Dual energy CT of the chest: how about the dose?

OBJECTIVE: New generation Dual Source computed tomography (CT) scanners offer different x-ray spectra for Dual Energy imaging. Yet, an objective, manufacturer independent verification of the dose required for the different spectral combinations is lacking. The aim of this study was to assess dose and image noise of 2 different Dual Energy CT settings with reference to a standard chest scan and to compare image noise and contrast to noise ratios (CNR). Also, exact effective dose length products (E/DLP) conversion factors were to be established based on the objectively measured dose. MATERIALS AND METHODS: An anthropomorphic Alderson phantom was assembled with thermoluminescent detectors (TLD) and its chest was scanned on a Dual Source CT (Siemens Somatom Definition) in dual energy mode at 140 and 80 kVp with 14 x 1.2 mm collimation. The same was performed on another Dual Source CT (Siemens Somatom Definition Flash) at 140 kVp with 0.8 mm tin filter (Sn) and 100 kVp at 128 x 0.6 mm collimation. Reference scans were obtained at 120 kVp with 64 x 0.6 mm collimation at equivalent CT dose index of 5.4 mGy*cm. Syringes filled with water and 17.5 mg iodine/mL were scanned with the same settings. Dose was calculated from the TLD measurements and the dose length products of the scanner. Image noise was measured in the phantom scans and CNR and spectral contrast were determined in the iodine and water samples. E/DLP conversion factors were calculated as ratio between the measured dose form the TLDs and the dose length product given in the patient protocol. RESULTS: The effective dose measured with TLDs was 2.61, 2.69, and 2.70 mSv, respectively, for the 140/80 kVp, the 140 Sn/100 kVp, and the standard 120 kVp scans. Image noise measured in the average images of the phantom scans was 11.0, 10.7, and 9.9 HU (P > 0.05). The CNR of iodine with optimized image blending was 33.4 at 140/80 kVp, 30.7 at 140Sn/100 kVp and 14.6 at 120 kVp. E/DLP conversion factors were 0.0161 mSv/mGy*cm for the 140/80 kVp protocol, 0.0181 mSv/mGy*cm for the Sn140/100 kVp mode and 0.0180 mSv/mGy*cm for the 120 kVp examination. CONCLUSION: Dual Energy CT is feasible without additional dose. There is no significant difference in image noise, while CNR can be doubled with optimized dual energy CT reconstructions. A restriction in collimation is required for dose-neutrality at 140/80 kVp, whereas this is not necessary at 140 Sn/100 kVp. Thus, CT can be performed routinely in Dual Energy mode without additional dose or compromises in image quality.

Saving dose in triple-rule-out computed tomography examination using a high-pitch dual spiral technique.

OBJECTIVE: High radiation doses remain a drawback of current triple-rule-out computed tomography (CT) protocols. With dual source CT, a new high-pitch dual spiral technique offers the possibility to acquire an Electrocardiography (ECG)-gated-synchronized dataset of the whole chest in less than 1 second. The aim of this study was to compare the dose of such a protocol to a standard, nongated chest scan and to a conventional, retrospectively ECG-gated triple-rule-out protocol. Also, the efficacy and dose of this dual spiral protocol was to be compared in patients examined with this high-pitch technique and matched controls scanned with the conventional technique. MATERIALS AND METHODS: An anthropomorphic Alderson Rando phantom was equipped with thermoluminescent detectors and scanned with the high-pitch protocol (Siemens Somatom Definition Flash; 2 x 120 kVp, 426 mAseff, 128 x 0.6 mm collimation, pitch 3.2), the nongated chest scan (same scanner; 120 kVp, 160 mAseff, 128 x 0.6 mm, pitch 1.2; equivalent Computed Tomography Dose Index (CTDI) of 7.12 mGy), and the conventional gating technique (Siemens Somatom Definition; 120 kVp, 560 mAseff with ECG pulsing interval at 30%-70% of the R-R cycle, 64 x 0.6 mm, pitch 0.3). Noise was measured in air, central and peripheral soft tissue of the phantom. Conversion factors were determined based on the measured dose and the dose-length products of the scanner. The protocol was then applied with ethics committee approval in 31 patients suffering from acute chest pain. The 120 mL of contrast material (Ultravist 370, Bayer Schering Pharma) was applied at 5 mL/s. Dose was calculated based on the dose-length products and the conversion factor. Image quality was assessed by 2 readers for aorta, pulmonary arteries, and coronary arteries. The results were compared with matched controls scanned with the conventional ECG gating technique and non-ECG gated thorax scans. RESULTS: The dose determined with thermoluminescent dosimeters measurements amounted to 2.65, 2.68, and 19.27 mSv, respectively, for the dual spiral technique, the standard chest scan, and the conventional retrospective technique. There was no significant difference in image noise. Respective conversion factors were 0.0186, 0.0188, and 0.0180 mSv/mGy x cm. In the patient examinations, dose was 4.08 +/- 0.81 mSv with the high-pitch protocol compared with 20.4 +/- 5.3 mSv in the matched controls with the conventional technique, and 4.40 +/- 0.83 mSv for the non-ECG gated thorax scan. Scan times were 0.7 +/- 0.1 seconds for the high-pitch scan and 15 +/- 3 seconds for the conventional chest pain scan. Aorta and pulmonary arteries were depicted in diagnostic quality in both groups. About 84.7% of coronary artery segments were rated as diagnostic in the high-pitch exams (95.4% below 65 bpm and only 72.8% in higher heart rates), whereas 92.9% were diagnostic with the conventional approach. CONCLUSION: The high-pitch dual spiral technique requires only about one-fifth of the dose of conventional ECG gated triple-rule-out protocols, thus matching that of a standard nongated chest scan. With less than 1 second, the scan time is very short. This protocol can be recommended for patients with unclear chest pain with rhythmic heart rates below 65 bpm.