SERUM INORGANIC-PHOSPHATE IN PROTEIN-ENERGY MALNUTRITION

We retrieved a series of measurements made 35 years ago of the concent ration of inorganic phosphate (P) in the serum from 56 cases of severe protein-energy malnutrition at the Tropical Metabolism Research Unit, Jamaica. There is no record of whether or not the cases were randomly selected. The samples were obtained within 4 days of admission and ex cept in 3 cases there was no follow-up. The average age was 12 months. The children have been classified retrospectively from the notes as m arasmus (11 cases), kwashiorkor (22 cases) and marasmic kwashiokor (23 cases). In all 11 children died (fatality rate 20%), eight of them fr om the group with marasmic kwashiorkor. Weight-for-age, length-for-age and weight-for-length have been calculated as Z-scores. Nearly all se rum phosphate concentrations were low (mean 1.41 mmol.l-1, SD 0.444, r ange 0.50-2.45) compared with the normal value at this age of about 2 mmol.l-1. The serum P was significantly less depressed in the marasmic children (P = 0.042), but there was no relation between serum P and a ny of the anthropometric measurements, nor with outcome (death or surv ival). There was, however, a significant relationship with the degree of oedema. Death was related to age - the children who died were young er (mean difference 3.8 months; P = 0.01; 95% confidence interval 0.23 -6.43). It took about 3 weeks of feeding a milk-based diet for serum p hosphate to reach normal levels. There have been few previous measurem ents of serum P in malnutrition. We agree with previous authors that t he low serum values are evidence of phosphate depletion and suggest th at phosphate might be added to the electrolyte solutions used in the e arly stages of recovery. However, reports of adverse effects indicate that this should be done with great care.