基于血常规炎性指标构建衰弱/衰弱前期发生风险列线图模型研究

BackgroundFrailty is a common geriatric syndrome linked to negative clinical outcomes. Current assessments of frailty predominantly depend on various scaleslacking a standardized gold standard. Chronic inflammation is a key pathophysiological mechanism of frailtyand examination of inflammatory markers in routine blood tests is easy and simple-to#2;use. Neverthelessthe correlation between these inflammatory markers and frailty has not been fully elucidated. ObjectiveTo explore the correlation between inflammatory markers in routine blood tests and frailty in the elderlyanalyze the influencingfactors of frailty and construct a risk prediction model for the risk of frailty or pre-frailty. MethodsElderly individuals receiving physical examinations in the Healthcare CenterBeijing Friendship HospitalCapital Medical University from August 2020 to September 2022 were recruited. Baseline characteristics and laboratory test results were collected. The frailty was assessed using the Simple Frailty QuestionnaireFRAIL. Univariate and multivariate Logistic regression analyses were applied to identify risk factors for frailty/pre-frailty in the elderly. A nomogram was then createdfollowed by an internal validation of its performance via Bootstrap. Finallythe receiver operating characteristicROCcurvescalibration curve and decision curve analysisDCA were used to evaluate the identification abilityaccuracy and clinical applicability of the nomogram. ResultsA total of 554 elderly individuals were included in the studyof whom 21338.4% were identified as frail or pre-frail. Multivariate logistic regression analysis showed that age-adjusted Charlson Comorbidity IndexACCIOR=1.4295%CI=1.21-1.66Mini Nutritional Assessment Short-FormMNA-SFOR=0.7195%CI=0.61-0.83hemoglobin/red cell distribution width ratioHRROR=0.4495%CI=0.23-0.86and medication of multiple drugsOR=0.5495%CI=0.36-0.81 were independent influencing factors of frailty or pre-frailty in the elderlyP<0.05. The predictive nomogram was established by employing the above-mentioned variables. The area under the curveAUC of the nomogram for identifying frailty or pre-frailty in the elderly was 0.71995%CI=0.675-0.764. The nomogram had a high goodness-of-fit after internal validation using the Bootstrap resampling method. The nomogram was found with a high goodness-of-fit by the Hosmer-Lemeshow testP>0.05. DCA showed that when the threshold probability of patients ranging from 0.15 to 0.95the nomogram resulted in higher net benefit of predicting the risk of frailty or pre-frailty. ConclusionAgecomorbiditiesmedication of multiple drugsmalnutritionand HRR are influencing factors of frailty or pre-frailty in the elderly. The constructed predictive nomogram shows strong discriminationconsistencyand clinical utilityoffering valuable guidance for the early screening of frailty or pre-frailty.