Introduction: Digital substraction angiography (DSA) remains as gold standard for intracranial aneurysm (IA) diagnosis. Use of computed tomographic angiography (CTA) to diagnose IAs continues to increase, however, previous data comparing CTA and DSA estimated CTA sensitivity at 93% and specificity at 77%. A prospectively acquired cohort of 112 patients with CTA diagnosis of IA who underwent confirmatory DSA was studied to re-evaluate the diagnostic accuracy of CTA for unruptured IAs.
Methods: Between 2007-2010 112 patients (75 females; mean age 53.2 years) with one or more unruptured aneurysms by CTA, who later underwent four-vessel DSA were identified. DSA and CTA results were compared to determine accuracy of CTA in diagnosing unruptured IAs. Results are reported as “inaccuracy rate” of CTA. Aneurysms missed by CTA but diagnosed by DSA (false-CTA-negatives) were recorded separately and analyzed by parent vessel and aneurysm size. Cases diagnosed by CTA but ruled out by DSA(false-CTA-positives) were further classified into completely negative or negative with an infundibulum.
Results: While CTA identified 117 anterior and 15 posterior circulation aneurysms, DSA identified 127 anterior and 7 posterior circulation aneurysms. DSA detected more very small(0-5 mm, 91 vs. 85), small(6-10 mm, 29 vs. 28), and medium(11-15 mm, 7 vs. 11) aneurysms than CTA. CTA detected more large aneurysms(16-24 mm) than DSA(3 vs. 1). In 27 cases, CTA identified an IA later ruled out by DSA(false-CTA-positive). Within this group DSA ruled out 18 aneurysms previously identified by CTA, and an infundibulum but not an aneurysm was identified in 9. Most “false-CTA-positive” aneurysms(33%) were located along the anterior communicating artery(ACoA), as well as in the posterior circulation(33%) and were classified as very small(63%) or small(3.7%). DSA identified additional aneurysms In 20 patients not found on CTA. The most common location of "false-CTA-negative" aneurysms was the ICA with 75.9%, mostly located in the cavernous ICA, followed by MCA aneurysms with 24.1%. Most false-CTA-negative aneurysms were classified as very small (96.6%).
Conclusions: CTA results must be carefully evaluated as previously reported accuracy rates differ from those found in this study. Larger studies are needed to further elucidate this phenomenon.
Patient Care: Based on this data DSA should remain as the most sensitive and specific modality for diagnosis of unruptured intracranial aneurysms. CTA results must be carefully evaluated prior to proceeding with surgical treatment in patients with unruptured aneurysms.
Learning Objectives: To understand the sensitivity and specificity of CTA and DSA in the diagnosis of unruptured intracranial aneurysms. To evaluate the PPV and NPV of CTA in the diagnosis of intracranial aneurysms.