天津医药

天津医药 ›› 2021, Vol. 49 ›› Issue (4): 427-431.doi: 10.11958/20202825

• 应用研究 • 上一篇    下一篇

18F-FDG PET/MR与PET/CT在颅内肿瘤诊断中的初步对比研究#br#

王佳1，许远帆2，宫健1，李德鹏1，张晓宏1△   

  1. 1天津全景医学影像诊断中心（邮编300191）；2杭州全景医学影像诊断中心
  • 收稿日期:2020-10-15 修回日期:2021-02-05 出版日期:2021-04-15 发布日期:2021-04-16
  • 通讯作者: 张晓宏 E-mail:zhangxh102@163.com
  • 作者简介:王佳（1983），女，本科，主治医师，主要从事PET/CT诊断工作。E-mail：981448227@qq.com

A preliminary comparative study of PET/CT and PET/MR in the diagnosis of intracranial tumors

WANG Jia1, XU Yuan-fan2, GONG Jian1, LI De-peng1, ZHANG Xiao-hong1△   

  1. 1 Tianjin Universal Medical Imaging Diagnostic Center, Tianjin 300191, China; 2 Hangzhou Universal Medical Imaging Diagnostic Center
  • Received:2020-10-15 Revised:2021-02-05 Published:2021-04-15 Online:2021-04-16
  • Contact: ZHANG Xiao-hong E-mail:zhangxh102@163.com

PDF (PC)

4304

摘要： 目的　比较一体化PET/MR与PET/CT在颅内肿瘤诊断中的差异，以及对临床治疗方案的影响。方法　32例颅内肿瘤患者，同日先后行18F-脱氧葡萄糖（18F-FDG）PET/CT全身显像及PET/MR颅脑局部显像检查。用Pearson相关性分析、Bland-Altman分析及t检验对两者最大标准摄取值（SUVmax）及靶区/本底比值（T/B）进行比较。参照病理及随访结果，运用χ2检验比较PET/CT与PET/MR的病灶检出率。结果　32例患者共78个病灶均经病理或临床随访证实，其中76个病灶为恶性肿瘤，包括胶质瘤7例共10个病灶，淋巴瘤3例共11个病灶，脑转移瘤20例共55个病灶，其余2例为脑膜瘤、垂体微腺瘤各1例，均为良性。PET/MR较PET/CT多发现31个病灶，差异有统计学意义（χ2=40.266，P＜0.01）。78个病灶中有40个病灶18F-FDG摄取明显高于脑实质，病灶的SUVmax-MR与SUVmax-CT有良好的相关性（r=0.799，P＜0.05），PET-MR影像T/B值较PET-CT略高，均值差为0.21（95%CI：-0.433~0.850）；另38个病灶18F-FDG摄取接近脑实质，PET/MR比PET/CT多检出28个，两者检出率差异有统计学意义（χ2=44.335，P＜0.01）。36个病灶周围有明显水肿，两者检出率差异无统计学意义（χ2=2.060，P＞0.05），但其中3个病灶PET/CT检查仅发现水肿而未发现明确病灶；42个无水肿病灶中，PET/MR多发现29个，两者差异有统计学意义（χ2=44.290，P＜0.01）。62个最大径＜3.0 cm的病灶中，PET/MR比PET/CT多检出31枚，两者检出率差异有统计学意义（χ2=41.338，P＜0.01）。结论　一体化PET/MR对颅内病灶的检出率明显高于PET/CT。运用全身PET/CT加颅脑局部PET/MR联合显像可明显弥补PET/CT对颅内病灶检出率的不足，指导临床选择治疗方案。

关键词: 磁共振成像, 体层摄影术, X线计算机, 正电子发射断层显像术, 氟脱氧葡萄糖F18, 颅内肿瘤, 检出率

Abstract: Objective　To compare the differences between PET/CT and integrated PET/MR in the diagnosis of intracranial tumors, and the effects on clinical  therapy strategy. Methods　Thirty-two patients with intracranial tumors detected by sequentially 18F-FDG PET/CT imaging and PET/MR intracranial imaging on the same day were enrolled. Pearson correlation analysis, Bland-Altman analysis and t test were used to compare the maximum standard uptake value (SUVmax) and target area/back ground ratio (T/B) of PET/CT and PET/MR. With reference to the pathological and follow-up results, the χ2 test was used to compare the differences in detection rates of PET/CT and PET/MR. Results　A total of 78 lesions (32 patients) were confirmed by pathology or clinical follow-up including 10 glioma, 11 lymphoma, 55 metastases, 1 meningiomas and 1 pituitary tumor. All lesions were benign. PET/MR found 31 more lesions compared with PET/CT, and there was a significant difference between them (χ2=40.266，P＜0.01). In the 78 lesions, the 18F-FDG uptake of 40 lesions were significantly higher than intracranial parenchyma, with good agreement in SUVmax-MR and SUVmax-CT (r=0.799，P＜0.05). For other 38 lesions with comparable or lower uptake of intracranial parenchymal, PET/MR detected 28 more lesions than PET/CT, and the detection rate was statistically significant (χ2=44.335, P＜0.01). There were 36 lesions with obvious edema found in both examinations (χ2=2.060，P＞0.05), although 3 only found edema with no clear lesions in PET/CT. Among the 42 lesions without edema, 29 more lesions were found in PET/MR, which was significant compared with that of PET/CT (χ2=44.290， P＜0.01). For 62 lesions with Dmax＜3.0 cm, 31 more lesions were detected by PET/MR than PET/CT (χ2=41.338, P＜0.01). Conclusion　The detection rate of integrated PET / MR is significantly higher than that of PET/CT. Sequentially 18F-FDG PET/CT imaging and PET/MR intracranial imaging can obviously make up the deficiency of detection rate of PET/CT for intracranial lesions, guiding clinical selection of treatment options.

Key words: magnetic resonance spectroscopy, tomography, X-Ray computed, positron-emission tomography, fluorodeoxyglucose F18, intracranial tumors, detection rate

中图分类号: