BackgroundVitamin D deficiency is known to be highly prevalent in older persons. However, the prevalence in the subgroup of frail older hospitalized patients is not clear. We sought to investigate the prevalence and predictors of vitamin D deficiency in frail older hospitalized patients.Methods217 consecutively geriatric hospitalized patients with routine measurements of 25-hydroxyvitamin D [25 (OH)D] at hospital admission were analyzed retrospectively, including information of previous vitamin D supplementation and the geriatric assessment. Serum 25 (OH)D concentrations < 20 ng/ml and between 20 and 29.99 ng/ml were classified as deficient and insufficient, respectively, whereas concentrations ≥30 ng/ml were considered as desirable. A stepwise binary logistic regression model was performed to assess the simultaneous effects of age, gender and geriatric assessments on the prevalence of low vitamin D concentration.ResultsMean age of the cohort was 81.6 ± 8.0 years (70.0% females). Mean serum 25(OH)D was 12.7 ± 12.9 ng/ml. Of 167 (77%) subjects without known previous vitamin D supplementation, only 21 (12.6%) had serum concentrations ≥20 ng/ml and only 8 (4.2%) had desirable serum concentrations ≥30 ng/ml. In total population, 146 (87.4%) participants were vitamin D deficient. Despite vitamin D supplementation, 22 of 50 participants (44.0%) were vitamin D deficient and only 19 (38.0%) had desirable concentrations of ≥30 ng/ml. In a stepwise logistic regression analysis, only previous intake of vitamin D supplementation and high Geriatric Depression Scale score (GDS-15) were significantly associated with vitamin D deficiency.ConclusionsIn the group of frail older hospitalized patients without previous vitamin D supplementation, the prevalence of inadequate vitamin D concentrations is extremely high. Therefore, usefulness of the routine measurement of vitamin D status before initiating of supplementation appears to be questionable in this patient group.

Very little is known about the effect of malnutrition on short-term changes of body composition, particularly muscle, among older hospitalized patients. We sought to investigate the association of malnutrition as assessed by the Global Leadership Initiative on Malnutrition (GLIM) criteria with changes of thigh muscle mass and muscle strength among older patients during hospitalization. Forty-one patients (age range 66–97 years, 73% female) participated in this prospective longitudinal observational study. Nutritional status was evaluated using the GLIM criteria on admission and at discharge. Functional status and mid-thigh magnetic resonance imaging (MRI) measurements of cross-sectional area (CSA) were conducted on admission and before discharge. In all, 17% were malnourished and 83% had no malnutrition. Mean mid-thigh muscle CSA declined by 7.0 cm2 (−9%) in malnourished patients during hospitalization (p = 0.008) and remained unchanged among non-malnourished patients (−1%, p = 0.390). Mean mid-thigh CSA of subcutaneous and intermuscular fat did not change significantly during hospitalization in both groups. Malnourished subjects lost 10% of handgrip strength (−1.8 kg) and 12% of knee extension strength (−1.5 kg) during hospitalization. However, the magnitude of both changes did not differ between groups. In a stepwise multiple regression analysis, malnutrition and changes in body weight during hospitalization were the major independent risk factors for the reduction of muscle CSA. Malnutrition according to the GLIM criteria was significantly and independently associated with acute muscle wasting in frail older patients during 2-week hospitalization.

We sought to identify any association between whole blood thiamine level and functional status in older hospitalized patients. Methods: This cross-sectional study retrospectively analyzed the results of routine measurements of whole blood thiamine levels of 233 older patients who were consecutively hospitalized to a geriatric acute care ward. Nutritional status, depression, and the participants' cognitive impairment were evaluated using the Mini Nutritional Assessment-Short Form, Depression in Old Age Scale, and Montreal Cognitive Assessment, respectively. Activities of daily living were determined using Barthel Index (BI) on admission and at the time of discharge. Diagnoses of dementia and delirium were derived from the patients' medical records. Results: Of 233 participants (mean age 82.1 [7.1]), 47.0% and 39.0% were at risk of malnutrition and malnourished, respectively. There was no thiamine deficiency (<20 ng/mL) in total population. Nearly all patients (95%) were screened with impaired cognitive function, in which 36% and 9% had the diagnosis of dementia and delirium, respectively. Patients with dementia (P ¼ .040) and delirium (P ¼ .002) demonstrated lower mean thiamine blood levels compared to patients without. Mean blood vitamin B 1 was higher in patients with functional recovery (change in BI !5 points during hospitalization; P ¼ .018). In a binary logistic regression analysis, blood vitamin B 1 , weight loss, and female gender were the major independent risk factors for delirium but not for dementia. Conclusion: Despite the absence of thiamine deficiency, whole blood thiamine was lower in patients with dementia and delirium compared to those without. Higher thiamine levels were significantly associated with functional recovery during hospitalization.