癌因性疲乏气血两虚证患者聚类与富集探索性研究

Exploratory Study on Population Enrichment and Classification in Cancer-related Fatigue Patients with Qi and Blood Deficiency Syndrome through Cluster Analysis

目的客观划分癌因性疲乏(CRF)气血两虚证人群,探索其潜在的人群富集特征,为实现人群精准辨识提供思路。方法采用横断面研究,于2019年10月-2022年4月纳入CRF气血两虚证患者611例。采集患者临床症状(是否面色淡白或萎黄、少气懒言、头晕、自汗、心悸)及血常规[白细胞(WBC)、红细胞(RBC)、血红蛋白(HGB)、血小板(PLT)、中性粒细胞绝对值(NEUT)],并通过Piper疲乏调查量表(PFS)进行疲乏评分。采用R4.2.0进行数据处理与统计分析,借助K-mediods方法进行聚类分析。结果通过无监督划分聚类分析将目标人群分为气血两虚证气虚偏向、气血两虚证轻症、气血两虚证血虚偏向3类。疲乏程度上,前两类人群疲乏程度相似,血虚偏向人群疲乏程度最重;客观指标上,气虚偏向人群WBC、NEUT、PLT水平最低,血虚偏向人群HGB、RBC水平最低,气血两虚证轻症人群血常规各项指标水平均较高。结论通过聚类分析方法,初步在证候基础上实现人群富集分类,为后续的信息富集及人群精准辨识提供了思路。同时贫血指标和WBC、NEUT分别对血虚偏向和气虚偏向呈现出一定指向性,或可加入CRF气血两虚证诊断标准,辅助CRF人群的划分。

Objective To objectively classify cancer-related fatigue(CRF)patients with qi and blood deficiency syndrome;To explore possible enrichment characteristics;To provide ideas for precise population identification.Methods A cross-sectional study was conducted,611 CRF patients with qi and blood deficiency syndrome were included from October 2019 to April 2022.Clinical symptoms of patient were documented,including pale or withered-yellow complexion,shortness of breath with reluctance to talk,dizziness,spontaneous sweating,and palpitations.Additionally,blood routine were conducted,including white blood cell(WBC)count,red blood cell(RBC)count,hemoglobin(HGB),platelet(PLT)count,and neutrophil count(NEUT).The degree of fatigue was assessed using the Piper Fatigue Scale(PFS).Data analysis was performed using R 4.2.0.Clustering analysis was performed using the K-mediods method.Results Through unsupervised clustering analysis,the target population was divided into three categories:qi and blood deficiency syndrome leaning towards qi deficiency,qi and blood deficiency syndrome with mild symptom,qi and blood deficiency syndrome leaning towards blood deficiency.In terms of fatigue severity,the first two groups showed comparable levels,while the third group exhibited the most severe fatigue.Regarding objective indicators,the group leaning towards qi deficiency had the lowest WBC count,NEUT count and PLT count,and the group leaning towards blood deficiency had the lowest HGB count and RBC count.The group of qi and blood deficiency syndrome with mild symptom exhibited the highest level of objective indicators among the three.Conclusion By using clustering analysis methods,the study has preliminarily achieved population enrichment and classification based on syndromes,providing ideas for subsequent information enrichment and accurate identification of populations.Concurrently,the anemia index,WBC count and NEUT count demonstrate a inclination towards blood deficiency syndrome and qi deficiency syndrome,respectively.These findings indicate that these indicators could be incorporated into the syndrome diagnostic criteria for CRF to aid in the classification of the CRF population.