Detection of intracranial aneurysms using three-dimensional multidetector-row CT angiography: Is bone subtraction necessary?

Abstract

Purpose

The aim of this study was to evaluate the usefulness of three-dimensional CT angiography (3D CTA) with bone subtraction in a comparison with 3D CTA without bone subtraction for the detection of intracranial aneurysms.

Materials and methods

Among 337 consecutive patients who had intracranial aneurysms detected on 3D CTA, 170 patients who underwent digital subtraction angiography (DSA) were included in the study. CTA was performed with a 16-slice multidetector-row CT (MDCT) scanner. We created the 3D reconstruction images with and without bone subtraction by using the volume rendering technique. Three neuroradiologists in a blinded fashion interpreted both 3D CTA images with and without bone subtraction. The diagnostic accuracy of both techniques was evaluated using the alternative free-response receiver operating characteristic (ROC) analysis. The sensitivity and positive predictive value were also evaluated.

Results

A total of 200 aneurysms (size: 2–23 mm) were detected in 170 patients. The area under the receiver operating characteristic curve (Az) for 3D CTA with bone subtraction (mean, Az = 0.933) was significantly higher than that for 3D CTA without bone subtraction (mean, Az = 0.879) for all observers (P < 0.05). The sensitivity of 3D CTA with bone subtraction for three observers was 90.0, 92.0 and 92.5%, respectively, while the sensitivity of 3D CTA without bone subtraction was 83.5, 83.5 and 87.5%, respectively. No significant difference in positive predictive value was observed between the two modalities.

Conclusions

3D CTA with bone subtraction showed significantly higher diagnostic accuracy for the detection of intracranial aneurysms as compared to 3D CTA without bone subtraction.

Introduction

Conventionally, digital subtraction angiography (DSA) has been considered the reference standard for the detection and evaluation of intracranial aneurysms. Recently, three-dimensional (3D) rotational angiography has provided an improved detection rate and characterization of intracranial aneurysms [1], [2]. Nevertheless, DSA has still several disadvantages, such as invasiveness, time-consumption, a relatively high cost and some neurological complications [3], [4]. Therefore, CT angiography (CTA) has been increasingly used and has become more important as a first diagnostic technique in patients with suspected intracranial aneurysms, especially for the evaluation of patients with subarachnoid hemorrhage in an emergency setting [5], [6], [7], [8], [9], [10].

As CT technology has evolved and various subtraction and postprocessing techniques have been developed, subtraction CTA with good image quality for the detection of intracranial aneurysms has become possible [6], [11], [12], [13], [14], [15], [16]. However, several problems, such as increased radiation dose due to a second scan, image degradation due to inevitable patient movement, venous contamination and bone artifacts still remain with the use of subtraction CTA [6], [11], [15]. There have been several reports on the potential usefulness of subtraction CTA, but most previously published studies have not included a large population and were concerned about evaluation of subtracted image quality or a comparison with DSA for the diagnosis of intracranial aneurysms [6], [12], [13], [14], [15], [16]. The purpose of this retrospective study was to evaluate the usefulness of 3D CTA with bone subtraction in a comparison with 3D CTA without bone subtraction for the detection of intracranial aneurysms in a large patient population.