Table 2.
Repeatability and Reproducibility of Diffusion Tensor Imaging of Healthy and Unhealthy studies of Cervical Spinal Cord
| Authors (year) | Subject No. | Statistical analysis | C-spine levels examined | Region of interest | Repeatability | Reproducibility |
|---|---|---|---|---|---|---|
| Smith et al. 18 | 9 Healthy participants | BA | Between C2-C7 | Right, left, and posterior columns | Intrarater Lateral columns (Rt and Lt) using BA differences LOA [95% CI], (DBA%) MD: −0.58,0.60 [-0.22,0.24], (0.9%); AD: −0.97, 0.81 [-0.04, 0.03], (3.8%); RD: −0.48, 0.59 [-0.16, 0.27], (8.4%); FA: −0.18, 0.12 [-0.09, 0.03], (4.6%) Intrarater Dorsal columns using BA differences LOA [95% CI], (DBA%) MD: −0.30, 0.32 [0.11, 0.14], (1.3%); AD: −0.35, 0.36 [-0.13, 0.14], (0.1%); RD: −0.28, 0.33 [-0.09, 0.14], (3.6%); FA: −0.09, 0.11 [-0.03, 0.05], (1.6%). |
Inter-rater Lateral columns (Rt and Lt) using DBA differences LOA [95% CI], (DBA%) MD: −0.25, 0.23 [-0.01, 0.08], (1.0%); AD: −0.17, 0.17 [-0.06, 0.07], (0.2%); RD: −0.31, 0.27 [-0.00, 0.09], (2.9%); FA: −0.10, 0.14 [-0.03, 0.07], (3.0%) Interrater Dorsal columns using BA differences LOA [95% CI], (DBA%) MD: −0.20, 0.16 [-0.09, 0.05], (1.7%); AD: −0.20, 0.17 [-0.1, 0.05], (1.1%); RD: −0.20, 0.18 [-0.09, 0.06], (2.8%); FA: −0.07, 0.09 [-0.02, 0.04], (1.5%) |
| Mohamed, Feroze B., et al. 19 | 5 Children with SCI | ICC | Between C1-C7 | Whole WM and GM area |
Intrascans, ICC [95% CI]
MD: 0.77 [0.0–0.98]; AD: 0.80 [0.0–0.98]; RD: 0.77 [0.0–0.98]; FA: 0.97 [0.0–0.98]. |
- |
| Barakat, N., et al, 22 | 25 Healthy paediatric | ICC | Between C1-T1 at each disk and mid-level of cervical vertebral body | Whole WM and GM area | Intrascans C1-T1 ICC [95% CI] MD-ICC: 0.95 [0.92–0.98]; AD-ICC: 0.97 [0.92–0.99]; RD-ICC: 0.91 [0.83–0.98]; FA ICC: 0.87 [0.78–0.97] Intrascans Per cord level: ICC [95%, CI not reported] C1: MD = 0.68, AD = 0.73, RD = 0.70, FA = 0.59); Mid dens: MD = 0.70, AD = 0.70, RD = 0.69, FA = 0.55); Mid C2 body: MD = 0.71, AD = 0.73, RD = 0.73, FA = 0.55) C2-C3: MD = 0.76, AD = 0.76, RD = 0.76, FA = 0.56); Mid C3: MD = 0.76, AD = 0.77, RD = 0.76, FA = 0.61) C3-C4: MD = 0.76, AD = 0.76, RD = 0.74, FA = 0.58); Mid C4: MD = 0.73, AD = 0.75, RD = 0.71, FA = 0.59); C4-C5: MD = 0.72, AD = 0.71, RD = 0.69, FA = 0.63); Mid C5: MD = 0.68, AD = 0.65, RD = 0.64, FA = 0.57); C5-C6: MD = 0.61, AD = 0.59, RD = 0.61, FA = 0,58); Mid C6: MD = 0.39, AD = 0.51, RD = 0.39, FA = 0.62); C6-C7: MD = 0.51, AD = 0.48, RD = 0.46, FA = 0.68) Mid C7: MD = 0.54, AD = 0.56, RD = 0.45; FA = 0.70); C7-T1: MD = 0.57, AD = 0.61, RD = 0.48, FA = 0.43) |
- |
| Mulcahey et al. 21 | 10 Paediatric patients with SCI | ICC | Between C1-T1 at each disk and mid-level of cervical vertebral body | Whole cord at each level of the CSC | Intrascans Per cord level: ICC [95% CI]: C1: MD = 0.94[0.80–0.99], AD = 0.90[0.65–0.97], RD = 0.92[0.72–0.98], FA = 0.50[−.018 to 0.85]; Mid dens: MD = 0.94[0.77–0.98], AD = 0.86[0.51–0.96], RD = 0.91[0.69–0.98], FA = 0.62[0.07–0.89]; Mid C2 body: MD = 0.93 [0.74–0.98], AD = 0.87 [0.57–0.97], RD = 0.93 [0.74–0.98], FA = 0.60 [−0.05 to 0.88] C2-C3: MD = 0.89 [0.62–0.97], AD = 0.87 [0.51–0.97], RD = 0.93 (0.74–0.98), FA = 0.61 [−0.01 to 0.88]; Mid C3: MD = 0.87 [0.57–0.97], AD = 0.87 [0.52–0.97], RD = 0.88 [0.59–0.97], FA = 00.69 [0.14–0.91] C3-C4: MD = 0.87 [0.55–0.97], AD = 0.90 [0.64–0.97], RD = 0.82 [0.42–0.95], FA = 0.69 [0.12–0.91]; Mid C4: MD = 0.87 [0.54–0.97], AD = 0.92 [0.69–0.98], RD = 0.83 [0.45–0.96], FA = 0.75 [0.26–0.93] ; C4-C5: MD = 0.92 [0.71–0.98], AD = 0.89 [0.61–0.97], RD = 0.87 [0.54–0.97], FA = 0.71 [0.19–0.92]; Mid C5: MD = 0.90 [0.68–0.98], AD = 0.89 [0.61–0.97], RD = 0.88 [0.60–0.97], FA = 0.58 [−0.04 to 0.88] C5-C6: MD = 0.95 [0.82–0.99], AD = 0.94 [0.73–0.99], RD = 0.94[(0.79–0.99], FA = 0.82 [0.41–0.95]; Mid C6: MD = 0.93 [0.75–0.98], AD = 0.91 [0.64–0.97], RD = 0.93 [0.75–0.98], FA = 0.89 [0.61–0.97]; C6-C7: MD = 0.88 [0.46–0.97], AD = 0.88 [0.48–0.97], RD = 0.90 [0.56–0.97], FA = 0.82 [0.43–0.95] Mid C7: MD = 0.87 [0.33–0.97], AD = 0.82 [0.33–0.96], RD = 0.89 [0.40–0.98], FA = 0.81 [0.35–0.95]; C7-T1: MD = 0.80 [0.15–0.95], AD = 0.82 [0.27–0.95], RD = 0.84 (0.44–0.96], FA = 0.82 [0.36–0.96] |
- |
| Barakat et al. 20 | 10 Normal paediatric | ICC | Between C1-T1 at each disk and mid-level of cervical vertebral body | Free-hand and fixed-size ROIs methods on the entire cord | - | - |
| Taso, Manuel, et al. 23 | 3 Healthy volunteers | CV | At C2 and C5 | Whole WM, Lt and Rt WM pathways, and anterior GM |
Intrascan CV % [averaged for the whole cervical cord regions]
MD: 3.4%; AD: 3.5%; RD: 9.7%; FA: 5.1% |
- |
| By, Samantha, et al. 24 | 5 Healthy volunteers | BA | Between C2-C5 | Whole WM and GM regions |
WM region: Intrascan using BA differences
LOA [95% CI], (DBA%); WM at six directions MD: −0.18,0,24 [-0.02,0.08], (3.15); AD: −0.27,0.37 [-0.02,0.13], (2.48); RD: −0.21,0.26 [-0.03,0.08], (4.63); FA: −0.15,0.13 [-0.04,0.02], (1.09); WM at 15 directions: MD: −0.11,0.15 [-0.01,0.05], (2.28); AD: −0.19,0.26 [-0.02,0.09], (1.82); RD: −0.17,0.020 [-0.03,0.06], (3.1); FA: −0.12,0.12 [-0.03,0.03], (0.09); WM at 32 directions: MD: −0.31,0.20 [-0.11,0.01], (5.3) ; AD: −0.26,0.23 [-0.07,0.04], (0.75); RD: −0.38,0.24 [-0.14,0.00], (13.36); FA: −0.12,0.09 [0.0,0.07], (5.4) GM region: Intrascan, using BA differences LOA [95% CI], (DBA%): GM at six directions: MD: −0.16,0.17 [-0.03,0.04], (0.42) ;AD: −0.27,0.25 [-0.07,0.05], (0.48); RD: −0.15,0.18 [-0.03,0.05], (1.78); FA: −0.12,0.09 [-0.04,0.01], (2.84); GM at 15 directions: MD: −0.16,0.16 [-0.04,0.04], (0.07); AD: −0.30,0.35 [-0.05,0.11], (1.83) ; RD: −0.16,0.12 [-0.05,02], (2.54); FA: −0.10,0.12 [-0.01,0.04], (2.43); GM at 32 directions: MD: −0.26,0.22 [-0.08,0.03], (2.34); AD: −0.43,0.45 [-0.09,0.11], (0.8); RD: −0.24,0.17 [-0.09,0.01], (6.57); FA: −0.10,0.13 [-0.01,0.04], (2.53) |
- |
| Peterson, D. J., et al. 25 | 30 patients with acute cervical spine | ICC, CV | Extended from the foramen magnum to the C7-T1 | Whole WM, GM and other 24 individual tracts |
Intrareader of MD: (CV%, ICC) and FA: (CV%, ICC) GM: (8.1%,0.66), (6.1% ,0.83); WM: (5.6%,0.86), (3.8%,0.93) ; WM regions: Ventral: (12.0%,0.75), (10.5%,0.72); Lateral: (10.4%,0.71), (6.2% ,0.84) ; Dorsal: (11.3%,0.71), (7.7%, 0.80) ; Ventral WM tracts: Ventral reticulospinal tract: (18.3%,0.75), (18.2%, 0.47); Ventral corticospinal tract: (13.5%,0.69), (15.2%, 0.56) ; Lateral vestibulospinal tract: (17.0%,0.62), (14.0%,0.58) ; Spino-olivary tract: (13.9%,0.74), (12.7% ,0.68) ; Tectospinal tract: (13.2%,0.58), (10.5%, 0.69) ; Lateral WM tracts: Ventral spinocerebellar tract: (13.5%,0.72), (11.4%, 0.72);Ventrolateral reticulospinal tract: (14.6%,0.66), (12.1%, 0.69) ; Spinal lemniscus: (13.9%,0.67), (8.6%, 0.78) ; Rubrospinal tract: (18.0%,0.56), (10.9%, 0.66) ; Lateral corticospinal tract: (12.9%,0.69), (8.2%, 0.80) ;Dorsal WM tracts: Fasciculus cuneatus: (11.6%,0.70), (8.0%, 0.79) Fasciculus gracilis: (12.5%,0.66), (8.9%, 0.75) All individual WM tracts: (14.5%,0.75), (10.8%, 0.81) |
Interreader of MD: (CV%, ICC) and FA: (CV%, ICC) GM: (3.8%,0.93), (4.8% ,0.90) ; WM: (5.3%,0.88), (3.0%,0.95) WM: regions: Ventral: (7.6%,0.90), (6.8% ,0.88); Lateral: (9.2%, 0.78), (4.5%, 0.92) ; Dorsal: (9.4%,0.80), (5.6%,0.89) ; Ventral WM tracts: Ventral reticulospinal tract: (11.9%,0.82), (13.0%, 0.73) ; Ventral corticospinal tract: (10.6%, 0.81), (11.2%, 0.76); Lateral vestibulospinal tract: (10.0%,0.87), (9.4% ,0.81); Spino-olivary tract: (9.7%,0.87), (8.3% ,0.87) ; Tectospinal tract: (9.7%,0.85), (6.5%, 0.88) Lateral WM tracts: Ventral spinocerebellar tract: (11.3%,0.81), (8.8%, 0.83); Ventrolateral reticulospinal tract: (10.3%,0.83), (9.3% ,0.81) Spinal lemniscus: (9.8%,0.84), (5.4% ,0.91) Rubrospinal tract: (16.1%,0.65), (8.3%, 0.81) Lateral corticospinal tract: (11.6%,0.75), (6.4%,0.88); Dorsal WM tracts: Fasciculus cuneatus: (9.8%,0.79), (5.8%, 0.89); Fasciculus gracilis: (10.4%,0.77), (6.6%, 0.86) All individual WM tracts: (11.0%, 0.86), (7.8%,0.90) |
| Martin, A. R., et al. 26 | 17 Healthy volunteers and 9 patients with degenerative cervical myelopathy (DCM) | CV | Between C1-C7: Rostral(C1-C3), middle (C4-C5), caudal (C6-C7) levels | Whole cord, WM, GM over both right, left lateral corticospinal tract, fasciculus gracilis, fasciculus cuneatus and spinal lemniscus. |
Intrasubjects of FA across restrocaudal level: (CV %): Healthy
Rostral(C1-C3): (2.5%): Mid cervical or MCL (C4-C5): (3.0%) Caudal (C6-C7): (2.2%); DCM: Rostral(C1-C3): (2.8%); Mid cervical or MCL (C4-C5): (5.0%); Caudal (C6-C7): (4.6%) |
- |
| Guan, Li, et al. 27 | 10 Healthy volunteers | CV, BA | Between C1 and C7 | Whole cord | - | Interscan: CV [averaged for the whole regions] and BA not reported as numbers: FA:7.1%, AD:7.6%; RD:16.6%,; MD:9.7%. |
| Lee, Eugene, et al. 28 | 34 patients with cervical spondylotic myelopathy (CSM) | ICC | Between C1-T1 at each disk level | Whole cord | Intraobserver Per cord level: FA ICC [95% CI] for mean ROI method for observer1,2,3 and 4, respectively. C1/2: 0.460 [0.149–0.688], 0.558 [0.275–0.752], 0.986 [0.973–0.993] and 0.995 [0.990–0.997]; C2/3: 0.892 [0.794–0.944], 0.919 [0.845–0.959], 0.977 [0.955–0.987], and 0.992 [0.985–0.996]; C3/4: 0.948 [0.898–0.974], 0.843 [0.709–0.919], 0.991 [0.982–0.996], and 0.992 [0.983–0.996] ; C4/5: 0.910 [0.828–0.954], 0.959 [0.920–0.979], 0.945 [0.894–0.972], and 0.994 [0.989–0.997] ; C5/6: 0.635 [0.383–0.800], 0.807 [0.647–0.899], 0.964 [0.930–0.982], and 0.980 [0.961–0.990] C6/7: 0.582 [0.308–0.767], 0.813 [0.657–0.902], 0.946 [0.895–0.973], and 0.990 [0.980–0.995]; C7/T1: 0.834 [0.692–0.913], 0.735 [0.532–0.858], 0.980 [0.961–0.990], and 0.995 [0.989–0.997] Intraobserver Per cord level: FA ICC [95% CI] for manual ROI method C1/2: 0.541 [0.253–0.741], 0.524 [0.230–0.730], 0.969 [0.940–0.985], and 0.997 [0.9932–0.998]; C2/3: 0.798 [0.632–0.894], 0.888 [0.788–0.943], 0.944 [0.891–0.972], and 0.993 [0.986–0.996]; C3/4: 0.914 [0.834–0.956], 0.868 [0.753–0.932], 0.989 [0.978–0.995], and 0.984 [0.968–0.992] C4/5: 0.923 [0.851–0.961], 0.917 [0.840–0.957], 0.984 [0.968–0.992], and 0.990 [0.980–0.995]; C5/6: 0.888 [0.788–0.943], 0.783 [0.607–0.885], 0.981 [0.963–0.991], and 0.994 [0.988–0.997]; C6/7: 0.660 [0.418–0.814], 0.683 [0.452–0.828], 0.966 [0.932–0.983], and 0.992 [0.984–0.996] ; C7/T1: 0.814 [0.659–0.903], 0.711 (0.494–0.844], 0.972 [0.944–0.986], and 0.983 [0.966–0.992] Intraobserver Per cord level: FA ICC [95% CI] for Sagittal ROI method C1/2: 0.810 [0.653–0.901], 0.684 [0.453–0.828], 0.889 [0.790–0.943], and 0.645 [0.396–0.805]; C2/3: 0.646 [0.398–0.806], 0.816 [0.663–0.904], 0.926 [0.856–0.962], and 0.973 [0.948–0.987]; C3/4: 0.891 [0.794–0.944], 0.942 [0.886–0.970], 0.913 [0.834–0.956], and 0.965 [0.931–0.982]; C4/5: 0.797 [0.631–0.893], 0.808 [0.650–0.899], 0.865 [0.746–0.930], and 0.974 [0.949–0.987]; C5/6: 0.659 [0.418–0.814], 0.840 [0.703–0.917], 0.934 [0.872–0.966], and 0.887 [0.785–0.942] C6/7: 0.518 [0.223–0.727], 0.813 [0.658–0.902], 0.924 [0.854–0.961], and 0.977 [0.955–0.989]; C7/T1: 0.798 [0.632–0.894], 0.815 [0.661–0.903], 0.919 [0.844–0.959], and 0.924 [0.854–0.961] |
Interobserver Per cord level: FA ICC [95% CI] for mean ROI method first and second measurements of observer 1 and 2 &first and second measurements of observer 3 and 4 C1/2: 0.775 [0.595–0.881], 0.510 [0.213–0.721] & 0.690 [0.462–0.832],0.647 [0.400–0.807] C2/3: 0.925 [0.855–0.962], 0.906 [0.819–0.952] & 0.652 [0.406–0.810], 0.700 [0.477–0.838] C3/4: 0.954 [0.909–0.977], 0.868 [0.752–0.932] & 0.682 [0.451–0.827], 0.646 [0.398–0.806] C4/5: 0.948 [0.898–0.974], 0.932 [0.869–0.966] & 0.663 [0.423–0.816], 0.757 [0.566–0.871] C5/6: 0.783 [0.608–0.885], 0.580 [0.305–0.765] & 0.641 [0.391–0.803], 0.634 [0.380–0.799] C6/7: 0.596 [0.327–0.776], 0.854 [0.727–0.924] & 0.461 [0.150–0.689], 0.555 [0.271–0.750] C7/T1: 0.892 [0.795–0.945], 0.675 [0.440–0.823] & 0.678 [0.445–0.825], 0.694 [0.469–0.834] Interobserver Per cord level: FA ICC [95% CI] for manual ROI method first and second measurements of observer 1 and 2 &first and second measurements of observer 3 and 4 C1/2: 0.744 [0.546–0.864], 0.635 [0.382–0.799] & 0.551 [0.266–0.747], 0.579 [0.304–0.765] ; C2/3: 0.930 [0.864–0.964], 0.789 [0.618–0.889] & 0.747 [0.551–0.865], 0.709 [0.491–0.843] ; C3/4: 0.899 [0.807–0.948], 0.923 [0.851–0.961] & 0.792 [0.622–0.890], 0.800 [0.636–0.895] ; C4/5: 0.967 [0.936–0.984], 0.913 [0.834–0.956] & 0.756 [0.565–0.870], 0.773 [0.591–0.880] ; C5/6: 0.843 [0.708–0.918], 0.838 [0.700–0.916] & 0.594 [0.325–0.774], 0.607 [0.342–0.782] ; C6/7: 0.728 [0.520–0.854], 0.617 [0.357–0.789] & 0.478 [0.1716–0.700], 0.570 [0.291–0.759]; C7/T1: 0.861 [0.740–0.928], 0.799 [0.635–0.894] & 0.715 [0.501–0.847], 0.679 [0.447–0.826] Interobserver Per cord level: FA ICC [95% CI] for Sagittal ROI method first and second measurements of observer 1 and 2 &first and second measurements of observer 3 and 4 C1/2: 0.809 [0.651–0.900], 0.781 [0.604–0.884] & 0.479 [0.173–0.701], 0.430 [0.113–0.668] C2/3: 0.667 [0.428–0.818], 0.814 [0.660–0.903] & 0.275 [-0.065–0.558], 0.302 [-0.036–0.578] C3/4: 0.924 [0.853–0.961], 0.907 [0.823–0.953] & 0.532 [0.241–0.735], 0.624 [0.366–0.793] C4/5: 0.795 [0.628–0.892], 0.830 [0.689–0.912] & 0.422 [0.102–0.662], 0.410 [0.088–0.654] C5/6: 0.676 [0.442–0.824], 0.771 [0.589–0.879] & 0.222 [-0.121–0.517], 0.404 [0.081–0.650]; C6/7: 0.649 [0.402–0.808], 0.780 [0.604–0.884] & 0.595 [0.326–0.775], 0.628 [0.372–0.795]; C7/T1: 0.762 [0.575–0.874], 0.764 [0.578–0.875] & 0.084 [-0.257–0.406], 0.157 [-0.186–0.467] |
AD, axial diffusivity; CI, Confidence Interval; CV, coefficient of variation; DBA, Bland-Altman differences; FA, fractional anisotropy; ICC, intra-class correlation; Lt, left; MD, mean diffusivity; RD, radial diffusivity; Rt, right.