第一部分老年2型糖尿病患者肌少症与估计肾小球滤过率的关系

中文摘要

背景：肌少症是衰老相关性疾病，可导致老年人预后不良，如能量平衡失调、跌倒、心血管疾病和死亡率增加。肌少症患者机体能量消耗下降，内脏脂肪增加，导致机体慢性炎症反应。对于终末期肾脏病患者，肾脏替代治疗成本较高，终末期肾脏病并发症较多，大多预后不良。研究发现，终末期肾脏病患者肌少症与估算肾小球滤过率（estimated Glomerular Filtration Rate，eGFR）有关，然而肌少症与eGFR轻度下降的关系研究较少。

目的：探讨老年2型糖尿病（Type 2 diabetes mellitus，T2DM）患者肌少症与eGFR的关系。

内容与方法：选取188例老年人，采用双能X线测量骨骼肌量（appendicular skeletal muscle mass，ASM），通过ASM/ht²、ASM/wt、ASM/BMI公式分别计算骨骼肌量。根据CKD-EPI_{Cr-Cys C}公式分为两组：eGFR正常组（n = 63例，eGFR≥90ml/min·1.73m²）和eGFR轻度下降组（n = 125例，60≤eGFR＜90ml/min·1.73m²）。

结果：eGFR正常组ASM、ASM/ht²、ASM/wt、ASM/BMI均高于eGFR轻度下降组，差异有统计学意义（P ＜ 0.05）。Spearman相关分析显示，eGFR与ASM（r = 0.231，P = 0.001），ASM/ht²（r = 0.201，P = 0.006），ASM/wt（r = 0.313，P = 0.000），ASM/BMI（r = 0.293，P = 0.000），HDL（r = 0.208，P = 0.004）呈正相关；与年龄（r = -0.405，P = 0.000），体重（r = -0.240，P = 0.001），BMI（r = -0.250，P = 0.001），肌酐（r = -0.614，P = 0.000），胱抑素C（r = --0.928，P = 0.000），尿酸（r = -0.277，P = 0.000），舒张压（r = -0.254，P = 0.000），腹围（r = -0.144，P = 0.048）呈负相关。Logistic回归分析结果显示，eGFR与ASM/wt（OR = 0.763，95%CI：0.667~0.874）呈正相关，与高龄（OR = 1.331，95%CI：1.190~1.488）、高尿酸血症（OR = 1.006，95%CI：1.000~1.013）、舒张压高（OR = 1.089，95%CI：1.041~1.139）呈负相关。

结论：老年T2DM患者eGFR的下降主要与高龄、高尿酸血症、舒张压高及肌少症有关。

第二部分 老年住院患者认知障碍与估计肾小球滤过率的关系

中文摘要

背景：肾功能减退是老年住院患者常见的功能障碍之一，包括生理性减退及病理性减退，可导致心血管事件增加、死亡率升高。终末期肾脏病的治疗方式中，药物治疗效果较差、肾脏替代医疗费用高、透析患者生活质量明显下降。近年来有研究发现，在晚期慢性肾脏病甚至终末期肾脏病患者中，认知障碍、焦虑抑郁的发病率较高，认知障碍与估算肾小球滤过率（estimated glomerular filtration rate，eGFR）关系的探索较少。

目的：老年住院患者认知障碍与eGFR的关系。

内容与方法：选取633例患者，根据CKD-EPI公式，分为eGFR正常组257例（eGFR≥90ml/min·1.73m²）及eGFR轻度下降组376例（60ml/min·1.73m²≤eGFR＜90ml/min·1.73m²）。通过简易智力状态检查量表（Mini-mental State Examination，MMSE）、蒙特利尔认知评估量表（Montreal Cognitive Assessment，MoCA）评估认知功能，焦虑抑郁状态通过汉密尔顿抑郁量表（Hamilton Depression Scale，HAMD）、汉密尔顿焦虑量表（Hamilton Anxiety Scale，HAMA）评估。

结果：eGFR轻度下降组MoCA、MMSE低于eGFR正常组，差异有统计学意义（P ＜ 0.05，表 1），而HAMD、HAMA无统计学差异（P ＞ 0.05，表 1）。Spearman相关分析显示，eGFR与MoCA评分（r = 0.347，P = 0.000）、MMSE评分（r = 0.276，P = 0.000）、血白蛋白（r = 0.171，P = 0.000）、舒张压（r = 0.104，P = 0.009）、血红蛋白（r = 0.139，P = 0.000）呈正相关，血同型半胱氨酸（r = -0.327，P = 0.000）、血肌酐（r = -0.684，P = 0.000）、尿素（r = -0.269，P = 0.000）、糖化血红蛋白（r = -0.116，P = 0.004）、年龄（r = -0.407，P = 0.000）、诊断数量（r = -0.268，P = 0.000）、脑卒中史（r = -0.146，P = 0.000）、高血压史（r = -0.128，P = 0.001）、糖尿病史（r = -0.161，P = 0.000）、冠心病史（r = -0.113，P = 0.004）呈负相关，eGFR与HAMD评分、HAMA评分不相关（P ＞ 0.05）（表 2）。Logistic回归分析显示，eGFR与MoCA评分高（OR = 0.951，95%CI：0.919~0.984）呈正相关，与增龄（OR = 1.114，95%CI：1.080~1.150）、高同型半胱氨酸血症（OR = 1.067，95%CI：1.037~1.097）、诊断数量多（OR = 1.063，95%CI：1.013~1.115）、有脑卒中史（OR = 1.628，95%CI：1.100~2.409）呈负相关（P < 0.05，表3）。

结论：老年住院患者认知障碍与eGFR降低有关。

第三部分 老年住院患者认知障碍与冠心病的关系研究

中文摘要

背景：冠心病是导致老年人预后不佳、死亡率升高的主要原因之一。认知功能在维持老年患者日常生活活动能力中发挥着重要的作用。近年来有研究发现老年冠心病患者可能与认知障碍有一定的相关性。因此，我们提出假设，老年住院患者认知障碍可能与冠心病的风险增加有关，为冠心病及认知障碍的研究增加新的思路。

目的：探索老年住院患者认知障碍与冠心病的关系。

内容与方法：选取678例研究对象，认知状态通过蒙特利尔认知评估量表（Montreal Cognitive Assessment，MoCA）简易智力状态检查量表（Mini-mental State Examination，MMSE）、焦虑抑郁状态通过汉密尔顿焦虑量表（Hamilton Anxiety Scale，HAMA）、汉密尔顿抑郁量表（Hamilton Depression Scale，HAMD），日常生活活动能力（Activity of Daily Living，ADL）通过Barthel指数评估，通过CKD-EPI公式计算估算肾小球滤过率（estimated Glomerular Filtration Rate，eGFR）。分为冠心病组（n = 203例）与非冠心病组（n = 478例）。

结果：冠心病组MoCA、MMSE、Barthel低于非冠心病组，而HAMA评分、HAMD评分、高于非冠心病组，差异具有统计学意义（P ＜ 0.05，表1）。Spearman相关性分析结果显示，冠心病病程与年龄（r = 0.234，P = 0.000）、HAMD评分（r = 0.086，P = 0.025）、HAMA评分（r = 0.128，P = 0.001）、Urea（r = 0.106，P = 0.006）、HbA1c（r = 0.139，P = 0.000）、住院时间（r = 0.081，P = 0.034）、高血压病程（r = 0.212，P = 0.000）、糖尿病病程（r = 0.160，P = 0.000）、高血压（有/无）（r = 0.103，P = 0.007）、糖尿病（有/无）（r = 0.131，P = 0.001）呈正相关，而与MMSE评分（r = -0.092，P = 0.017）、MoCA评分（r = -0.157，P = 0.000）、入院Barthel评分（r = -0.122，P = 0.004）、出院Barthel评分（r = -0.114，P = 0.003）、ALT（r = -0.068，P = 0.079）、Alb（r = -0.130，P = 0.001）、eGFR（r = -0.189，P = 0.000）、Hb（r = -0.114，P = 0.003）、体重（r = -0.086* ，P = 0.025）呈负相关。多因素Logistic回归提示，冠心病与MoCA评分高（OR = 0.962，95%CI：0.931~0.994）呈正相关（P < 0.05），而与高龄（OR = 1.031，95%CI：1.004~1.058）、HAMA评分高（OR = 1.042，95%CI：1.014~1.071）、高血压病病程长（OR = 1.024，95%CI：1.010~1.038）、糖尿病病程长（OR = 1.036，95%CI：1.013~1.060）呈负相关（P < 0.05，表2）。

结论：老年住院患者的冠心病可能与认知障碍、焦虑状态有关。

第四部分 老年住院患者衰弱及估计肾小球滤过率的联合作用

与非计划再住院的关系

中文摘要

背景：衰弱导致老年人跌倒、残疾、死亡风险增加。肾功能减退是老年人高发的增龄性疾病之一，是老年患者反复就诊的主要原因之一。目前关于衰弱与肾功能的研究大多集中在终末期肾脏病患者中。本研究通过研究老年住院患者衰弱状态及估计肾小球滤过率（estimated glomerular filtration rate，eGFR）的联合作用与非计划再住院的关系，为老年患者衰弱及肾功能减退的研究增加新的思路。

目的：探讨老年住院患者衰弱及eGFR的联合作用与非计划再住院的关系。

内容与方法：选取400例患者，通过衰弱表型（Fried）评估衰弱，根据CKD-EPI公式计算eGFR。根据Fried评分将研究对象分为非衰弱组（Fried ≤ 2分）及衰弱组（Fried ≥ 3分），采用COX回归分析衰弱状态与非计划再住院的关系。根据eGFR分为eGFR正常组（eGFR ≥ 60ml/min·1.73m²）及eGFR下降组（eGFR ＜ 60ml/min·1.73m²），采用COX回归分析eGFR与非计划再住院的关系。再根据Fried评分及eGFR，将患者分为四组：分为A组（无衰弱状态+eGFR正常）、B组（无衰弱状态+eGFR下降）、C组（衰弱状态+eGFR正常）、D组（衰弱状态+eGFR下降），采用COX回归分析衰弱状态及eGFR的联合作用与非计划再住院的关系。

结果：随访生存分析提示，出院6个月内随访期间，共有97例非选择性再入院，其中A组发生率为21%（45/212），B组发生率为26%（18/70），C组发生率为24%（16/68），D组发生率为36%（18/50）。Kaplan-Meier曲线也显示了相同的趋势（整体=11.213，P = 0.011）。A组（ = 10.398，P = 0001）和C组（ = 5.876，P = 0.015）无事件生存率均高于D组，而B组（ = 2.208，P = 0.137）无事件生存率与D组之间无统计学差异。

COX回归方法分析衰弱和eGFR下降各自与6个月内再住院的关系，结果提示在调整混杂因素后，eGFR下降与老年住院患者6个月内再住院相关，而衰弱与再住院不相关。eGFR下降的患者6个月内再住院的风险是eGFR正常患者的1.777倍（95%CI：1.001~3.154），而衰弱患者与非衰弱患者6个月内再住院的风险无统计学差异。

  多因素COX回归分析显示，在调整了年龄、性别、吸烟、尿酸、甘油三酯、高密度脂蛋白、糖尿病、冠心病后，以A组为参比，D组患者再住院风险是A组患者的2.295倍（95%CI: 1.096 ~ 4.810）,差异有统计学意义。B组患者再住院风险是A组患者的1.401倍（95%CI: 0.665 ~ 2.953）, C组患者再住院风险是A组患者的91.8%（95%CI: 0.403~2.092）,但差异没有统计学意义。

结论：老年住院患者eGFR下降与6个月内非计划再住院相关，而衰弱与再住院不相关。老年住院患者衰弱状态及eGFR的联合作用与非计划再住院有关。

PART 1: Association Between Sarcopenia and estimated glomerular

filtration rate in Elderly with Diabetes

Background: Sarcopenia is an aging related disease, which can lead to poor prognosis in the elderly, such as energy imbalance, falls, cardiovascular disease and increased mortality. The decrease of energy consumption and increase of visceral fat in patients with sarcopenia play an important role in chronic inflammation. Microinflammatory response is closely related to cardiovascular mortality and total mortality related to renal dysfunction. Renal dysfunction is a common aging disease in the elderly, which is characterized by decreased excretion of metabolic waste and imbalance of electrolyte and acid-base. In China, about 1/3 patients with chronic kidney disease (CKD) have irreversible damage of renal function at the first visit. For patients with end-stage renal disease, the effect of drug treatment is poor, so hemodialysis, peritoneal dialysis, kidney transplantation and other renal replacement methods are often used to maintain part of renal function. However, the cost of renal replacement therapy is high, there are many complications of end-stage renal disease, and most of them have poor prognosis. In recent years, studies have found that sarcopenia is associated with estimated glomerular filtration rate (eGFR) in patients with end-stage renal disease. However, there are few studies on the relationship between sarcopenia and eGFR in people with better renal function.

Objective: To investigate the relationship between sarcopenia and eGFR in elderly patients with type 2 diabetes mellitus.

Content and methods: 188 cases of elderly were selected. The skeletal muscle mass (ASM) was measured by dual energy X-ray, and the muscle mass was calculated by ASM/ht², ASM/wt and ASM/BMI formula. According to CKD EPI_{cr-Cys C} formula, they were divided into two groups: normal eGFR group (n = 63 cases, eGFR ≥ 90ml / min · 1.73m²) and mild eGFR decrease group (n = 125 cases, 60 ≤ eGFR < 90ml / min · 1.73m²).

Results: ASM, ASM/ht², ASM/wt and ASM/BMI in normal eGFR group were significantly higher than those in mild eGFR group (P < 0.05). Spearman correlation analysis showed that eGFR was positively correlated with ASM (r = 0.231, P < 0.001), ASM/ht² (r = 0.201, P < 0.006), ASM/wt (r = 0.313, P < 0.000), ASM/BMI (r = 0.293, P < 0.000) and HDL (r = 0.208, P < 0.004); It was negatively correlated with age (r = -0.405, P < 0.000), body weight (r = -0.240, P < 0.001), BMI (r = -0.250, P < 0.001), creatinine (r = -0.614, P < 0.000), Cystatin C (r = -0.928, P < 0.000), uric acid (r = -0.277, P < 0.000), diastolic blood pressure (r = -0.254, P < 0.000) and abdominal circumference (r = -0.144, P < 0.048). Logistic regression analysis showed that eGFR was positively correlated with ASM/wt (OR = 0.763, 95% CI: 0.667 ~ 0.874), and negatively correlated with old age (OR = 1.331, 95% CI: 1.190 ~ 1.488), hyperuricemia (OR = 1.006, 95% CI: 1.000 ~ 1.013) and high diastolic blood pressure (OR = 1.089, 95% CI: 1.041 ~ 1.139).

Conclusion: The decline of eGFR in elderly patients with type 2 diabetes is mainly related to old age, hyperuricemia, high diastolic blood pressure and sarcopenia.

PART 2 Association Between Cognitive Impairment, and estimated

 glomerular filtration rate in Elderly Inpatients

Background: Renal dysfunction is one of the common dysfunction in elderly inpatients, including physiological and pathological decline, which can lead to cardiovascular events and death. In the treatment of end-stage renal disease, the effect of drug treatment is poor, the cost of renal replacement medicine is high, and the quality of life of dialysis patients is significantly decreased. In recent years, studies have found that in patients with advanced chronic kidney disease and even end-stage renal disease, the incidence rate of cognitive impairment, anxiety and depression is higher. However, there are few studies on the relationship between estimated glomerular filtration rate (eGFR) and cognitive function in patients with better renal function.

Objective: To investigate the relationship between estimated glomerular filtration rate (eGFR) and cognitive function in elderly inpatients with better renal function.

Content and methods: According to CKD-EPI formula, 633 patients were divided into normal eGFR group (257 cases) (eGFR ≥ 90ml / min · 1.73m2) and mild eGFR decrease group (376 cases) (60ml / min · 1.73m2 ≤ eGFR < 90ml / min · 1.73m2). All subjects received laboratory examination and physical examination. Cognitive function was assessed by mini mental state examination (MMSE), Montreal Cognitive Assessment (MoCA), and depression was assessed by Hamilton Depression Scale (HAMD), anxiety was assessed by Hamilton Anxiety Scale (HAMA).

Results: The MOCA score, MMSE score, serum albumin, hemoglobin and diastolic blood pressure of the eGFR decreased group were significantly lower than those of the eGFR normal group, while the education level, age, serum homocysteine, serum creatinine, diagnosis number, stroke, hypertension, diabetes and coronary heart disease of the eGFR decreased group were significantly higher than those of the eGFR normal group (P < 0.05, table 1), There was no significant difference in HAMD score and HAMA score between the two groups (P > 0.05, table 1). Spearman correlation analysis showed that eGFR was positively correlated with MOCA score (r = 0.347, P = 0.000), MMSE score (r = 0.276, P = 0.000), serum albumin (r = 0.171, P = 0.000), diastolic blood pressure (r = 0.104, P = 0.009) and hemoglobin (r = 0.139, P = 0.000), serum homocysteine (r = -0.327, P = 0.000), serum creatinine (r = -0.684, P = 0.000), serum creatinine (r = -0.684, P = 0.000) and hemoglobin (r = 0.139, P = 0.000) Urea (r = -0.269, P = 0.000), glycosylated hemoglobin (r = -0.116, P = 0.004), age (r = -0.407, P = 0.000), number of diagnoses (r = -0.268, P = 0.000), history of stroke (r = -0.146, P = 0.000), history of hypertension (r = -0.128, P = 0.001), history of diabetes (r = -0.161, P = 0.000), history of coronary heart disease (r = -0.113, P = 0.004) were negatively correlated, eGFR was not correlated with HAMD score and HAMA score (P > 0.05) (Table 2). Logistic regression analysis showed that eGFR was positively correlated with high MOCA score (OR = 0.951, 95% CI: 0.919 ~ 0.984), age (OR = 1.114, 95% CI: 1.080 ~ 1.150), hyperhomocysteinemia (OR = 1.067, 95% CI: 1.037 ~ 1.097), more diagnoses (OR = 1.063, 95% CI: 1.013 ~ 1.115), and history of stroke (OR = 1.628, 95% CI: 1.037 ~ 1.097), 95% CI: 1.100 ~ 2.409) was negatively correlated (P < 0.05, table 3).

Conclusions: The decline of eGFR in elderly inpatients may be related to old age, stroke history, hyperhomocysteinemia, more diagnosis and cognitive dysfunction.

PART 3 Association between cognitive impairment and coronary heart disease  in elderly inpatients

Background: Coronary heart disease is one of the main causes of poor prognosis and high mortality in the elderly. Cognitive function plays an important role in maintaining activities of daily living in elderly patients. Some research indicated that elderly patients with coronary heart disease may have a certain correlation with cognitive impairment. Therefore, we hypothesized that cognitive impairment may be associated with the increased risk of coronary heart disease, thus adding new ideas for the study of risk factors of coronary heart disease.

Objective: To explore the association of cognitive impairment and coronary heart disease in elderly inpatients.

Content and methods: 678 subjects were selected. Their cognitive state was assessed by Montreal Cognitive Assessment (MoCA) Mini Mental State Examination (MMSE), depression was evaluated by Hamilton Depression Scale (HAMD), and anxiety was evaluated by Hamilton Anxiety Scale (HAMA), Barthel index was used to assess the activities of daily living (ADL) and estimated glomerular filtration rate (eGFR) was calculated by CKD-EPI formula. According to the patients' previous history of myocardial infarction, angina pectoris, coronary angiographic stenosis or coronary revascularization, and confirmed by participant interviews and medical records, the subjects were divided into coronary heart disease group (n = 203) and non-coronary heart disease group (n = 478). Multivariate logistic regression analysis was used to analyze the risk factors of coronary heart disease.

Results: Moca, MMSE and Barthel in CHD group were lower than those in non CHD group, while HAMA and HAMD scores were higher than those in non CHD group, the difference was statistically significant (P < 0.05, table 1). Spearman correlation analysis showed that the course of coronary heart disease was related to age (r = 0.234, P = 0.000), HAMD score (r = 0.086, P = 0.025), HAMA score (r = 0.128, P = 0.001), urea (r = 0.106, P = 0.006), HbA1c (r = 0.139, P = 0.000), length of hospital stay (r = 0.081, P = 0.034), course of hypertension (r = 0.212, P = 0.000), duration of hospital stay (r = 0.081, P = 0.034) The course of diabetes (r = 0.160, P = 0.000), hypertension (with / without) (r = 0.103, P = 0.007), diabetes (with / without) (r = 0.131, P = 0.001) were positively correlated with MMSE score (r = -0.092, P = 0.017), MOCA score (r = -0.157, P = 0.000), Barthel score at admission (r = -0.122, P = 0.004), Barthel score at discharge (r = -0.114, P = 0.000), 001), eGFR (r = - 0.189, P = 0.000), HB (r = - 0.114, P = 0.003) and body weight (r = - 0.086, P = 0.025) were negatively correlated（Table 2）.Multivariate Logistic regression analysis showed that, the duration of CHD was positively correlated with higher MOCA score (OR = 0.962, 95% CI: 0.931-0.994) (P < 0.05), and negatively correlated with older age (OR = 1.031, 95% CI: 1.004-1.058), higher HAMA score (OR = 1.042, 95% CI: 1.014-1.071), longer duration of hypertension (OR = 1.024, 95% CI: 1.010-1.038) and longer duration of diabetes (OR = 1.036, 95% CI: 1.013-1.060) (Table 3).

Conclusions: Coronary heart disease in elderly inpatients may be related to old age, long course of hypertension, long course of diabetes, eGFR, anxiety and cognitive function. Early maintenance of cognitive function may have certain significance for the prevention and treatment of coronary heart disease.

PART 4 Association Between the Combined Effect of Frailty and

Estimated Glomerular Filtration Rate with Non-elective

Hospital Readmission in Elderly Inpatients

Background: Frailty increases the risk of falls, disability and death in the elderly. Renal dysfunction is one of the age-related diseases with high incidence in the elderly, and it is one of the main reasons for the repeated treatment of elderly patients. At present, most of the research on the relationship between weakness and renal function is focused on patients with end-stage renal disease. This study aims to study the relationship between the elderly inpatients' weakness and estimated glomerular filtration rate (eGFR) and unplanned rehospitalization, so as to provide new ideas for the research of elderly patients' weakness and renal dysfunction.

Objective: To explore the association between the combined effect of frailty and estimated glomerular filtration rate with non-elective hospital readmission in elderly inpatients.

Contents and methods: 400 cases of elderly patients were selected. The Fried scale is used to assess frailty, CKD-EPI formula is used to calculate eGFR. The subjects were divided into two groups: non-frailty group (Fried ≦ 2) and frailty group (Fried ≧ 3). The association between frailty and non-elective hospital readmission was analyzed using Cox regression model. The subjects were divided into normal eGFR group (eGFR ≧ 60 ml / min · 1.73m²) and mild eGFR decrease group (eGFR < 60 ml / min · 1.73m²). The relationship between eGFR and non-elective hospital readmission was analyzed by Cox regression model. Finally, according to the Fried score and eGFR, the patients were divided into four groups: group A (no frailty + eGFR normal), group B (no frailty + eGFR decreased), group C (frailty + eGFR normal), and group D (frailty + eGFR decreased). Cox regression analysis was used to analyze the association between the combined effect of asthenia and eGFR and non-elective hospital readmission.

Results: A total of 400 subjects were included in this study, including 248 males (62%), 118 patients with asthenia (29.5%), and 120 patients with decreased eGFR (30%).

There were significant differences in age, creatinine, total protein, albumin, uric acid, total bilirubin, lactate dehydrogenase, alanine aminotransferase, total cholesterol, low density lipoprotein cholesterol and hypertension among different groups (P < 0.05).

The results of follow-up survival analysis showed that there were 97 cases of non-selective readmission within 6 months after discharge. The incidence of group A was 21% (45 / 212), group B was 26% (18 / 70), group C was 24% (16 / 68), and group D was 36% (18 / 50).

Kaplan Meier curve also showed the same trend (overall  = 11.213, P = 0.011). The event free survival rates of group A ( = 10.398, P =, 0001) and group C ( = 5.876, P = 0.015) were higher than those of group D, while there was no significant difference between group B ( = 2.208, P = 0.137) and Group D.

Cox regression was used to analyze the association of frailty and decreased eGFR with readmission within 6 months. The results suggest that after adjusting for confounding factors, the decline of eGFR is associated with readmission within 6 months, while frailty is not associated with readmission. The risk of rehospitalization within 6 months in patients with decreased eGFR was 1.777 times higher than that in patients with normal eGFR (95% CI: 1.001~3.154), while the risk of rehospitalization within 6 months in frailty patients and non-frailty patients had no significant difference.

Multivariate Cox regression analysis showed that after adjusting for age, gender, smoking, uric acid, triglyceride, high density lipoprotein, diabetes and coronary heart disease, the risk of readmission in group D was 2.295 times higher than that in group A (95% CI: 1.096-4.810), with statistical significance. The risk of rehospitalization in group B was 1.401 times of that in group A (95% CI: 0.665-2.953), while that in group C was 91.8% (95% CI: 0.403-2.092), but the difference was not statistically significant.

Conclusions: The decline of eGFR is associated with non-elective hospital readmission in elderly inpatients within 6 months, while frailty is not associated with rehospitalization. The combined effect of asthenia and eGFR in elderly inpatients is related to non-elective hospital readmission.