Clinical implications of the relationship between low calcium concentration dialysate and sudden death.

In this issue of CJASN, Pun et al. (1) increase our understanding about the risk of sudden death associated with lower calciumdialysate concentrations. Formany years, we have known that lower calcium dialysate concentrations are associated with a prolongation of the QT interval and QT dispersion (2)—both risk factors for sudden death (3). These changes in cardiac conduction were noted by Näppi et al. (2) when the dialysate calcium concentration was decreased from 1.5 mmol/L (3 mEq/L) to 1.25 mmol/L (2.5 mEq/L). The use of a lower calcium dialysate has also been associated with a statistically significant lowering of BP during the dialysis treatment (4). Because increasedQT dispersion and lower BPs are associated with lower dialysate calcium concentration, one might suspect that low calcium dialysate would be associated with an increased risk of intradialytic cardiac arrest, and indeed, Pun et al. (5) had previously shown that lower calcium dialysate was associated with an increased risk of sudden death during the dialysis procedure. In this edition of CJASN, Pun et al. (1) extended their findings to show that the higher the gradient between serum calcium and dialysate calcium, themore likely it is that sudden death will occur. Pun et al. (1) performed a case-control study of 510 patients who suffered cardiac arrest during the hemodialysis procedure and 1560 matched controls. Both cases and controls were obtained from a database of 43,200 patients who received hemodialysis between 2002 and 2005. The study revealed that low calcium dialysate (,2.5 mEq/L) was associated with a twofold increased risk of sudden death during the dialysis procedure (95% confidence interval51.4–2.9). Higher corrected serum calcium and an increasing serum dialysate calcium gradient were also significant risk factors. Suddendeath events duringdialysis are uncommon, estimated to occur between four and seven times per 100,000 dialysis treatments (5); therefore, the risk of sudden death during the dialysis treatment is very low. However, sudden death and cardiac events are the most significant cause of mortality in the dialysis population. The increased risk of sudden death occurs at the end of the weekend interval immediately before dialysis or within 12 hours of the start of a dialysis session (6). Although the dialysate calcium concentration is unlikely to be a risk factor for deaths occurring at the end of the weekend interval, it could verywell contribute to the large number of deaths and cardiac events occurring in the hours after the dialysis session. In a study from our center, approximately one-third of all sudden deaths occurred within 12 hours of the start of the dialysis treatment (6). There are several possible reasons why the high serum-to-dialysate calcium gradient may cause suddendeath. Themost likely possibility is that the larger serum-to-dialysate calcium gradient results in a larger calcium shift with increasing QT dispersion and risk of sudden death. Another possibility is that hemodialysis patients with higher serum calcium levels—who have been reported to have decreased survival compared with patients with normal serum calcium levels (7)—have increased cardiac dysfunction at baseline that puts them at risk of sudden death. It is not possible to tell with certainty which one of these factors is responsible. However, the statistical analysis from the study by Pun et al. (1) indicated that it was likely the gradient that was more important. This studymust also be interpreted in light of recent findings showing an increased risk of cardiac arrest during dialysis with higher dialysate bicarbonate concentrations. Alkalemia is well known to lower serum calcium, and the combination of increased alkalemia with a high calcium gradient could lead to an even higher calcium gradient and risk of sudden death. How should this study affect our clinical care? It is important to appreciate the complex interplay between serum calcium levels, body calcium stores, and calcium balance. Patients with normal serum calcium levels may have high, normal, or low body stores of calcium, and patients with high serum calcium levels may actually have low body stores of calcium. It would seem that management of serum calcium levels and maintenance of neutral calcium balance might use different techniques. Lower dialysate calcium concentrations seem to increase the risk of sudden death, and their use may not be preferred. Kidney Disease: Improving Global Outcomes (KDIGO) guidelines suggest a dialysate calcium concentrate of 2.5 mEq/L (8). Preferred mechanisms to lower serum calcium levels may include the use of noncalcium containing binders, cinacalcet, or vitamin Section on Nephrology, Wake Forest School of Medicine, WinstonSalem, North Carolina