血浆游离甲氧基肾上腺类物质对嗜铬细胞瘤和副神经节瘤的诊断价值

Value of Plasma Free Metanephrine and Normetanephrine in the Diagnosis of Pheochromocytoma and Paraganglioma

【目的】探讨血浆游离甲氧基肾上腺素类物质(包括甲氧基肾上腺素MN、甲氧基去甲肾上腺素NMN,合称MNs)在嗜铬细胞瘤和副神经节瘤(PPGL)中诊断的价值及影响因素。【方法】选择2014年12月-2020年12月中山大学孙逸仙纪念医院疑诊PPGL的病人,采用高效液相色谱-串联质谱法(LC-MS/MS)检测患者手术前血浆游离MNs水平,用受试者工作曲线(ROC)评价NMN和MN的诊断能力,并分析出现假阳性结果的影响因素。【结果】研究纳入108例病理确诊PPGL的病人,1 523例诊断明确的非PPGL病人作为对照组。两组病人血浆MN中位数[0.54(0.17~4.48)nmol/L vs. 0.15(0.11~0.21)nmol/L,P<0.001],NMN中位数[7.48(2.12~15.01)nmol/L vs. 0.32(0.22~0.46)nmol/L,P<0.001]差异具有统计学意义。血浆MN如以0.395 nmol/L为切点,其灵敏性为60.2%,特异性为97.8%;NMN切点值为1.105 nmol/L,其诊断灵敏性为87.0%,特异性为98.7%,二者联合时灵敏性和特异性分别为91.7%、99.8%。血浆MN、NMN及两者联合诊断PPGL的ROC曲线下面积及95%置信区间分别为0.800(0.743,0.858)、0.959(0.932,0.985)、0.970(0.944,0.996)。分析假阳性情况,发现在对照组出现3%的假阳性病例,其肾小球滤过率(eGFR)中位数较真阴性组更小[74.42(51.04~96.96)mL/min vs. 88.51(72.80~101.83)mL/min,P=0.001]。在212例eGFR <60 mL/min的患者中,假阳性率明显增加,为8%。将血MNs参考区间上限增加25%时,特异性增加至96.7%;参考区间上限增加50%时,特异性为98.6%。【结论】血浆游离MNs诊断PPGL具有很高的可靠性,其阴性预测值极高,接近100%。同时测定MN和NMN,可进一步提高诊断灵敏性和特异性。极少数可出现假阳性,需注意肾功能不全的影响。在eGFR<60 mL/min的患者中,假阳性率明显升高,参考区间上限需要增加25%~50%,以保证较高的特异性。

【Objective】 To explore the value of plasma free metanephrines,including metanephrine(MN) and normetanephrine(NMN),collectively referred to as MNs in the diagnosis of pheochromocytoma and paraganglioma(PPGL).【Methods】The study included 1 631 patients suspected of PPGL from December 2014 to December 2020 in SunYatsen Memorial Hospital. In these subjects,Plasma free MNs were measured by liquid chromatography-tandem mass spectrometry(LC-MS/MS)before surgery. Receiver operating characteristic(ROC)curves were used to determine the sensitivity and specificity of plasma NMN and MN in the diagnosis of PPGL.【Results】Of the screened patients,108 patients had pathologically confirmed PPGL and 1 523 patients had definitive diagnoses other than PPGL as a control group. The median value of plasma MN[0.54(0.17~4.48)nmol/L vs. 0.15(0.11~0.21)nmol/L,P<0.001]and NMN[7.48(2.12~15.01)nmol/L vs. 0.32(0.22~0.46)nmol/L,P<0.001]were significant higher in the PPGL group than in the control group. Using an upper cut-off of 0.395 nmol/L for MN,the sensitivity was 60.2% and the specificity was 97.8%;when using a cut-off of1.105 nmol/L for NMN,the diagnostic sensitivity was 87.0%,and the specificity was 98.7%. The area under the ROC curve and 95% confidence interval of plasma MN and NMN combined were 0.800(0.743,0.858),0.959(0.932,0.98)and 0.970(0.944,0.996),respectively. Analyzing the false-positive results,it was found that in the control group,3% of the false-positive cases appeared. The estimated glomerular filtration rate(eGFR)were significantly lower in the false-positive group than in the true-negative group[74.42(51.04~96.96)mL/min vs. 88.51(72.80~101.83)mL/min,P=0.001].In 212 patients with eGFR lower than 60 mL/min,the false positive rate was increased to 8%. If the upper limit of the reference interval was increased by 25%,the specificity was increased to 96.7%;when the upper limit of the reference interval was increased by 50%,the specificity was 98.6 %.【Conclusions】Plasma free MNs has a great reliability in the diagnosis of PPGL. The negative predictive value is extremely high and close to 100%. Combination analysis of plasma MN and NMN can further improve diagnostic performance. However,a few false positive cases may appear and might be influenced by impaired renal function. In patients with eGFR lower than 60 mL/min,the false positive rate is significantly increasing,which need a 25%-50% increase in the expected upper limit of a reference range to guarantee high specificity.