A Comparison of Sequential and Spiral Scanning ...

Peer Review

A Comparison of Sequential and Spiral Scanning Techniques in Brain CT Ivana Pace, BSc(Hons) Francis Zarb, PhD Purpose To evaluate and compare image quality and radiation dose of sequential computed tomography (CT) examinations of the brain and spiral CT examinations of the brain imaged on a GE HiSpeed NX/I Dual Slice 2CT scanner. Methods A random sample of 40 patients referred for CT examination of the brain was selected and divided into 2 groups. Half of the patients were scanned using the sequential (n  20) technique; the other half were scanned using the spiral (n  20) technique. Radiation dose data—both the computed tomography dose index (CTDIvol) and the dose length product (DLP)—were recorded on a checklist at the end of each examination. Using the European Guidelines on Quality Criteria for Computed Tomography, 4 radiologists conducted a visual grading analysis and rated the level of visibility of 6 anatomical structures that are considered necessary according to the mentioned criterion to produce images of high quality. Results The mean CTDIvol and DLP values were statistically significantly higher (P  0.05) with the sequential scans (CTDIvol: 22.06 mGy; DLP: 304.60 mGy cm) than with the spiral scans (CTDIvol: 14.94 mGy; DLP: 229.10 mGy cm). The mean image quality rating scores for all criteria of the sequential scanning technique were statistically significantly higher (P  0.05) in the visual grading analysis than those of the spiral scanning technique. Discussion In this local study, the sequential technique was preferred over the spiral technique for both overall image quality and differentiation between grey and white matter in brain CT scans. Other similar studies counter this finding. Conclusion The radiation dose seen with the sequential CT scanning technique was significantly higher than that seen with the spiral CT scanning technique. However, image quality with the sequential technique was statistically significantly superior (P  0.05).

T

he introduction of computed tomography (CT) was a ground-breaking development in structural brain imaging, enabling better evaluation of pathologies that in the past were difficult to detect.1 CT is now the first-line modality of choice for evaluating the brain and the most often performed CT examination in many hospitals.2 The increased frequency in CT use has raised concerns regarding the amount of radiation exposure the patient receives during the examination. During a CT examination of the brain, the radiation dose is equally distributed throughout the field being scanned. To minimize radiation dose effects, the dose should be optimized and the effects monitored by establishing diagnostic reference levels1,3 using radiation dose RADIOLOGIC TECHNOLOGY, March/April 2015, Volume 86, Number 4

descriptors—the computed tomography dose index (CTDI) and the dose length product (DLP). 4 Radiation dose optimization is achieved by reducing the radiation dose as much as possible, with image quality still meeting the requirements for adequate diagnosis (diagnostic efficacy). The European Guidelines on Quality Criteria for Computed Tomography can be referred to during radiation dose optimization in CT examinations as it provides a list of anatomical structures that should be visualized to obtain an accurate diagnosis.5 CT images of the brain can be obtained using either sequential or spiral CT scanning techniques. In sequential scanning, the CT table moves through the rotating gantry, which images thin slices of the brain. Because the table advances only after each slice is scanned, this 1

Peer Review A Comparison of Sequential and Spiral Scanning Techniques in Brain CT

technique is time consuming, prone to potential misregistration and motion artifacts, and has limited availability of the overlapping images used for postprocessing. 6,7 In spiral scanning, the CT table moves through the gantry at a constant speed as scanning occurs, resulting in faster scan times, continuous data acquisition, and continuous radiation.1,8 Studies comparing image quality between the 2 CT scanning techniques have returned contradictory outcomes. A 1998 study by Kuntz et al found similar image quality between the 2 scanning techniques. A similar 1998 study by Bahner et al found that the sequential scanning technique returned better image quality, especially of small structures with low tissue contrast.9,10 Straten et al showed contradictory results with regard to radiation dose. Quality images achieved with the spiral technique had an increase in patient radiation dose, especially when using multislice CT scanners.11 Conversely, findings by Abdeen et al indicated much higher DLP values, representing higher radiation doses during sequential scans.12 Malta, one of the world’s smallest and most densely populated countries, has one general public hospital equipped with 2 CT units. Most brain CT examinations are performed using a dual-slice CT scanner in sequential mode, but approximately a third are performed in spiral mode for situations requiring faster scanning to reduce the potential for motion artifacts. No local research exists comparing the image quality and radiation dose of the 2 scanning techniques, resulting in an unspecific standard scanning protocol for brain CT in that Malta facility. The aim of this study was to compare and evaluate sequential CT examinations of the brain with spiral CT examinations of the brain performed at the local general hospital using a high-speed dual-slice CT scanner. Radiation dose was compared according to CTDIvol and DLP, and image quality was evaluated by 4 radiologists who visually graded anatomical structures as outlined in the European Guidelines on Quality Criteria for Computed Tomography.

Methods

Before collecting the data, permission was sought and obtained from the University of Malta Research 2

and Ethics Committee board, and from the Medical Imaging Department and the Data Protection Officer of the hospital where the data were collected. This study comprised 2 elements. The first was the CT images of those patients referred for brain CT examinations as part of their diagnosis at the local hospital. Pediatric patients and patients with pathologies that might affect the evaluation of the anatomical structures were excluded from the study. The second constituted the radiologists who worked at the hospital. Those radiologists eligible to participate in the study had a minimum of 2 years of experience in CT image reporting. An equal number (n  20) of brain CT examinations from each scanning technique were randomly selected, and the participation of each radiologist depended solely on his or her availability and willingness to participate during the time of data collection. Radiation dose data were recorded on a checklist incorporating details of the scanning technique used: Scan parameters:  slice thickness  interslice distance  pitch  scan length  gantry tilt  reconstruction algorithm Radiation dose descriptors:  CTDIvol DLP Exposure factors:  mA  kV The data were collected prospectively, as both CTDIvol and DLP were not available retrospectively on the CT unit used in the study. Image quality was evaluated by 4 radiologists who rated 6 anatomical structures according to their level of visibility on a scale of 1 to 5 (1 indicating that the anatomical structure was not at all visible, 5 indicating excellent and well-defined visibility) as outlined in the European Guidelines on Quality Criteria for Computed Tomography (see Box). The image quality data set for evaluation consisted of 40 brain CT examinations equally divided between the sequential (n  20) and the spiral (n  20) scanning techniques. Two images of each scanning RADIOLOGIC TECHNOLOGY, March/April 2015, Volume 86, Number 4

Peer Review Pace, Zarb

Box

Table 1

European Guidelines on Quality Criteria for 5 Computed Tomography

Sequential and Spiral Scan Protocols

1. Visually sharp reproduction of the border between white and grey matter.

Dose-Related Parameters

2. Visually sharp reproduction of the basal ganglia.

Slice thickness (mm)

3. Visually sharp reproduction of the ventricular system.

Interslice thickness/gap (mm)

4

14

4. Visually sharp reproduction of the cerebrospinal fluid space around the mesencephalon.

Pitch factor (mm/rotation)

n/a

n/a

Average scan length (cm)

13.84

15.34

5. Visually sharp reproduction of the cerebrospinal fluid space over the brain.

kV

120

120

Smart mA (max)

160

100

Rotation time (s)

1

1

5.5 – 20 (mean : 13)

No tilt

Standard

Standard

technique were replicated to facilitate intrarater and interrater reliability testing. These CT examinations were anonymized and evaluated blindly by radiologists at a workstation (GE Advantage) with a 3-megapixel monitor (Barco Voxar 3D-TeraRecon AquariusNET). The quantitative data collected were analyzed using the Statistical Package for the Social Sciences software (SPSS, version 17.0). Data analysis included Pearson correlation, Cronbach α, the 2-tailed independentsample t-test, and visual grading analysis using the Mann-Whitney U test. All statistical measures were considered significant at the 95% confidence interval.

Results

Standard CT scanning protocol When a brain CT is performed at the local general hospital using the sequential technique, 2 different protocols are used: one for the posterior fossa, and another for the cerebrum. Only one protocol is used with the spiral technique (see Table 1). Observer reliability Interobserver reliability results showed that all correlations were positive, indicating conformity among the 4 radiologists. Pearson’s correlation (r) was calculated, and the range for the 4 observers varied from r  0.341 to 0.767. Moreover, the result showed that all correlations were statistically significant (P  0.05), indicating that interobserver reliability is not attributed to chance and that a 100% consistency existed among the observers overall. RADIOLOGIC TECHNOLOGY, March/April 2015, Volume 86, Number 4

Sequential Technique

Spiral Technique

Posterior fossa Cerebrum

Gantry tilt (degrees) Reconstruction algorithm

4

7

Furthermore, Pearson’s correlation (r) also was calculated for intraobserver reliability, and the range of the 4 observers varied from r  0.11 to 0.90. All correlations were positive, indicating compliance among the observers. Consequently, Cronbach α ranged from   0.19 to 0.95, indicating satisfactory internal consistency in the observers’ rating of the images. Radiation dose – CTDIvol and DLP The purpose of the independent-sample t-test was to compare the difference between the means of 2 independent groups.13 Table 2 shows the difference between the means of the 2 radiation dose descriptors for each scanning technique. The results indicate a statistically significant difference between the 2 scanning techniques for both CTDIvol and DLP, since P  0.05. The mean scores for both CTDIvol and DLP obtained from the sequential technique images are significantly higher than the mean scores obtained from the spiral technique images. Image quality criteria The image-quality rating scores provided during the VGA of the image sets evaluated by the 4 radiologists were analyzed using the Mann Whitney U test, which compared the mean ranking scores between the 2 techniques. Descriptive statistics were used to present the 3

Peer Review A Comparison of Sequential and Spiral Scanning Techniques in Brain CT

Table 2 Comparison of Radiation Dose Indicators Between Sequential and Spiral CT Brain Examinations Dose Indicators

Scanning Technique

N

Max. Value

Mean Score

CTDIvol (mGy)

Sequential

20

20.21

23.69

22.06

0.97

Spiral

20

12.52

16.35

14.94

1.07

Sequential

20

282.90

342.45

304.60

15.01

Spiral

20

191.51

258.64

229.10

18.43

DLP (mGy cm)

Min. Value

average rating of the image-quality scores. The mean rating scores provided for the images obtained with the sequential scanning technique with respect to all the criteria were statistically significantly higher than the mean rating scores provided for the images obtained with the spiral scanning technique, since P  0.05 for all the 5 criteria of significance. Thus, this result indicates a preference toward images produced by the sequential technique with regard to image quality (see Table 3).

Discussion

The radiation dose results in this study match the findings of Abdeen et al and Alberico et al showing a reduction in radiation dose for the spiral technique.12,14 DLP is an important measure, as it gives an overall radiation dose for a given scanning protocol and incorporates both CTDIvol and the scan length of the examination.15 The DLP values in this local study for both scanning techniques (304.60 mGy cm for sequential and 229.10 mGy cm for spiral) and in Abdeen et al (1005 mGy cm for spiral) were found to be lower than the DLP recommended in the European Guidelines on Image Quality Criteria for Computed Tomography for a general brain CT (1050 mGy cm). This variation indicates that the radiation dose patients receive is within the recommended diagnostic reference levels for brain CT. Consequently, the DLP values obtained in this local study are lower than the values seen in the Abdeen et al study. However, a lower DLP was obtained in both studies with the spiral technique. It is noteworthy that 16- and 64-slice CT systems were used in the Abdeen 4

Std. Deviation

P Value  0.05

 0.05

Table 3 Comparison of the Mean Rating Score in Image Quality Criteria Between Sequential and Spiral CT Brain Examinations Mann-Whitney U Test (P Value)

Mean Rating Score Criteria

Sequential

Spiral

1

4.02

3.20

 0.05

2

4.19

3.03

 0.05

3

4.63

4.17

 0.05

4

4.55

4.22

 0.05

5

4.58

4.23

 0.05

et al study. A difference in scanning parameters also was noted.12 In comparison to other studies,12,14 lower DLP values in the local study could be attributed to the use of Smart-mA, a type of automatic exposure control system that adjusts the required optimal mA and thereby reduces the radiation dose the patient receives as the tube rotates around his or her head. The spiral technique reaches a lower maximum mA (100 mA) when compared to the maximum mA reached when scanning with the sequential technique (160 mA). This difference also might contribute to the lower DLP value reached when scanning with the spiral technique. Straten et al concluded that the only disadvantage of the spiral technique is using it with a “relatively old scanner, like the 4-section CT scanner,” as it delivers a higher radiation dose. However, this disadvantage cannot be related to the local study—even though a RADIOLOGIC TECHNOLOGY, March/April 2015, Volume 86, Number 4

Peer Review Pace, Zarb

dual-slice CT unit was used—because the radiation dose was found to be lower than with the sequential technique.11 However, it is important to consider that strategies to reduce radiation dose often compromise image quality. In contrast to the local study, where all 5 anatomical criteria scored lower in image quality with the spiral technique when compared with the sequential technique, the study conducted by Straten et al showed a preference for the spiral technique with regard to improved overall image quality, especially in the case of brain tissue near the skull. In the local study, overall results showed a clear preference for the sequential technique, with the visualization of the basal ganglia scoring much higher. In the study by Abdeen et al, as in this local study, the sequential technique was preferred over the spiral technique for both overall image quality and differentiation between grey and white matter; contrary to the local study, however, this finding was not statistically significant in the Abdeen et al study. There was a statistically significant difference in the visualization of the basal ganglia in both studies. The reasons Abdeen et al gave for this preference were the tube currents and the different peak voltages used (sequential: 120 kVp cerebrum, 140 kVp posterior fossa, 150 mA cerebrum, 170 mA posterior fossa; spiral: 120 kV and 300 mA).12 In contrast to this local study, the tube current used in the spiral technique (100 mA) was much lower than the one used in the sequential technique (160 mA), while the same peak voltage (120 kV) was used, thus providing a possible reason the spiral technique scored significantly lower in image quality in the local study. The primary limitation of the local study was the use of only one CT scanner, a dual-slice CT scanner, thus omitting other scanners within the same imaging department, at other local hospitals, and within the public and private sectors that contribute to a large number of brain CT examinations performed locally. Another limitation was that the scanning parameters used were from the local protocol, which could have led to the differences in results between the 2 scanning techniques. In this local study, the sequential technique produces higher image quality; however, this does not mean that the images produced using the spiral technique are RADIOLOGIC TECHNOLOGY, March/April 2015, Volume 86, Number 4

insufficient for diagnosis. Further research to evaluate the diagnostic efficacy of the images produced by the current spiral technique used in the site of the local study is recommended, as is investigating ways to optimize the scanning protocols.

Conclusion

The results of this study can be generalized to brain CT examinations using a GE HiSpeed NX/I Dual Slice CT scanner in line with the local CT protocol. The spiral technique was found to have a significantly lower radiation dose over the sequential technique. Statistical analysis showed that no matter how large the sample, important differences in both CTDIvol and DLP would still be obtained, and the difference between the 2 techniques would remain the same. The difference in image quality between the 2 scanning techniques was significant for all 5 anatomical criteria. This difference showed that sequential scanning was the preferred option with regard to image quality. In the case of the visualization of the basal ganglia, the difference in the mean score was highest, while the least difference between the scanning techniques was seen in the visualization of the ventricular system. Further study is recommended to adapt the results to other types of CT scanner units and other effective measures to interpret image quality and detect abnormalities such as visual grading analysis and receiver operator characteristics. Ivana Pace, BSc(Hons), is a diagnostic radiographer in the medical imaging department at Mater Dei Hospital in Msida, Malta. Pace can be reached at [email protected]. Francis Zarb, PhD, is a lecturer, in the Department of Radiography, Faculty of Health Sciences at the University of Malta in Msida. Zarb can be reached at [email protected]. Received November 23, 2013; accepted after revision January 24, 2014. Reprint requests may be mailed to the American Society of Radiologic Technologists, Communications Department, at 15000 Central Ave SE, Albuquerque, NM 87123-3909, or e-mailed to [email protected]. © 2015 American Society of Radiologic Technologists

5

Peer Review A Comparison of Sequential and Spiral Scanning Techniques in Brain CT

References

1. Kalender W. Computed Tomography: Fundamentals, System Technology, Image Quality, Applications. Erlangen, Germany: Publicis Publishing; 2005. 2. Lee B, Newberg A. Neuroimaging in traumatic brain imaging. NeuroRx. 2005;2(2):372-383. 3. Kalra MK, Maher MM, Toth TL, et al. Strategies for CT radiation dose optimization. Radiology. 2004;230(3):619628. 4. Smith AB, Dillon WP, Gould R, Wintermark M. Radiation dose-reduction strategies for neuroradiology CT protocols. AJNR AM J Neuroradiol. 2007;28(9):1628-1632. 5. European Commission. European Guidelines on Quality Criteria for Computed Tomography. Report EUR 16262. Brussels: Belgium; 1999. 6. Hall EJ, Brenner DJ. Cancer risks from diagnostic radiology. Br J Radiol. 2008;81(965): 362-378. doi:10.1259 /bjr/01948454. 7. Saunders J, Ohlerth S. Veterinary computed tomography. In: Schwarz T, Saunders J, eds. CT Physics and InstrumentationMechanical Design. 3rd ed. West Sussex, UK: John Wiley & Sons Ltd; 2008:1-8. 8. Radaideh MM, West OC, Silverman PM. Using multi-slice spiral CT scanner: the principles you need. The University of Texas MD Anderson Cancer Center. 2011. www.uth.tmc. edu/radiology/media/documents/exhibits/radaideh/multi_ slice_scanner.pdf. Accessed November 2, 2011. 9. Kuntz R, Skalej M, Stefanou A. Image quality of spiral CT versus conventional CT in routine brain imaging. Eur J Radiol. 1998;26(3):235-240. 10. Bahner ML, Reith W, Zuna I, Engenhart-Cabillic R, van Kaick G. Spiral CT vs incremental CT: is spiral CT superior in imaging of the brain? Eur Radiol. 1998;8(3):416-420. 11. Straten M, Venema HW, Majoie CBLM, Freling NJM, Grimbergen CA, Heeten GJ. Image quality of multisection CT of the brain: thickly collimated sequential scanning versus thinly collimated spiral scanning with image combining. AJNR AM J Neuroradiol. 2007;28(3):421-427. 12. Abdeen N, Chakraborty S, Nguyen T, et al. Comparison of image quality and lens dose in helical and sequentially acquired head CT. Clin Radiol. 2010;65(11):868-873. doi:10.1016/j.crad.2010.02.016. 13. Polit DF, Beck CT. Nursing Research: Principles and Methods. 8th ed. Philadelphia, PA: Lippincott Williams & Wilkins; 2007. 14. Alberico RA, Loud P, Pollina J, Greco W, Patel M, Klufas R. Thick-section reformatting of thinly collimated helical CT for reduction of skull base-related artifacts. AJR AM J Roentgenol. 2000;175 (5):1361-1366. 6

15. McCollough CH, Primak AN, Braun N, Kofler J, Yu L, Christner J. Strategies for reducing radiation dose in CT. Radiol Clin North Am. 2009;47(1):27-40. doi:10.1016/j .rcl.2008.10.006.

RADIOLOGIC TECHNOLOGY, March/April 2015, Volume 86, Number 4