Letters to the Editor

Abstract

22 December 2006 Dear Editor, RE: HYPONATRAEMIA AND SEIZURES IN ONCOLOGY PATIENTS ASSOCIATED WITH HYPOTONIC INTRAVENOUS FLUIDS We read with interest the article by Duke et al. in the December 2005 issue of the Journal, describing three cases of hyponatraemia and seizures in oncology patients. Only one of the three patients described had leukaemia and this patient was not receiving hyper-hydration for chemotherapy at the time of his hyponatraemic episode.1 In our unit, hyper-hydration is most commonly used for patients with acute lymphoblastic leukaemia (ALL) or T-cell non-Hodgkin's lymphoma (NHL) being treated with cyclophosphamide or high-dose methotrexate. We therefore conducted an audit of hyponatraemia in patients beginning treatment for ALL or T-cell NHL at the Mater Children's Hospital between January 2001 and December 2005. Forty-five patients commenced treatment during the audit period, 42 for ALL and three for T-cell NHL. Biochemistry results for these patients were reviewed to identify all episodes of serum sodium less than 130 mmol/L. Charts were then examined to determine the context in which the episode of hyponatraemia (sodium <60; 130 mmol/L) occurred. Almost one-third (31.1%, n = 14) of our patients had an episode of hyponatraemia (sodium <60; 130 mmol/L). Five patients (11.1%) had a sodium level of 125 mmol/L or less, with the lowest sodium recorded being 122 mmol/L in two patients. Table 1 summarises the phase of treatment during which these episodes occurred. A detailed analysis as to whether episodes of hyponatraemia were due to vincristine alone was not undertaken. The 14 patients identified in our audit had received a collective total of 93 courses of hyper-hydration (47 for high-dose methotrexate and 46 for cyclophosphamide), with the base fluid in each instance being either 3.3% dextrose + 0.3% sodium chloride or 4% dextrose + 0.18% sodium chloride. Surprisingly, only one episode of hyponatraemia occurred during hyper-hydration for chemotherapy. This occurred in a patient who had received 3 days of hyper-hydration for high-dose methotrexate. Two patients (4.4%) died during their hyponatraemic episodes, both from overwhelming infection. There were no seizures in the other 12 patients. Our findings highlight the potential for hyponatraemia in patients with ALL or T-cell NHL, with almost one-third of our patients having an episode of hyponatraemia less than 130 mmol/L. While there was only one episode in the setting of hyper-hydration for chemotherapy in our audit, we agree with Duke et al. that the use of hypotonic fluids exposes patients to the risk of excessive free water accumulation and iatrogenic hyponatraemia. The periods of induction and re-induction therapy would appear to pose a particular risk. The use of agents such as vincristine, that may cause the syndrome of inappropriate antidiuretic hormone, may be responsible for this increased risk. Careful monitoring of sodium and fluid balance is of critical importance. When a patient requires parenteral fluids, particularly during induction and re-induction, the use of hypotonic solutions should be avoided. The authors wish to thank Dr Philip Sargent, Consultant Intensivist, Department of Paediatric Intensive Care, Mater Children's Hospital, Brishane, Queensland, Australia, for his expert opinion and advice.