【摘要】 目的 探討0.35 T MRI各方位各序列掃描對肩袖撕裂的診斷價值。 方法 對2010年3月-2011年4月就診的38例肩關節疼痛患者，分別在斜冠狀位、斜矢狀位及橫軸位行T1加權像（T1 weighted image，T1WI）、T2加權像（T2 weighted image，T2WI）、質子密度加權像（proton density weighted image，PWI）及脂肪抑制像（turbo inversion recovery magnitude，TIRM）掃描。將38例患者的岡上肌及岡下肌肌腱合并為A組、肩胛下肌及小圓肌肌腱合并為B組進行研究，分別比較A、B組采用各方位的掃描表現；再將A組患者分為斜冠狀位T1WI+T2WI組與斜冠狀位PWI+TIRM組，將B組患者分為橫軸位T1WI+T2WI組與橫軸位PWI+TIRM組，分別比較各序列組的掃描表現。 結果 A組患者在3個方位掃描像以及斜冠狀位T1WI+T2WI組與PWI+TIRM組之間診斷準確度差異有統計學意義（Plt;0.05），其斜冠狀位診斷肩袖撕裂的靈敏度為88.9%，特異度為81.8%，準確度為86.8%，為最佳掃描方位，而PWI+TIRM組診斷肩袖撕裂的靈敏度為88.9%，特異度為90.9%，準確度為89.5%，為優選掃描序列；B組患者在3個方位掃描像以及橫軸位T1WI+T2WI組與PWI+TIRM組之間診斷準確度差異有統計學意義（Plt;0.05），其橫軸位診斷肩袖撕裂的靈敏度為87.5%，特異度為86.4%，準確度為86.8%，為最佳掃描方位，而PWI+TIRM組診斷肩袖撕裂的靈敏度為100%，特異度為83.3%，準確度為94.7%，為優選掃描序列。 結論 低場MRI診斷岡上肌、岡下肌肌腱撕裂以斜冠狀位PWI及TIRM掃描序列為首選，診斷小圓肌、肩胛下肌肌腱撕裂則以橫軸位PWI及TIRM掃描序列為主。【Abstract】 Objective To explore the clinical value of 0.35 T MRI diagnosing rotator cuff tears with different scan sequence and patient position. Methods From March 2010 to April 2011, there were 38 patients with shoulder pain were separately scanned by MRI at the position of oblique coronal, oblique sagittal and transaxial planes. Otherwise, the MRI images completed with T1 weighted, T2 weighted, PDWI and TIRM technique. The 38 cases were divided into two groups (group A: to study the supraspinatus and infraspinous tendons of the 38 cases; group B: to study the musculus teres minor and musculus subscapularis tendons of the 38 cases). Afterwards, the diagnostic results were compared among images at different patient positions. Furthermore, the images at oblique coronal plane of T1WI+T2WI and PWI+TIRM technique in group A were compared; on the other hands, the images at transaxial plane of T1WI+T2WI and PWI+TIRM technique in group B were compared. Results The difference of diagnostic accuracy in group A at different patient positions and scan sequences were statistical significant (Plt;0.05), and oblique cornal plane was the best patient position with sensitivity of 88.9%, specificity of 81.8% and accuracy of 86.8%; at the same time, the PWI+TIRM sequence was better sequence with sensitivity of 88.9%, specificity of 90.9% and accuracy of 89.5%. The difference of diagnostic accuracy in group B at different patient positions and scan sequences were statistical significant (Plt;0.05), and transaxial plane was the best patient position with sensitivity of 87.5%, specificity of 86.4% and accuracy of 86.8%; at the same time, the PWI+TIRM sequence was better sequence with sensitivity of 100%, specificity of 83.3% and accuracy of 94.7%. Conclusion In low field MRI, the oblique cornal plane with PWI+TIRM sequence are a first-line method for diagnosing supraspinatus tendon tears or infraspinous tendon tears; on the other hands, the transaxial plane with PWI+TIRM sequence are a first-line method for diagnosing musculus teres minor hurt or musculus subscapularis hurt.
Objective To determine the influence and significance of combinative assessment of 64 multi-slice spiral computed tomography (MSCT) with serum amyloid A protein (SAA) or C-reactive protein (CRP) on the selection of operative procedures of rectal cancer under the multi-disciplinary team. Methods Prospectively enrolled patients diagnosed definitely as rectal cancer at West China Hospital of Sichuan University from February to August 2009 were randomly assigned into two groups. In one group named MSCT+SAA group, both 64 MSCT and SAA combinative assessment were made for the preoperative evaluation. In another group named MSCT+CRP group, both MSCT and CRP combinative assessment were made for preoperative evaluation. Furthermore, the preoperative staging and predicted operation procedures were compared with postoperative pathologic staging and practical operation procedures, respectively, and the relationship between the choice of operation procedures and clinicopathologic factors was analyzed. Results All 165 patients were randomly assigned into MSCT+SAA group (n=83) and MSCT+CRP group (n=82). The baseline characteristics of two groups were statistically similar (Pgt;0.05). For MSCT+SAA group, the accuracies of preoperative staging T, N, M and TNM were 74.7%， 68.7%, 100% and 66.3%, respectively. For MSCT+CRP group, the accuracies of preoperative staging T, N, M and TNM were 72.0%, 86.6%, 100% and 81.7%, respectively. There were statistically significant differences in the accuracies of N staging and TNM staging between two groups (P＜0.05). However, there was no statistically significant difference of the accuracy of prediction to operative procedures between two groups (90.4% vs. 95.1%, Pgt;0.05). The pathological T staging (P＜0.001), N staging (P＜0.001), TNM staging (P＜0.001), preoperative serum level of SAA (P=0.010), serum level of CRP (P=0.042), and distance of tumor to the dentate line (P=0.011) were associated with the operative procedures. Conclusion Combinative assessment of MSCT+CRP could improve the accuracy of preoperative staging and operative procedures prediction, which may be superior to MSCT+SAA.
Objective To establish the optimal morphological criteria combined with fibrinogen level for evaluation of lymph node metastasis in colorectal cancer. Methods A consecutive series of 690 patients who underwent curative surgery for colorectal cancer, were examined by abdominopelvic enhanced multi-slice spiral computed tomography (MSCT) scan. If regional lymph nodes appeared, the maximal long-axis diameter (MLAD), maximal short-axis diameter (MSAD), and axial ratio (MSAD/MLAD) were recorded. At each lymph node size cut-off value, the following were calculated: accuracy, sensitivity, specificity, positive predictive value (PPV) and negative predictive value (NPV). Moreover, preoperative plasma level of fibrinogen was retrospectively examined to identify metastatic or inflammatory lymph node combined with MSCT image. Both modalities, MSCT plus fibrinogen and MSCT alone, were compared based on the pathologic findings. Results The study population consisted of 100 patients with regional lymph nodes show. No significant difference was found between metastatic and inflammatory lymph nodes in imaging characteristics (Pgt;0.05). The best cut-off value of MSAD was 6 mm for lymph node metastasis with the sensitivity of 46.8%, specificity of 68.4%, accuracy of 55.0%, PPV of 70.7% and NPV of 44.1%. The best cut-off value of MLAD was 8 mm with the sensitivity of 43.5%, specificity of 63.2%, accuracy of 51.0%, PPV of 65.9% and NPV of 40.7%. Using hyperfibrinogenemia (FIB ≥3.5 g/L) to identify small metastatic lymph node, of which MSAD lt;6 mm or MLAD lt;8 mm, showed statistical diagnostic value (Kappa=0.256, P=0.047). Compared with MSAD (6 mm) alone, MSAD (6 mm) combined with hyperfibrinogenemia had a higher sensitivity (79.0% vs. 46.8%, Plt;0.001), but a similar accuracy (66.0% vs. 55.0%, Pgt;0.05) and a lower specificity (44.7% vs. 68.4%, P=0.037). MLAD (8 mm) combined with hyperfibrinogenemia led to a greater diagnostic value in sensitivity (80.6% vs. 43.5%, Plt;0.001) and accuracy (66.0% vs. 51.0%, P=0.031) than MLAD (8 mm) alone, with a no-significantly decreasing specificity (42.1% vs. 63.2%, Pgt;0.05). Conclusions This present study recommend MSAD ≥6 mm or MLAD ≥8 mm as the optimal criteria for preoperative N staging in colorectal cancer. Moreover, the sensitivity and even accuracy could be improved by combining hyperfibrinogenemia for lymph node metastasis identification.