Division of Nephrology and Hemodialysis, Humanitas Clinical and Research Hospital, via Manzoni 56, 20089, Rozzano, MI, Italy.
Division of Biostatistics, Humanitas Clinical and Research Hospital, via Manzoni 56, 20089, Rozzano, MI, Italy.
J Nephrol. 2018 Oct;31(5):757-765. doi: 10.1007/s40620-018-0505-3. Epub 2018 Jun 23.
Malnutrition is an important risk factor for cardiovascular mortality in hemodialysis (HD) patients. However, current malnutrition biomarkers seem unable to accurately estimate the role of malnutrition in predicting cardiovascular risk. Our aim was to investigate the role of the Subjective Global Assessment-Dialysis Malnutrition Score (SGA-DMS) compared to two well-recognized comorbidity scores-Charlson Comorbidity Index (CCI) and modified CCI (excluding age-factor) (mCCI)-in predicting cardiovascular events in HD patients.
In 86 maintenance HD patients followed from June 2015 to June 2017, we analyzed biohumoral data and clinical scores as risk factors for cardiovascular events (acute heart failure, acute coronary syndrome and stroke). Their impact on outcome was investigated by linear regression, Cox regression models and ROC analysis.
Cardiovascular events occurred in 26/86 (30%) patients during the 2-year follow-up. Linear regression showed only age and dialysis vintage to be positively related to SGA-DMS: B 0.21 (95% CI 0.01; 0.30) p 0.05, and B 0.24 (0.09; 0.34) p 0.02, respectively, while serum albumin, normalized protein catabolic rate (nPCR) and dialysis dose (Kt/V) were negatively related to SGA-DMS: B - 1.29 (- 3.29; - 0.81) p 0.02; B - 0.08 (- 1.52; - 0.35) p 0.04 and B - 2.63 (- 5.25; - 0.22) p 0.03, respectively. At Cox regression analysis, SGA-DMS was not a risk predictor for cardiovascular events: HR 1.09 (0.9; 1.22), while both CCI and mCCI were significant predictors: HR 1.43 (1.13; 1.87) and HR 1.57 (1.20; 2.06) also in Cox adjusted models. ROC analysis reported similar AUCs for CCI and mCCI: 0.72 (0.60; 0.89) p 0.00 and 0.70 (0.58; 0.82) p 0.00, respectively, compared to SGA-DMS 0.56 (0.49; 0.72) p 0.14.
SGA-DMS is not a superior and significant prognostic tool compared to CCI and mCCI in assessing cardiovascular risk in HD patients, even it allows to appraise both malnutrition and comorbidity status.
营养不良是血液透析（HD）患者心血管死亡率的一个重要危险因素。然而，目前的营养不良生物标志物似乎无法准确估计营养不良在预测心血管风险中的作用。我们的目的是研究主观整体评估-透析营养不良评分（SGA-DMS）与两种公认的合并症评分-Charlson 合并症指数（CCI）和改良 CCI（不包括年龄因素）（mCCI）相比，在预测 HD 患者心血管事件中的作用。
在 86 名接受维持性 HD 治疗的患者中，我们分析了生物化学数据和临床评分作为心血管事件（急性心力衰竭、急性冠状动脉综合征和中风）的危险因素。通过线性回归、Cox 回归模型和 ROC 分析研究了它们对结局的影响。
在 2 年的随访中，26/86（30%）患者发生了心血管事件。线性回归显示只有年龄和透析龄与 SGA-DMS 呈正相关：B 0.21（95%CI 0.01；0.30）p 0.05，B 0.24（0.09；0.34）p 0.02，而血清白蛋白、校正蛋白分解率（nPCR）和透析剂量（Kt/V）与 SGA-DMS 呈负相关：B -1.29（-3.29；-0.81）p 0.02；B -0.08（-1.52；-0.35）p 0.04 和 B -2.63（-5.25；-0.22）p 0.03。在 Cox 回归分析中，SGA-DMS 不是心血管事件的风险预测因子：HR 1.09（0.9；1.22），而 CCI 和 mCCI 都是显著的预测因子：HR 1.43（1.13；1.87）和 HR 1.57（1.20；2.06）在 Cox 调整模型中也是如此。ROC 分析报告 CCI 和 mCCI 的 AUC 相似：0.72（0.60；0.89）p 0.00 和 0.70（0.58；0.82）p 0.00，而 SGA-DMS 为 0.56（0.49；0.72）p 0.14。
与 CCI 和 mCCI 相比，SGA-DMS 在评估 HD 患者的心血管风险方面并不是一种优越且具有显著意义的预后工具，即使它可以评估营养不良和合并症的状况。
