双源结石能谱CT对尿酸结石的预测价值

Predictive value of dual-source stone energy spectrum CT for uric acid stones

目的探讨基于双源结石能谱CT的结石成分预测方法对尿酸结石的预测价值。方法回顾性分析上海市第六人民医院2020年7月至2022年7月收治的204例泌尿系结石患者的临床资料,男159例,女45例。年龄(50.7±14.3)岁。上尿路(肾、输尿管)结石187例,下尿路(膀胱、尿道)结石17例。结石长径(13.4±5.1)mm。所有患者术前行双源结石能谱CT检查,分别测量150 kV和100kV电压下的CT值,并根据计算的双能比(Ratio值=100kV的CT值除以150kV的CT值)预测结石成分:Ratio值≤1.15为尿酸结石,1.16~1.35为胱氨酸结石,1.36~1.55为草酸钙结石,>1.55为羟基磷灰石。本组204例中,行经皮肾镜取石术33例、输尿管软镜碎石术171例。术后结石标本行红外光谱分析结石成分。采用受试者工作特征(ROC)曲线分析术前双源结石能谱CT预测尿酸结石的效能。结果204例患者,术前CT预测尿酸结石28例;非尿酸结石176例,其中草酸钙结石136例,羟基磷灰石38例,胱氨酸结石2例。术后红外光谱分析结果为尿酸结石26例;非尿酸结石178例,其中草酸钙结石129例,羟基磷灰石47例,胱氨酸结石2例。与术前CT结果相比,尿酸结石中有2例假阳性,无假阴性。ROC曲线显示,在100kV电压下CT值预测尿酸结石的敏感性为96.2%,特异性为99.5%,ROC曲线下面积(AUC)为0.995,最佳预测值为620HU;在150kV电压下CT值预测尿酸结石的敏感性为96.2%,特异性为85.5%,AUC为0.916,最佳预测值为597.5HU。采用Ratio值预测尿酸结石的敏感性为100.0%,特异性为97.9%,AUC为0.996,最佳预测值为1.14。在100kV和150kV电压下尿酸结石的CT值分别为(558.58±77.07)HU和(521.12±83.54)HU,草酸钙结石的CT值分别为(1335.26±301.82)HU和(878.21±200.21)HU,羟基磷灰石的CT值分别为(1104.09±203.61)HU和(710.38±178.44)HU,胱氨酸结石的CT值分别为(684.5±67.18)HU和(573.5±67.10)HU,尿酸结石与其他成分比较差异均有统计学意义(P<0.01)。尿酸结石、草酸钙结石、羟基磷灰石、胱氨酸结石的Ratio值分别为1.08±0.06、1.52±0.08、1.62±0.40、1.19±0.02,尿酸结石与其他成分结石比较差异均有统计学意义(P<0.01)。结论双源结石能谱CT的CT值和Ratio值能有效区分尿酸结石与其他类型结石,该方法预测尿酸结石的准确性、敏感性、特异性均较高。

Objective To investigate the predictive value of stone composition prediction method based on dual-source stone energy spectrum CT for uric acid stones.Methods The clinical data of 204 patients with urinary stones,159 males and 45 females,admitted to Shanghai Sixth Peoples Hospital from July 2020 to July 2022 were retrospectively analyzed.The average age was(50.7±14.3)years.There were 187 cases of upper urinary tract(kidney,ureter)stones and 17 cases of lower urinary tract(bladder,urethra)stones.All patients underwent preoperative dual-source stone energy spectroscopy CT,measuring CT values at 150 kV and 100 kV,respectively,and the calculated dual energy ratio(Ratio)was used to predict stone composition.Of the 204 cases in this group,33 cases underwent percutaneous nephrolithotomy and 171 cases underwent ureteroscopic lithotripsy.Postoperative stone specimens were analyzed for stone composition by infrared spectroscopy.Subject work characteristic(ROC)curves were used to analyze the efficacy of preoperative dual-source stone energy spectrum CT to predict uric acid stones.Results In 204 patients,preoperative CT predicted 28 cases of uric acid stones and 176 cases of non-uric acid stones,including 136 cases of calcium oxalate stones,38 cases of hydroxyapatite,and 2 cases of cystine stones.Postoperative IR spectral analysis detected 26 cases of uric acid stones and 178 cases of non-uric acid stones,including 129 cases of calcium oxalate stones,47 cases of hydroxyapatite,and 2 cases of cystine stones.Compared with the preoperative CT results,there were 2 false positives and no false negatives in the classification of uric acid stones.The ROC curve showed that the sensitivity of the CT value for predicting uric acid stones at 100 kV was 96.2%,the specificity was 99.5%,and the area under the ROC curve(AUC)was 0.995,with the best prediction value of 620 HU.The sensitivity of the CT value for predicting uric acid stones at 150 kV was 96.2%,specificity was 85.5%,AUC was 0.916,and the best predictive value was 597.5 HU.The sensitivity of using Ratio values for predicting uric acid stones was 100.0%,specificity was 97.9%,AUC was 0.996,and the best predictive value was 1.14.The CT values for uric acid stones at 100 kV and 150 kV were(558.58±77.07)HU and(521.12±83.54)HU,CT values for calcium oxalate stones were(1335.26±301.82)HU and(878.21±200.21)HU,CT values for hydroxyapatite were(1104.09±203.61)HU and(710.38±178.44)HU,respectively.The CT values of cystine stones were(684.5±67.18)HU and(573.5±67.10)HU,respectively,and the differences between uric acid stones and other components were statistically significant(P<0.01).The Ratio values of uric acid stones,calcium oxalate stones,hydroxyapatite,and cystine stones were 1.08±0.06,1.52±0.08,1.62±0.40,and 1.19±0.02,respectively,and the dfferences were statistically significant(P<0.01)when comparing uric acid stones with other components.Conclusion The CT and Ratio values of dual-source stone energy spectrum CT can effectively distinguish uric acid stones from other types of stones,and the accuracy,sensitivity,and specificity of this method for predicting uric acid stones are high.