Authors' reply

EDITOR,--The constraints on the number of words in a short report prevented us from giving the lactate concentrations. In the 31 patients with a positive response to the subanaerobic threshold exercise test, concentrations exceeded the upper 99% confidence interval established in normal control volunteers in the paper by Nashef and Lane,1 cited by Alan S Hutchison, in 28 subjects before exercise and in 29 subjects immediately after and in 29 subjects 30 minutes after exercise. Before November 1990 the plasma lactate concentration was measured with a Sigma kit, as detailed by Nashef and Lane. Since then the concentration has been measured with a Stat lactate analyser (Yellow Springs Instrument, Ohio). The table shows the precision data between batches for two periods in which these methods were used; the data are based on internal quality control samples run with each batch of assays.

To some extent we share Hutchison's reservations regarding the validity of the confidence intervals for lactate concentrations based on historical controls. Nevertheless, the subanaerobic threshold exercise test largely eradicates variations related to age, sex, and weight.2 We would have been more concerned if the majority of patients with the chronic fatigue syndrome had shown abnormal results: the upper limit of the confidence interval might have been artificially low. However, results were normal in two thirds of cases.

The point concerning the baseline lactate concentrations is pedantic. A raised resting concentration is abnormal and might indicate defective mitochondrial function, for example.2 The issue of lactate responses to exercise in glycogenoses is not really pertinent. Here, one is looking for a partial or complete lack of increase in lactate concentration after ischaemic exercise of a limb. Control of factors such as the intensity and duration of work and procedures for collecting samples must be much more stringent.3