MNA predictive value in the follow-up of geriatric patients

L.M. Donini, C. Savina, A. Rosano*, M.R. De Felice**, L. Tassi**, L. De Bernardini**, A. Pinto, A.M. Giusti, C. Cannella Istituto di Scienza dell’Alimentazione, Università degli Studi di Roma “La Sapienza” (Italy) * Istituto Nazionale di Statistica ISTAT (Rome - Italy) ** Istituto Geriatrico “Villa delle Querce” – Nemi (Rome, Italy)

Mailing Address: Lorenzo M. Donini Università degli Studi di Roma "La Sapienza" Istituto di Scienza dell'Alimentazione P.le Aldo Moro, 5 00185 Rome (Italy) Fax: +3906/4991-0699 e-mail: Lorenzomaria.Donini @ uniroma1.it

Running title MNA in the follow-up of elderly patients

Abstract Objective: The aim of this study is to verify, in a sample of elderly subjects admitted to long-term care, the impact of malnutrition, according to the Mini Nutritional Assessment (MNA), on mortality and on the occurrence of Adverse Clinical Events in a 3-12 months follow-up study. Subjects: The survey included all patients admitted to a geriatric hospital--“Villa delle Querce”, Nemi (Rome, Italy)-between January 1997 and April 2000, whose nutritional status we were able to monitor for over 3 months. The study comprised 167 elderly subjects, of which 125 women (74.9%) aged 83.3±8 years (60-95 years), and 42 men (25.1%) aged 79.6±9 years with an average follow-up period of 7.5 months. Methods: Upon admission and at every check we evaluated each subject’s cognitive functions, functional status, comorbidity, frailty, nutritional status (anthropometric and biochemical indices; MNA). During the follow-up we recorded Adverse Clinical Events. We calculated the predictive value of MNA, we correlated variations in MNA scores with variations of nutritional parameters. Results: MNA’s predictive ability both upon admission and upon discharge was found to be excellent. The MNA score was found to be correlated—although not to a very high degree—with variations nutritional parameters. Even more than malnutrition, a low MNA score was found to be predictive of a greater incidence of Adverse Clinical Events during hospitalisation and of higher mortality.

Key words: MNA, malnutrition

Introduction A good nutritional status is crucial for “successful ageing,” and malnutrition is related with an increased incidence of morbidity and mortality. The relationship between poor nutritional status and impaired immune functions, pressure sores and impaired muscle function is well established (1). For these reasons, nutritional status risk assessment has become an essential component of the Comprehensive Geriatric Evaluation, which should be performed in every elderly patient at the admission in hospital. By the same token, the screening tool for the assessment of nutritional status should be able to detect with enough accuracy even slight variations in the patient’s nutritional status. Many studies have shown that the Mini Nutritional Assessment (MNA) (2) has an excellent predictive value of malnutrition in elderly populations, both free-living and institutionalised (3-7) The aim of this study is to verify, in a sample of elderly subjects admitted to long-term care, the impact of malnutrition, according to the MNA, on mortality and on the occurrence of Adverse Clinical Events in a 3-12 months follow-up study. Subjects and methods 1.

Subjects The survey included all patients admitted to a geriatric hospital--“Villa delle Querce”, Nemi (Rome, Italy)-between January 1997 and April 2000, whose nutritional status we were able to monitor for over 3 months. The study comprised 167 elderly subjects, of which 125 women (74.9%) aged 83.3±8 years (60-95 years), and 42 men (25.1%) aged 79.6±9 years (61-97 years), for a period of time ranging between 3 and 12 months according to cases (7.5 months on average) (Table 1). Patients were evaluated twice: within the first 48 hrs after admission, and at discharge. For patients institutionalised in long-term care, the second evaluation was conducted after 12 months. The hospital has 140 long-term care beds, with a turnover of about 300 admissions per year. The cases considered here represent about one third of total admissions in the above mentioned period. Data regarding other subjects was not processed since the patients’ observation period was below 3 months. The characteristics of the whole population were previously presented elsewhere (8).

2.

Methods

2.1 Upon admission and at every check we evaluated each subject’s: -

cognitive functions by the Short Portable Mental Status Questionnaire-SPMSQ (9)

-

functional status by ADL-Activities of Daily Living (10)

-

comorbidity by a modified version of the Geriatric Index of Co-morbidity (11, 12) which is computed on the basis of severity of 15 of the most frequent chronic conditions. Five comorbidity classes were defined: - Class I: patients with no symptomatic diseases - Class II: patients with symptomatic diseases under satisfactory control - Class III: patients with only one uncontrolled disease - Class IV: patients with two or more uncontrolled diseases - Class V: patients with one or more diseases at their greater severity - frailty: we considered age, cognitive functions, functional and nutritional status (13). For each item we distinguished three levels: age (< 75; 75-80; > 80); cognitive functions (normal, mild impaired, severely impaired); functional status (independent, dependent in 1-2 ADLs, dependent in more than two ADLs); nutritional status (albumin > 35, 30-35, < 30 g/l). We assigned a frailty score to each level: 0 points for best scenarios (age < 75; normal cognitive functions...); up to 2 points for worst scenarios (age > 80; severely impaired cognitive functions…). Frailty classes were defined as follows: - A: 0-2 points - B: 3-4 points - C: 5-8 points

During the follow-up we recorded Adverse Clinical Events (ACE) following the definition of Bernardini et al (14): “any acute or sub-acute change in health status, detected by specific signs, symptoms and/or laboratory findings, that suggests acute or sub-acute illness.”. ACEs were classified into 4 levels of severity: Class A: requiring no intervention, < 1 day of monitoring, without residual functional impairment Class B: requiring therapeutic intervention, 1-7 days of monitoring, without residual functional impairment Class C: requiring therapeutic intervention, 8-21 days of monitoring, without residual functional impairment

Class D: requiring therapeutic intervention and carrying a residual functional impairment 2.2 Nutritional status Upon admission and at discharge we performed on each subject: i.

an evaluation of nutritional anthropometric [stature, weight, arm circumference (AC), triceps skinfold thickness (TSF)] and biological parameters [albumin, transferrin, cholesterol, cholinesterase, lymphocytes, C-reactive protein (CRP), mucoprotein]. We then calculated the Body Mass Index [BMI = weight (Kg) / (stature (m))²] and the mid-upper arm muscle circumference [MAMC (cm) = AC (cm) – (π * TSF (cm))].

Lower limits of normality for nutritional parameters were defined as: MAMC = 20.2 cm for men and = 18.6 cm for women; TSF = 10 mm for men and = 13.2 mm for women; albumin = 31 g/l, transferrin = 2 g/l, lymphocytes = 1500 #/mm³ (15, 16). The anthropometric measurements were performed by a single trained operator according to the Standard Manual for Anthropometric Measures (17). The subjects were measured barefoot with lightweight clothing. Weight was measured to the nearest 0.1 kg on a SECA (Hamburg, Germany) weighing scale, and stature was measured on a wall-mounted stadiometer to the nearest 0.5 cm (SECA, Hamburg, Germany). To measure stature, the barefoot subjects were requested to stand straight on a horizontal surface, heels together, looking straight forward. Circumferences were measured to the nearest 0.1 cm with a cloth tape, and skinfold thickness was measured to the nearest 0.2 mm with a Harpenden skinfold caliper (British Indicators Ltd, St Albans, Herts, UK) on the dominant arm. The concentrations of biological parameter in serum were determined by routine methods with conventional commercial kits obtained from ABX Italia (Rome, Italy). Peripheral venous blood was collected from antecubital vein after an overnight fast. The laboratory tests were carried out using a COBAS-MIRA analyser at the Laboratory of the Istituto Geriatrico “Villa delle Querce” – Nemi (Rome, Italy). ii. A MNA (2) including a comprehensive anthropometric assessment, data about general condition and dietary habits, and a self-evaluation of health and nutritional status. According to their MNA scores, patients were classified into three risk categories: 1. malnourished: MNA < 17 2. at risk of malnutrition: 17 24 iii. A second MNA that we called “proportional and objective” (MNA-PO) (8), in which we: - replaced the self-perception of health and nutrition status with an objective assessment; - replaced the total MNA score with the ratio of this value with the maximum of points that each subject could obtain without including the items to which we failed to obtain a response (BMI for bedbound patients, arm and calf circumferences due to subcutaneous edema). Similarly, the cut-off points (17 and 24) were substituted by the ratio of these values with the maximum of points obtainable by a complete MNA (30). Patients were classified as “malnourished” below 0.56, “at risk of malnutrition” from 0.56 to 0.79, and “well-nourished” from 0.8 up. All patients received feeding assistance and changes in food consistency according to their ability to feed themselves and to the efficiency of their chewing. In 86 cases (51%) we resorted to modular energetic and/or protein oral supplements; in 4 cases (2.4%) a diet was prescribed; artificial nutrition was not necessary in any cases. 2.3. Data analysis We correlated: -

admission and discharge values of MNA and MNA-PO scores with corresponding values of anthropometric (BMI, TSF, MAC) and biological indices (albumin, transferrin, cholesterol, cholinesterase, lymphocytes, C-reactive protein (CRP), mucoprotein).

-

variations of MNA and MNA-PO scores with variations of results of nutritional indices. The above variations were calculated both as a difference between initial and final data, and as percentages [(initial data – final data / initial data) x 100]

-

admission values of MNA and MNA-PO scores with the clinical outcome (ACEs occurrence, mortality)

Linear regression analysis and correlation coefficient (Pearson’s r for absolute differences and Spearman’s ρ for percentage differences) were used to test the association between anthropometric or biological indices and MNA score. Student’s t test was used to assess differences in group means and Pearson’s χ2 to compare the observed frequencies of categories to the expected frequencies if the null hypothesis were true. Statistical significance was set at the p < 0.05 level. The data was analysed with the BMDP New System Statistical Package (BMDP Statistical Software Inc., Los Angeles, CA, USA) and SPSS for Windows (SPSS Inc 1989-1999). Results The overall characteristics of the 167 subjects included in the study are displayed in table 1. The average number of diseases per patient was about 4 (range 1 – 9). The most frequent disorders were cardiovascular diseases and cognitive deterioration. A high percentage of patients (about 80%) had severe co-morbidity (classes III, IV and V: patients with “uncontrolled diseases” or “diseases at their greater level of severity”). Over 70% of the subjects were dependent in more than two ADLs, while 50% of men and 80.8% of women had neurological disorders. Therefore, 42.8% of men and 70.4% of women had severe (class C) frailty, i.e. “a pronounced tendency to homeostatic imbalance to coincide with pathological, psychological or social occurrences” (12). In our sample, prevalence of malnutrition upon admission was 67.7% (113 out of 167 subjects) according to MNA and 49.7% (75 out of 167 patients) according to MNA-PO (tab. 2). MNA scores were correlated (Pearson’s r) with nutritional indices both at admission and discharge (tab. 3). A better correlation was always found for MNA-PO and at discharge rather than at admission. The best correlations were observed at admission with MAC (r = 0.4; p < 0.01) and at discharge with MAC (r = 0.46; p < 0.01), albumin (r = 0.47; p < 0.01), transferrin (r = 0.46; p < 0.01) and negatively with mucoprotein (r = - 0.48, p < 0.01) values. Differences between the initial and final values of MNA scores and nutritional parameters appeared to be correlated (Pearson’s r) in most cases (tab. 4). The same can be said, although to a lesser degree, of percentage variations of the same values (Spearman’s ρ) (tab. 5). Differences and percentage variations of MNA-PO always appeared better correlated to similar variations of nutritional parameters. The best correlations were observed with the variations in blood albumin (r = 0.43, p = 0.000), cholinesterase (r = 0.52, p = 0.000), transferrin (r = 0.31; p = 0.000), and negatively with mucoprotein (r = - 0.3, p = 0.005). There was poor correlation with variations of anthropometric parameters. Among them, the best correlation was found for TSF: r = 0.28, p = 0.000. Nutritional status checks conducted after 3-12 months from admission (7.5±4 months on average) showed an improvement in the nutritional status in 83 subjects (49.7%) and a deterioration in 63 subjects (37.7%) according to MNA score; an improvement in the nutritional status in 85 subjects (50.9%) and a deterioration in 74 subjects (44.3%) according to MNA-PO score. The prevalence of “malnutrition” and “risk of malnutrition” judgements was nevertheless basically unchanged. At the same time, we observed a relevant increase in the share of subjects with normal parameters (+28.6% for TSF, +30.6% for transferrin, +4.3% for albumin), and/or a relevant increase in the share of subjects in which we observed a normalization of inflammation indices (+10.6% for CRP, +49.6% for mucoprotein) (Tables 2). The way we performed the nutritional intervention did not affect the changes in nutritional status. The share of subjects who showed an improvement in their nutritional status did not change whether the chosen intervention was feeding assistance, changes in food consistency, oral supplementation, or a prescribed diet. Artificial nutrition was not necessary in any cases. In patients who showed a deterioration of the nutritional status, this was accompanied by a worsening of comorbidity, of functional status, of cognitive status and of frailty level, and by an increase in stress protein (CRP and mucoprotein) in connection with the onset (in 28 cases, 16.8%) of acute events in the 3 months prior to check. During the observation period we had the onset of ACE (C or D type) (14) in 28 subjects (16.8%) and 10 subjects (6%) died. In the case of ACE (Tables 6, 7) their onset appeared to be correlated to a higher level of comorbidity (χ2= 11.9; p = 0.008) and frailty score (6.3 ± 0.7 vs. 4.9 ± 2; t = 2.5, p = 0.01) upon admission. The initial nutritional status was not influential: in subjects who incurred in ACE, anthropometric parameters and levels of blood albumin were better than in the remaining sample. MNA class was nevertheless predictive of ACE occurrence: although it did not achieve statistical significance, we observed that when the MNA assessment worsened the incidence of ACE during follow-up increased. ACE occurrence was 15.9 and 19.3% among patients rated as malnourished by MNA and MNA-PO respectively, as compared to 20 and 15.7% among patients at risk of malnutrition by MNA or MNA-PO and 0-7.1% among the well-nourished patients by MNA and MNA-PO respectively. As previously mentioned, during the observation period ACE occurred in 28 subjects (16.8%). In these subjects MNA score worsened (-2.9±5) as compared to the remaining sample (+1.3±4) (t = 5; p = 0.000). The same trend was found for MNA-PO: -0.29±0.6 for patients who incurred in ACEs and +0.11±0.7 for the remaining patients (t = 2.8; p = 0.005). This was confirmed by a parallel negative variation of the MNA assessment which followed each

ACE. In 29.4% of patients who shifted to a lower MNA assessment class, and in 38% of subjects who showed similar developments in MNA-PO, this deterioration of the nutritional status was preceded by an ACE. Instead, among patients who sowed an improvement of their nutritional status during hospitalisation, only in 3 cases (9.7%) did an ACE occur. Here, too, the data was confirmed by MNA. An ACE occurred in only 2 (5.1%) of the subjects who had an improvement in MNA score, and in 1 (5%) of those who showed similar development in MNA-PO. As to deaths (Tables 8, 9), they were more frequent in subjects with serious cognitive deterioration, with marked inflammation signs (mucoprotein increase), with high dependency and high frailty levels. In this case, too, the initial nutritional status did not seem to play a decisive role. The 10 patients who died during follow-up had better anthropometric parameters (TSF, MAC, MAMC) than other subjects; lower transferrin levels (1.67 ± 0.3 g/l vs. 1.99 ± 0.6 g/l; t = 2.8; p = 0.02); higher mucoprotein (160.5 ± 52 mg/l vs. 101.2 ± 32 mg/l; t = 4.2, p = 0.000); a substantively better nutritional status. MNA class was predictive of the mortality risk with 9 (8% for MNA, 10.8% for MNA-PO) deaths among the patients rated as malnourished (113 subjects at the MNA, 83 subjects at the MNA-PO) as compared to 1 death (2% for MNA and 1.4% for MNA-PO) among patients at risk of malnutrition (50 subjects at the MNA, 70 at the MNA-PO) and no deaths among well-nourished patients (4 subjects at MNA, 14 subjects at MNA-PO). MNA scores appeared to be lower in deceased patients (MNA: 10.4 ± 4 vs. 15 ± 4, t = 3.3, p = 0.000; MNA-PO: 0.39 ± 0.1 vs. 0.55 ± 0.1, t = 3.2, p = 0.002). Statistically significant differences were observed in the sections “Global evaluation” and “Diet evaluation,” where the scores of patients who then died were remarkably lower than those of patients who survived (1.8 ± 1 vs 3.9 ± 1 for Global evaluation - t = 4.8, p = 0.000; 4.2 ± 2 vs 6.7 ± 2 for Diet evaluation – t = 3.9, p = 0.000) (Table 9). Discussion Patients who will benefit from nutritional intervention should be accurately diagnosed. Clinical research has focused on the necessity of early diagnosis of malnutrition (M) when a nutritional intervention is still effective. For this reason, nutritional status assessment has become a part of the comprehensive geriatric assessment that must be performed in every elderly patients upon admission to acute, sub-acute or long-term care. M is considered to be an important clinical problem in hospitalised geriatric patients, but no standard diagnostic criteria are available. The concept of malnutrition and the tools used in the different studies vary, and there is no clear consensus on the definition of M and its management. In a recent paper, Joosten E et al (18) found that in the same sample the prevalence of M varied between 6.5 and 85% using different sets of criteria for the diagnosis. This raises questions about the validity of comparing results of different studies and drawing conclusions for clinical practice. The high prevalence of malnutrition (68% according to MNA and 50% to MNA-PO) upon admission in our population is probably justified by its general characteristics (Tables 1, 2). Severe comorbidity, mildly or severely impaired cognitive functions, functional impairment, including immobility and eating dependence—all with a high prevalence in our sample--are considered to be risk factors for malnutrition (19-21). In the past few years MNA became a standard reference for the assessment of nutritional risk in geriatric practice: 18 items grouped in 4 domains allow a rapid, effective and cheap evaluation of nutritional risk for M (3-5, 22). In our protocol we introduced MNA into the comprehensive geriatric evaluation that we perform in all our patients upon admission. Following the results of MNA, for patients rated “at risk of malnutrition” or “malnourished”, we successively proceed to a complete evaluation of nutritional status by performing the analysis of anthropometric (BMI, TSF, MAC, MAMC) and biological indices (albumin, transferrin, cholesterol, cholinesterase, lymphocytes, C-reactive protein (CRP), mucoprotein). In a previous paper (7) we concluded that MNA is a valid tool for the evaluation of nutritional status in the elderly population since it is highly sensitive and easy to use. Unfortunately, we had a large number of false positive judgements. Other studies, particularly those conducted to develop the MNA, gave different results (2, 4, 5, 22). In our view these different results were due to physiopathological differences (severity of condition, comorbidity) and in frailty levels (cognitive status, functional status) of the population under examination, as well as the patients’ location (at home, in nursing homes, in hospitals for acute diseases, in rehabilitation…). When we examined the characteristics of the patients with a good nutritional status but with a low MNA score ( 90 Number Diseases (per patient) By system Cardiovascular Respiratory CNS Bone-joint Metabolic, endocrinol. Digestive Urinary Co-morbidity Class I Class II Class III Class IV Class V Number Drugs taken (per patient) Class A Frailty Class B Class C Independent Functional status Dependent ADL in 1-2 ADLs Dependent in > 2 ADLs Normal Cognitive functions Mildly SPMSQ impaired Severely impaired

M

F

42 79.6±9

125 83.3±8

7 sub’s

16.7%

11 sub’s

8.8%

15 16 4

35.7 38.1 9.5

33 65 16

26.4 52.0 12.8

3.9±1

4.1±2

13 sub’s 6 7 5 5

31.0% 14.3 16.6 11.9 11.9

42 sub’s 9 24 20 14

33.6% 7.2 19.2 16.0 11.2

5 1 0 sub’s 7 16 18 1

11.9 2.4 0% 16.7 38.1 42.8 2.4

13 3 0 sub’s 27 61 24 10

10.4 2.4 0% 21.6 49.6 20.0 8.8

3.8±2

4.3±2

9 sub’s 15 18 8 sub’s 4

21.5% 35.7 42.8 19.1% 9.5

8 sub’s 29 88 6 sub’s 10

6.4% 23.2 70.4 4.8% 8.0

30

71.4

109

87.2

21 sub’s 17

50.0% 40.5

24 sub’s 45

19.2% 36.0

4

9.5

56

44.8

Legend: SPMSQ = Short Portable Mental Status Questionnaire; ADL = Activities of Daily Living.

TABLE 2 Nutritional status trend of the studied population

Admission MNA

sub’s 4

Well-nourished Risk of Malnutrition Malnutrition score

MNA-PO

TSF

MAC

MAMC

Albumin

Transferrin

Cholesterol

Cholinesterase

Lymphocytes CRP Mucoprotein

Well-nourished Risk of Malnutrition Malnutrition score Normal Reduced M: < 10 mm F: < 13.2 mm mm Normal Reduced ( 35 g/l 30-35 < 30 g/l > 2 g/l 150 mg/dl 3000 UI/l 1500 #/mm³ 7 < 120 mg/l > 120 mg/l

% 2.4

50 29.9 113 67.7 14.7±4

Discharge sub’s 6

% 3.6

χ =49.4

p 0.000

1.6

0.1

2

48 28.7 113 67.7 15.3±5

14 8.4 70 41.9 83 49.7 0.54±0.2 42 25.1 125 74.9

13 7.8 79 47.3 75 44.9 0.55±0.2 54 32.3 113 67.7

χ =32.8

0.000

1.3 χ =116

0.2 0.000

10.6 ± 5 81 48.5 86 51.5 22.7 ± 4 96 57.5 71 42.5

11.2 ± 6 91 54.5 76 45.5 22.4 ± 3 86 51.5 81 48.5

t=0.6 χ2=94.2

0.5 0.000

t=1.8 χ2=78.1

0.08 0.000

19.4 ± 2 84 50.3 55 32.9 28 16.8 34.9 ± 5 72 43.1 95 56.9 2 ± 0.6 127 76.1 40 23.9 176 ± 41 124 74.3 43 25.7 3801 ± 1318 135 80.8 32 19.2 94 56.3 73 43.7 91 54.5 76 45.5

18.8 ± 2 88 52.7 57 34.1 22 13.2 35.6 ± 4 94 56.3 73 43.7 2 ± 0.6 120 71.9 47 28.1 178 ± 40 126 75.5 41 24.5 4021 ± 1406 140 85.7 27 14.3 104 61.9 63 38.1 136 81.0 31 19.0

t=3.7 χ2=24.9

0.000 0.000

t=1.8 χ2=

0.07

t=1 χ2=28.5

0.3 0.000

t=0.7 χ2=22.8

0.5 0.000

t=2.1 χ2=75.6

0.03 0.000

χ2=17.6

0.001

χ2=2.6

0.1

104 ± 37

107 ± 46

t=0.3

0.74

2

2

Legend: MNA = Mini Nutritional Assessment; MNA-PO = Mini Nutritional Assessment-“Proportional and Objective Score”; TSF = triceps skinfold thickness; MAC = mid-upper arm circumference; MAMC = mid-upper arm muscle circumference.

TABLE 3 Correlation of MNA and MNA-PO scores with anthropometric and biological parameters of nutritional status (Pearson’s r) at admission and discharge

MNA Pearson’s r TSF MAC MAMC Albumin Transferrin Cholesterol Cholinesterase CRP Mucoprotein

Admission

Discharge

0.25 ** 0.36 ** 0.35 ** .021 ** 0.21 ** 0.07 0.28 ** - 0.16 * - 0.26 **

0.32 ** 0.43 ** 0.38 ** 0.45 ** 0.38 ** 0.20 * 0.35 ** - 0.17 * - 0.41 **

MNA-PO Admission Discharge 0.32 ** 0.40 ** 0.36 ** 0.28 ** 0.26 ** 0.09 0.34 ** - 0.25 ** - 0.30 **

0.35 ** 0.46 ** 0.40 ** 0.47 ** 0.46 ** 0.23 ** 0.43 ** - 0.24 ** - 0.48 **

Legend: MNA = Mini Nutritional Assessment; MNA-PO = Mini Nutritional Assessment-“Proportional and Objective Score”; ** = p < 0.01, * = p < 0.05

TABLE 4 Correlation of the differences (admission-discharge) of the MNA and MNA-PO scores with the similar differences of anthropometric and biological parameters of nutritional status (Pearson’s r)

Differences admission-discharge TSF MAC MAMC Albumin Transferrin Cholesterol Cholinesterase Mucoprotein

MNA r 0.25 0.24 0.12 0.40 0.26 0.16 0.50 -0.22

MNA-PO p 0.001 0.002 0.122 0.000 0.001 0.06 0.000 0.04

r 0.28 0.28 0.16 0.43 0.31 0.16 0.52 - 0.30

p 0.000 0.000 0.04 0.000 0.000 0.05 0.000 0.005

Legend: TSF = triceps skinfold thickness; MAC = mid-upper arm circumference; MAMC = mid-upper arm muscle circumference; MNA = Mini Nutritional Assessment; MNA-PO = Mini Nutritional Assessment-“Proportional and Objective Score”.

TABLE 5 Correlation of the percent variations (admission-discharge) of the MNA and MNA-PO scores with the similar differences of anthropometric and biological parameters of nutritional status (Spearman’s ρ)

Percent variations admission-discharge TSF MAC MAMC Albumin Transferrin Cholesterol Cholinesterase Mucoprotein

MNA ρ 0.16 0.08 0.03 0.25 0.13 0.09 0.35 -0.16

MNA-PO ρ 0.18 0.11 0.04 0.30 0.22 0.12 0.41 - 0.25

p 0.04 0.28 0.73 0.000 0.09 0.27 0.000 0.12

p 0.02 0.17 0.63 0.000 0.01 0.15 0.000 0.02

Legend: TSF = triceps skinfold thickness; MAC = mid-upper arm circumference; MAMC = mid-upper arm muscle circumference; MNA = Mini Nutritional Assessment; MNA-PO = Mini Nutritional Assessment-“Proportional and Objective Score”.

TABLE 6 Overall characteristics of the studied population upon admission according to Adverse Clinical Events occurring during the follow-up

Adverse Clinical Events occurring no ACEs ACEs 139 83.2% 28 16.8%

Subjects Sex

M F

Age (years) Diseases (number per patient) Class I Co-morbidity Class II Class III Class V Class V Drugs taken (number per patient) Class A Frailty Class B Class C Points Independent Functional Status ADL Dependent in 1-2 ADLs Dependent in > 2 ADLs Lost functions to ADL (number per patient) Cognitive Status Normal SPMSQ Mildly impaired Severely impaired

31 73.8% 108 86.4% 82.7 ± 9 4±1 0 sub’s 32 66 29 11 4±2 15 sub’s 36 88 4.9 ± 2 12 sub’s 13

11 26.2% 17 13.6% 80.9 ± 6 4.3 ± 2 0 sub’s 2 11 13 0 4.5 ± 2 2 sub’s 8 18 6.3 ± 0.7 2 sub’s 1

p

χ2= 3.6

0.06

t = 1.03 t = 1.1 χ2= 11.9

0.3 0.3 0.008

t = 1.4 χ2= 0.48

0.2 0.92

t = 2.5 χ2=1.1

0.01 0.57

114

25

4.6 ± 2

4.6 ± 2

t = 0.2

0.8

35 sub’s 55 48

10 sub’s 7 11

χ2=2.4

0.3

Legend: ACE = Adverse Clinical Events; SPMSQ = Short Portable Mental Status Questionnaire; ADL = Activities of Daily Living.

TABLE 7 Overall characteristics of the studied population upon admission according to survival during the follow-up

Subjects Sex

M F

Age (years) Diseases (number per patient) Class I Co-morbidity Class II Class III Class V Class V Drugs taken (number per patient) Class A Frailty Class B Class C Points Independent Functional Status ADL Dependent in 1-2 ADLs Dependent in > 2 ADLs Lost functions to ADL (number per patient) Cognitive Status Normal SPMSQ Mildly impaired Severely impaired

Surviving 157 94% 42 100% 115 92% 82.5 ± 8 4.1 ± 1 0 sub’s 33 74 39 11 4±2 17 sub’s 44 96 4.9 ± 2 14 sub’s 14

Survival Dead 10 6% 0 0% 10 8% 81.4 ± 5 3.6 ± 2 0 sub’s 1 3 3 3 4.8 ± 2 0 sub’s 0 10 6.3 ± 0.7 0 sub’s 0

129

10

4.5 ± 2

45 sub’s 61 50

p χ2= 3.6

0.06

t = 0.39 t = 1.02 χ2= 1.5

0.7 0.31 0.7

t = 1.3 χ2= 6.1

0.2 0.1

t = 2.5 χ2= 2.1

0.01 0.3

5.9 ± 3

t = 7.6

0.000

0 sub’s 1 9

χ2= 13.9

0.001

Legend: ACE = Adverse Clinical Events; SPMSQ = Short Portable Mental Status Questionnaire; ADL = Activities of Daily Living.

TABLE 8 Nutritional status of the studied population upon admission according to Adverse Clinical Events occurring during the follow-up

Anthropometric Global Diet Subjective Objective MNA score MNA-PO score MNA Well-nourished Risk of malnutrition Malnutrition MNA-PO Well-nourished Risk of malnutrition Malnutrition Normal TSF Reduced: M < 10; F < 13.2 mm mm Normal MAC Reduced: < 22 cm cm MAMC Normal Reduced: M < 20.2; F < 18.6 cm cm > 35 g/l Albumin 30-35 < 30 g/l > 2 g/l Transferrin 150 mg/dl Cholesterol 3000 UI/l 7 < 120 mg/l Mucoprotein > 120 mg/l MNA

Adverse Clinical Events occurrence no ACEs ACEs p 2.6 ± 2 2.7 ± 1 t = 0.17 0.87 3.8 ± 1 3.6 ± 2 t = 0.92 0.36 6.7 ± 2 6.2 ± 2 t = 1.25 0.22 1.7 ± 0.7 1.7 ± 0.8 t = 0.14 0.89 1.8 ± 0.9 1.8 ± 0.8 t = 0.1 0.92 14.8 ± 4 14.1 ± 5 t = 0.78 0.44 0.54 ± 0.2 0.52 ± 0.2 t = 0.64 0.52 2 4 sub’s 0 sub’s 0.54 χ = 1.2 40 10 95 18 13 sub’s 1 sub’s 0.5 χ2= 1.4 59 11 67 16 30 sub’s 12 sub’s 0.02 χ2= 10.3 109 16 9.9 ± 5 13.6 ± 7 t = 3.4 0.001 72 sub’s 9 sub’s 0.006 χ2= 10.4 67 19 22.3 ± 3 24.6 ± 3 t = 3.3 0.01 75 sub’s 10 sub’s 0.05 χ2= 6 64 18 19.2 ± 2 69 sub’s 44 26 34.6 ± 5 59 sub’s 79 1.98 ± 0.6 95 sub’s 29 178 ± 41 90 sub’s 38 3773 ± 1295 82 sub’s 48 76 sub’s 28 104.2 ± 36

20.4 ± 2 15 sub’s 11 2 36.5 ± 4 10 sub’s 16 1.93 ± 0.4 8 sub’s 20 171.9 ± 37 20 sub’s 6 3939 ± 1445 14 sub’s 12 15 sub’s 5 103.5 ± 34

t = 2.4 χ2= 2.3

0.02 0.31

t=2 χ2= 0.17

0.04 0.68

t = 0.3 χ2= 0.33

0.74 0.56

t = 0.68 χ2= 2.4

0.5 0.12

t = 0.46 χ2= 1.5

0.5 0.48

χ2= 0.03

0.86

t = 0.09

0.93

Legend: ACE = Adverse Clinical Events; MNA = Mini Nutritional Assessment; MNA-PO = Mini Nutritional Assessment “Proportional and Objective Score”; TSF = triceps skinfold thickness; MAC = mid-upper arm circumference; MAMC = mid-upper arm muscle circumference.

TABLE 9 Nutritional status of the studied population upon admission according to survival during the follow-up

Anthropometric Global Diet Subjective Objective MNA score MNA-PO score MNA Well-nourished Risk of malnutrition Malnutrition MNA-PO Well-nourished Risk of malnutrition Malnutrition Normal TSF Reduced: M < 10; F < 13.2 mm mm Normal MAC Reduced: < 22 cm cm MAMC Normal Reduced: M < 20.2; F < 18.6 cm cm > 35 g/l Albumin 30-35 < 30 g/l > 2 g/l Transferrin 150 mg/dl Cholesterol 3000 UI/l 7 < 120 mg/l Mucoprotein > 120 mg/l MNA

Surviving 2.6 ± 2 3.9 ± 1 6.7 ± 2 1.7 ± 0.7 1.8 ± 0.9 15 ± 4 0.55 ± 0.1 4 sub’s 49 104 14 sub’s 69 74 37 sub’s 120

Survival Dead 2.6 ± 1 1.8 ± 1 4.2 ± 2 1.7 ± 0.6 1.8 ± 0.9 10.4 ± 4 0.39 ± 0.2 0 sub’s 1 9 0 sub’s 1 9 5 sub’s 5

10.2 ± 5 81 sub’s 76 22.5 ± 3 32 sub’s 54

16 ± 6 0 sub’s 10 25.2 ± 2 3 sub’s 7

19.3 ± 2 78 sub’s 54 25 34.8 ± 4 67 sub’s 89 1.99 ± 0.6 108 sub’s 35 176.3 ± 41 105 sub’s 39 3389 ± 1646 92 sub’s 54 89 sub’s 29 101.2 ± 32

20.1 ± 0.9 6 sub’s 1 3 36.3 ± 6 2 sub’s 6 1.67 ± 0.3 7 sub’s 2 181.3 ± 29 5 sub’s 5 3389 ± 1646 4 sub’s 6 2 sub’s 4 160.5 ± 52

t = 0.55 t = 4.8 t = 3.9 t = 0.1 t = 0.02 t = 3.28 t = 3.22 χ2= 2.5

p 0.96 0.000 0.000 0.93 0.98 0.001 0.002 0.3

χ2= 6.95

0.03

χ2= 9.9

0.02

t = 3.5 χ2= 10

0.01 0.07

t = 2.4 χ2= 8.2

0.02 0.02

t = 1.04 χ2= 3

0.8 0.22

t = 0.99 χ2= 1

0.32 0.32

t = 2.8 χ2= 0.02

0.02 0.88

t = 0.36 χ2= 2.4

0.71 0.12

t = 1.02 χ2= 4.7

0.3 0.1

χ2= 3

0.22

t = 4.25

0.000

Legend: ACE = Adverse Clinical Events; MNA = Mini Nutritional Assessment; MNA-PO = Mini Nutritional Assessment “Proportional and Objective Score”; TSF = triceps skinfold thickness; MAC = mid-upper arm circumference; MAMC = mid-upper arm muscle circumference.

References 1. Volkert D, Hubsch S, Oster P, Schlierf G: Nutritional support and functional status in undernourished geriatric patients during hospitalisation and 6 month follow up. Aging Clin Exp Res 1996, 8, 386 2. Guigoz Y, Vellas B, Garry PJ: The Mini Nutritional Assessment: a practical assessment tool for grading the nutritional state of elderly persons (Third Edition). Facts, research and intervention in gerontology 1997. Serdi Publishing Company – Paris (France) pag. 15-60 3. Gazzotti C, Albert A, Pepinster A, Petermans J: Clinical usefulness of the Mini Nutritional Assessment scale in geriatric medicine. J Nutr Health Aging 2000, 4, 176 4. Lebreton B, Hazif-Thomas C, Thomas P: Etude du statut nutritionnel des résidents en long séjour par les marqueurs biologiques, anthropométriques et diététiques. Age & Nutrition 1997, 8, 22 5. Vellas B, Guigoz Y, Garry PJ, Nourhashemi F, Bennahum D, Lauque S, Albarede JL: The Mini Nutritional Assessment and its use in grading the nutritional state of elderly patients. Nutrition 1999, 15, 116 6. Salva A, Bolibar I, Munnoz M, Sacristan V: Results of the Mini Nutritional Assessment in nursing home residents. JAGS 1996, 44, P11 7. Donini LM, De Felice MR, Tagliaccica A, Palazzotto A, De Bernardini L, Cannella C: Valeur prédictive du MNA en long séjour gériatrique. Age & Nutrition 2000, 11, 1, 3-11 8. Donini LM, De Felice MR, Tassi L, De Bernardini L, Pinto A, Giusti AM, Cannella C: A "Proportional and Objective Score"for the Mini Nutritional Assessment in Long-Term Geriatric Care (J Nutr Health Aging in press, 2002, 1) 9. Pfeiffer E: Short Portable Mental Status Questionnaire JAGS 1975, 23, 433 10. Katz S: Progress in the development of the index of ADL. Gerontologist, 1970,1,20-30 11. Palmalee PA, Thura PD, Katz IR, Lawton MP: Validation of the cumulative illness rating scale in a geriatric residential population. JAGS 1995, 43, 130 12. Rozzini R, Frisoni GB, Bertozzi B, Barbisoni P, Franzoni S, Trabucchi M: Gli indici di comorbilità. Giorn Geront 1996, 44, 735 13. Trabucchi M, Franzoni S, Bertozzi B, Barbisoni P, Rozzini R: Metodologie di classificazione dei ricoveri ospedalieri degli anziani alternative ai DRG. Giorn Geront 1996, 44, 75 14. Bernardini B, Meinecke C, zaccarini C, Bongiorni N, Fabbrini S, Gilardi C, Bonaccorso O, Guaita A: adverse clinical events in dependent long-term nursing home residents. JAGS 1993, 41, 105 15. Jelliffe DB: The assessment of the nutritional status of the community. WHO Monograph Ser. N° 53, Genève 1966 16. Linee Guida per l’impiego della nutrizione parenterale ed enterale nei pazienti adulti ospedalizzati. Riv It Nutr Parent Ent 1995, 13, S2, 2-7 17. Lohman TG, Roche AF, Martorell R: Manuale di riferimento per la standardizzazione antropometrica. EDRA, Milano 1992 18. Joosten E, Vanderelst B, Pelemans W: The effect of different diagnostic criteria on the prevalence of malnutrition in a hospitalised geriatric population. Aging Clin Exp Res 1999, 11, 390 19. Allison SP: Définition et origine de la malnutrition. Cah Nutr Diét 2000, 35, 3, 161 20. Lerebours E, Déchelotte P: Prévention et traitement de la dénutrition Cah Nutr Diét 2000, 35, 3, 177 21. Mowé M, Bohmer T, Kindt E: Reduced nutritional status in eledrly population is probable before disease and possibly contributes to the development of disease. Am J Clin Nutr 1994, 59, 317 22. Lauque S, Baudouin M, Nourhashemi F, Guyonnet S, Guigoz Y: Letter to the Editor. J Nutr Health Aging 1998, 2, 61 23. Compan B, Di Castri A, Plaze JM, Arnaud-Battandier F: Epidemiological study of malnutrition in elderly patients in acute, subacute and long-term care using the MNA. J Nutr Health Aging 1999, 3, 3, 146 24. Lauque S, Arnaud-Battandier F, Mansourian R, Guigoz Y, Paintin M, Nourhashemi F, Vellas B: Effectiveness of protein-energy oral supplementation in malnourished nursing home resident. Age and Aging 1999 (cited in Compan B et al: J Nutr Health Aging 1999, 3, 3, 146) 25. Consensus Conference: “L’integrazione dei Servizi Geriatrici”, CNR-Progetto Finalizzato “Invecchiamento”, Ostuni (Bari), Italy, april 1995 26. Beck AM, Ovesen L, Osler M: The “Mini Nutritional Assessment” and the “Determine your Nutritional Health” checklist as predictors of morbidity and mortality in an elderly Danish population. Br J Nutr 1999, 81 (1) 31-6 27. Frisoni GB, Franzoni S, Rozzini R, Ferrrucci L, Boffelli S, Trabucchi M: Food intake nd mortality in the frail elderly. J Gerontol 1995, 50A, M203 28. Lipski PS, Torrance A, Kelly PJ, James OFW: A study of nutritional deficits of long stay geriatric patients. Age and Ageing 1993, 22, 244 29. Sullivan DH, Walls RC: Impact of nutritional status on morbidity in a population of geriatric rehabilitation patients JAGS 1994, 42, 471 30. Morley JE: Anorexia of aging. Am J Clin Nutr 1997, 66, 760 31. Berlowitz DR, Brandeis GH, Anderson J, Du W, Brand H: Effect of pressure ulcers on the survival of long term care residents. J Gerontol 1997, 52A, M106 32. Griep MI, Mets TF, Collys K, ponjaert-Kristoffersen I, Massart DL: Risk of malnutrition in retirement homes elderly persons measured by the MNA. J Gerontol 2000, 55A, M57

33. Lesourd B: Evaluation de l’état nutritionnel du sujet âgé. Cah Nutr Diét 1999, 34, 387 34. Ferry M: L’alimentation des personnes âgées. La Lettre Scientifique de l’Institut Français pour la Nutrition. n° 19, 1993, 19