A mathematical analogue to calculate the counts from 32P in the choroidal blood during the 32P uptake test.

In a previous paper' the authors described their experience of the 32P uptake test in the diagnosis of intraocular malignancies and discussed some of the factors affecting the uptake. The 32p test is based on the differential accumulation of radioactive phosphorus between neoplastic and non-neoplastic tissue following an intravenous injection of the isotope. Counts are recorded from a site on the surface of the eye, overlying the lesion, and from a control site on the eye. The 32p uptake is the difference between the test and control counts expressed as a percentage of the control count. In a recent review of the literature2 threshold values for a positive test were shown to have varied between 30 and 150% according to different investigators, but now a figure of around 60% is most frequently used. Several investigators have discussed the reliability of the test and stressed the need for accurate localisation of the probe.3 Various matters such as the rate of phosphorus uptake in uveal melanomas7 and the relation between 32P counts and histological features8'0 have been investigated. An aspect of the 32p test which has received little attention relates to the source of the control counts. An uptake of 100% for example implies that there are half as many counts recorded over control tissue as from a site which overlies a malignant tumour. The counts from 32p in normal tissue are therefore not negligible in comparison with those from a malignant tumour. The activity recorded on the surface of the eye depends on the distribution of isotope in the tissue below the detector. A knowledge of this distri-