10.3389/fpsyg.2012.00132

Vollständiger Abstract

Worum geht es in dieser Arbeit?

<h4>Background</h4>The incidence of malnutrition is relatively high among patients with chronic obstructive pulmonary disease (COPD). The main objective of this study was to explore and compare the discriminative ability of phase angle (PhA) and geriatric nutritional risk index (GNRI) tools for assessing malnutrition among hospitalized patients with COPD who underwent BIA assessment and had available serum albumin data based on the GLIM criteria.<h4>Methods</h4>This single-center retrospective study included 121 patients with hospitalized COPD with BIA measurement and valid serum albumin. All patients underwent body composition analysis and biochemical measurements. Based on the global leadership initiative on malnutrition (GLIM) criteria, patients were categorized into COPD patients with malnutrition and COPD patients without malnutrition. Bootstrap resampling with 1,000 iterations and Hosmer-Lemeshow test evaluate the stability and reliability of the two tools and malnutrition in COPD. More importantly, the exploratory diagnostic efficacies and net clinical benefit of PhA and GNRI tools were assessed for identifying COPD patients with malnutrition using ROC, DeLong test, Decision-curve analysis (DCA).<h4>Results</h4>Using the GLIM malnutrition criteria as the reference, a diagnostic efficacy analysis was conducted using the preset cutoff values of PhA = 4.25° and GNRI = 98. The sensitivity of PhA was 0.691, the specificity was 0.625, the accuracy was 0.669, and the K value was 0.298. The sensitivity, specificity, accuracy and K value were 0.753, 0.700, 0.736 and 0.431 for GNRI. The ROC results showed that the AUC of PhA and GNRI were 0.656 (95% CI: 0.543 ~ 0.769) and 0.785 (95% CI: 0.699 ~ 0.870), respectively. The DeLong test indicated that there was a statistically significant difference in the AUC between the two tools (<i>p</i> = 0.035). DCA showed that net clinical benefits of the two tools were similar at low thresholds. As the threshold increased, the net clinical benefit of PhA decreased rapidly and turned negative, while GNRI maintained a positive net clinical benefit in a wider threshold range.<h4>Conclusion</h4>GNRI and PhA showed differing discriminative profiles for GLIM-defined malnutrition. The valid threshold range differed between the two tools. GNRI yielded net clinical benefit over a wider threshold span, while positive net clinical benefit for PhA was restricted to low-risk thresholds.

Abstract: PubMed · Datensatz