Treatment of Hyponatremic Encephalopathy—Reply

In Reply: Dr Smith is concerned about the absence of patient randomization. Given the incidence of in-hospital hyponatremic encephalopathy (about 0.15%),¹ (ie, all hospitalized patients from 1990 to 1992) it would be very difficult for a single group of investigators to obtain enough patients to perform a randomized study. Nevertheless, the absence of randomization need not introduce bias because we included every patient meeting the inclusion criteria during the study interval. However, the suggestion that the seizures, respiratory arrest, and neurological damage might not have been associated with the hyponatremia is untenable in the 1990s, as this association was well described in the 1950s and confirmed in hundreds of published cases thereafter.^2- 4 Smith suggests that the difference between group 3 and groups 1 and 2 appears to result from a failure to correct hyponatremia, rather than the mode of management. Our study indicates that failure to treat hyponatremic encephalopathy before the onset of respiratory insufficiency carries a significant morbidity and that fluid restriction is ineffective in correcting encephalopathic symptoms.