ABSTRACT
Objective To explore the value of 6 commonly-used quantitative measures on crosssectional MR images to diagnose the patellofemoral instability.Methods Clinical data of 32 patients with patellofemoral instability(34 knees) confirmed by arthroscopy surgery were retrospective analyzed.Knee MR imaging at full extension of the knee was performed in them.Their MR images were compared with those of 23 asymptomatic volunteers.Trochlear facet asymmetry,lateral trochlear inclination,trochlear depth,patellar tilt angle,lateral patellar displacement,and tibial tuberosity-trochlear groove distance(TTTG) were measured in transverse fat-suppressed turbo spin-echo proton density-weighted MR images.Independent sample t test or Mann-Whitney U test were used to compare the differences of these measurements.ROC was used to calculate the area under curve(AUC) and to define the diagnostic threshold value of each measure.Results Trochlear facet asymmetry,lateral trochlear inclination,and trochlear depth in patients were 45.0％±9.2％,10.9°±5.0° and(3.7±1.3) mm,while these measurements were 68.0％±10.5％,21.1°±3.4° and (6.3 ± 1.0) mm in volunteers.These measures were significantly lower in patients than those in volunteers (t values were 10.123,10.862 and 9.835,P＜0.01).Patellar tilt angle and lateral patellar displacement were significantly higher in patients than those in volunteers(24.8°±9.0° vs.12.3°±5.2°,7.36 mm vs.-3.93 mm,t value and Z value were-7.657 and-6.953,P＜0.01).TTTG showed no significant difference between the 2 groups(P=0.798).AUC of trochlear facet asymmetry,lateral trochlear inclination,trochlear depth,patellar tilt angle,and lateral patellar displacement were 0.957,0.957,0.947,0.921 and 0.961,respectively.The recommended diagnostic threshold values for these 5 measures were＜55％,＜15°,＜5 mm,＞17° and＜3 mm,respectively.The sensitivity for these five measurements were 91.2％ (31/34),85.3％ (29/34),91.2％ (31/34),82.4％ (28/34) and 79.4％ (27/34),respectively.The specificity for these five measures were 91.3％ (42/46),97.8％ (45/46),89.1％ (41/46),89.1％ (41/46) and 100.0％ (46/46),respectively.AUC of TTTG was 0.520.Conclusions At full extension of the knee,trochlear facet asymmetry,lateral trochlear inclination,trochlear depth,patellar tilt angle and lateral patellar displacement are all effective in evaluating patellofemoral instability.The diagnostic value of TTTG is relatively poor.
ABSTRACT
Objective To evaluate the contrast-enhanced 3D sampling perfection with application-optimized contrasts by using different flip angle evolutions-short TI inversion recovery sequence (SPACE-STIR) for the imaging of the post-ganglionic segments of the brachial plexus.Methods Forty-three patients with suspected brachial plexus lesions were examined with 3D SPACE-STIR and contrast-enhanced 3D SPACE-STIR prospectively.Signal-to-noise ratios (SNR),contrast-to-noise ratios (CNR),and the conspicuousness of roots,trunks,divisions and cords of the brachial plexus of the two 3D sequences were retrospectively compared.Statistical analysis was performed by using student t-test and Wilcoxon rank sum test.Results Compared with 3D SPACE-STIR,contrast-enhanced 3D SPACE-STIR provided the similar SNRs (left,37.41 ±7.34 vs 36.27 ±7.66,t =1.574,P =0.123,right,43.85 ±9.56 vs 42.34 ±9.74,t =1.937,P =0.073),but significantly higher nerve-to-muscle CNRs (left,24.01 ± 6.31 vs 26.39 ± 6.95,right,29.31 ± 7.84 vs 31.77 ± 8.85,t =-3.278,-3.278,both P ＜ 0.01) and nerve-to-lymph gland CNRs(left,-0.84 ± 10.51 vs 15.35 ±8.02,right,-8.47 ± 10.85 vs 19.30 ± 10.35,t =-15.984,-15.651,both P ＜0.01).The conspicuousness of roots and trunks on contrast-enhanced 3D SPACE-STIR was significantly better than that on 3D SPACE-STIR (Z =-3.606,-4.472,P ＜ 0.01),while the conspicuousness of divisions and cords was similar(Z =-1.732,-1.414,P =0.083,0.157).The signal intensity of neoplastic lesions on contrast-enhanced 3D SPACE-STIR tended to decrease rapidly,thus the lesion conspicuousness was worse than that on 3D SPACE-STIR.Conclusions Contrast-enhanced 3D SPACE-STIR has obvious advantages in displaying normal brachial plexus and revealing non-neoplastic lesions of the brachial plexus,but may be insufficient for the diagnosis of neoplastic lesions of the brachial plexus.