Luo Wen, Wang Yi, Zhou Zhiyong, Li Hongying
Department of Preventive health care, the Ninth People's Hospital of Chongqing, Chongqing 400700, China.
Department of General Surgery, the Ninth People's Hospital of Chongqing, Chongqing 400700, China.
Zhonghua Wei Chang Wai Ke Za Zhi. 2016 May;19(5):575-9.
To evaluate the clinical effectiveness of Onodera prognostic nutrition index (OPNI) in the predictive value of nutrition risk.
In a prospective cohort study from July 2014 to June 2015 in the Department of General Surgery of the Ninth People's Hospital of Chongqing, NRS2002 and OPNI were conducted in 200 patients undergoing gastrointestinal elective operation. OPNI was calculated with serum albumin (Alb) and peripheral lymphocyte (TLC) [OPNI=Alb(10(9)/L)+5×TLC(10(9)/L)]. By using the results of NRS2002 as the golden standard for diagnosis of nutrition risk (A NRS2002 score≥3 was deemed as nutritional risk and a nutritional care plan should be initiated. A NRS2002 score <3 was deemed as no nutritional risk), the effectiveness of OPNI was evaluated by the receiver operator characteristic(ROC) curve. The sensitivity, specificity, positive and negative predictive values, Youden indexes and area under ROC curve(AUC) of different diagnostic cut-off points of OPNI were analyzed to determine the optimal operating point (OOP). Kappa test was used to estimate the consistency of different cut-off points for OPNI with NRS2002 in defining nutrition risk.
A total of 103 patients were of NRS2002 ≥3 group, and 97 of NRS2002 <3 group. The overall OPNI was 45.4±7.4. When OOP was 45.8, the AUC of OPNI was 0.914 (95% CI: 0.873 to 0.954); the sensitivity, specificity, Youden indexes were 85.4%, 85.6%, 0.711; the positive predictive value and negative predictive value were 85.3% and 83.7%, respectively. According to this OOP, the subjects were divided into the OPNI ≥45.8 group(n=102) and OPNI <45.8 group (n=98). Compared with OPNI ≥45.8 group, OPNI <45.8 group were older [(66.5±12.1)years vs. (57.0±15.3) years, t=-4.905, P=0.000], and had lower BMI[(20.4±3.0) kg/m(2) vs. (21.7±3.0) kg/m(2), t=3.069, P=0.002], lower albumin[(34.7±4.7)10(9)/L vs.(43.6±3.4)10(9)/L, t=15.542, P=0.000] and lower TLC[(1.0±0.5)10(9)/L vs.(1.6±0.7)10(9)/L, t=7.254, P=0.000], respectively. Kappa test indicated that when using OPNI=45.8, the diagnostic value of OPNI on nutrition risk was consistence with NRS2002(Kappa=0.691, P=0.000).
OPNI can be used as a relatively simple and reliable method for clinical screening and assessment of nutrition risk.
评估小野寺预后营养指数（OPNI）在营养风险预测中的临床有效性。
在2014年7月至2015年6月重庆第九人民医院普通外科进行的一项前瞻性队列研究中，对200例行胃肠道择期手术的患者进行了NRS2002和OPNI评估。OPNI通过血清白蛋白（Alb）和外周血淋巴细胞（TLC）计算得出[OPNI = Alb（g/L）+ 5×TLC（×10⁹/L）]。以NRS2002结果作为诊断营养风险的金标准（NRS2002评分≥3被视为存在营养风险，应启动营养护理计划；NRS2002评分<3被视为无营养风险），通过受试者工作特征（ROC）曲线评估OPNI的有效性。分析OPNI不同诊断切点的敏感性、特异性、阳性和阴性预测值、约登指数及ROC曲线下面积（AUC），以确定最佳临界点（OOP）。采用Kappa检验评估OPNI不同切点与NRS2002在定义营养风险方面的一致性。
NRS2002≥3组共103例患者，NRS2002<3组97例。总体OPNI为45.4±7.4。当OOP为45.8时，OPNI的AUC为0.914（95%CI：0.873至0.954）；敏感性、特异性、约登指数分别为85.4%、85.6%、0.711；阳性预测值和阴性预测值分别为85.3%和83.7%。根据此OOP，将受试者分为OPNI≥45.8组（n = 102）和OPNI<45.8组（n = 98）。与OPNI≥45.8组相比，OPNI<45.8组年龄更大[（66.5±12.1）岁 vs.（57.0±15.3）岁，t = -4.905，P = 0.000]，BMI更低[（20.4±3.0）kg/m² vs.（21.7±3.0）kg/m²，t = 3.069，P = 0.002]，白蛋白更低[（34.7±4.7）g/L vs.（43.6±3.4）g/L，t = 15.542，P = 0.000]，TLC更低[（1.0±0.5）×10⁹/L vs.（1.6±0.7）×10⁹/L，t = 7.254，P = 0.000]。Kappa检验表明，当OPNI = 45.8时，OPNI对营养风险的诊断价值与NRS2002一致（Kappa = 0.691，P = 0.000）。
OPNI可作为临床筛查和评估营养风险的一种相对简单可靠的方法。
