Department of Nuclear Medicine, Saitama Medical University International Medical Center, 1397-1 Yamane, Hidaka, 350-1298, Japan.
Department of Uro-Oncology, Saitama Medical University International Medical Center, 1397-1 Yamane, Hidaka, 350-1298, Japan.
Ann Nucl Med. 2021 Mar;35(3):338-346. doi: 10.1007/s12149-020-01568-2. Epub 2021 Jan 5.
Technological innovations in single-photon emission computed tomography (SPECT) have enabled a more accurate quantitative evaluation of the uptake, and the standardized uptake value (SUV) can be measured as a semi-quantitative value, as in positron emission tomography. Nevertheless, the reliability of the SUV of bone SPECT has not been well established. The purpose of this study is to evaluate the test-retest repeatability of the SUV of bone SPECT/CT in clinical settings.
This prospective study recruited patients with prostate cancer planning to receive bone SPECT/CT for the evaluation of bone abnormality between August 2017 and September 2019. Bone images were acquired twice by an integrated SPECT/CT scanner (Symbia Intevo, Siemens) within a 4- to 10-day interval. The maximum SUV (SUVmax) and peak SUV (SUVpeak) were calculated for the volumes of interests on the normal bone areas, degeneration/fracture lesions, and metastatic lesions. To determine repeatability, we calculated statistical indicators, including intraclass correlation coefficient (ICC), repeatability coefficient (RC), and mean absolute percentage difference (MAPD). For the ICC, the 95% confidential interval (CI) was also calculated, and an ICC of ≥ 0.8 was defined as an almost perfect correlation.
Twelve male patients were enrolled in the study (58-86 years; median, 71 years), and a total of 229 volumes of the interest were included in the analyses. The ICCs were 0.968 [95% CI (0.959, 0.975)] for SUVmax and 0.976 [95% CI (0.969, 0.981)] for SUVpeak. The RCs of the relative difference were 30.7% for SUVmax and 27.6% for SUVpeak, and the MAPDs (± standardized deviation) of all lesions were 12.3 ± 9.9% for SUVmax and 11.5 ± 8.3% for SUVpeak. The RCs and the MAPDs showed comparable value with the previous report regarding repeatability studies on PET.
An almost perfect correlation was demonstrated by repeated SUVmax and SUVpeak measured by quantitative integrated SPECT/CT. The quantitative values could be reliable indicators in patient management.
单光子发射计算机断层扫描(SPECT)技术的创新使摄取的更准确的定量评估成为可能,并且可以像正电子发射断层扫描(PET)一样测量标准化摄取值(SUV)作为半定量值。然而,骨 SPECT 的 SUV 的可靠性尚未得到很好的证实。本研究旨在评估临床环境中骨 SPECT/CT 的 SUV 测试-再测试重复性。
这项前瞻性研究招募了 2017 年 8 月至 2019 年 9 月期间计划接受骨 SPECT/CT 检查以评估骨异常的前列腺癌患者。在 4-10 天的时间内,由集成的 SPECT/CT 扫描仪(Symbia Intevo,Siemens)两次采集骨图像。为了计算 SUVmax 和 SUVpeak,我们在正常骨区域、退行性/骨折病变和转移性病变的感兴趣区域上计算最大 SUV(SUVmax)和峰值 SUV(SUVpeak)。为了确定重复性,我们计算了统计指标,包括组内相关系数(ICC)、重复性系数(RC)和平均绝对百分比差异(MAPD)。对于 ICC,还计算了 95%置信区间(CI),ICC≥0.8 被定义为几乎完美相关。
本研究共纳入 12 名男性患者(58-86 岁;中位数,71 岁),共纳入 229 个感兴趣区域。SUVmax 的 ICC 为 0.968 [95%CI(0.959,0.975)],SUVpeak 的 ICC 为 0.976 [95%CI(0.969,0.981)]。SUVmax 的相对差异 RC 为 30.7%,SUVpeak 的 RC 为 27.6%,所有病变的 MAPD(±标准差)SUVmax 为 12.3±9.9%,SUVpeak 为 11.5±8.3%。RC 和 MAPD 与之前关于 PET 重复性研究的报告具有可比价值。
通过定量 SPECT/CT 重复测量 SUVmax 和 SUVpeak 显示出几乎完美的相关性。定量值可能是患者管理的可靠指标。