-
颅内原发性中枢神经系统淋巴瘤(primary central nervous system lymphoma,PCNSL)是一种罕见的中枢神经系统恶性肿瘤,属于侵袭性结外非霍奇金淋巴瘤,病理类型多为弥漫性大B细胞淋巴瘤[1]。虽然其发病率低,但恶性程度很高[2-3]。目前,颅内PCNSL的初步诊断依赖CT、MRI,但因颅内PCNSL与颅内其他常见占位性疾病(如胶质瘤、恶性肿瘤的颅内转移病灶)在CT、MRI的影像学表现上存在诸多相似之处[4-6],因此误诊率仍然较高。18F-FDG PET/CT显像是融合解剖与功能信息的影像学模式,综合解剖结构信息和葡萄糖代谢状态来分析病灶特点[7-8]。我们回顾性分析经病理学结果证实的颅内PCNSL患者的18F-FDG PET/CT影像资料,总结影像学特征,并分析颅内PCNSL与脑胶质瘤和脑转移瘤的18F-FDG PET/CT影像的鉴别诊断阈值,以提高对该病的认识。
-
45例颅内PCNSL患者经病理学结果证实均为弥漫性大B细胞淋巴瘤,病灶数共70个,多位于脑组织较深的部位,近中线分布多见于脑室周围,体部未见病灶。45例患者中,单发病灶较常见(34/45,75.6%);70个病灶中,病灶位于幕上者多见(61/70,87.1%)、位于幕下者仅9个(9/70,12.9%)。52例脑胶质瘤患者的病灶数共98个,其中,位于幕上者32个(32/98,32.7%)、位于幕下者66个(66/98,67.3%),经病理学结果证实均为高级别脑胶质瘤。60例脑转移瘤患者的病灶数共129个,其中,位于幕上者15个(15/129,11.6%)、位于幕下者114个(114/129,88.4%),经病理学结果证实原发肿瘤均为肺癌(表1)。
组别 例数 单发病灶分布情况 多发病灶分布情况 PCNSL 45 幕上(额叶、顶叶、颞叶、枕叶、基底节区、脑室旁、胼胝体、丘脑);幕下(小脑) 幕上(额叶、颞叶、顶叶、枕叶、基底节区、侧脑室后角及内侧);幕下(小脑蚓部、中脑、脑桥) 脑胶质瘤 52 幕上(额叶、颞叶、顶叶、枕叶、基底节区);幕下(小脑、中脑、脑干) 幕上(额叶、颞叶、枕叶、基底节区、脑室);幕下(小脑、脑桥) 脑转移瘤 60 幕上(额叶、颞叶、顶叶、枕叶);幕下(小脑、脑干) 幕上(额叶、颞叶、顶叶、枕叶、基底节区、侧脑室);幕下(小脑、脑桥) 注:表中,PCNSL:原发性中枢神经系统淋巴瘤 表 1 3组患者的病灶分布特征
Table 1. Distribution characteristics of lesions in the three groups
-
颅内PCNSL的实质区呈结节状和团块状,CT表现为等或稍高密度,密度均匀,占位效应不明显,病灶周边水肿区不明显,病灶内继发征象囊变、坏死少见,中线移位、脑室受压较少见,肿瘤实质区对18F-FDG的摄取明显增高(图1)。脑胶质瘤的实质区病灶形态不规则,CT表现为稍低或等密度,占位效应较明显,病灶周边水肿区较明显,肿瘤实质病灶内伴继发征象囊变、出血,个别见钙化,中线移位及脑室受压可见,病灶边缘对18F-FDG的摄取增高,内部囊变、坏死,出血区的摄取减低。脑转移瘤的实质区病灶呈结节状,CT表现为稍低密度或混杂密度,几乎所有的病灶周边均可见中重度水肿,中线移位及脑室受压多见,肿瘤实质病灶边缘对18F-FDG的摄取呈不同程度增高(表2)。
图 1 颅内原发性中枢神经系统淋巴瘤患者(男性,80岁)的18F-FDG PET/CT图像
Figure 1. 18F-FDG PET/CT image of intracranial primary central nervous system lymphoma patient(male,80 years old)
组别 例数 密度 囊变、坏死 水肿 中线移位 脑室受压 18F-FDG摄取程度 18F-FDG高摄取的分布情况 PCNSL 45 等或稍高 少见 轻、重度 少见 少见 显著升高 绝大多数均匀 脑胶质瘤 52 稍低或等 可见 轻、重度 可见 可见 显著升高 不均匀 脑转移瘤 60 稍低或混杂 多见 中、重度 多见 多见 不同程度升高 明显不均匀 注:表中,PCNSL:原发性中枢神经系统淋巴瘤;FDG:氟脱氧葡萄糖 表 2 3组患者的病灶形态特征
Table 2. Morphological characteristics of lesions in the three groups
-
对3组患者病灶的半定量分析结果(表3)显示,颅内PCNSL病灶的SUVmax(t=13.02、11.07,均P=0.000)和T/WM(t=10.13、13.88,均P=0.000)明显高于脑胶质瘤和脑转移瘤病灶,且差异有统计学意义。颅内PCNSL的病灶大小和CT值与脑胶质瘤相比,差异无统计学意义(t=0.61、1.40,P=0.546、0.165);但颅内PCNSL的病灶大小和CT值高于脑转移瘤,差异有统计学意义(t=4.82、2.32,P=0.000、0.023)。
组别 例数 病灶大小(cm) CT值(HU) SUVmax T/WM PCNSL 45 3.08±0.12a 37.59±0.83a 26.42±1.17ab 4.37±0.10ab 脑胶质瘤 52 2.98±0.11 36.02±0.76 9.96±0.48 2.99±0.09 脑转移瘤 60 2.34±0.09 35.57±0.43 11.97±0.58 2.60±0.08 注:表中,a:与脑转移瘤相比,差异均有统计学意义(t=2.32~13.88,均P<0.05);b:与脑胶质瘤相比,差异均有统计学意义(t=13.02、10.13,均P=0.000)。FDG:氟脱氧葡萄糖;PET:正电子发射断层显像术;CT:计算机体层摄影术;PCNSL:原发性中枢神经系统淋巴瘤;SUVmax:最大标准化摄取值;T/WM:肿瘤与对侧相应部位脑组织SUVmax的比值 表 3 3组患者的病灶18F-FDG PET/CT半定量分析结果(
)$\bar x \pm s $ Table 3. Results of the semi-quantitative analysis of lesions in the three groups (
)$\bar x \pm s $ -
ROC曲线分析结果显示,当SUVmax=15.8时,颅内PCNSL与脑胶质瘤鉴别诊断的AUC最大(0.982),灵敏度为95.6%,特异度为94.2%(图2中A);而当SUVmax=16.8时,颅内PCNSL与脑转移瘤鉴别诊断的AUC最大(0.946),灵敏度为93.3%,特异度为85.0%(图2中B)。当T/WM=3.395时,颅内PCNSL与脑胶质瘤鉴别诊断的AUC最大(0.918),灵敏度为86.7%,特异度为73.1%(图2中C);而当T/WM=3.220时,颅内PCNSL与脑转移瘤鉴别诊断的AUC最大(0.973),灵敏度为97.8%,特异度为83.3%(图2中D)。可以认为T/WM=3.395为颅内PCNSL与脑胶质瘤鉴别诊断的最佳阈值,而T/WM=3.220为颅内PCNSL与脑转移瘤鉴别诊断的最佳阈值。
18F-FDG PET/CT在颅内原发性中枢神经系统淋巴瘤诊断中的价值
Diagnostic value of 18F-FDG PET/CT in intracranial primary central nervous system lymphoma
-
摘要:
目的 探讨18F-FDG PET/CT显像视觉和半定量分析在颅内原发性中枢神经系统淋巴瘤(PCNSL)诊断中的价值。 方法 回顾性分析2011年5月至2018年12月于郑州大学第一附属医院行18F-FDG PET/CT检查的45例颅内PCNSL患者[男性26例、女性19例,年龄(57.49±2.54)岁]的影像资料,并分别与52例脑胶质瘤和60例脑转移瘤患者进行对比,采用视觉分析3组患者的病灶分布和形态特征,半定量分析最大标准化摄取值(SUVmax)、肿瘤与对侧相应部位脑组织SUVmax的比值(T/WM),评价18F-FDG PET/CT在颅内PCNSL诊断中的价值。2组间半定量结果的比较采用独立样本t检验和校正后t检验,2组间诊断效能的比较及鉴别诊断阈值的判断采用受试者工作特征(ROC)曲线进行分析。 结果 视觉分析结果显示,颅内PCNSL以高摄取18F-FDG的幕上、单发、结节状或团块状病灶为多见,水肿、占位效应及瘤内继发改变不明显。半定量分析结果显示,与脑胶质瘤和脑转移瘤相比,颅内PCNSL的SUVmax最高(脑胶质瘤:9.96±0.48;脑转移瘤:11.97±0.58;颅内PCNSL:26.42±1.17,t=13.02、11.07,均P=0.000),T/WM亦最高(脑胶质瘤:2.99±0.09;脑转移瘤:2.60±0.08;颅内PCNSL:4.37±0.10,t=10.13、13.88,均P=0.000),且差异均有统计学意义。当SUVmax=15.8时,颅内PCNSL与脑胶质瘤鉴别诊断的ROC曲线下面积(AUC)最大(0.982),而当SUVmax=16.8时,颅内PCNSL与脑转移瘤鉴别诊断的AUC最大(0.946);T/WM=3.395为颅内PCNSL与脑胶质瘤鉴别诊断的最佳阈值,T/WM=3.220为颅内PCNSL与脑转移瘤鉴别诊断的最佳阈值。 结论 18F-FDG PET/CT显像可以作为传统影像学诊断颅内PCNSL的有效补充,同时在颅内PCNSL与脑胶质瘤和脑转移瘤的鉴别诊断方面具有显著优势。 -
关键词:
- 中枢神经系统肿瘤 /
- 神经淋巴瘤病 /
- 神经胶质瘤 /
- 正电子发射断层显像术 /
- 体层摄影术,X线计算机 /
- 氟脱氧葡萄糖F18
Abstract:Objective To investigate the diagnostic value of fluorine-18 fluorodeoxyglucose (18F-FDG) PET/CT visual and semi-quantitative analyses in intracranial primary central nervous system lymphoma (PCNSL). Methods PET/CT images of 45 patients with PCNSL who underwent 18F-FDG PET/CT examination in the Department of Nuclear Medicine of the First Affiliated Hospital of Zhengzhou University from May 2011 to December 2018 (26 males and 19 females, 57.49±2.54 years old) were retrospectively reviewed and compared with 52 cases of gliomas and 60 cases of brain metastases to evaluate the value of 18F-FDG PET/CT in the diagnosis of intracranial PCNSL. The lesion distribution and morphological characteristics of the 3 groups of patients were visually analyzed, and the maximum standardized uptake value (SUVmax) and the ratio of SUVmax of tumor to white matter (T/WM) were semi-quantitatively analyzed. The mean comparison between the two groups was performed using independent sample t test and adjusted t test. The comparison of the diagnostic efficacy between the two groups and the judgment of the differential diagnosis threshold were performed using receiver operating characteristic (ROC) curve analysis. Results Visually, intracranial PCNSL showed a very high uptake of 18F-FDG in single, focal nodule or mass lesions mostly located in the supratentorial brain. The space-occupying effects of edema, as well as cystic degeneration, were not obvious in PCNSL. Semi-quantitative analysis showed that intracranial PCNSL had the highest SUVmax (gliomas: 9.96±0.48, brain metastases: 11.97±0.58, PCNSL: 26.42±1.17) and T/WM (gliomas: 2.99±0.09, brain metastases: 2.60±0.08, PCNSL: 4.37±0.10) among the three types of tumors with statistical differences (t=13.02 and 11.07, t=10.13 and 13.88, all P=0.000). In the differential diagnosis of intracranial PCNSL and glioma, the area under the ROC curve (AUC) analysis reached the largest value at the SUVmax of 15.8. The AUC for PCNSL and metastatic tumor peaked at the SUVmax of 16.8. The T/WMs of 3.395 and 3.220 were considered the optimal thresholds for the differential diagnosis of intracranial PCNSL from gliomas and brain metastases, respectively. Conclusion 18F-FDG PET/CT imaging can effectively complement the traditional diagnosis of intracranial PCNSL, especially in the differential diagnosis of PCNSL from gliomas and brain metastases. -
表 1 3组患者的病灶分布特征
Table 1. Distribution characteristics of lesions in the three groups
组别 例数 单发病灶分布情况 多发病灶分布情况 PCNSL 45 幕上(额叶、顶叶、颞叶、枕叶、基底节区、脑室旁、胼胝体、丘脑);幕下(小脑) 幕上(额叶、颞叶、顶叶、枕叶、基底节区、侧脑室后角及内侧);幕下(小脑蚓部、中脑、脑桥) 脑胶质瘤 52 幕上(额叶、颞叶、顶叶、枕叶、基底节区);幕下(小脑、中脑、脑干) 幕上(额叶、颞叶、枕叶、基底节区、脑室);幕下(小脑、脑桥) 脑转移瘤 60 幕上(额叶、颞叶、顶叶、枕叶);幕下(小脑、脑干) 幕上(额叶、颞叶、顶叶、枕叶、基底节区、侧脑室);幕下(小脑、脑桥) 注:表中,PCNSL:原发性中枢神经系统淋巴瘤 表 2 3组患者的病灶形态特征
Table 2. Morphological characteristics of lesions in the three groups
组别 例数 密度 囊变、坏死 水肿 中线移位 脑室受压 18F-FDG摄取程度 18F-FDG高摄取的分布情况 PCNSL 45 等或稍高 少见 轻、重度 少见 少见 显著升高 绝大多数均匀 脑胶质瘤 52 稍低或等 可见 轻、重度 可见 可见 显著升高 不均匀 脑转移瘤 60 稍低或混杂 多见 中、重度 多见 多见 不同程度升高 明显不均匀 注:表中,PCNSL:原发性中枢神经系统淋巴瘤;FDG:氟脱氧葡萄糖 表 3 3组患者的病灶18F-FDG PET/CT半定量分析结果(
)$\bar x \pm s $ Table 3. Results of the semi-quantitative analysis of lesions in the three groups (
)$\bar x \pm s $ 组别 例数 病灶大小(cm) CT值(HU) SUVmax T/WM PCNSL 45 3.08±0.12a 37.59±0.83a 26.42±1.17ab 4.37±0.10ab 脑胶质瘤 52 2.98±0.11 36.02±0.76 9.96±0.48 2.99±0.09 脑转移瘤 60 2.34±0.09 35.57±0.43 11.97±0.58 2.60±0.08 注:表中,a:与脑转移瘤相比,差异均有统计学意义(t=2.32~13.88,均P<0.05);b:与脑胶质瘤相比,差异均有统计学意义(t=13.02、10.13,均P=0.000)。FDG:氟脱氧葡萄糖;PET:正电子发射断层显像术;CT:计算机体层摄影术;PCNSL:原发性中枢神经系统淋巴瘤;SUVmax:最大标准化摄取值;T/WM:肿瘤与对侧相应部位脑组织SUVmax的比值 -
[1] 朱雄增, 李小秋. 解读2008年恶性淋巴瘤WHO分类—B细胞淋巴瘤[J]. 临床与实验病理学杂志, 2010, 26(2): 125−130. DOI: 10.3969/j.issn.1001−7399.2010.02.001.
Zhu XZ, Li XQ. Inerpretation of the 2008 WHO classification of malignant lymphoma—B celllymphoma[J]. J Clin Exp Pathol, 2010, 26(2): 125−130. DOI: 10.3969/j.issn.1001−7399.2010.02.001.[2] Chukwueke UN, Nayak L. Central Nervous System Lymphoma[J]. Hematol/Oncol Clin North Am, 2019, 33(4): 597−611. DOI: 10.1016/j.hoc.2019.03.008. [3] Fuentes-Raspall R, Solans M, Auñon-Sanz C, et al. Incidence and survival of primary central nervous system lymphoma (PCNSL): results from the Girona cancer registry (1994−2013)[J]. Clin Transl Oncol, 2018, 20(12): 1628−1630. DOI: 10.1007/s12094−018−1890−8. [4] Suh CH, Kim HS, Park JE, et al. Primary Central Nervous System Lymphoma: Diagnostic Yield of Whole-Body CT and FDG PET/CT for Initial Systemic Imaging[J]. Radiology, 2019, 292(2): 440−446. DOI: 10.1148/radiol.2019190133. [5] 吴慧. 原发性中枢神经系统淋巴瘤磁共振成像特征[J]. 影像研究与医学应用, 2019, 3(6): 80−81. DOI: 10.3969/j.issn.2096−3807.2019.06.049.
Wu H. Magnetic resonance imaging features of primary central nervous system lymphoma[J]. J Imaging Res Med Appl, 2019, 3(6): 80−81. DOI: 10.3969/j.issn.2096−3807.2019.06.049.[6] 曾雪明. 颅内肿瘤MRI分析(附667例报告)[D]. 南宁: 广西医科大学, 2010. DOI: 10.7666/d.y1683074.
Zeng XM. MRI analysis of intracranial tumor (report of 667 cases)[D]. Nanning: Guangxi Medical University, 2010. DOI: 10.7666/d.y1683074.[7] Yang YH, He MZ, Li T, et al. MRI combined with PET-CT of different tracers to improve the accuracy of glioma diagnosis: a systematic review and meta-analysis[J]. Neurosurg Rev, 2019, 42(2): 185−195. DOI: 10.1007/s10143−017−0906−0. [8] Zhang Q, Gao X, Wei GH, et al. Prognostic Value of MTV, SUV<sub>max</sub> and the T/N Ratio of PET/CT in Patients with Glioma: A Systematic Review and Meta-Analysis[J]. J Cancer, 2019, 10(7): 1707−1716. DOI: 10.7150/jca.28605. [9] 杨柳. 18F-FDG PET/CT在原发性中枢神经系统淋巴瘤诊断中的应用价值[D]. 郑州: 郑州大学, 2019.
Yang L. The value of 18F-FDG PET/CT in the diagnosis of central nervous system lymphoma[D]. Zhengzhou: Zhengzhou University, 2019.[10] 白人驹, 张雪林. 医学影像诊断学[M]. 3版. 北京: 人民卫生出版社, 2001: 36−42.
Bai RJ, Zhang XL. Medical imaging diagnostics[M]. 3rd ed. Beijing: People's Medical Publishing House, 2001: 36−42.[11] 张晶晶, 刘保平, 阮翘, 等. 椎管内原发性弥漫性大B细胞淋巴瘤<sup>18</sup>F-FDG PET/CT显像一例[J]. 中华核医学与分子影像杂志, 2018, 38(5): 355−356. DOI: 10.3760/cma.j.issn.2095−2848.2018.05.012.
Zhang JJ, Liu BP, Ruan Q, et al. <sup>18</sup>F-FDG PET/CT imaging in intra-spinal canal primary diffuse large B cell lymphoma: a case report[J]. Chin J Nucl Med Mol Imaging, 2018, 38(5): 355−356. DOI: 10.3760/cma.j.issn.2095−2848.2018.05.012.[12] 崔向丽, 贾文清, 初君盛, 等. NCCN2009颅内原发淋巴瘤诊疗指南(译文)[J]. 中国神经肿瘤杂志, 2009, 7(4): 311−314. DOI: CNKI:SUN:CJNO.0.2009−04−018.
Cui XL, Jia WQ, Chu JS, et al. NCCN2009 Guidelines for Diagnosis and Treatment of Intracranial Primary Lymphoma (Translation)[J]. Chin J Neuro-Oncol, 2009, 7(4): 311−314. DOI: CNKI:SUN:CJNO.0.2009−04−018.[13] Sethi TK, Reddy NM. Treatment of newly diagnosed primary central nervous system lymphoma: current and emerging therapies[J]. Leuk Lymphoma, 2019, 60(1): 6−18. DOI: 10.1080/10428194.2018.1466296. [14] Kong ZR, Jiang CD, Zhu RZ, et al. 18F-FDG-PET-based radiomics features to distinguish primary central nervous system lymphoma from glioblastoma[J/OL]. Neuroimage Clin, 2019, 23: 101912[2019-10-28]. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6702330. DOI: 10.1016/j.nicl.2019.101912. [15] Jara H. Primary Central Nervous System Lymphoma: Lessons and Opportunities from 2 Decades of CT and PET/CT[J]. Radiology, 2019, 292(2): 447−448. DOI: 10.1148/radiol.2019191147. [16] Liu DL, Kong ZR, Wang YK, et al. Quantitative and Visual Characteristics of Primary Central Nervous System Lymphoma on <sup>18</sup>F-FDG-PET[J]. Interdiscip Sci: Comput Life Sci, 2019, 11(2): 300−306. DOI: 10.1007/s12539−019−00333−y.