Octreotide-scintigraphy is a disease-activity parameter in Graves' ophthalmopathy.

Abstract

It is thought that immunosuppressive treatment of Graves' ophthalmopathy should be restricted to patients with active eye disease, but assessing disease activity is difficult. Octreotide scintigraphy has been claimed to differentiate active from inactive disease. Here we study the intraobserver variability and diagnostic accuracy of the quantitative measurement of orbital octreotide uptake. Twenty-two consecutive patients with moderately severe ophthalmopathy were treated with retrobulbar radiotherapy. Pretreatment octreotide scintigraphic data were related to the response at six months after radiotherapy, using Receiving-Operator-Characteristic curves. Octreotide uptake was measured at 4 and 24 h after i.v. injection of approximately 3 mCi (= 111 MBq; range 75-150 MBq) 111Indium-DTPA-Octreotide with a neuro-SPECT camera. Counts were measured in fixed regions-of-interest in 4 transversal slices of the orbit, the temporal and the occipital area. Measurements were done twice and intraobserver variability was analysed by coefficients of variations (CV). Uptake is expressed as orbital/background ratio. The nature of the temporal uptake was studied by matching an octreoscan with a technetium scan and MRI. Intra-observer variability of measuring octreotide uptake was acceptable, and the coefficient of variation slightly better using the orbital/occipital ratio (11%), than the orbital/temporal ratio (16%). From matching studies it appears that the temporal uptake takes place, in part, in the parotid gland. The orbital/occipital ratio was used to predict the outcome of radiotherapy. Mean (+/- SD) uptake on the 4 h scan was higher in responders (2.2 +/- 0.66) than in nonresponders (1.7 +/- 0.39; P = 0.04). From the Receiving-Operator-Characteristic curve we determined a cut-off value of 1.85, which yielded a positive predictive value of 92% and a negative predictive value of 70%. The 24 h scan could not predict a response. Quantitative measurement of orbital octreotide uptake is possible. Using the orbital/occipital ratio on the 4 h scan, the octreoscan seems useful in predicting response to subsequent radiotherapy. The 24 h scan seems not to be useful in predicting therapeutic outcome.