocardiogram taken six hours after admission when the serum K+ had came down to 4.3 mEq./L. showed marked decrease in the ST segment elevation. However, slight J point and early ST segment elevation and some T wave tenting persisted (Fig. 1, middle panel). Subsequent ECGs, two and three days later, were completely within normal limits (Fig. 1, lower panel). Slight QRS prolongation, which had been noted on the admission ECG, was absent in subsequent tracings. The frontal plane QRS axis remained unchanged throughout the patient’s course. The cardiac enzymes remained grossly within normal limits. The patient made an uneventful recovery and was placed back on NPH insulin. Ten days later a treadmill exercise test was performed using the Bruce protocol. The patient reached the predicted level of 6 to 7 mets and a maximal heart rate of 150 beats/minute (87 per cent of maximum predicted for the age). The ECG did not show abnormal ST segment changes or arrhythmia and the patient did not complain of chest pain. The test was considered negative for ischemic response. Cardiac catheterization was performed two weeks after the admission and selective coronary arteriograms were obtained in multiple views. Both the left and right coronary arteries and their major branches were visualized and appeared completely free of any obstructive lesions (Fig. 2). Left
636
ventricular function of 69 per cent.
was also normal
with
an ejection
fraction
Discussion
Increased plasma potassium may occur spontaneously or iatrogenically. Some of the more frequent clinical causes include renal failure, untreated diabetic ketoacidosis, adrenocortical insufficiency, and over-enthusiastic therapeutic use of potassium salts or drugs with high potassium content. An increase in the concentration of extracellular potassium lowers the resting membrane potential, leading to a reduction of the upstroke velocity, amplitude, and duration of the action potential. This effect has been recorded in atrial, ventricular, and Purkinje fibers and may be manifested on the ECG by changes in the amplitude and duration of the P wave and QRS complex.“, 6 The mechanism of T wave and ST segment changes is not well understood and has not been well correlated with action potential abnormalities.
May, 1978, Vol. 95, No. 5
EC’G changes
Fig. 2. Coronary arteriograms. right coronary artery in the left
Table
I. Laboratory
A, Left coronary anterior
oblique
artery in the left anterior oblique (60 degree) (40 degree) projection, showing all branches
Serum glucose (rngm.%)
Time
s. bicarb. (mEq./L.)
A B
11/B/76 11/B/76
1:20 7:10
A.M. A.M.
8.3 4.3
7.09 7.35
840 348
6.8 14.1
C
11/10/76
790
A.M.
4.0
7.44
452
25.8
The hyperkalemia of ketoacidosis differs from that of other causes in that there is a shift of potassium out of the cell, resulting in an altered ratio of intracellular to extracelluar potassium.‘. R This may contribute in part to the severity of the ST segment elevations in the setting of diabetic ketoacisosis. It is also possible that the associated severe acidosis and marked hyperglycemia present in our patient may have added to the effect of hyperkalemia in producing the observed ECG changes. Electrocardiographic abnormalities resembling myocardial infarction in the presence of severe hyperkalemia have been reported by several Q waves as observers in the past. 1-SPathological well as ST-T changes resembling acute injury pattern have been described separately. The appearance of pathological Q waves has generally been attributed to altered initial QRS forces secondary to intraventricular conduction abnormalities resulting from hyperkalemia.3 Levine and associate9 reported four cases of acute renal failure with hyperkalemia and ST-T patterns suggesting myocardial infarction. In all
Heart
projection and B, widely patent.
data
Date
American
caused by hyperkalemia
Journal
ECG Injury pattern and tented T waves Slight ST elevation and tented T waves Normal
of these patients lowering of the serum potassium level by hemodialysis was associated with return of the electrocardiogram towards normal. However, in two of the four patients, fibrinous pericarditis was noted to be present at autopsy and this may have had an effect on the ECG of these two patients. In the third patient a left bundle branch block pattern was present which, of course, renders the ECG diagnosis of acute myocardial infarction difficult. The fourth patient was not autopsied so that neither coronary artery disease nor pericarditis, both of which can be associated with renal failure, were ruled out. Gelzayd and Holzmar? reported a case of diabetic ketoacidosis with hyperkalemia presenting with ECG abnormalities similar to that of our patient. However, neither coronary arteriography nor exercise testing was reported as performed in their patient. Lichstein and colleagues’ reported two patients with hyperkalemia induced fascicular block, one of whom had marked ST segment elevation in Leads V, and V, similar to those of our patient. The resemblance of the ST-T changes to those of
639
Chawla
et al.
acute anteroseptal myocardial infarction is evident from their Fig. 1 although the authors do not call attention to this point. The pattern is described as showing right bundle branch block and left anterior hemiblock. However, these conduction abnormalities, unlike left bundle branch block, do not by themselves explain the observed ST-T changes. The findings in our patient reemphasize the previously reported observation that hyperkalemia may produce marked ST-T changes resembling those of acute myocardial infarction. To the best of our knowledge, there is no previous report in the literature of a patient with the hyperkalemia-induced ECG changes of acute myocardial infarction in whom coronary artery disease was judged unlikely on the basis of negative exercise electrocardiography or ruled out by normal selective coronary arteriograms. We would conclude that hyperkalemia should be considered in the differential diagnosis of ST-T changes resembling acute myocardial infarction, especially when these occur in those clinical settings where hyperkalemia is frequently observed. Summary
A patient is described with severe diabetic ketoacidosis and hyperkalemia who presented with an ECG resembling an acute anterior wall myocardial infarction, Treatment of hyperkalemia resulted in prompt return of the ECG towards normal. Subsequent work-up including exercise testing and selective coronary arteriography ruled out any significant coronary artery
640
disease suggesting that the ECG changes were probably caused by hyperkalemia. While similar changes have rarely been described in the past, this would appear to be the first such case in whom coronary artery diseasewas ruled out by a negative exercise testing and coronary arteriography . We thank assistance.
Miss
Suzanne
Adams
for her expert
secretarial
REFERENCES 1.
2.
3.
4.
5.
6.
7.
8.
Ettinger, P. O., Regan, T. J., and Oldewurtel, H. A.: Hyperkalemia, cardiac conduction, and the electrocardiogram: A review, AM. HEART J. 88:360, 1974. Gelzayd, E. A. and Holzman, D.: Electrocardiographic changes of hyperkalemia simulating acute myocardial infarction: Report of a case, Dis. Chest 51:211, 1967. Arnsdorf, M. F.: Electrocardiogram in hyperkalemia: Electrocardiographic pattern of anteroseptal myocardial infarction mimicked by hyperkalemia-induced disturbance of impulse conduction, Arch. Intern. Med. 136:1161, 1976. Lichstein, E., Gupta, P. K., and Grumwald, A. A.: Hyperkalemic fascicular block: Electrocardiogram of the month, Chest 70:290, 1976. Levine, H. D., Wanzer, S. H., and Merrill, J. P.: Dialyzable currents of injury in potassium intoxication resembling acute myocardial infarction or pericarditis, Circulation 13:29, 1956. Fisch, C., Knoebel, S. B., Feigenbaum, H., and Greenspan, K.: Potassium and the monophasic action potential: Electrocardiogram, conduction and arrhythmias, Progr. Cardiovasc. Dis. 8:387, 1966. Cohen, H. C., Rosen, K. M., and Pick, A.: Disorders of impulse conduction and impulse formation caused by hyperkalemia in man: Case report, AM. HEART J. 89:501, 1975. Nadler, C. S., Bellet, S., and Lanning, M.: Influence of the serum potassium and other electrolytes on the electrocardiogram in diabetic acidosis, Am. J. Med. 5:838, 1948.
May, 1978, Vol. 95, No. 5