血浆凝血酶调节蛋白检测的临床研究

　　作者：卢兴国　　作者单位：浙江大学医学院附属第二医院血液科， 杭州 310009

　　【摘要】为了探讨血浆凝血酶调节蛋白(PTM)检测的临床价值，用ELISA法测定979例患者的PTM，并选择60名健康人作为对照。结果表明：对照组PTM 水平为20.40±7.72 μg/L，无性别和年龄差异。在疾病组中，原发性慢性肾小球疾病肾功能衰竭(CRF)组PTM水平高于无CRF组，败血症组PTM水平高于非败血症组，多脏器功能衰竭(MOF)组PTM水平高于无MOF组(P<0.01);以>70、>50和>40 μg/L为标准，分别预示CRF、败血症和MOF的灵敏度为85.7%、86.6%和77.8%，特异性为82.4%、89.5%和77.3%，阳性预示值为77.8%、76.5%和73.7%。系统性红斑狼疮(SLE)尿蛋白阳性组PTM水平高于阴性组;糖尿病并发症组的PTM水平高于无并发症组，并发微血管病变组的PTM水平高于大血管病变组(P均<0.01);以PTM高于正常上限值(>35.54 μg/L)为标准，预示SLE尿蛋白阳性临床肾损害、糖尿病并发症和微血管病变的灵敏度为77.8%、53.4%和71.2%，特异性为92.3%、97.1%和97.1%，阳性预示值为93.3%、98.6%和97.9%。急性白血病(AL)和多发性骨髓瘤(MM)初诊 时PTM升高，两病并发肾衰时极度升高(P<0.01)。动态检测多发伤、脑卒中急性期和恢复期、AL和MM化疗前后、癌症术前后PTM水平与病情变化相关。以微血管病变为主要疾病的PTM 水平高于大血管病变疾病 (P<0.01)， 以高于正常上限值为标准，微血管病变疾病的灵敏度为77.7%、特异性71.2%，阳性预示值75.6%。结论：PTM水平是评估微血管病变疾病的良好指标，也是预警或评估疾病严重程度及其演变或疗效观察的有用指标。

　　【关键词】 血浆凝血酶调节蛋白 疾病严重程度/并发症 微血管病变 大血管病变

　　Thrombomodulin (TM), a thrombinbinding glycoprotein expressed on the endothelial cell surfaces in various tissues, is involved in negative regulation of coagulation through the activation of protein C[1]. The soluble form of TM, which was arised by proteolytic cleavage from membrane TM on endothelial cells, can be detected in human plasma and urine. Since the plasma TM concentration is elevated in a variety of diseases accompanied by endothelial injury, soluble TM is believed to be a good indicator of endothelial damage or activation[2,3]. Plasma TM is not specific for diagnosis of single disease, but it may be helpful to demonstrate the presence of microangiopathy and to assess the severity or complication of diseases.

　　Materials and Methods

　　Patients and healthy controls

　　979 patients admitted to the second hospital of Zhejiang University from 1996 to 2004 were included in this study. The patients were 542 men and 437 women with mean age of 47.6±18.5 years old (range, 15 to 91 years). Sixty healthy volunteers with mean age of 38.6±16.2 years old (range, 12 to 77 years) were included in control group. The sex and age of the healthy volunteers, the characteristics of the patients including disease stage, complications and severity are summarized in table 1 and 2.

　　Methods

　　The levels of plasma TM were measured by enzymelinked immunosorbent assay (ELISA) kit according to manufacture′ s instruction (STAGO, France). Venous blood was collected into 3.2% sodium citrate (9 parts of blood∶ 1 part of anticoagulant) using a 10ml syringe with a 21gauge needle. Plasma was obtained by centrifuging the blood at 1 500× g for 30 minutes. The plasma samples were frozen and stored at -70℃ until the assay.

　　Statistical analysis

　　The results were expressed as mean±SD, and POMS statistical package was used for multivariate analysis, student′s t test and correlation analysis. Statistical significance was defined as P value<0.05 for a twotailed test.

　　Results

　　PTM level in control group

　　As shown in Table 1, the PTM level in healthy control group was 20.40±7.72 μg/L, no significant difference was observed between different age and sex. Here we defined PTM level higher than its normal upper limit (1.96 ±SD, i.e. 35.54 μg/L) as PTM positive criterion, and the positive rate of the control group was less than 5%.

　　Table 1. PTM levels of the different sex and age in healthy controls(±SD)

　　nPTM(μg/L) female2521.16±8.33male3519.86±7.33≤25 years old1819.50±8.2926-45 years old2020.35±7.7846-77 years old2221.41±7.60

　　PTM levels in relation with the severity and complications of diseases

　　Cardiocerebral vascular disease The diagnosis was made based on the criteria proposed at the 4th National Conference on Cerebrovascular Disease. Cerebral infarction within 3 days showed significant higher PTM level than the transient ischemic attack (P<0.01), but only 32.1% of the PTM were positive, similar to those of cerebral hemorrhage (33.9%) and coronary heart disease (12.9%). In patients with essential hypertension, PTM level is correlated with hypertension stages (Table 2). Among 137 patients suffering from stroke, 27 were accompanied by diabetes, 28 by hypertension or coronary heart disease, 6 by renal diseases and 76 without underlying diseases, PTM levels of these patients were 31.37±12.06 μg/L, 32.47±12.08 μg/L, 60.00±23.52 μg/L and 31.51±14.13 μg/L respectively. Patients with renal diseases had significantly higher PTM than other patients(P<0.01). No correlation was found between PTM and blood glucose (r=0.0360), cholesterol (r=-0.0587) and triglyceride (r=0.0542) among 280 cardiocerebral vascular diseases or diabetes. PTM was not correlated with fibrinogen (r=0.087) and fibrin degradation product (FDP) (r=0.026) in 81 cases (P>0.05). 26 patients with cerebral infarction and 22 patients with cerebral hemorrhage for at least 14±1 days had been followed up PTM dropped when the disease was remitted (cerebral infarction: 31.67±11.50 μg/L vs 23.08±9.37 μg/L, P<0.01, and cerebral hemorrhage: 33.61±7.24 μg/L vs 27.94±5.83 μg/L, P =0.02).

　　Primary chronic glomerular disease The diagnosis of primary chronic glomerular disease was confirmed by clinical and/or pathological findings. Patients with chronic renal failure (CRF) had significantly higher PTM level than those without CRF (Table 2). 100% and 55.9% of their PTM were positive respectively. PTM over 70 μg/L would be considered as a CRF risk factor with sensitivity 85.7%, specificity 82% and positive predictive value 77.8%. PTM was found to positively correlated with blood urea nitrogen (BUN) (r=0.5926) and serum creatinine (Scr) (r=0.4781) in 105 patients, blood (2microglobulin ((2MG) (r=0.5559) and urine (2MG (r=0.4409) in 62 patients. There was negative correlation between PTM and creatine clearance rate (Ccr) in 64 patients (r=-0.5844, P<0.01).

　　Infections PTM level of patients with septicemia was higher than that of patients without septicemia (Table 2), PTM>50 μg/L suggested presence of septicemia with sensitivity 86.6%, specificity 89.5%, and positive predictive value 76.5%. There was no correlation between PTM level and leukocyte or neutrophil count (r=0.2543 and 0.2333, P>0.05).

　　Multiple trauma PTM level of patients with multiple trauma was higher than that of healthy control. These patients with multiple organ failure (MOF) had significantly higher PTM level than those without MOF (Table 2). It was found in 36 patients followedup that high PTM level dropped when remission was achieved in 20 patients (23.63±7.20 μg/L vs 34.64±10.10 μg/L, P<0.01),but it rose higher when disease got worse in 11 patients (54.83±21.20 μg/L vs 40.64±15.90 μg/L,P<0.01).It was extremely high in 5 patients with disseminated intravascular coagulation (DIC) (126.25±62.06 μg/L).

　　Systemic lupus erythematosus (SLE) The diagnosis was made according to the criteria (1982 edition) of American Rheumatism Association among 60 patients with primary or relapsed SLE. The patients with albuminuria had obviously been higher PTM than those without albuminuria (Table 2). PTM positive indicated renal insufficiency with sensitivity 77.8%,

　　Table 2. PTM levels of the severity, complication and stage in different diseases(±SD)

　　Group nPTM(μg/L) Healthy control6020.40±7.72 Macroangiopathy coronary heart disease3128.44±11.49** cerebral infarction7829.44±14.17* cerebral hemorrhage5934.02±12.19* essential hypertension7432.91±18.99** stageⅢ2943.00±25.69*Δ stageⅡ2429.41±8.33** stageⅠ2122.95±6.78 transient ischemic attacks1323.23±7.67Microangiopathy Primary chronic glomerular disease11781.54±56.54* with renal failure49124.10±61.47*ΔΔ without renal failure6851.13±35.68* Multiple trauma4041.30±16.30* multiple organ failure1848.70±18.71*# without multiple organ failure2236.33±11.32* Infection septicemia1579.93±29.17*## nonsepticemia3835.52±14.61* Systemic lupus erythematosus62 41.77±34.04* with albuminuria3655.81±38.49*× without albuminuria2622.26±8.55Other diseases Diabetes mellitus14633.25±12.43* without complication3522.11±6.58 with complication11138.75±11.72*×× microangiopathy66 43.10±13.28*§ macroangiopathy4531.89±8.25** Cancer18833.47±14.25* metastasis86 41.68±16.96*§§ nonmetastasis10226.54±11.14* Hematologic malignancies acute leukemia(AL)6535.78±17.21*※ multiple myeloma(MM)3536.14±10.05*※ with renal failure18105.33±51.35*

　　*P<0.01，**P<0.05 vs healthy controls;ΔP<0.01 vs stage II and I;ΔΔP<0.01 vs without renal failure;# P<0.01 vs without multiple organ failure; ## P<0.01 vs nonsepticemia;×P<0.01 vs without albuminuria;××P<0.01 vs without complication;§P<0.01 vs macroangiopathy;§§P<0.01 vs nonmetastasis;※P<0.01 vs with renal failure

　　specificity 92.3% and positive predictive value 93.3%. BUN, Scr and β2MG were positively correlated with PTM levels (r=0.4913, 0.6368, 0.8641, 0.8251, P<0.01) .

　　Diabetes The diagnosis was made based on 1997 ADA criteria. 146 diabetic patients were divided into groups with or without complications. The group with complications showed higher PTM levels than that without complications (Table 2). PTM levels of DM with microangiopathy (renal 44 and retinal 22) were higher than those of macroangiopathy including 24 strokes and 21 coronary heart diseases. PTM positive would predict the presence of complication and microangiopathy with sensitivity 53.4% and 71.2%, specificity 97.1% and 97.1%, and positive prediction 98.6% and 97.9% respectively.

　　Tumors Acute leukemia and multiple myeloma had higher PTM levels, and PTM levels were extremely high when renal failure developed (Table 2), PTM positive rate were 44.6%, 54.3% and 100% respectively, and we also found a positive correlation between PTM level and marrow blast or plasma cells (r=0.4051 and 0.4845, P<0.05). PTM levels were followed up in AL patients undergoing chemotherapy, and were fluctuated with remission in 27 patients (21.11±10.04 μg/L vs 35.78±17.21 μg/L, P<0.01) or nonremission in 15 patients (40.53±17.14 μg/L) of the primary disease, PTM rose 200% times higher when DIC developed (108.50±60.74 μg/L, P<0.01). In MM, PTM dropped when remission was achieved in 5 patients (24.20±7.85 vs 38.10±8.51 μg/L), and it was still high in 10 patients with nonremission (39.40± 8.17 μg/L). The metastatic cancers had higher PTM than the nonmetastaic cancers. PTM level >35.54 μg/L would indicate cancer metastasis with sensitivity 71.6%, specificity 66.7% and positive predictive value 72.3%. In 33 cancer patients, PTM level dropped in 24 patients who were better 14 days after surgery (18.45±9.96 μg/L vs 28.29±11.74 μg/L, P<0.01), and rose again in case cancer relapsed in half a year (34.50±12.57 μg/L).

　　PTM levels in microangiopathy and macroangiopathy

　　The macroangiopathy including cardiovascular and cerebrovascular diseases showed mild increase of PTM, the average PTM level of macroangiopathy was less than 35 μg/L. However, in diseases characterized by microangiopathy such as chronic glomerular disease, systemic lupus erythematosus (SLE) with albuminuria, septicemia and multiple organ failure (MOF), disseminated intravascular coagulation (DIC) and diabetic microangiopathy, PTM level moderately increased with an average level higher than 40 μg/L. There was significant difference between micro and macroangiopathy (Table 2). If PTM level was defined higher than the normal upper limit (35.54 μg/L) in microangiopathy, the sensitivity was 77.9%, the specificity of 71.4% and the positive predictive value of 75.6%.

　　Discussion

　　Previous studies showed that there was PTM level variation among population of different ages and sex[2], but the difference was not demorstrated in our study. PTM is regarded as a reliable marker for endothelial injury, but little is known about its clinical significance. We studied various diseases and found that the evelation of PTM was well correlated with the disease severity or the incidence of complications. In terms of predicting sensitivity, specificity and positive predictive value, PTM in CRF, SLE with renal damage and septicemia was high, but PTM in DM with complication showed less sensitivity. Uehara et al[4] consider that a fluctuation in the blood glucose level is significantly related to vascular endothelial cells damage in DM. Therefore, PTM can be considered as a promising marker to evaluate disease progression or complications. However, different PTM limits were used for different diseases; usually PTM exceeding its normal upper limit (35.54 μg/L) would indicate endothelial damage. In addition, PTM over 40 μg/L may imply disease progression or complications, and PTM over 70 μg/L indicates the presence of critical conditions such as CRF, DIC and septicemia. The PTM levels were follwedup in some diseases and found that high PTM would drop at remission; or it would rise again when relapsed. It was regarded that the PTM might be a useful index for reflecting the severity and evaluating the curative effect in different diseases. Some reported that TM was also involved in development of DM nephropathy[5]; PTM could be better than the other makers in indicating diabetes albuminuria[6]; PTM level reflected activity of rheumatiod arthritis and could be helpful in monitoring disease status and response to therapy[7]. PTM obviously dropped and the survival periods were prolonged after therapy in mouses with sepsis[8]. High level of PTM indicated high death rate in patients with severe acute respiratory syndrome (SARS)[9]. Dynamic monitoring of plasma TM would be useful in reflecting the seveity and evaluating therapeutic response of thrombotic thrombocytopenic purpura (TTP).

　　PTM is well correlated with renal function, both primary and secondary CRF or nephropathy show significant PTM increase, elevated PTM may be a response to both glomerular capillary injury and azotemia. Patients with cerebrovascular disease are often accompanied by diabetes, hypertension, coronary heart disease. Our study demonstrated that PTM of cerebrovascular disease was not affected by occurrence of DM, hypertension or coronary heart diseases, and was different from these three disease. Neutrophils and their products may interact with endothelium and lead to PTM release[10], no association was found between the white blood cell or neutrophils and PTM in patients with infections. While in patients with AL and MM, positive correlation was demonstrated between PTM with blast or plasma cells in bone marrow, indicating that infiltration by blast cells may cause endothelium damage and result in PTM increase. PTM level was found nothing to do with blood fattiness and glucose in patients with cardiocerebral vascular diseases, but it was reported that smoking and obesity would affect PTM levels[11]. The sensitivity of PTM to endothelial injury caused by hypoxaemia was lower than von Willebrand factor and tissue plasminigen activator[12].

　　The relationship between PTM and vascular damage is not well understood. In our study, the macroangiopathyinvolved diseases such as coronary heart disease, cerebral infarction and cerebral hemorrhage showed mild increase of PTM, in those microangiopathyinvolved diseases including primary chronic glomerular disease, SLE with albuminuria, diabetic microangiopathay, MOF and DIC, PTM increased significantly, suggesting that PTM is a good index in evaluating sensitivity, specificity and positive predictive value when used for assessment of microangiopathy. It is proposed that PTM over 40 μg/L as a risk factor of microangiopathy.

　　【参考文献】

　　1 Liaw PC. Endogenous protein C activation in patients with severe sepsis. Crit Care Med, 2004; 32: S214-S218

　　2Blann AD, Daly RJ, Amiral J. The influence of age, gender and ABO blood group on soluble endothelial cell markers and adhesion molecules. Br J Haematol，1996; 92: 498-500

　　3Yokota H, Naoe Y, Nakabayashi M, et al. Cerebral endothelial injury in severe head injury: the significance of measurements of serum thrombomodulin and the von Willebrand factor. J Neurotrauma, 2002;19: 1007-1015

　　4Uehara S,Gotoho K, Handa H. separation and characterzation of the molecular species of thrombomodulin in the plasma of diabetic patients. Thromb Res， 2001;104:325-332

　　5丁新，盛宏光，龚家雷.血栓调节蛋白在糖尿病肾病中的关系探讨. 中华新医学，2004，5：1171-1172

　　6Hirano T, Ookubo K, Kashiwazaki K, et al. Vascular endothelial markers, von Willebrand factor and thrombomodulin index, are specififically elevated in type 2 diabetic patients with nephropathy : comparision of primary renal disease. Clin Chim Acta，2000;299:65-75

　　7Cobankara U, Ozatli D, Kiraz S, et al. Successful treatment of rheumatiod arthritis is associated with a reduction in serum sEselectin and thrombomodulin level. Clin rheumatol，2004; 23:430-434

　　8袁修学.葛根素对脓毒症大鼠凝血分子标志物水平的影响. 中国病理生理杂志，2005;21：1662

　　9Liu ZH, Wei R, Wu YP, et al. Elevated plasma tissuetype plasminogen activator(tPA) and soluble thrombomodulin in patients suffering from severe acute respiratory syndrome (SARS) as a possible index for prognosis and treatment strategy. Biomed Environ Sci, 2005; 18: 260-264

　　10Macgregor IR, Perrie AM, Donnelly SC, et al. Modulation of human endothelial thrombomodulin by neutrophils and their release products. Am J Respir Crit Care Med，1997;155:47-52

　　11Rosc D, Drewniak W, KinaszRozycka I, et al. Thrombomodulin, von Willebrand factor and tissue plasminogen activator in the blood plasma of obese women and men. Pol Merkuriusz, 2003; 15: 518-520

　　12de PS Soares R, Maeda NY, Bydlowski SP, et al.Markers of endothelial dysfunction and severity of hypoxaemia in the Eisenmenger syndrome. Cardiol Young, 2005; 15: 504-513Figure 1.