Abstract
目的
初步建立腮腺CT影像报告与数据系统(Parotid Imaging Reporting and Data System,PI-RADS),并探讨其临床应用价值。
方法
纳入2013年1月至2016年12月间因腮腺肿物就诊于北京大学口腔医院并进行手术治疗的病例,回顾性评估所有病例的影像资料,获取相关影像特征,评估肿瘤恶性风险概率,并分为6个等级(1级,正常腮腺; 2级,基本确定为良性病变或肿瘤; 3级,无明确恶性病变证据但不能确定为良性病变; 4级,怀疑为恶性肿瘤病变但证据不充分; 5级,恶性肿瘤影像征象较充分; 6级,有恶性肿瘤病理学证据)。
结果
共纳入腮腺肿物病例897例次,其中良性病变905例次、恶性肿瘤98例次,影像诊断为2级、3级、4级和5级的病变中,恶性肿瘤的构成比分别为0.4%、5.7%、35.5%和96.7%,随PI-RADS分级呈逐渐增高趋势(Z=-15.579,P<0.001)。相邻等级[2级与3级(χ 2=12.048,P=0.001)、3级与4级(χ 2=75.231,P<0.001)、4级与5级(χ 2=32.266,P<0.001)]之间的恶性构成比差异有统计学意义。Cohen’s Kappa检验表明两位研究者分级诊断具有中度一致性(κ=0.614,P<0.001,95%CI: 0.569~0.695)。
结论
应用影像诊断分级方法对腮腺肿瘤性疾病的诊断和临床治疗有一定的帮助。
Keywords: 腮腺肿瘤, 体层摄影术, X线计算机, 诊断, 鉴别, 治疗
Abstract
Objective
To establish a Parotid Imaging Reporting and Data System (PI-RADS) for CT diagnosis of the parotid gland neoplasms and to investigate the clinical applicable value and feasibility of PI-RADS.
Methods
Patients who had been diagnosed with primary parotid gland neoplasms and had received surgical treatments in Peking University School and Hospital of Stomatology during the period of January 2013 to December 2016 were included in this study. The diagnoses were confirmed by the postoperative pathological examinations in all the patients. The CT imaging data of all patients were retrospectively reviewed and analyzed by two readers in consensus. Imaging characteristics related to the parotid neoplasms were extracted and quantified. Based on comprehensive analysis of the imaging characteristics, the probabilities of the benign and malignant neoplasms were evaluated and classified into six grades, PI-RADS 1-6 (PI-RADS 1: normal parotid gland; PI-RADS 2: confidently benign lesions; PI-RADS 3: probably benign lesions without confirmed evidence of malignancy; PI-RADS 4: suspected malignancy without sufficient evidence of malignancy; PI-RADS 5: confidently malignant lesions; PI-RADS 6: lesions with confirmed pathological evidence of malignancy).
Results
A total of 897 patients with 1 003 parotid lesions were included. The lesions included 905 benign and 98 malignant lesions. The proportions of the malignancies in PI-RADS 2, PI-RADS 3, PI-RADS 4 and PI-RADS 5 according to the two readers in consensus were 0.4%, 5.7%, 35.5% and 96.7% respectively. The overall Cohen’s Kappa test showed medium consistency between the two independent researchers (κ=0.614, P<0.001, 95%CI: 0.569-0.695). Pearson Chi-square test showed that the proportions of malignancies increased with the diagnostic PI-RADS grades (Cochran-Armitage trend test, Z=-15.579, P<0.001). The results of Pearson Chi-square tests showed significant differences between the grades [PI-RADS 2 and 3 (χ 2=12.048, P=0.001); PI-RADS 3 and 4 (χ 2=75.231, P<0.001); PI-RADS 4 and 5 (χ 2=32.266, P<0.001)].
Conclusion
PI-RADS can be used to evaluate the risk of malignancy and will be helpful to improve the imaging diagnosis and clinical treatment of paro-tid gland neoplasms.
Keywords: Parotid neoplasms, Tomography, X-ray computed, Diagnosis, differential, Therapy
唾液腺肿瘤是常见的口腔疾病[1],约占全身肿瘤的3%[2],其中80%发生于腮腺[3]。腮腺肿瘤病理学种类繁多[4],多形性腺瘤和Warthin瘤等良性肿瘤占80%左右。恶性腮腺肿瘤的临床表现有较大差异[5],部分恶性肿瘤的临床和影像表现缺少特异性,在术前难以准确判断其性质[6]。
超声、CT和磁共振成像等为常用的腮腺肿瘤影像学检查方法[7]。CT通过分析腮腺肿瘤的大小、部位、边界和强化特征等能够初步判断其性质[7],对于Warthin瘤等部分肿瘤可以较准确地诊断,但对于多形性腺瘤和部分恶性肿瘤的鉴别则存在一定的困难。因此,传统影像诊断方法较难实现规范化诊断报告。
甲状腺肿瘤等疾病的临床实践证明,应用影像报告与数据系统对肿瘤的恶性概率作出等级评估有利于肿瘤的诊疗[8,9],能够提高肿瘤影像学诊断对临床的指导意义。腮腺肿瘤的临床实践中尚未见应用影像报告与数据系统,因此,本研究拟初步建立一个基于CT的腮腺影像报告与数据系统(Parotid Imaging Reporting and Data System,PI-RADS),探讨和研究腮腺肿瘤恶性概率等级评估方法,以进一步提高腮腺肿瘤临床诊疗的规范化水平和科研水平。
1. 资料与方法
1.1. 研究对象
纳入2013年1月至2016年12月间因原发性腮腺肿物就诊于北京大学口腔医院的病例,所有患者术前均行CT检查和手术治疗,且术后病理诊断明确。排除腮腺复发性肿物、发育异常、涎石症及化脓性炎症疾病。
1.2. CT检查
所有患者于仰卧位进行检查(GE,Brightspeed, Optima CT520, USA)。共120例患者(131例肿物)进行CT平扫检查,411例患者(475例肿物)进行单期相增强CT检查,366例患者(397例肿物)进行多期相增强CT检查。单期相或多期相增强CT检查于平扫完成后,经肘静脉注射对比剂(碘帕醇,3.7 mg I/mL,1.5~2.0 mL/kg,2.0~3.0 mL/s),于注射完成后即刻、30 s和5 min进行增强第一至三相扫描。平扫及增强第一相扫描范围自颅底层至胸锁关节层,增强二和三相扫描于病变范围扫描。扫描参数:管电压120~140 kV,管电流200~380 mA,层厚1.25 mm,螺距1.65:1。图像重建参数为:标准重建方式,层厚1.25 mm,层间距1.25 mm,重建视野20 cm×20 cm。所有影像资料以DICOM格式存储于影像PACS系统。
1.3. 腮腺肿物影像诊断分级系统
影像分析由两名经验丰富的口腔颌面医学影像专科医师共同商讨决定。于影像诊断系统中评价肿瘤部位、大小、形态、边界、密度、强化特点(时相、程度、分布)、肿物单发或多发情况、钙化情况、囊性变情况、周围及颈部淋巴结评估、邻近组织是否被侵犯评估等(表1,图1)。由两名口腔颌面医学影像专科医师独立进行分级诊断,并进行一致性评价。
1.
腮腺肿物PI-RADS诊断结果与影像学征象
Correlations between PI-RADS grades of the parotid gland neoplasms and radiologic signs
| Items | PI-RADS 2 | PI-RADS 3 | PI-RADS 3b | PI-RADS 4 | PI-RADS 5 | PI-RADS 6 | Benign | Malignancy |
| PI-RADS, Parotid Imaging Reporting and Data System. The numbers of neoplasms were calculated here. | ||||||||
| Size | ||||||||
| Small (<2.0 cm) | 30 | 82 | 7 | 7 | 2 | 0 | 118 | 10 |
| Medium (2.0-3.0 cm) | 158 | 453 | 4 | 49 | 14 | 1 | 624 | 55 |
| Large (>3.0 cm) | 56 | 102 | 1 | 20 | 14 | 3 | 163 | 33 |
| Single/multiple | ||||||||
| Single on one side | 139 | 584 | 0 | 62 | 19 | 3 | 730 | 77 |
| Multiple on one side | 33 | 25 | 5 | 11 | 9 | 1 | 66 | 18 |
| Multiple on two sides | 72 | 28 | 7 | 3 | 2 | 0 | 109 | 3 |
| Boundary | ||||||||
| Absolutely clear | 153 | 153 | 0 | 3 | 0 | 0 | 309 | 0 |
| Majorly clear | 68 | 261 | 2 | 21 | 4 | 0 | 345 | 11 |
| Partially ill-defined | 17 | 182 | 1 | 26 | 3 | 1 | 210 | 20 |
| Majorly ill-defined | 6 | 38 | 7 | 18 | 8 | 3 | 37 | 43 |
| Ill-defined | 0 | 3 | 2 | 8 | 15 | 0 | 4 | 24 |
| Shape | ||||||||
| Round | 54 | 269 | 5 | 21 | 7 | 0 | 334 | 22 |
| Oval | 154 | 162 | 3 | 13 | 3 | 0 | 322 | 13 |
| Scalloped | 31 | 182 | 2 | 27 | 6 | 2 | 222 | 28 |
| Irregular | 5 | 24 | 2 | 15 | 14 | 2 | 27 | 35 |
| Dominant density on plain CT | ||||||||
| Fat | 7 | 1 | 0 | 0 | 0 | 0 | 8 | 0 |
| Liquid | 30 | 59 | 0 | 4 | 12 | 0 | 91 | 2 |
| Soft-tissue (partial fluid) | 66 | 239 | 1 | 36 | 18 | 3 | 311 | 46 |
| Soft-tissue | 141 | 338 | 11 | 36 | 0 | 1 | 495 | 50 |
| Enhancement degree on first enhanced phase | ||||||||
| No significant | 25 | 42 | 0 | 3 | 0 | 0 | 68 | 2 |
| Slightly | 10 | 122 | 0 | 12 | 4 | 0 | 140 | 8 |
| Medium | 29 | 203 | 5 | 32 | 13 | 1 | 236 | 47 |
| Obvious | 89 | 114 | 7 | 21 | 11 | 3 | 217 | 28 |
| Significant | 67 | 53 | 0 | 5 | 1 | 0 | 119 | 7 |
| Enhancement pattern | ||||||||
| Gradual | 20 | 294 | 5 | 30 | 11 | 1 | 333 | 28 |
| Fast in-out | 124 | 128 | 1 | 23 | 7 | 2 | 254 | 31 |
| Basically no enhancement | 4 | 4 | 0 | 2 | 1 | 0 | 8 | 3 |
| Gradual in-out | 9 | 51 | 2 | 10 | 2 | 0 | 61 | 13 |
| Enhancement distribution | ||||||||
| Membrane enhancement | 5 | 34 | 0 | 7 | 0 | 0 | 40 | 6 |
| Even | 152 | 182 | 12 | 24 | 15 | 1 | 346 | 36 |
| Uneven to even | 17 | 198 | 0 | 21 | 5 | 1 | 223 | 20 |
| Uneven | 14 | 84 | 0 | 17 | 8 | 2 | 99 | 26 |
| No enhancement | 32 | 36 | 0 | 4 | 0 | 0 | 72 | 0 |
| Calcification | ||||||||
| No | 239 | 602 | 12 | 59 | 26 | 4 | 861 | 81 |
| Small granular or capsule | 5 | 31 | 0 | 10 | 1 | 0 | 38 | 9 |
| Large irregular | 0 | 4 | 0 | 7 | 3 | 0 | 6 | 8 |
| Cystic percentage | ||||||||
| 0 | 177 | 488 | 12 | 49 | 26 | 2 | 683 | 71 |
| ≤25% | 19 | 53 | 0 | 10 | 2 | 2 | 72 | 14 |
| >25%, ≤50% | 7 | 17 | 0 | 6 | 2 | 0 | 23 | 9 |
| >50%, ≤75% | 5 | 33 | 0 | 5 | 0 | 0 | 40 | 3 |
| >75% | 36 | 46 | 0 | 6 | 0 | 0 | 87 | 1 |
| Adjacent structure destruction | ||||||||
| No | 237 | 609 | 7 | 64 | 16 | 3 | 867 | 69 |
| Move | 5 | 19 | 0 | 5 | 0 | 0 | 26 | 3 |
| Poor boundary | 2 | 8 | 5 | 6 | 9 | 0 | 12 | 18 |
| Significant | 0 | 1 | 0 | 1 | 5 | 1 | 0 | 8 |
1.
PI-RADS 2级(A)、3级(B)、4级(C)、5级(D)诊断的典型病例
A characteristic example of a PI-RADS 2 (A),3 (B), 4 (C), 5 (D) tumor
A, a parotid gland neoplasm showing all regular and well-defined oval boundary, characteristic enhancement feature of benign nature (enhancement phase 1), indicating a PI-RADS 2 tumor. It was proved as Warthin tumor. B, a parotid gland neoplasm showing basically regular and well-defined round boundary, with mild enhancement (enhancement phase 3), indicating a PI-RADS 3 tumor. It was proved as pleomorphic adenoma. C, a parotid gland neoplasm showing seemingly well-defined boundary, with mild enhancement (enhancement phase 1), indicating a PI-RADS 4 tumor. It was proved as mucoepidermoid carcinoma. D, a paro-tid gland neoplasm showing all poorly-defined boundary, with erosion into adjacent structures and mild enhancement (enhancement phase 2), indicating a PI-RADS 5 tumor. It was proved as adenoid cystic carcinoma.
所有腮腺肿物的影像评价分为PI-RADS 1~6级:PI-RADS 1级为正常腮腺;PI-RADS 2级确定为良性病变;PI-RADS 3级良性病变或肿瘤可能性较大,无明确恶性肿瘤证据,但不能排除恶性肿瘤可能性;PI-RADS 4级无充分恶性证据但恶性肿瘤可能性增加;PI-RADS 5级为有确定的恶性肿瘤征象;PI-RADS 6级为有病理学恶性肿瘤证据。诊断分级具体方法详见表2。
2.
腮腺CT影像报告与数据系统的诊断标准
The diagnostic criteria of PI-RADS
| PI-RADS | Criteria |
| PI-RADS, Parotid Imaging Reporting and Data System. | |
| 1: Normal parotid gland | There are definitely no masses in the parotid gland |
| 2: More likely benign | (1) Multiple bilateral or unilateral masses presented regular arc border and oval contour with enhancement and without enlarged or necrotic lymph nodes around the parotid gland and neck; (2) A single tumor may also be considered if it is typical (e.g. oval, with obvious and uniform enhancement in the first enhancement phase, with smooth clear border or a low-density fat envelope) |
| 3: Indeterminate | (1) Single, round, no enhancement or uniform enhancement, most of the boun-dary clear, for example majorly clear or partially ill-defined; (2) Granular with uneven enhancement but definite boundary; (3) Multiple without enhancement, and without enlarged or necrotic lymph nodes around the parotid gland and neck; (4) Small tumors, that means, all tumors diameter less than 1 cm are included, regardless of their enhancement features |
| 3b: Inflammation including Sjögren syndrome and IgG4 |
In multiple gland, lesions are diffuse, with enhancement, without necrotic lymph nodes around the neck |
| 4: Probably malignant | (1) Single lesion, the shape is slightly irregular, part of the boundary is indistinctly (the boun-dary marked 3 or 4) or the membrane enhancement with villous edges; (2) The enhancement type is medium enhancement even slightly enhancement, solid lesion with little or without cystic degeneration; (3) Soft tissue density or with partial fluid especially with calcification, a small amount of unilateral diffuse; (4) Have large masses of irregular calcification with swelling lymph nodes but without necrosis lymph nodes |
| 5: Highly suggestive malignancy | (1) Necrotic lymph nodes are seen in the neck; (2) The boundary is unclear, the shape is irregular; (3) The border of the surrounding muscles, bones or fat tissues are not clear, which means destruction |
| 6: Already had malignant diagnosis | Had pathological result to pronounced it is a malignant neoplasm |
| 7: Unsatisfied illustration on CT | The density of neoplasm and the parotid is too nearly to find out what the neoplasm is really like. But couldn’t exclude that a neoplasm may exist |
1.4. 统计分析
应用SPSS 20.0和SAS 3.7进行统计分析。计数资料应用卡方检验、Cochran-Armitage趋势检验等方法进行统计,计量资料采用单因素方差分析(one way ANOVA), P<0.05为差异有统计学意义。计算kappa系数评价诊断一致性,kappa系数0~0.20为较低一致性,0.21~0.40为一般一致性,0.41~0.60为中等一致性,0.61~0.80为高度一致性,0.81~1.00为几乎完全一致。
2. 结果
2.1. 病例资料
共纳入腮腺肿物患者897例,包括单发肿物832例,多发肿物64例;共1 003个肿物,其中良性肿物905个,恶性肿物98个。男性465例,女性432例,年龄4~85岁,平均年龄46.0岁。
2.2. PI-RADS诊断与病理结果
两位研究者共同诊断结果中PI-RADS 2级、3级、3b级、4级、5级和6级的构成比分别为24.3%、63.5%、1.2%、7.6%、3.0%和0.4%。各分级诊断中的病理结果详见表3。
3.
腮腺肿物病理诊断与PI-RADS诊断结果的相关性
Correlations between pathological results of the parotid gland neoplasms and PI-RADS grades
| Items | PI-RADS 2 | PI-RADS 3 | PI-RADS 3b | PI-RADS 4 | PI-RADS 5 | PI-RADS 6 | PI-RADS total |
| PI-RADS, Parotid Imaging Reporting and Data System; NOS, nonspecific. The numbers of neoplasms were calculated here. | |||||||
| Warthin tumor | 176 | 72 | 0 | 5 | 0 | 0 | 253 |
| Pleomorphic adenoma | 4 | 367 | 0 | 23 | 1 | 0 | 395 |
| Basal cell adenoma | 11 | 81 | 0 | 2 | 0 | 0 | 94 |
| Myoepithelioma | 0 | 3 | 0 | 0 | 0 | 0 | 3 |
| Oncocytoma | 4 | 4 | 0 | 0 | 0 | 0 | 8 |
| Cystadenoma | 0 | 6 | 0 | 0 | 0 | 0 | 6 |
| Keratocystoma | 0 | 0 | 0 | 1 | 0 | 0 | 1 |
| Cyst | 28 | 14 | 0 | 0 | 0 | 0 | 42 |
| Lipoma | 4 | 1 | 0 | 0 | 0 | 0 | 5 |
| Schwannoma | 0 | 9 | 0 | 0 | 0 | 0 | 9 |
| Vascular malformation | 2 | 9 | 0 | 2 | 0 | 0 | 13 |
| Eosinophilic lymphogranuloma | 0 | 1 | 2 | 0 | 0 | 0 | 3 |
| Inflammation | 13 | 31 | 9 | 12 | 0 | 0 | 65 |
| Calcified epithelioma | 1 | 1 | 0 | 0 | 0 | 0 | 2 |
| Non-sebaceous lymphoadenoma | 0 | 0 | 0 | 1 | 0 | 0 | 1 |
| Lymphoepithelial lesions | 0 | 0 | 0 | 3 | 0 | 0 | 3 |
| Nodular fasciitis | 0 | 1 | 0 | 0 | 0 | 0 | 1 |
| Myofibromatosis | 0 | 1 | 0 | 0 | 0 | 0 | 1 |
| Mucoepidermoid carcinoma | 1 | 6 | 0 | 6 | 6 | 1 | 20 |
| Adenocarcinoma, NOS | 0 | 2 | 0 | 3 | 4 | 0 | 9 |
| Acinic cell carcinoma | 0 | 13 | 0 | 2 | 0 | 0 | 15 |
| Salivary duct carcinoma | 0 | 0 | 0 | 1 | 3 | 0 | 4 |
| Adenoid cystic carcinoma | 0 | 4 | 0 | 0 | 3 | 0 | 7 |
| Polymorphous adenocarcinoma | 0 | 3 | 0 | 5 | 1 | 0 | 9 |
| Uncertainty adenocarcinoma | 0 | 0 | 1 | 2 | 5 | 0 | 8 |
| Epithelial-myoepithelial carcinoma | 0 | 1 | 0 | 0 | 0 | 0 | 1 |
| Clear cell carcinoma | 0 | 0 | 0 | 0 | 1 | 0 | 1 |
| Oncocytic carcinoma | 0 | 0 | 0 | 2 | 0 | 0 | 2 |
| Sarcoma | 0 | 1 | 0 | 0 | 0 | 1 | 2 |
| Lymphoepithelial carcinoma | 0 | 1 | 0 | 2 | 2 | 2 | 7 |
| Malignant lymphoma | 0 | 4 | 0 | 3 | 4 | 0 | 11 |
| Malignant melanoma | 0 | 0 | 0 | 1 | 0 | 0 | 1 |
| Metastatic solitary fibroma | 0 | 1 | 0 | 0 | 0 | 0 | 1 |
| Total | 244 | 637 | 12 | 76 | 30 | 4 | 1 003 |
PI-RADS 2级、3级、4级和5级的病变中,恶性肿瘤构成比分别为0.4%、5.7%、35.5%和 96.7%。Cochran-Armitage趋势检验表明,随PI-RADS诊断分级增高,恶性肿瘤构成比有增高趋势(Z= -15.579,P<0.001)。PI-RADS相邻等级之间恶性肿瘤构成比差异均有统计学意义[2级与3级(χ2=12.048,P=0.001),3级与4级(χ2=75.231,P<0.001),4级与5级(χ2=32.266,P<0.001)]。
多形性腺瘤(395例)诊断为PI-RADS 2级、3级、3b级、4级、5级和6级的构成比分别为1.0%、92.9%、0、5.8%、0.3%、0。Warthin瘤(253例)诊断为PI-RADS 2级、3级、3b级、4级、5级和6级的构成比分别为69.6%、28.5%、0、2.0%、0、0。黏液表皮样癌(20例)诊断为PI-RADS 2级、3级、3b级、4级、5级和6级的构成比分别为5%、30%、0、30%、30%、5%。腺样囊性癌(7例)诊断为PI-RADS 2级、3级、3b级、4级、5级和6级的构成比分别是0、 57.1%、 0、 0、 42.9%、 0。腺泡细胞癌(15例)诊断为PI-RADS 2级、3级、3b级、4级、5级和6级的构成比分别是0、 86.7%、 0、13.3%、 0、 0。
2.3. PI-RADS诊断一致性评价
两位读片者的分级诊断判读结果具有中等一致性,κ=0.614(95%可信区间为0.569~0.695), P<0.001。
3. 讨论
3.1. 影像报告与数据系统的现状
影像报告与数据系统已经在多个器官的影像诊断中广泛应用。1986年美国放射学会(American College of Radiology)最早提出乳腺影像报告与数据系统(Breast Imaging Reporting and Data System, BI-RADS),根据乳腺结节的B超表现,将其分为5个等级,在世界范围内广泛应用,目前已更新至第5版。2009年Horvath等[8]提出甲状腺影像报告与数据系统(Thyroid Imaging Reporting and Data System, TI-RADS),随后相继出现了妇科影像报告与数据系统、肝脏影像报告与数据系统和前列腺影像报告与数据系统[9]。2015年,Abdel Razek等[10]提出的腮腺B超影像报告与数据系统,至今在我国仍未普及应用。
影像学诊断对良恶性肿瘤的判断存在不准确性,肿瘤性疾病的最终诊断大多需依靠组织病理学诊断。基于恶性肿瘤并非一定在影像中表现出恶性征象的临床事实,影像报告与数据系统采用分级诊断的思想对恶性风险概率进行预测,以更好地发挥影像诊断指导临床决策的功能。
3.2. PI-RADS的临床应用价值讨论
传统影像诊断方式对肿瘤性质的判断可以分为确定性诊断和非确定性诊断(可能性诊断和排除性诊断),非确定性诊断所占比例较多,不能有效体现影像诊断对于肿瘤恶性风险的评估。PI-RADS分级诊断方法与传统影像诊断方式相比,有以下优点。(1)有利于临床诊断报告的规范化:传统影像诊断方法用文字描述良性或者恶性肿瘤的确定性和可能性,而PI-RADS则用4个递增等级评价肿瘤的恶性概率,更易于实现规范化的诊断报告,体现影像诊断对肿瘤恶性风险的评估结果。(2)对传统诊断中可以确定良恶性的病例仍具有较好的准确性:BI-RADS与TI-RADS中,2级均为确定良性或恶性概率极小的肿瘤,本研究的分级系统中,PI-RADS 2级的恶性肿瘤构成比为0.4%,阴性预测值为99.6%;BI-RADS与TI-RADS中,5级为恶性概率较大的肿瘤,TI-RADS 5级的恶性肿瘤概率大于80%,BI-RADS 5级的恶性肿瘤概率大于90%,本研究中PI-RADS 5级的恶性肿瘤概率为96.7%。(3)有利于交界性肿瘤的临床诊断:BI-RADS与TI-RADS中,3级和4级用于评价逐渐增加的恶性概率,腺泡细胞癌等恶性肿瘤经常因缺少较特异的影像表现而类似于良性肿瘤,因此,建议将缺少影像特异性表现且无明显恶性征象的肿瘤(如多形性腺瘤等)诊断为PI-RADS 3级,而将肿瘤表现出一定的恶性征象但仍不充分者诊断为PI-RADS 4级。(4)有利于腮腺炎症性疾病的诊断:根据腮腺疾病的临床特色,本研究建议将一些需与腮腺肿瘤作鉴别诊断的炎症性或免疫性疾病(如干燥综合征结节型、IgG4相关唾液腺炎症性肿块等)诊断为PI-RADS 3b级。(5)有利于增进医患沟通:通过规范诊疗行为,减少了医患纠纷。
本研究虽初步提出了PI-RADS的基本概念,但仍需进行全面深入研究,为其临床应用提供理论基础。首先,需要全面建立基于超声和磁共振成像等常用影像诊断方法的影像报告与数据系统,进一步规范腮腺肿瘤临床影像诊断流程和影像报告方式;目前,基于磁共振的腮腺肿瘤诊断分级系统尚待研究。其次,要进一步全面确定每种影像指标对诊断分级的准确性,进一步筛选影像指标,以利于PI-RADS的推广应用。再次,本研究采用综合分析多种影像指标(如边界、形态、大小、CT值和强化方式等)确定PI-RADS分级的方法,进一步研究需形成确定的诊断思维流程,以利于提高应用PI-RADS诊断的一致性。最后,PI-RADS与传统影像诊断方式相结合,有利于二者取长补短,将会进一步规范临床工作。此外,本研究为回顾性病例研究,纳入的PI-RADS 4级肿瘤病例数量受限,尚需多中心临床研究以进一步验证PI-RADS的临床应用价值。
3.3. PI-RADS与病理类型的相关性
本研究结果提示腮腺肿瘤中良性肿瘤比例为90.2%,良性肿瘤中多形性腺瘤和Warthin瘤合计约占71.6%,与既往研究结果接近[11]。因此,明确多形性腺瘤和Warthin瘤的影像表现是最为重要的鉴别诊断要点。Warthin瘤的CT表现具有特征性,增强CT诊断Warthin瘤的准确性较高,部分病例可做确定性诊断。本研究中诊断为PI-RADS 2级的病例中约72.1%为Warthin瘤。Warthin瘤中少部分由于囊性变、部位变异或未作增强检查等原因可能会导致诊断准确性降低。多形性腺瘤的影像学表现类型较多,且其影像学表现与组织学类型有密切相关性[12],上皮细胞为主型、黏液成分丰富型和梗死型多形性腺瘤的影像学表现差别较大[13]。部分腮腺恶性肿瘤的影像表现接近多形性腺瘤,因此,在影像诊断中如果仅把肿瘤分类为良性和恶性,诊断结果经常导致与病理结果不符,而影像分级诊断方法允许影像医师在4个不同等级的恶性风险中做出评价,更适用于腮腺肿瘤性疾病的影像诊断。
本研究中PI-RADS 2级诊断结果以具有特异性良性表现或良性可能性较大的肿瘤为主,如脂肪瘤、Warthin瘤等。PI-RADS 3级诊断结果包括影像缺少确定性恶性表现的肿瘤,以多形性腺瘤为主。腺泡细胞癌、黏液表皮样癌等一些低度恶性肿瘤在影像中可以表现为边界基本清楚的病变,且缺少确定性的恶性表现,与多形性腺瘤表现难以区分,将此类肿瘤疾病诊断为PI-RADS 3级,可以很好地提示临床肿瘤的恶性风险。PI-RADS 4级肿瘤的恶性风险增加,但仍缺少确定性恶性证据,如明确的较大范围边界不清、周围结构侵犯等,目前,本研究尚缺少足够的数据支持对PI-RADS 4级做更进一步的划分。本研究结果中PI-RADS 5级以腺样囊性癌和黏液表皮样癌为常见,表明随着影像诊断分级增高,恶性肿瘤构成比逐渐增加。
肿瘤边界不清、侵犯毗邻结构、伴有恶性淋巴结、伴有明确的恶性肿瘤临床症状等可以作为影像诊断恶性肿瘤的较为确定的依据。大块不规则钙化影像可能与多形性腺瘤恶变相关;肿瘤的强化特征与具体的病理学类型相关,而不是与良恶性有直接关系。强化特征包括强化的范围性、时期性和强度,常见的唾液腺来源肿瘤中,同等条件下,强化程度由高至低依次是基底细胞腺瘤、Warthin瘤、黏液表皮样癌、腺样囊性癌、多形性腺瘤,因此,参考强化特征对基底细胞腺瘤和Warthin瘤进行预测的准确性较高,但对其他肿瘤的预测准确性欠佳。
腮腺尚有多种肿瘤为交界性肿物,如细胞密集型多形性腺瘤[14]或病理建议需要密切随访的淋巴上皮性病变,其在影像表现上可能也有所差异。因此,建议将缺少影像特异性和确切恶性证据的腮腺肿瘤诊断为PI-RADS 3~4级。
3.4. PI-RADS与临床治疗策略的相关性
PI-RADS与治疗策略的相关性尚需进一步深入研究。由于可以根据肿瘤的恶性概率做出PI-RADS分级诊断,其对指导临床治疗有一定意义。PI-RADS 2级的腮腺肿物治疗可考虑定期随诊观察或择期手术,手术术式可考虑选择保守的腮腺部分切除术等。PI-RADS 3级和4级的腮腺肿瘤由于恶性可能性增加,建议择期手术治疗。PI-RADS 3b级通常为易与良性肿瘤、恶性肿瘤混淆的炎症性或免疫性肿块,如干燥综合征结节型、IgG4相关性唾液腺炎、嗜酸性淋巴肉芽肿等,这些疾病多需要临床进一步相关辅助检查以确诊。PI-RADS 5级的腮腺肿瘤建议手术治疗,考虑术前活检和术后辅助治疗。
综上,应用PI-RADS对腮腺肿瘤的临床和影像诊断具有一定的指导意义,有利于提高腮腺肿瘤的临床诊疗水平。
References
- 1.张 震康, 俞 光岩. 口腔颌面外科学. 北京: 北京大学医学出版社; 2013. [Google Scholar]
- 2.Abdullah A, Rivas FF, Srinivasan A. Imaging of the salivary glands. Semin Roentgenol. 2013;48(1):65–74. doi: 10.1053/j.ro.2012.08.002. [DOI] [PubMed] [Google Scholar]
- 3.俞 光岩, 高 岩, 孙 勇刚. 口腔颌面部肿瘤. 北京: 人民卫生出版社; 2002. [Google Scholar]
- 4.Seethala RR, Stenman G. Update from the 4th edition of the World Health Organization classification of head and neck tumours: Tumors of the salivary gland. Head Neck Pathol. 2017;11(1):55–67. doi: 10.1007/s12105-017-0795-0. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.White SC, Pharoah MJ. Oral radiology principles and interpretation. 3rd ed. St. Louis, Missouri: Mosby Elsevier. 2014. [Google Scholar]
- 6.马 绪臣. 口腔颌面医学影像学. 北京: 北京大学医学出版社; 2014. [Google Scholar]
- 7.马 绪臣, 李 铁军. 口腔颌面部疾病CT诊断与鉴别诊断. 北京: 北京大学医学出版社; 2019. [Google Scholar]
- 8.Horvath E, Majlis S, Rossi R, et al. An ultrasonogram reporting system for thyroid nodules stratifying cancer risk for clinical ma-nagement. J Clin Endocrinol Metab. 2009;94(5):1748–1751. doi: 10.1210/jc.2008-1724. [DOI] [PubMed] [Google Scholar]
- 9.Barentsz JO, Richenberg J, Clements R, et al. ESUR prostate MR guidelines 2012. Eur Radiol. 2012;22(4):746–757. doi: 10.1007/s00330-011-2377-y. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Abdel Razek AA, Ashmalla GA, Gaballa G, et al. Pilot study of ultrasound parotid imaging reporting and data system (PIRADS): Inter-observer agreement. Eur J Radiol. 2015;84(12):2533–2538. doi: 10.1016/j.ejrad.2015.09.001. [DOI] [PubMed] [Google Scholar]
- 11.马 大权, 俞 光岩. 唾液腺病学. 北京: 人民卫生出版社; 2014. [Google Scholar]
- 12.Kim H, Kim SY, Kim YJ, et al. Correlation between computed tomography imaging and histopathology in pleomorphic adenoma of parotid gland. Auris Nasus Larynx. 2018;45(4):783–790. doi: 10.1016/j.anl.2017.09.013. [DOI] [PubMed] [Google Scholar]
- 13.Ito FA, Jorge J, Vargas PA, et al. Histopathological findings of pleomorphic adenomas of the salivary glands. Med Oral Patol Oral Cir Bucal. 2009;14(2):E57–61. [PubMed] [Google Scholar]
- 14.黄 敏娴, 马 大权, 俞 光岩, et al. 复发性涎腺多形性腺瘤的临床与病理分析. 现代口腔医学杂志. 2008;22(1):1–4. [Google Scholar]