Determination of subdural hematoma volume: Comparative analysis of calculation methods

Abstract

Background. Accurate calculation of the volume of subdural hematomas of traumatic, vascular, and other origins during diagnosis is complicated which leads to difficulties in selection of optimal tactics of neurosurgical treatment. Aim. To preform comparative analysis of accuracy of different methods of subdural hematoma volume calculations to increase the quality of evaluation of the results of radiological methods in selection of surgical tactics.Materials and methods. Data from preoperative computed tomography of 20 patients (15 men and 5 women between the ages of 20 and 60 years) with confirmed diagnosis of subdural hematoma who were examined and treated at the Neurosurgery Clinic of the Kirov Military Medical Academy were used. Four groups were formed based of the calculation method: two groups using formulas for volume of ellipsoid or convexo-concave lens (the shape most accurately representing the shape of subdural hematoma); electronic algorithm of volume calculation (Gamma MultiVox D2 software, Gammamed-Soft, Russia); control method (manual sectional calculation of areas of the formed structures with subsequent summation).Results. Deviations between the obtained values varied between +18 and -16 %. Deviations for calculations using ellipsoid volume formula were +46 and -19 %; using convexo-concave lens formula +38 and -35 %, respectively. Electronic calculation algorithm (Gamma MultiVox D2) showed the best accuracy compared to other methods including the control method.Conclusions. Comparative analysis of the accuracy of the studied methods of calculation of subdural hematoma volume showed that median results are statistically similar which allows for selection and use of these methods in accordance with the neurosurgeon's preferences.Study of the algorithm based on ellipsoid volume shows low specificity of this method, comparatively high deviations of the results from the true value with a trend toward overprediction.Calculation using volume of convexo-concave lens showed intermediate result if the shape of the hematoma corresponded to the ideal chape of a figure enclosed between two segments of a sphere. In cases where the shape did not correspond to the convexo-concave lens, both overprediction and underprediction of hematoma volume were observed.Calculations using the Gamma MultiVox D2 software showed the highest accuracy, lowest range of deviation from the control data, best versatility, independence from both the shape and location of the subdural hematoma, as well as selection of a section used in other algorithms for the measurements.It should be noted that accuracy of determination of subdural hematoma volume by the studied methods is directly proportional to time necessary for calculation.