Selection, presentation, and interpretation of biochemical data in renal failure

There is more information in biochemical laboratory data than is generally appreciated; much can be gained by display and manipulation of data. Patients with renal disease are often under regular clinic review so that many will have sets of sequential data. Display of these data, which requires a computer if it is to be routine, will reveal patterns and define trends which could be otherwise missed or inadequately assessed. Trends are most useful if they are recti-linear and several transforms of the data should be tried to achieve this aim; for plasma creatinine the usual transforms are the reciprocal and the logarithm. The choice of transform is based on usefulness and not on any prior assumption about the underlying pathophysiology. More frequently, however, all that is available is a single measurement, typically the simple plasma creatinine rather than the more complex creatinine clearance. The problem is to balance simplicity against precision and decide what the needs and the reliability of the interpretations of the data are. There is an advantage in manipulation of the data and comparison with standard data; and the possibilities were discussed for the plasma creatinine; these tables are ideally suited to a laboratory computer. It was suggested that complex calculations so easily achieved with a computer should be avoided unless they bring real gains, since complexity tends to hide the basic principles involved.