Association of Malnutrition-Inflammation Score with quality of life and mortality in hemodialysis patients: a 5-year prospective cohort study

Abstract

Background: The Malnutrition-Inflammation Score (MIS), an inexpensive and easy-to-assess score of 0 to 30 to examine protein-energy wasting (PEW) and inflammation, includes 7 components of the Subjective Global Assessment, body mass index, and serum albumin and transferrin concentrations. We hypothesized that MIS risk stratification of hemodialysis (HD) patients in predicting outcomes is better than its components or laboratory markers of inflammation.

Study design: 5-Year cohort study.

Setting & participants: We examined 809 stable HD outpatients and followed them for up to 5 years (October 2001 to December 2006).

Predictors: MIS and other nutritional and inflammatory markers.

Outcomes & measurements: Prospective all-cause mortality, health-related quality of life using the 36-Item Short Form Health Survey (SF-36), and tests of body composition.

Results: The MIS correlated with logarithm of serum interleukin 6 level (r = +0.26; P < 0.001), logarithm of C-reactive protein level (r = +0.16; P < 0.001), and several measures of nutritional status. Patients with a higher MIS had lower SF-36 scores. After multivariate adjustment for case-mix and other measures of PEW, HD patients in the second (3 to 4), third (5 to 7), and fourth (>or=8) quartiles of MIS had worse survival rates than those in the first (0 to 2) quartile (P < 0.001). Each 2-unit increase in MIS was associated with a 2-fold greater death risk, ie, adjusted death hazard ratio of 2.03 (95% confidence interval, 1.76 to 2.33; P < 0.001). Cubic spline survival models confirmed linear trends. Adding MIS to the constellation of age, sex, race/ethnicity, and vintage significantly improved the area under the receiver operating characteristic curve developed for predicting mortality (0.71 versus 0.67; P < 0.001).

Limitations: Selection bias and unknown confounders.

Conclusions: In HD patients, the MIS is associated with inflammation, nutritional status, quality of life, and 5-year prospective mortality. The mortality predictability of the MIS appears equal to serum interleukin 6 and somewhat greater than C-reactive protein levels. Controlled trials are warranted to examine whether interventions to improve the MIS can also improve clinical outcomes in HD patients.

Figure 1

Distribution of the baseline malnutrition-inflammation score in 809 maintenance hemodialysis patients

Figure 2. Standardized SF-36 quality of life scores in the quartiles of malnutrition-inflammation score in 688 maintenance hemodialysis patients

* P<0.001; † 0.001≤P<0.05

Figure 3. Mortality predictability of malnutrition-inflammation score in 809 maintenance hemodialysis patients (Oct 2001 – Jan 2007)

Panel A. Unadjusted Panel B. adjusted for Case-mix variables Panel C. adjusted for Case-mix and MICS variables Panel D. adjusted for case-mix, MICS, and inflammation Case-mix variables: age, gender, race/ethnicity, diabetes mellitus, log vintage, primary insurance, marital status, dialysis dose (Kt/V), and kidney residual urine (KRU) MICS variables: erythropoietin dose, creatinine, hemoglobin, phosphorus, normalized protein catabolic rate (nPCR), bicarbonate, calcium, ferritin, WBC, lymphocyte percentage, and vitamin D dose Inflammatory variables: C-reactive protein, Interleukin-6, Tumor necrosis factor-α

Figure 4. Mortality predictability of quartiles of baseline malnutrition-inflammation score (MIS) in 809 maintenance hemodialysis patients (Oct 2001 – Jan 2007)

Case-mix variables: age, gender, race/ethnicity, diabetes mellitus, log vintage, primary insurance, marital status, dialysis dose (Kt/V), and kidney residual urine (KRU) MICS variables: erythropoietin dose, creatinine, hemoglobin, phosphorus, normalized protein catabolic rate (nPCR), bicarbonate, calcium, ferritin, WBC, lymphocyte percentage, and vitamin D dose Inflammatory variables: Log C-reactive protein, Log Interleukin-6, Log Tumor necrosis factor-α Quartile 1: 0 to 2; Quartile 2: 3 & 4; Quartile 3: 5 to 7; Quartile 4: ≥8

Figure 5

Kaplan-Meier proportion of surviving after 5 years of observation according to the quartiles of baseline malnutrition-inflammation score (MIS) in 809 HD patients (Oct 2001 – Jan 2007)

Figure 6. Receiver operating characteristic (ROC) curves of probabilities obtained from hazard regression models including (right) C-reactive protein (CRP), (middle) Interleukin-6 (IL-6), and (left) malnutrition-inflammation score (MIS) as independent variables and all-cause mortality as dependent (reference) variable

Footnote: values in parentheses are 95% confidence interval of the calculated area under the ROC curves.

Figure 6. Receiver operating characteristic (ROC) curves of probabilities obtained from hazard regression models including (right) C-reactive protein (CRP), (middle) Interleukin-6 (IL-6), and (left) malnutrition-inflammation score (MIS) as independent variables and all-cause mortality as dependent (reference) variable

Footnote: values in parentheses are 95% confidence interval of the calculated area under the ROC curves.