Abstract

AIM: To correlate splenic volume calculations obtained through 1.Cavalieri method based on point counting; and 2. the formula based approaches, which are presupposed to be more practical. MATERIAL AND METHOD: The study was conducted on T2 weighted axial MR images of upper abdomen in 36 patients with no evidence of splenomegaly and/or mass lesion. MR imaging was performed by Philips Achieva 1.5 Tesla MR equipment with 7 mm slice thickness and 1.5 mm gap. Splenic volume was calculated through Cavalieri method based on point counting and two different formula based approaches (A x B x C x k, and A x B x C + k).. To eliminate the negative effect of slice thickness in Cavalieri method, a mathematical correction was performed to keep the coefficient of error below 0.05%. Interobserver reliability of the measurements and the final calculations were statistically correlated. FINDINGS: The interobserver correlation in Cavalieri method was signifıcantly high (R=+0,96). Although the reproducibility of the linear measurements used in formula based approaches was high (R> 0.94), the correlation with the formula and point counting-based calculations was fair-to-good (R = 0.80). CONCLUSION: Formula based approaches, which are presupposed to be more practical, can roughly estimate the spleen volume. In borderline cases of splenomegaly, or in cases whom the spleen size estimations require more precision, Cavalieri method offers a more rational approach.