[Comparison of planimetric CT‑based volumetry with simplified models for determining the size of intrapelvic hematomas due to pelvic fractures in emergency room diagnostics].

BACKGROUND

Bleeding in the pelvis can lead to a circulatory problem. The widely used whole-body computed tomography (WBCT) scan in the context of treatment in the trauma resuscitation unit (TRU) can give an idea of the source of bleeding (arterial vs. venous/osseous); however, the volume determination of an intrapelvic hematoma by volumetric planimetry cannot be used for a quick estimation of the blood loss. Simplified measurement techniques using geometric models should be used to estimate the extent of bleeding complications.

OBJECTIVE

To determine whether simplified geometric models can be used to quickly and reliably determine intrapelvic hematoma volume in fractures type Tile B/C during emergency room diagnostics or whether the time-consuming planimetric method must always be used.

MATERIAL AND METHODS

Retrospectively, 42 intrapelvic hemorrhages after pelvic fractures Tile B + C (n = 8:B, 34:C) at two trauma centers in Germany were selected (66% men, 33% women; mean age 42 ± 20 years) and the CT scans obtained during the initial trauma scan were analyzed in more detail. The CT datasets of the included patients with 1-5 mm slice thickness were available for analysis. By area labelling (ROIs) of the hemorrhage areas in the individual slice images, the volume was calculated by CT volumetrically. Comparatively, volumes were calculated using simplified geometric figures (cuboid, ellipsoid, Kothari). A correction factor was determined by calculating the deviation of the volumes of the geometric models from the planimetrically determined hematoma size.

RESULTS AND DISCUSSION

The median planimetric bleeding volume in the total collective was 1710 ml (10-7152 ml). Relevant pelvic bleeding with a total volume > 100 ml existed in 25 patients. In 42.86% the volume was overestimated in the cuboid model and in 13 cases (30.95%) there was a significant underestimation to the planimetrically measured volume. Thus, we excluded this volume model. In the models ellipsoid and measuring method according to Kothari, an approximation to the planimetrically determined volume could be achieved with a correction factor calculated via a multiple linear regression analysis. The time-saving and approximate quantification of the hematoma volume using a modified ellipsoidal calculation according to Kothari makes it possible to assess the extent of bleeding in the pelvis after trauma if there are signs of a C-problem. This measurement method, as a simple reproducible metric, could be embedded in trauma resuscitation units (TRU) in the future.