Occurrence of Hypokalemia in Suspected Acute Myocardial Infarction and Its Relation to Clinical History and Clinical Course

Transcription

1 Clin. Cardiol. 11, (1988) Occurrence of Hypokalemia in Suspected Acute Myocardial Infarction and Its Relation to Clinical History and Clinical Course J. HERLITZ, M.D., Ph.D., A. HJALMARSON, M.D., Ph.D., A. BENOTSON, R.N. Department of Medicine I, Sahlgren s Hospital, University of Goteborg, Goteborg, Sweden Summary: In 1350 patients with suspected acute myocardial infmtion, serum potassium levels during the first 3 days in hospital was correlated to clinical history and clinical course. A higher incidence of hypokalemia was observed in women, in patients with hypertension, and in those on chronic diuretic treatment. Patients with anterior infarction had a higher incidence of hypokalemia than those with inferior infarction, as did patients with large as compared with small infarcts. No clear difference was observed between patients whose infarction was confirmed and those in whom the diagnosis was not confirmed. Independent predictors for hypokalemia were treatment with diuretics before admission to hospital, infarct size, and female sex. Hypokalemia during the first 3 days of hospitalization was associated with the occurrence of severe ventricular arrhythmias during hospitalization, but not with survival during a 5-year follow-up. Key words: suspected acute myocardial infarction, potassium Introduction Hypokalemia has been reported to occur in a high percentage of patients with suspected acute myocardial infarction,1-3 and has been shown to increase the risk of severe ventricular arrhythmia^.^.^ Increased serum concentrations of catecholamines in acute myocardial Address for reprints: Johan Herlitz, M.D. Department of Medicine 1 Sahlgrenska Hospital Goteborg, Sweden Received: March 12, 1988 Accepted: May 26, 1988 have been suggested to be the main contributor to hypokalemia. The aim of this study was to relate serum levels of potassium in the initial phase of suspected acute myocardial infarction to the clinical history and clinical course of the syndrome. The main results from this study have been reported previously.9 Patients All patients participated in a double-blind trial evaluating the effect of the betal-selective blocker metoprolol during the first 3 months in suspected acute myocardial infarction (AMI). They were admitted to the coronary care units (CCU) at Sahlgren s Hospital and Ostra Hospital in Goteborg and at Skovde Hospital. Inclusion criteria for participation were: (a) age < 75 years; (b) chest pain with a duration of more than 30 min and/or electrocardiographic (ECG) signs indicating acute myocardial infarction; (c) onset of symptoms within the previous 48 h. The cardiovascular exclusion criteria on admission to hospital were: heart rate < 45 beatshin; systolic blood pressure < 100 mmhg; AV block (PR interval s); severe congestive heart failure (des auscultated > 10 cm above the lung bases). There were some further exclusion criteria which have been presented el~ewhere.~ As soon as possible after admission, intravenous metoprolol 5 mg was administered and scheduled for t.i.d. dosage, followed by oral treatment 50 mg q.h.d. for 2 days and then 100 mg b.i.d. for 3 months. Placebo was given according to the same protocol. After the first 3 months, all patients received metoprolol, considering established contraindications, with the exception of those in whom there was no suspicion of ischemic heart disease. Among 5-year survivors, about 50% were still on beta blockers. Methods For the purposes of this study the definition of myocardial infarction was the presence of at least two of the fol-

2 J. Herlitz et al.: Hypokalemia in AM1 679 TABLE I Occurrence of hypokalemia in the emergency ward in various subgroups Age 5 61 years (median) (690)" > 61 years (628) Sex Men (997) Women (321) Clinical history Previous infarction Yes (292) No (1026) Previous angina pectoris Yes (458) No (855) Previous hypertension Yes (391) No (927) On chronic treatment Diuretics Yes (247) No (1067) Beta blockers Yes (334) No (984) Initial heart rate 5 70 beatslmin (median) (666) > 70 beatslmin (651) Infarct size Heat-stable LDmax I 15 pkatll (median) (382) Heat-stable LDmax > 15 pkatll (371) Infarct site Anterior infarction (328) Inferior infarction (285) Delayb I 3 h (median) (719) > 3 h (585) " = Figures in parentheses = number of patients. = Delay between onset of symptoms and arrival in hospital. Hypokalemia (%) P c 0.01 c in hospital to the occurrence of severe ventricular arrhythmias during hospitalization. Hypokalemia in Relation to 5-Year Survival As shown in Table IV, the 5-year mortality rate was similar in patients with hypokalemia in the emergency ward and during the first 3 days in hospital, when compared with patients with no hypokalemic episodes. Discussion Hypokalemia is a fairly common observation in patients with acute myocardial infarction. In this study, we have tried to describe not only the total occurrence of hypokalemia in a population with suspected acute myocardial infarction, but also to relate its occurrence to the clinical history, to infarct development, the estimated size and site of the infarction, and to the prognosis. The patient population is collected from an early intervention study with metoprolol in suspected acute myocardial infarction. Although a large proportion of patients were excluded from participation, the main exclusion criterion was administrative. With the exception of patients with the largest and most complicated infarcts, who were excluded because of hemodynamic reasons, the series can be regarded as relatively representative of patients with acute myocardial infarction.

3 680 Clin. Cardiol. Vol. 1 I. October 1988 lowing parameters: (a) chest pain of at least 15 min duration; (b) at least two values of serum aspartate aminotransferase (ASAT) above the normal range (0.7 pkat/l) with simultaneously lower activity of serum alanine aminotransferase (ALAT); (c) development of Q waves and/or initial ST elevation followed by T-wave inversion in at least two leads in a 12-lead standard ECG. Routine Procedures A standard 12-lead ECG was recorded once daily during the first 3 days in hospital. Blood samples were collected for analysis of serum activity of ASAT and ALAT once daily during the first 3 days in hospital and for analysis of heat-stable lactate dehydrogenase (LDH) activity every 12 h for h after arrival at the hospital. The activity of ASAT, ALAT, and total LDH was determined according to the Scandinavian Committee on Enzymes. lo The heat-labile fraction of LDH was inactivated according to Brydon and Smith." Blood samples for serum potassium were analyzed immediately after arrival at the hospital in the emergency ward, and then each morning during the following 3 days. This meant that the second blood sample for potassium analysis was collected between 0 and 24 h after the start of blind treatment. The serum concentration of potassium was analyzed according to the flame photometric method routinely used in the participating hospitals. Infarct size was defined as the maximum recorded activity of serum heat-stable LDH. Indication for treatment with lidocaine was ventricular fibrillation or prolonged ventricular tachycardia (> 60 s duration) or ventricular tachycardia associated with symptoms. Hypokalemia was defined as a serum potassium concentration <3.5 mmol/l. Mortality surpassing 5 years was assessed according to a central register of deaths. All patients reported as living according to this register were contacted either by letter or by phone. Statistical Methods Pitman's nonparametric test was used. A two-tailed test was applied. Results are expressed as mean f standard error of the mean (SEM). In the multivariate analysis, a stepwise logistic regression was used. Because of the large number of p values created, no formal significance level is stated. Results Of the 1395 patients participating in the trial, 809 (58%) developed a myocardial infarction during the first 3 days. The serum potassium level was recorded at least once in 1350 patients, twice in 1257 patients, and on one occasion in 63 patients. Occurrence of Hypokalemia in the Emergency Ward In all, hypokalemia was observed in 70 patients (5.3 %). The mean value of serum potassium concentration was 4.1 fo.o1 mmol/l. In patients developing an early infarction, hypokalemia was observed in 6.3% of the cases versus 4.0% in those who did not develop early infarction (p = 0.09). Subgroups. As shown in Table I, more women exhibited hypokalemia than men, and patients with hypertension and those on chronic treatment with diuretics had a higher incidence of hypokalemia compared with the remaining patients. Patients with larger infarcts tended to exhibit hypokalemia more often than those with smaller infarcts. Occurrence of Hypokalemia During the First 3 Days in Hospital In these analyses, only patients randomized to placebo were included. In all, hypokalemia was observed in 7.8% of the patients. In patients developing infarction it was observed in 9.0%, and in those without infarction in 6.1 % (p ). Subgroups: As shown in Table 11, evaluation of serum potassium levels over the first 3 days in hospital (including the emergency ward) compared to observation in the emergency ward only showed a similar pattern. Thus 20% of the patients on chronic treatment with diuretics showed hypokalemia at some time during the first 3 days in hospital. Patients with hypertension exhibited a higher incidence of hypokalemia, as did women. Patients with anterior infarcts had a higher incidence of hypokalemia than those with inferior infarcts. Multivariate Analysis In a stepwise logistic regression procedure, the following variables were included: chronic treatment with diuretics, hypertension, infarct size, sex, infarct development, infarct site, and initial heart rate. Independent variables associated with hypokalemia in the emergency ward were: chronic treatment with diuretics, infarct size, and sex. Independent variables associated with hypokalemia during the first 3 days in hospital were: chronic treatment with diuretics and infarct size. Hypokalemia in Relation to Severe Ventricular Arrhythmias During Hospitalization As shown in Table 111, patients with hypokalemia in the emergency ward had a higher occurrence of ventricular fibrillation, previously treated ventricular tachycardia, and requirement of lidocaine during hospitalization as compared with patients with no hypokalemic episodes. Similar observations were made when relating the occurrence of hypokalemia at any time during the first 3 days

4 J. Herlitz er al.: Hypokalemia in AM1 68 I TABLE I1 Occurrence of hypokalemia during the first 3 days in hospital in patients randomized to placebo Hypokalemia (%) Age I 61 years (median) (365) 6.3 > 61 years (314) 9.6 Sex Men (518) Women (161) Clinical history Previous infiuction Yes (156) No (523) Previous angina pectoris Yes (236) No (442) Previous hypertension Yes (203) No (476) On chronic treatment Diuretics Yes (126) No (550) Beta blockers Yes (176) No (503) Initial heart rate I 70 beatshin (343) > 70 beatslmin (336) Infarct size Heat-stable LDmax 5 15 pkat/l (197) Heat-stable LDmax > 15 pkat/l (195) Infarct site Anterior infarction (167) Inferior infarction (149) Delay I h (375) 8.0 > 3 h (294) 7.1 P 0. I6 0. I Figures in parentheses = number of patients. * Delay between onset of symptoms and arrival in hospital. TABLE 111 Occurrence of hypokalemia in the emergency ward in relation to occurrence of severe ventricular arrhythmias during hospitalization in patients randomized to placebo < 3.5 mmolll mmol/l n = 30 n = 634 Ventricular fibrillation (96) Previously treated ventricular tachycardia (%) Required lidocaine (%)

5 682 Clin. Cardiol. Vol. II, October 1988 TABLE IV Five-year mortality in relation to early hypokalemia in suspected acute myocardial infarction < 3.5 mmolll mmolll n = 70 n = year mortality in relation to hypokalemia in the emergency ward 5-year mortality in relation to hypokalemia during the first 3 days in hospital" Only patients in the placebo group were included. Hypokalemia was observed in 9% of the patients with a confirmed infraction. In previous studies this figure has varied between 9% and 21 % Its occurrence in patients with a nonconfirmed infarction was lower, although it did not significantly differ from those with a confirmed infarction. Most likely there is also an increase in circulatory catecholamines in patients with an unconfirmed myocardial infarction. Patients with anterior infarction and those with larger infarcts tended to have more hypokalemic episodes than those with inferior and those with smaller infarcts. Such findings raise the possibility that the association between infarct size and the occurrence of severe ventricular arrhythmias, which has previously been could, to some extent, depend on the more frequent development of hypokalemia in patients with large infarcts. Our results fit with the hypothesis that a higher sympathetic tone increases the risk for development of hypokalemia. On the other hand, a higher initial heart rate did not significantly increase the risk for development of hypokalemia. As previously reported,i4 patients on chronic treatment with diuretics were at much greater risk for development of hypokalemia. This explains its more frequent occurrence in hypertensive patients. A sex difference was also observed, indicating a more frequent occurrence in women. In a multivariate analysis, it was observed that the independent contributors to hypokalemia were chronic pretreatment with diuretics, infarct size, and female sex. Several previous studies have shown an association between the Occurrence of hypokalemia and severe ventricular arrhythmias in acute myocardial infarcti~n.~,~ Similar tendencies were observed in the present study. Whether the occumnce of hypokalemia during suspected acute myocardial infarction increases the risk for future malignant arrhythmias or indicates a less favorable prognosis has, as far as we know, not been investigated previously. In this study we observed that patients with hypokalemia in the emergency ward and during the first few days in hospital had a prognosis very similar to patients without hypokalemia. References 1. Beck OA, Hochrein H: Serumkaliumspiegel und Herzrhythmusstorung bei akutem Myocardinfarkt. Z Kardiol66, 187 (1977) 2. Dyckner T, Helmers C, Lundman T, Wester PO: Initial serum potassium level in relation to early complications and prognosis in patients with acute myocardial infarction. Acra Med Scad 197, 207 (1975) 3. Morgan DB: Hypokalaemia and diuretics. Roy Soc Med Intern Congr Symp Series 44, 3 (1981) 4. Nordrehaug JE, von der Lippe G: Hypokalaemia and ventricular fibrillation in acute myocardial infarction. Br Heart J 50, 525 (1983) 5. Nordrehaug JE, von der Lippe G: Serum potassium concentrations are inversely related to ventricular, but not to atrial, arrhythmias in acute myocardial infarction. Eur Heart J 7, 224 ( 1986) 6. Bertel 0, Buhler FR, Baitsch G, Ritz R: Plasma adrenaline and noradrenaline in patients with acute myocardial infarctionrelationship to ventricular arrhythmias of varying severity. Chest 82, 64 (1982) 7. Goldstein DS: Plasma norepinephrine as an indicator of sympathetic neural activity in clinical cardiology. Am J Cardiol48, 1147 (1981) 8. Karlsberg RP. Cryer PE, Roberts R: Serial plasma catecholamine response early in the course of clinical acute myocardial relationship to infarct extent and mortality. Am Heart J (1981) 9. Hjalmarson A, Elmfeldt D, Herlitz J, Holmberg S, Mhlek I, Nyberg G, Rydkn L, Swedberg K, Vedin A, Waagstein F, Waldenstrom A, Waldenstrom J, Wedel H, Wilhelmsen L, Wilhelmsen C: Effect on mortality of metoprolol in acute myocardial infarction. Lancer ii, 823 (1981) 10. Scandinavian Committee on Enzymes. Recommended methods for the determination of four enzymes in blood. Scand J Clin Lab Invest 33, 291 (1974) 11. Brydon WG, Smith AF: An appraisal of routine methods for the determination of the anodal isoenzymes of lactate dehydmgenase. Clin Chim Acra 43, 361 (1973) 12. Chapman BL: Relation of d iac complications to S OT level in acute myocardial infarction. Br Heart J 34, 890 (1972) 13. Herlitz J, Hjalmarson A, Swedberg K, Waagstein F, Waldenstriim J: Relationship between infarct size and incidence of severe ventricular arrhythmias in a double blind trial with metoprolol in acute myocardial infarction. Inr J Cardiol6, 47 ( 1984) 14. Brass EP, Thompson WL: Drug-induced electrolyte abnormalities. Drugs 24, 207 (1982)