E1906. Effect of Demographic and Anthropometric Factors on Spleen Size: A Prospective Study in American Adults Without Known Splenic Abnormality
  1. Robert Petrocelli; University of Pittsburgh Medical Center
  2. Justin McCloskey; University of Pittsburgh Medical Center
  3. Juliana Tobler; University of Cincinnati
  4. Jacob Yousef; University of Pittsburgh Medical Center
  5. Peter Swain; University of Pittsburgh Medical Center
  6. Roberta Catania; University of Pittsburgh Medical Center
  7. Amir Borhani; University of Pittsburgh Medical Center
Data on normal spleen size in American adult population is lacking (Prassopoulos P, et al., Kaneko J, et al.). There is also inconsistency and ambiguity regarding measurement technique and threshold to define splenomegaly (Rezai P, et al., Lamb PM, et al.). We sought to describe normal range of spleen size in a large population of adults and to determine if it is affected by demographics and body habitus.

Materials and Methods:
ER patients undergoing abdominal CT between 2017-2019 were prospectively studied. Patients with splenic abnormality or relevant comorbidity (liver disease, hematologic disorders, malignancy, etc.) and patients with incomplete clinical data were excluded. 996 subjects (age 44.2 ± 15; 51.9% female; 80.1% Caucasian; BMI 30 ± 7.9) were included in final analysis. Spleen dimensions on 3 orthogonal planes, max craniocaudal length (maxCC), as well as spleen volume using semi-automated segmentation were measured. Subject demographics, height, and weight were recorded. Statistical analysis including Spearman’s correlation and Wilcoxon rank-sum test as well as multiple linear regression analysis were performed.

Mean volume, craniocaudal length, and maxCC length were 244±119 cc, 7.4±2 cm, and 10.5±1.6 cm, respectively. Estimated volume, using elliptical formula, strongly correlated with true splenic volume (?=0.87; p<0.0001) although there was non-significant tendency for underestimation. Among one-dimensional measures, maxCC had highest correlation with volume (?=0.8; p<0.0001). Splenic size and volume were significantly different between sexes (p= <0.0001). The 95th percentile of volume and maxCC length respectively were 499 cc and 13.8 cm in males, and 396 cc and 12.5 cm in females. Univariate analysis showed African American race, female sex, BMI, height, and weight as statically significant predictors of volume (with coefficients of -89, -59, 4.9, 4.3, and 2.3, respectively; p<0.0001). A multivariate model based on sex, race, height, weight, and age reliably predicted splenic volume (F[5, 990] =114.49; p<0.0001; R2=0.366).

Determination of splenomegaly has significant clinical importance. The threshold to use for this designation should be adjusted based on patient demographic and anthropometric factors. Splenic size was most notably affected by sex, race, and height. These factors should be considered when determining splenomegaly. Maximal craniocaudal length had the highest correlation with splenic volume and may be the best uni-dimensional measurement to replace true volume.