Application of T1-weighted BLADE sequence to abdominal magnetic resonance imaging of young children: a comparison with turbo spin echo sequence

Abstract

Background

The image quality of abdominal magnetic resonance imaging (MRI) in children who cannot hold their breath has been severely impaired by motion artifacts.

Purpose

To evaluate the usefulness of T1-weighted (T1W) BLADE MRI for axial abdominal imaging in children who cannot hold their breath.

Material and Methods

Two different BLADE sequences, with and without an inversion recovery (IR-BLADE), were compared to conventional turbo-spin echo (TSE) with a high number of excitations in 18 consecutive patients who cannot hold their breath. Overall image quality, motion artifact, radial artifact, hepatic vessel sharpness, renal corticomedullary differentiation, and lesion conspicuity were retrospectively assessed by two radiologists, using 4- or 5-point scoring systems. Signal variations of each sequence were measured for a quantitative comparison. The acquisition times of the three sequences were compared.

Results

IR-BLADE and BLADE showed significantly improved overall image quality and reduced motion artifact compared with TSE. IR-BLADE showed significantly better hepatic vessel sharpness and corticomedullary differentiation compared to both BLADE and TSE. Radial artifacts were only observed on IR-BLADE and BLADE. In nine patients with lesions, there were no significant differences in lesion conspicuity among three sequences. Compared to TSE, both IR-BLADE and BLADE showed decreased signal variations in the liver and muscle, and an increased signal variation through air. The mean acquisition times for IR-BLADE, BLADE, and TSE were comparable.

Conclusion

Compared to the TSE sequence, T1W IR-BLADE for pediatric abdominal MRI resulted in improved image quality, tissue contrast with a diminished respiratory motion artifact, and a comparable acquisition time.

Keywords

Magnetic resonance imaging PROPELLER inversion recovery spin echo imaging artifacts

Introduction

Recently, abdominal magnetic resonance imaging (MRI) has been performed more increasingly in children because of its excellent soft-tissue contrast and absence of radiation hazards (1 –3). Traditionally, the T1-weighted (T1W) turbo-spin echo (TSE) technique has been widely used as a fundamental T1W sequence (4,5) to achieve excellent T1 contrast, good image quality, and acceptable acquisition time. For T1W TSE imaging, some strategies are required to prevent motion artifacts and the resulting image degradation, because young children are often unable to lie down and hold their breath for the required time. For example, using respiratory gating and/or increasing the number of signals acquired (NSA) are the most commonly accepted strategies to reduce motion artifacts for T1W TSE imaging (1). However, the main drawback of respiratory gating is the prolonged imaging time for patients with an irregular breathing pattern. On the other hand, by increasing NSA, motion artifacts can be reduced because it averages redundant signals and, more specifically, a NSA of six provides acceptable motion reduction without significantly delaying the scanning time on abdominal MRI (1).

Periodically rotated overlapping parallel lines with enhanced reconstruction (PROPELLER), also known as its commercial name BLADE (Siemens Healthcare, Erlangen, Germany), was originally proposed by Pipe (6) in 1999 as a non-Cartesian k-space filling technique used to compensate for respiratory motion. The basic pulse sequence of BLADE sequence is also TSE sequence, not a gradient recalled echo (GRE)-based sequence. However, the difference between BLADE and conventional TSE is that BLADE is a hybrid of Cartesian and radial k-space filling (6). Recently, BLADE has been applied in various clinical settings in order to reduce motion artifacts (4,5). This multi-shot, fast spin-echo based sequence with radial k-space sampling via use of rotated blades, contains phase-encoding lines, which disperses motion artifacts throughout the radial section (5,7). Moreover, inherent oversampling of the central k-space corrects motion due to the in-plane rotation and translation (4,8). Hitherto, several studies have reported that the PROPELLER or BLADE techniques were able to reduce motion artifacts in T2-weighted (T2W) images of pediatric patients (9,10). However, to our knowledge, there have been only a few applications of the BLADE technique for T1W imaging, most of which concern head and neck or brain imaging (11 –13). As a relatively long echo time (TE) is required for BLADE, which is contrary to the purposes of T1 contrast, it can be difficult to obtain sufficient T1 contrast. Alternatively, the optimal T1 contrast can be achieved in conjunction with an inversion recovery (IR) technique (11). To date, to our knowledge, there has been no study examining the usefulness of the inversion recovery BLADE technique in T1W pediatric abdominal imaging. Therefore, the aim of the present study was to compare the following motion-reducing techniques employed in T1W pediatric abdominal imaging in both a qualitative and quantitative manner: BLADE; inversion recovery BLADE (IR-BLADE); and conventional TSE with a high NSA.

Material and Methods

This retrospective study was approved by our institutional review board of Seoul National University Hospital, with a waiver for obtaining informed patient consent.

Patients

From January 2015 to May 2015, 18 consecutive children, who were unable to voluntarily hold their breath, underwent abdominal MR examinations. The indications for abdominal MRI included the initial evaluation or post-treatment surveillance for the presence of neuroblastoma (n = 8), rhabdomyosarcoma (n = 2), Ewing sarcoma (n = 2), germ cell tumor (n = 1), hepatoblastoma (n = 1), lymphangioma (n = 1), hemangioma (n = 1), ganglioneuroblastoma (n = 1), and renal rhabdoid tumor (n = 1). Among the total of the 18 abdominal MR examinations, nine examinations identified lesions in the following: retroperitoneum (n = 3); liver (n = 3); subdiaphragmatic area (n = 2); and bladder (n = 1). The other nine examinations did not show any pathology.

MR examination

All MR examinations included the following three T1W sequences: axial IR-BLADE sequence; conventional BLADE sequence without IR; and T1W TSE with a high NSA. All MR examinations were conducted with a 1.5-T MR scanner (Avanto, Siemens Healthcare, Erlangen, Germany), with the children freely breathing. Inversion pulse at 860 ms was used for IR-BLADE for fat suppression, or nulling fat signals (4,11). No MR exams were performed with the use of contrast material. During MR scanning, 12/18 children were sedated using chloral hydrate (Pocral syrup); IV ketamine and/or midazolam were used, if necessary. The detailed imaging parameters for the three MR sequences are presented in Table 1.

Table 1.

MR parameters for BLADE techniques and TSE with a high NSA.

MR parameters	IR-BLADE	BLADE	TSE with 6 NSA
TR range (ms)	2120–3120	450–717	440–535
TE ranges (ms)	53 or 55	23 or 24	6.9
Flip angle (°)	150	140	180
Slice thickness (mm)	4–7	4–7	4–7
Spacing (mm)	4.4–8.4	4.4–8.4	4.4–8.4
Echo-train length	19	9	4
Field of view (cm)	18–26 × 18–26	18–26 × 18–26	18–26 × 15–19.5
Matrix	256 × 256	256 × 256	256 × 192
Band width (Hz/Px)	362	514	250
Parallel imaging factor (GRAPPA)	2	2	2
BLADE coverage (%)	135	175	–
Inversion time (ms)	860		–
Average acquisition time (min:s)	3:27 (range = 3:04–4:14)	3:46 (range = 2:08–6:05)	3:24 (range = 2:18–5:10)

BLADE, Siemens Healthcare (Erlangen, Germany) implementation for periodically rotated overlapping parallel lines with enhanced reconstruction (PROPELLER); GRAPPA, generalized autocalibrating partially parallel acquisition; IR, inversion recovery; MR, magnetic resonance; NSA, number of signal acquisitions; TE, echo time; TR, repetition time; TSE, turbo spin-echo.

Qualitative analysis

Two radiologists (YHC and KSC, with 12 and 3 years of experience as pediatric radiologists, respectively) assessed 54 axial imaging series (18 imaging series for each sequence of IR-BLADE, BLADE, and TSE with high NSA) in a randomized order. The observers were blinded to the clinical and imaging information such as imaging sequence, patient’s age, and diagnosis, and assessed all the images on the PACS (Gx; Infinitt healthcare, Seoul, Republic of Korea) in a stack mode, allowed to scroll though the images without restriction of reading time and window level setting adjustment. The scores from the two observers of each imaging assessment category were collected for IR-BLADE, BLADE, and TSE with a high NSA.

To compare the image quality of IR-BLADE, BLADE, and TSE with a high NSA, six imaging categories were evaluated with a 4- or 5-point scoring system: overall image quality; motion artifact; radial artifact; hepatic vessel sharpness; corticomedullary differentiation of kidneys; and lesion conspicuity. Two categories related to artifacts were evaluated with a 4-point scoring system: 0 = absent; 1 = mild; 2 = moderate; and 3 = severe. Radial artifacts are defined as “star-like” streaking artifacts that occur because of the dispersion of motion artifacts along the phase encoding direction, which is in a radial direction in the BLADE technique. Overall image quality, hepatic vessel sharpness, and renal corticomedullary differentiation were evaluated with a 5-point scoring system: 1 = unacceptable; 2 = poor; 3 = acceptable; 4 = good; and 5 = excellent. The lesion conspicuity for the 19 lesions was scored using a 5-point scale: 1 = non-delineable; 2 = blurred but visualized; 3 = acceptable; 4 = good; and 5 = excellent.

Quantitative analysis

At the vertebral level visualizing both the liver and the kidneys on imaging, three circular regions of interest (ROIs) of the same size (100 mm²) were drawn onto the image at the same level, corresponding to the liver parenchyma, paraspinal muscle, and background air. The coefficient of signal variation was calculated by a ratio of the SD to the mean signal value within the ROIs.

Statistical analysis

All statistical analyses were performed using IBM SPSS Statistics, Version 22 (SPSS Inc., Chicago, IL, USA). For qualitative analysis, the mean scores of the two observers for the three sequences were compared using the Friedman test. A P value < 0.05 was considered statistically significant. Mann–Whitney U test with Bonferroni correction for multiple comparisons was performed as post-hoc analysis. For subgroup analysis—by sedation status and lesion size (>2 cm)—Mann–Whitney U test, which is a non-parametric equivalent of the independent t-test, was performed. A P value < 0.017 was considered statistically significant for post-hoc analysis. For the analysis of the inter-observer agreement, Cohen’s kappa test was performed. The following kappa values are indicated with their respective ratings: 0–0.20 indicates poor agreement; 0.21–0.40, fair agreement; 0.41–0.60, moderate agreement; 0.61–0.80, good agreement; and 0.81–0.99, excellent agreement. For quantitative analyses, Wilcoxon signed rank test was performed to compare the coefficients of the signal variations for IR-BLADE, BLADE, and TSE with a high NSA. Because sample data were small (n = 18) and were not normally distributed when using Kolmogorov–Smirnov test, the Wilcoxon signed rank test, a non-parametric equivalent of the paired t-test, was used instead. In addition, the paired sampled test should be used because all the three sequences were obtained for the same patient.

Results

There were 18 patients (8 boys, 10 girls; mean age =4.4 ± 1.9 years; age range = 6 months–8 years) and most of the patients had difficulty holding their breath for a reasonable period of time. The qualitative analyses, as indicated by the average scores of observers 1 and 2, of IR-BLADE and BLADE showed improved overall image quality (IR-BLADE vs. BLADE vs. TSE, 4.08 ± 0.52 vs. 3.61 ± 0.44 vs. 3.11 ±0.47; P < 0.001) and reduced motion artifact (1.25 ± 0.43 vs. 1.19 ± 0.77 vs. 2.19 ± 0.64; P < 0.001) compared with TSE (Fig. 1 and Table 2). According to the averaged scores between observers 1 and 2, IR-BLADE showed superior edge sharpness of the hepatic vessels (3.69 ± 0.49 vs. 2.53 ± 0.47 vs. 2.17 ± 0.51; P < 0.001) and corticomedullary differentiation (4.14 ± 0.61 vs. 2.69 ± 0.49 vs. 2.94 ± 0.48; P < 0.001) compared to both BLADE and TSE (Fig. 2 and Table 2). Radial artifacts were only observed on IR-BLADE and BLADE, and all the BLADE images showed radial artifacts. The mean size of the total 16 lesions was 1.64 cm ± 1.54. In nine patients with lesions, there was no significant difference in lesion conspicuity among three sequences (2.89 ± 0.60 vs. 2.72 ± 0.71 vs. 2.39 ± 0.60; P = 0.273) (Fig. 3 and Table 2). Five out of nine patients with lesions have shown at least one lesion, which is >2 cm. In subgroup analysis by lesion size, IR-BLADE and TSE showed higher lesion conspicuity compared to BLADE (3.50 ± 0.71 vs. 3.30 ±0.44 vs. 2.63 ± 1.11, respectively) for small lesions (<2 cm), though statistically not significant (P = 0.223). In contrast, IR-BLADE and BLADE showed higher lesion conspicuity than in TSE (3.40 ±0.65 vs. 2.80 ± 0.27 vs. 2.20 ± 0.22, respectively) for large lesions (>2 cm), though statistically not significant (P = 0.150). In subgroup analysis by sedation status, there was no significant difference in motion artifacts between non-sedated and sedated group with all sequences (BLADE = 0.67 ± 0.52 vs. 1.47 ± 0.75, P = 0.082; IR-BLADE = 1.08 ± 0.20 vs. 1.33 ± 0.49, P = 0.335; TSE = 2.00 ± 0.64 vs. 2.58 ± 0.49, P = 0.067). For the assessment of the inter-observer agreement, the kappa values for the qualitative analyses were 0.21–0.40 (11.8%), 0.41–0.60 (64.7%), and 0.61–0.80 (23.5%), showing fair to good agreement between the two observers (Table 2).

Fig. 1.

A five-year-old girl with neuroblastoma (lesion not shown). (a) T1 IR-BLADE image shows some “star-like streaking” radial artifacts (arrows), which is not present in TSE with high NSA image. (b) Ghosting artifacts of anterior abdominal wall (short arrows) due to respiratory motion and pulsating aorta (long arrows) are shown on TSE with high NSA, which is not present on the T1 IR-BLADE image (a), which may possibly obscure the liver parenchymal lesion, if present. IR, inversion recovery; NSA, number of signal acquisitions; TSE, turbo spin-echo.

Table 2.

Comparison of subjective image quality among three sequences.

Image quality assessment category	IR-BLADE	BLADE	TSE with high NSA	IR-BLADE* vs. BLADE	IR-BLADE vs. TSE with high NSA*	BLADE vs. TSE with high NSA*	BLADE vs. IR-BLADE vs. TSE with high NSA^†
Overall image quality
Observer 1	3.67 ± 0.59	3.56 ± 0.62	3.06 ± 0.54
Observer 2	4.50 ± 0.62	3.67 ± 0.59	3.17 ± 0.62
Averaged score	4.08 ± 0.52	3.61 ± 0.44	3.11 ± 0.47	P = 0.0027	P < 0.0001	P = 0.0002	P < 0.001
Kappa value	0.58	0.5	0.558
Respiratory motion artifact
Observer 1	1.33 ± 0.49	1.28 ± 0.67	2.11 ± 0.68
Observer 2	1.17 ± 0.51	1.11 ± 1.08	2.28 ± 0.75
Averaged score	1.25 ± 0.43	1.19 ± 0.77	2.19 ± 0.64	P = 0.7695	P = 0.0001	P = 0.0021	P < 0.001
Kappa value	0.52	0.61	0.46
Radial artifact
Observer 1	0.72 ± 0.46	0.83 ± 0.62	0
Observer 2	0.33 ± 0.49	0.39 ± 0.61	0
Averaged score	0.53 ± 0.40	0.61 ± 0.53	0	P = 0.6250	–	–	P < 0.001
Kappa value	0.492	0.556	–
Hepatic vessel sharpness
Observer 1	3.56 ± 0.62	2.28 ± 0.46	2.22 ± 0.65
Observer 2	3.83 ± 0.62	2.78 ± 0.65	2.11 ± 0.47
Averaged score	3.69 ± 0.49	2.53 ± 0.47	2.17 ± 0.51	P < 0.0001	P < 0.0001	P = 0.0245	P < 0.001
Kappa value	0.404	0.407	0.686
Corticomedullary differentiation
Observer 1	4.00 ± 0.49	2.93 ± 0.50	2.83 ± 0.38
Observer 2	4.28 ± 0.89	3.00 ± 0.77	3.06 ± 0.64
Averaged score	4.14 ± 0.61	2.69 ± 0.49	2.94 ± 0.48	P < 0.0001	P < 0.0001	P = 0.0547	P < 0.001
Kappa value	0.438	0.382	0.52
Lesion conspicuity
Observer 1	3.33 ± 0.87	2.56 ± 0.88	2.33 ± 0.71
Observer 2	3.56 ± 0.73	2.89 ± 0.60	2.44 ± 0.53
Averaged score	3.44 ± 0.63	2.72 ± 0.71	2.39 ± 0.60	P = 0.375	P = 0.1875	P = 0.4609	P = 0.273
Kappa value	0.654	0.357	0.8

Values are given as mean ± SD.

Kappa values = 0–0.20, poor; 0.21–0.40, fair; 0.41–0.60, moderate; 0.61–0.80, good; 0.81–1.00, excellent inter-observer agreement.

*P values < 0.017 are considered significant (Bonferroni correction for multiple comparison) (bold), using Wilcoxon signed rank test.

^†P values < 0.05 considered significant (bold), using Friedman test.

BLADE, Siemens Healthcare (Erlangen, Germany) implementation for periodically rotated overlapping parallel lines with enhanced reconstruction (PROPELLER); IR, inversion recovery; NSA, number of signals acquired; TSE, turbo spin-echo.

Fig. 2.

A five-year-old girl with neuroblastoma (lesion not shown). (a) T1 IR-BLADE image (a) shows more prominent corticomedullary differentiation in both kidneys (arrows) and improved blood vessel edge contrast (arrowhead) compared with (b) TSE with high NSA image (b). IR, inversion recovery; NSA, number of signal acquisitions; TSE, turbo spin-echo.

Fig. 3.

An eight-year-old boy with recurred ganglioneuroblastoma. (a) T1 IR-BLADE image shows better lesion conspicuity of the right retroperitoneal mass (arrowheads) showing better contrast between right kidney and the mass, as well as reduced respiratory motion artifact, more prominent corticomedullary differentiation in both kidneys, and improved blood vessel sharpness, compared with (b) TSE with high NSA image. IR, inversion recovery; NSA, number of signal acquisitions; TSE, turbo spin-echo.

The quantitative analyses indicated that both IR-BLADE and BLADE showed decreased signal variation, compared to TSE with a high NSA, in the regions of interests (ROI) that were drawn onto the liver (2.85% vs 3.86% vs 6.53%, p < 0.001), and muscle (7.11% vs 8.60% vs 22.7%, p < 0.001), with an increased signal variation evident through the medium of air (55.7% vs 43.3% vs 40.4%, p = 0.006). The mean acquisition times for IR-BLADE, BLADE and TSE were 3 min 46 s, 3 min 27 s, and 3 min 24 s respectively, which had no significant difference (p = 0.64) (Table 1).

Discussion

The pre-contrast T1W sequence is fundamental in pediatric abdominal MRI because it provides important information about the anatomical structures and pathologies of solid organs such as the liver, pancreas and kidneys. Additionally, it is helpful in visualizing the fat component of tissues and identifying the presence of subacute hemorrhage in pathologies (4). The standard, abdominal T1W sequence can be obtained using spin-echo or GRE sequences. The T1W TSE sequence has superior signal-to-noise ratio (SNR) and image quality compared to the GRE sequence, and it is less affected by susceptibility artifacts, field inhomogeneity, and chemical shift effects. Therefore, a spin-echo sequence is preferred when obtaining T1W images of an abdominal mass or a focal parenchymal lesion (4). However, motion artifacts including respiratory and cardiac motion are the most important artifacts that may affect the overall image quality and subsequent diagnostic performance of a sequence. Furthermore, these factors cannot be fully removed by using sedative or even general anesthesia. As a result, several motion-reducing sequences were developed, one of which was the BLADE technique. The BLADE sequence involves the filling of the k-space with a number of radially oriented blades that consist of parallel k-space lines rotating around the center of the k-space, leading to its oversampling. Many previous studies successfully demonstrated reduced motion artifacts by using this BLADE technique in T2W imaging (9,11–14). However, the direct application of the BLADE technique used in T1W imaging results in a reduced T1 contrast. The minimal effective TE of the BLADE technique is relatively long because the minimal, achievable turbo factor is relatively high in the BLADE technique. In our study, the minimum turbo factor of the BLADE sequence was 7, and the minimum effective TE with the BLADE technique was 23 ms, which was much longer than the TE of routine TSE sequences (usually <10 ms). As a result, the sharpness of the hepatic vessels and the corticomedullary differentiation was not significantly different between the BLADE and TSE sequences in spite of the better overall image quality and reduced motion artifacts evident in the BLADE sequence of our study (Figs. 1 and 2). To overcome this loss of T1 contrast with the conventional BLADE sequence, the inversion recovery prepulse sequence was be applied to the BLADE sequence in this study. Several previous studies have reported on the usefulness of IR-BLADE in motion reduction to improve lesion detection (4,11,13).

Our study intended to compare the use of IR-BLADE, BLADE, and conventional TSE with a high NSA for pediatric abdominal T1W imaging. Our qualitative analyses showed improved overall image quality and reduced motion artifact for IR-BLADE and BLADE sequences when compared with conventional TSE, which had been mainly attributable to the radial sampling property. The IR-BLADE sequence showed improved sharpness of hepatic vessels and better corticomedullary differentiation compared with BLADE and TSE sequences because of its superior T1 contrast and better subjective image quality, respectively (Fig. 2). With regard to the lesion conspicuity, IR-BLADE showed slightly higher lesion conspicuity score compared to TSE, although it was insignificant. Higher lesion conspicuity was noted in the case where the retroperitoneal neuroblastoma was abutting on vascular structures such as IVC or renal veins, because vascular structures showed signal void in IR-BLADE, whereas vascular structures showed inhomogeneous high signal due to flow-related enhancement in TSE (Fig. 3). In subgroup analysis by lesion size, TSE showed higher lesion conspicuity compared to BLADE for small lesions (<2 cm), though statistically not significant, and BLADE sequences showed higher lesion conspicuity than in TSE for large lesions (>2 cm). More specifically, for small lesions, TSE was better than BLADE sequences in terms of image coarseness, because of its lower bandwidth, higher NSA, and Cartesian k-space sampling. Furthermore, TSE showed higher T1 contrast due to shorter TE compared to BLADE. For these reasons, small lesions, where spatial and contrast resolutions were more important, may be looked more conspicuous on TSE, compared to BLADE (Fig. 4). On the contrary, IR-BLADE and BLADE showed better lesion conspicuity in case of large lesions, because of the reduced motion artifacts of the BLADE technique (Fig. 3). In subgroup analysis by sedation status, all sequences showed no significant difference in motion artifacts for sedation. However, because the sample size of subgroup was too small, further study with larger sample size is warranted for the effect of lesion size in BLADE sequences. Radial artifacts appeared on IR-BLADE and BLADE sequences, because of the intrinsic characteristics of radial sampling (6,8). However, radial artifacts did not appear to significantly degrade the images when compared with the respiratory motion artifacts that were observed on the conventional TSE images (Fig. 1).

Fig. 4.

A six-month-old girl with hemangiomatosis. Note that the lesion conspicuity of a small nodular lesion (<2 cm) in segment 4 of the liver (arrow) was better in (a) IR-BLADE image and (b) TSE with high NSA image than in (c) BLADE image without IR. IR, inversion recovery; NSA, number of signal acquisitions; TSE, turbo spin-echo.

Our quantitative analyses showed that the IR-BLADE and BLADE sequences demonstrate a significantly decreased signal variation in the liver and muscle when compared to the conventional TSE sequence. This could be explained by averaging effect and small gradient moments with BLADE technique, reducing sensitivity to motion-induced signal variation. This is because signal variation due to the motion can arise as a result of tissue displacement during each TR intervals, and also as a result of spin phase variation induced by motion through magnetic field gradients between an excitation RF pulse and data sampling period . Oversampling of central k-space in BLADE reduces artifact in a manner similar to multiple averaging in conventional imaging and radial sampling scheme in BLADE reduce the in-plane gradient moments in the central region of k-space, minimizing the spin phase variation induced by gradient moments (6). Conversely, compared to TSE, the IR-BLADE and BLADE sequences showed significantly increased signal variation through the medium of air, which may be possibly due to the effects of the radial streaking artifacts that occur outside of the body (Fig. 2). It has been well-known that the acquisition time of the BLADE technique is longer than that of conventional TSE sequences (6,8,14,15). In our study, however, the acquisition times of the BLADE technique were similar to that of TSE with a high NSA, possibly because of the use of rotary phased-array coils, application of parallel imaging and rearrangement of the number of BLADEs. This may encourage the application of the axial T1W IR-BLADE technique for pediatric abdominal MRI instead of the use of conventional TSE with high NSA.

There were some limitations in this study. First, the present study was a retrospective study and the study population was relatively small. Using the “pwr” R package based on the results, 60 was the calculated sample size to obtain the same statistical power for parametric t-tests instead of non-parametric tests, which requires further study with a larger cohort. Second, although the inter-observer agreement was fair to good, the comparative evaluation of the BLADE techniques and the TSE sequence was based on subjective scoring. The other confounding factor was the presence of radial artifacts and the different tissue contrast on the BLADE images. As a result of these factors, the radiologists could easily identify the use of the BLADE techniques, even if they had reviewed the images in a randomized order and blinded fashion. Moreover, further study is warranted for contrast-enhanced T1W IR-BLADE for pediatric abdominal MRI as an alternative for conventional TSE with high NSA.

In conclusion, the IR-BLADE sequence has less respiratory motion artifacts and provides better corticomedullary differentiation and vessel sharpness, resulting in improved overall image quality over a comparable acquisition time compared to conventional TSE imaging with a high NSA for pediatric T1W abdominal MRI.

Footnotes

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

ORCID iD

Young Hun Choi

References

Pearce

Salotti

Little

, et al. Radiation exposure from CT scans in childhood and subsequent risk of leukaemia and brain tumours: a retrospective cohort study. Lancet 2012; 380:499–505.

Brenner

Hall

EJ.

Computed tomography–an increasing source of radiation exposure. N Engl J Med 2007; 357:2277–2284.

Frush

Applegate

Computed tomography and radiation: understanding the issues. J Am Coll Radiol 2004; 1:113–119.

Darge

Anupindi

Jaramillo

MR imaging of the abdomen and pelvis in infants, children, and adolescents. Radiology 2011; 261:12–29.

Chavhan

Babyn

Vasanawala

SS.

Abdominal MR imaging in children: motion compensation, sequence optimization, and protocol organization. Radiographics 2013; 33:703–719.

Pipe

JG.

Motion correction with PROPELLER MRI: application to head motion and free-breathing cardiac imaging. Magn Reson Med 1999; 42:963–969.

Morelli

Runge

, et al. An image-based approach to understanding the physics of MR artifacts. Radiographics 2011; 31:849–866.

Hirokawa

Isoda

Maetani

, et al. MRI artifact reduction and quality improvement in the upper abdomen with PROPELLER and Prospective Acquisition Correction (PACE) technique. Am J Roentgenol 2008; 191:1154–1158.

Lee

Choi

Cheon

, et al. Improved abdominal MRI in non-breath-holding children using a radial k-space sampling technique. Pediatr Radiol 2015; 45:840–846.

10.

Cheng

Zhang

Ruangwattanapaisarn

, et al. Free-breathing pediatric MRI with nonrigid motion correction and acceleration. J Magn Reson Imaging 2015; 42:407–420.

11.

Finkenzeller

Zorger

Kuhnel

, et al. Novel application of T1-weighted BLADE sequences with fat suppression compared to TSE in contrast-enhanced T1-weighted imaging of the neck: cutting-edge images? J Magn Reson Imaging 2013; 37:660–668.

12.

Alibek

Adamietz

Cavallaro

, et al. Contrast-enhanced T1-weighted fluid-attenuated inversion-recovery BLADE magnetic resonance Imaging of the brain: An alternative to spin-echo technique for detection of brain lesions in the unsedated pediatric patient? Acad Radiol 2008; 15:986–995.

13.

Naganawa

Satake

Iwano

, et al. Contrast-enhanced MR imaging of the brain using T1-weighted FLAIR with BLADE compared with a conventional spin-echo sequence. Eur Radiol 2008; 18:337–342.

14.

Deng

Miller

Salem

, et al. Multishot diffusion-weighted PROPELLER magnetic resonance imaging of the abdomen. Invest Radiol 2006; 41:769–775.

15.

Dietrich

Ulbrich

Zanetti

, et al. PROPELLER technique to improve image quality of MRI of the shoulder. AJR Am J Roentgenol 2011; 197:W1093–100.