Electrocardiography is unreliable in detecting potentially lethal hyperkalaemia in haemodialysis patients.

BACKGROUND: It is speculated, but unconfirmed, that the usual electrocardiographic manifestations of hyperkalaemia are less frequent and less pronounced in persons with end-stage renal disease (ESRD) than in those with normal renal function. We studied 74 consecutively selected stable haemodialysis patients to determine the prevalence of electrocardiographic changes of hyperkalaemia in stable persons with ESRD receiving haemodialysis. METHODS: Pre-dialysis serum potassium concentration and other electrolytes were measured and simultaneous 12-lead electrocardiogram obtained. RESULTS: The 74 study subjects (45 men, 29 women) comprised 63 blacks (85%), four Hispanics (6%), four whites (6%), and three Asians (4%) of mean+/-standard deviation age 55.5+/-14.7 years. Mean pre-dialysis potassium concentration was 4.9+/-0.71 mEq/l (range 3.3-6.7). No study subject evinced arrhythmia or any of the typical electrocardiographic changes associated with hyperkalaemia. There was no significant difference in T wave amplitude (F statistic=2.1; P=0.11) or T wave to R wave ratio (F statistic=2; P=0.12) between quartiles of serum potassium concentration. Also, T wave amplitude was equivalent in patients with serum potassium concentration >5.5 mEq/l (7.1+/-4.1 mm) or < or =5.5 mEq/l (5.2+/-3.5 mm) (P=0.13). Linear regression analysis showed that the total serum calcium concentration had an inverse relation with T wave amplitude (P=0.03) after adjustment for other factors (a high total serum calcium concentration was associated with a low T wave amplitude). CONCLUSION: Haemodialysis patients with hyperkalaemia may not exhibit the usual electrocardiographic sequella of hyperkalaemia, possibly due in part to fluctuations in serum calcium concentration. Thus, the absence of electrocardiographic changes in hyperkalaemic haemodialysis patients should be interpreted with caution.