Neutrophil gelatinase-associated lipocalin 作為貓隻腎臟疾病的生物標記

Abstract

Neutrophil gelatinase-associated lipocalin (NGAL) 為一個具有潛力的腎臟標誌物，在人醫和獸醫的研究中指出，NGAL不僅可早期預測急性腎損傷（acute kidney injury ,AKI），還可區別急性腎損傷和慢性腎臟病（chronic kidney disease ,CKD），以及作為腎臟疾病的預後評估。而NGAL在貓腎臟疾病中的應用則尚未有研究報告。本研究的目的是要確定貓隻NGAL和不同腎臟疾病之間的關連性。 從2014年9月至2016年3月，總共92隻氮血症貓隻納入實驗。根據獸醫國際腎臟醫學會（International Renal Interest Society, IRIS）分期系統，所有貓隻被分為五組:（1）健康的貓，（2）慢性腎病貓第二級，（3）慢性腎病貓第三級，（4）慢性腎病貓第四級（5）急性腎損傷貓隻第二到五級。就診時收集貓尿液和血液樣本，由實驗室建立的三明治夾心ELISA系統進行NGAL濃度測試，同時計算尿液NGAL和尿中肌酸酐比值（Urinary NGAL to creatinine ratio, UNCR），及記錄每個病患的血液和血清生化數值。 根據統計結果，氮血症貓的尿液NGAL（uNGAL）和UNCR皆顯著高於健康貓(uNGAL：氮血症貓隻Median [IQR] 1.8 [3.48] ng/mL vs. 健康貓隻0.95 [1.35]ng/mL; UNCR：氮血症貓隻Median [IQR] 2.58 [8.23]% vs. 健康貓隻0.23[0.59]%)，而兩組在血漿NGAL（pNGAL）的濃度間則無顯著差異（p=0.113）。不同於人類及犬隻之前的研究，貓的pNGAL無法預測AKI的預後，但uNGAL和UNCR在AKI和對照組之間有顯著的差異。在CKD組，IRIS分級越高的組別有越高的uNGAL濃度和UNCR值。對於所有的患貓，UNCR比值與uNGAL濃度，血漿肌酸酐，血容比，嗜中性球數和磷離子的濃度密切相關。然而，uNGAL濃度和UNCR與pNGAL濃度則沒有相關性。 根據ROC曲線（Receiving operative charaterstic, ROC）分析，uNGAL和UNCR可應用於預測CKD貓隻的臨床進展。uNGAL的ROC曲面下面積（Area under ROC, AUROC）為0.71,當最佳臨界點設在2.06 ng/mL時，有75％的靈敏度和76.9 ％的特異性。而UNCR的AUROC為0.78, 最佳臨界點設在4.08％時，有79.2%的靈敏度和76.9％的特異性。超過這些臨界點代表病貓在19-20天內病程有顯著的惡化。 根據實驗結果，uNGAL和UNCR在貓的腎臟疾病可作為一個有用的生物標誌物。此外，在CKD的病貓，較高濃度的uNGAL和UNCR，代表有較差的臨床進展。

Neutrophil gelatinase-associated lipocalin (NGAL) is emerging as a potential biomarker for early prediction of acute kidney injury (AKI), differentiation of AKI from chronic kidney disease (CKD), and the prognosis of AKI and CKD in human and canine medical studies. Nevertheless, the role of NGAL in feline kidney disease has never been studied. The purpose of this study was to determine the relationship between NGAL and different renal diseases in cats. From Sep 2014 to Mar 2016, 92 cats with azotemia were enrolled. According to the International Renal Interest Society (IRIS) staging system, the cats were categorized into five groups, (1) healthy cats, (2) cats with IRIS CKD stage II, (3) cats with IRIS CKD stage III, (4) cats with IRIS CKD stage IV, and (5) cats with AKI stages II-V. Urine and plasma samples of cats were collected at admission and tested for NGAL concentrations by our in-house sandwich ELISA system. The urinary NGAL (uNGAL) to creatinine ratio (UNCR) was also calculated. Simultaneously, hematologic and serum biochemical data of each case were also recorded. Statistically the uNGAL and UNCR of azotemic cats are significantly higher than those of healthy cats (i.e., uNGAL：Median [IQR] 1.8 [3.48] ng/mL in azotemic cats vs. 0.95 [1.35] ng/mL in healthy cats; UNCR：Median [IQR] 2.58 [8.23]% in azotemic cats vs. 0.23 [0.59]% in healthy cats), while plasma NGAL(pNGAL) had no statistical significance (p=0.113) between two groups. Unlike previous studies that increased pNGAL was associated with AKI in human medicine, the present findings indicated that only uNGAL and UNCR had significantly difference between AKI and healthy cats. Among CKD groups, cats with CKD grade 3, 4 had higher uNGAL and UNCR compared with cats with CKD grade 2 and healthy cats. For all patients, UNCR value was significantly correlated with uNGAL, plasma creatinine, hematocrit, segment and phosphorus concentrations. However, uNGAL concentration and UNCR were not associated with pNGAL concentration for all cases. According to the Receiving operative charaterstic (ROC) analysis, uNGAL and UNCR are useful to predict clinical progression for cats with CKD. Area under ROC (AUROC) for uNGAL in predicting the progression of CKD was 0.71 and the best cutoff value of 2.06 ng/mL had a sensitivity of 76.9% and specificity of 75％, whereas AUROC for UNCR in predicting the progression of CKD was 0.79 and the best cutoff value of 4.08% had a sensitivity of 76.9% and specificity of 79.2%. Above these values, patients experienced significantly faster deterioration within 19-20 days. In conclusion, uNGAL and UNCR could serve as useful biomarkers in cats with renal diseases. Furthermore, the CKD cats with higher concentration of uNGAL and UNCR tended to have poor clinical progression.