Plasma amino acid concentrations in elderly patients with protein energy malnutrition

Age and Ageing, Nov, 1997 by Anne Polge, Etiennette Bancel, Helene Bellet, Denise Strubel, Sophie Poirey, Pascale Peray, Christine Carlet, Bernard Magnan de Bornier

Introduction

The plasma amino acid pattern has been studied in healthy patients of different ages [1-7], but few studies provide information about very elderly people [8-11]. Protein energy malnutrition (PEM) occurs in about 50% of hospitalized old people. Both qualitative and quantitative changes in circulating amino acid concentrations have been reported in PEM. Factors affecting the concentration and the pattern of plasma amino acids include the amount and composition of dietary protein, muscle protein metabolism and the labile protein reserve in various tissues, particularly in the liver.

PEM has two main causes: a decrease in nutrient intake and an increase in catabolic reactions (i.e. a hypermetabolic state). In old patients both may be present at the same time or they may occur successively in a short time as a consequence of multiple pathology.

Decrease in food intake leads to a rise in gluconeogenesis. Hypermetabolic states, associated with infections and inflammatory diseases, turn protein anabolism towards acute phase protein synthesis in the liver, while anabolism of the whole organism is disturbed. Increases in catabolic reactions affect muscle protein [12] and provide phenylalanine, tyrosine and methionine -- the branched-chain amino acids [13-15], which are released without muscle metabolism -- glutamine (a regulator of muscle protein turnover [16] and an essential nutrient for cells of the immune [17] and intestinal systems [18]) and alanine, an important amino acid in gluconeogenesis.

We report an investigation into the plasma free amino acid pattern of healthy normal weight aged controls (80-100 years old) and elderly underweight people with PEM (80-100 years old).

Subjects and methods

Subjects

The 68 subjects investigated form two groups.

The first group comprises 44 control patients (20 men 84.5 [ or -] 2.7 years old, 24 women 85.4 [ or -] 3.2 years old), admitted to the geriatric unit of the Montpellier-Nimes University Hospital Centre. Selection was on the basis of benign disease without renal, hepatic or metabolic disorder, presenting symptoms being falls, loss of consciousness, immobility and psychiatric problems. Their clinical state and nutritional status were good. Nutritional status was determined using a food intake questionnaire. (Protein intake was in line with the recommendations of the NAS-NRC [19] and Young's published data [20].)

The second group comprises 24 elderly patients with PEM (10 men 86 [ or -] 5 years old, 14 women 88.4 [ or -] 5 years old) hospitalized in the same unit. Nine of these patients had anorexia with pure weight loss. Fifteen had anorexia and weight loss associated with underlying illness (e.g. cancer, pneumonia, Parkinsonism, dementia and the stress of recent surgery). Anthropometric and biological data were collected on admission, before beginning parenteral feeding.

Anthropometric measurements

The following anthropometric measurements were performed on all 68 subjects.

Weight and height/body mass index (BMI)

BMI was expressed as weight (kg)/height ([m.sup.2]) [21]. In bed-ridden patients, height is determined using Chumlea's formula [22] from knee height. (Men with malnutrition have BMI [is less than] 19.5 kg/[m.sup.2], women with malnutrition have BMI [is less than] 18.7 kg/[m.sup.2].)

Mid-arm circumference (MAC) [21]

Patients with malnutrition have MAC [is less than] 24.4 cm (men) and [is less than] 23.1 cm (women).

Body weight as a percentage of ideal weight (IBW)

This is determined using the weight table of Metropolitan Life Insurance Company [23]. (PEM is associated with percentage IBW [is less than] 90%.)

Biological data

Serum albumin and prealbumin, total lymphocyte count (TLC) and serum acute phase protein (C reactive protein, [[Alpha].sub.1]-glycoprotein acid) were measured in all patients. TLC is calculated as the percentage of lymphocytes multiplied by the total white blood cell count. PEM is associated with serum albumin [is less than] 32 g/l, serum prealbumin [is less than] 0.24 g/l and TLC [is less than] 1500 cells/[mm.sup.3].

Plasma amino acid profiles were performed on venous blood samples obtained in heparin-treated tubes, early in the morning, after an overnight fast. Plasma samples were determined using cation-exchange columns with ninhydrin detection in an high performance liquid chromatography system (model 6300, Beckman Instruments) with glucosaminic acid as internal standard.

Clinical data

Clinical features were extracted from medical files. Weight and information about the patients' functional abilities (mobility, continence, communication and appetite) were assessed and recorded by the nursing staff using a specially designed form. Assessment of malnutrition risk was recorded.

Statistical analysis

Statistical analysis was performed by means of the PC Statistical Analysis System (SAS Institute Inc.). For a comparison of quantitative variables (such as anthropometric and biological data) between the two groups (patients with PEM and control patients), Student's t-test was used when the distribution variable was normal. Otherwise, the Kruskal-Wallis test was used.