Filetype PDF | Posted on 04 Jan 2023 | 3 years ago
Authentication
digital resources
273x Filetype PDF File size 0.13 MB Source: www.nutricionhospitalaria.com
36. ASSESSING:01. Interacción 22/02/12 11:58 Página 590
Nutr Hosp. 2012;27(2):590-598
ISSN 0212-1611 • CODEN NUHOEQ
S.V.R. 318
Original
Assessing risk screening methods of malnutrition in geriatric patients;
Mini Nutritional Assessment (MNA) versusGeriatric Nutritional Risk
Index (GNRI)
1,4 2 3,4 1,4 1,4
P. Durán Alert , R. Milà Villarroel , F. Formiga , N. Virgili Casas and C. Vilarasau Farré
1Unidad de Dietética y Nutrición Clínica. Hospital. Hospital Universitario de Bellvitge. Hospitalet de Llobregat. Barcelona.
2
España. Departamento de Salud Pública. Facultad de Medicina. Universidad de Barcelona (UB). Barcelona. España.
3Unidad de Geriatría. Servicio de Medicina Interna. Hospital Universitario de Bellvitge. Hospitalet de Llobregat. Barcelona.
4
España. Institut de Recerca Biomédica Bellvitge (IDIBELL). Hospital Universitario de Bellvitge. Hospitalet de Llobregat.
Barcelona. España.
Abstract EVALUACIÓN DE LOS MÉTODOS DE CRIBAJE
Introduction:Elderly subjects are considered a vulner- DE RIESGO NUTRICIONAL EN PACIENTES
able group and they have more risk of nutritional prob- GERIÁTRICOS; MINI NUTRITIONAL
lems. The risk of malnutrition increases in hospitalized ASSESSMENT (MNA) VERSUSGERIATRIC
geriatric patients. NUTRITIONAL RISK ASSESSMENT (GNRI)
Objectives: To compare the correlation between MNA Resumen
and GNRI with anthropometric, biochemical and Barthel
Index in hospitalized geriatric patients and to test the Antecedentes: La población anciana esta considerada
concordance between MNA and GNRI and between Mini como un colectivo vulnerable a sufrir problemas nutricio-
Nutritional Assessment Short Form (MNA-SF) and nales. Entre estos, los ancianos hospitalizados tienen aun
MNA. un mayor riesgo a sufrir malnutrición.
Methods: It was a cross-sectional study on a sample of Objetivos: Los objetivos de este estudio fueron compa-
40 hospitalized geriatric patients. For determination rar el grado de correlación entre dos índices de cribaje
nutritional status we used MNA and GNRI; we evaluated nutricional, el Mini Nutritional Assessment (MNA) y el
the correlation between this both test with biochemical Geriatric Nutritional Risk Index (GNRI) con los paráme-
and anthropometric parameters and functional question- tros antropométricos, bioquímicos, el índice de Barthel y
naires. We used Pearson’s simple correlation model, one- ciertas patologías relacionadas con el estado nutricional
way ANOVA and multiple logistic regression to evaluate (infecciones y úlceras por presión).
the relationship between MNA and GNRI. Metodología: Se llevó a cabo un estudio transversal en
Results: According to MNA, 17 patients (42.5%) were una muestra de 40 pacientes hospitalizados en una uni-
malnourished and according to GNRI, 13 patients dad geriátrica de agudos. Para la determinación del
(32.5%) had high risk of nutritional complications. The estado nutricional se usaron los índices del MNA y el
concordance of MNA and GNRI was 39% and between GNRI. Se evaluó la correlación entre los parámetros bio-
MNA-SF and MNA was 81%. The most significant differ- químicos, antropométricos, parámetros funcionales y
ences were detected in weight, BMI, arm and calf circum- problemas nutricionales relacionados con la malnutri-
ference and weight loss parameters. Barthel index was ción (úlceras por presión y infecciones). Para el modelo de
significantly different in both tests. The MNA and GRNI correlación, se utilizó el grado de correlación de Pearson;
had significant correlations with albumin, total protein, para estudiar la relación entre los índices nutricionales
transferring, arm and calf circumference, weight loss and (MNA y GNRI) y los diferentes parámetros se utilizó un
BMI parameters. análisis de la variancia y un modelo de regresión logística.
Conclusions: In conclusion, it would be reasonable to Resultados: De acuerdo con el MNA, 17 pacientes
use GRNI in cases where MNA is not applicable, or even (42,5%) estaban desnutridos y de acuerdo con GNRI, 13
use GRNI as a complement to MNA in hospitalized pacientes (32,5%) tenían alto riesgo de complicaciones
nutricionales. La concordancia de la MNA y la GNRI fue
del 39% y entre MNA-SF y MNA fue de 81%. Las dife-
Correspondence: Raimon Milà Villarroel. rencias más significativas se detectaron en el peso, el
Profesor de Nutrición Comunitaria y Salud Pública. IMC, el brazo y circunferencia de la pantorrilla y los
Unidad de Bioestadística. Departamento de Salud Pública. parámetros de pérdida de peso. El MNA y GRNI mostró
Facultad de Medicina. Universidad de Barcelona. correlaciones significativas con la albúmina, proteínas
C/ Casanova, 143. totales, la transferencia, la circunferencia del brazo y de
08036 Barcelona (Spain).
E-mail: rmila@ub.edu la pantorrilla, con el % de pérdida de peso y el índice de
Recibido: 10-X-2011. masa corporal (IMC). Los pacientes malnutridos según el
1.ª Revisión: 23-XI-2011. MNA y los pacientes con riesgo elevado según el GNRI
Aceptado: 23-XI-2011. tenían mayor riesgo de sufrir úlceras por presión.
590
36. ASSESSING:01. Interacción 12/03/12 12:52 Página 591
elderly patients. There is no reason why they should be Conclusiones:en conclusión, sería razonable utilizar el
deemed incompatible, and patients could benefit from GNRI en los casos en que el MNA no fuera aplicable, o
more effective nutritional intervention. incluso utilizar GNRI como complemento al MNA en
(Nutr Hosp. 2012;27:590-598) pacientes ancianos hospitalizados. No hay ninguna razón
DOI:10.3305/nh.2012.27.2.5635 por la cual se deban considerar incompatibles, y los
pacientes podrían beneficiarse de una intervención nutri-
Key words: Elderly hospitalized. MNA. GNRI. Nutritional cional más efectiva.
Assessment. (Nutr Hosp. 2012;27:590-598)
DOI:10.3305/nh.2012.27.2.5635
Palabras clave: Ancianos hospitalizados. MNA. GNRI.
Evaluación nutricional.
Introduction tured to give greater weight to plasma albumin than to
patients’ weight and cut-off points are used to predict
The elderly are considered one of the most heteroge- health problems in the subsequent months.20
neous and vulnerable groups, with an increased risk of The aim of this study is to compare the correlation
1-4
imbalances, deficiencies and nutritional problems. between MNA and GNRI with anthropometric,
Physiological and social changes resulting from biochemical, functional status measure (Barthel Index)
advanced age, high consumption of drugs, chronic and nutritional relation complications (such as infection
illness and/or degenerative loss of mobility, psycholog- and bedsores) in a sample of older subjects admitted to
ical distress and loss of appetite are just some of the hospital. The second objective was to test the concor-
5-
factors that influence the nutritional status of this group. dance between these two methods of assessment and
11 The consequences of malnutrition in the group result in between MNA short form and complete MNA.
an increase in the prevalence of infections, longer-stay
hospitalizations and increased morbidity and mortality.
Malnutrition is not readily recognizable or distinguish- Materials and methods
able from the changes of the aging process, which means
12
that a significant percentage of cases are undiagnosed. We performed a single centre cross-sectional study
Indicators for diagnosing risk of malnutrition include on a sample of 40 consecutive acute geriatric patients
nutritional parameters, anthropometric, haematological, admitted during the three-month study period
biochemical and health conditions and associated (February 2010-April 2010). The study was performed
diseases.13 There are many indices for assessing nutri- at the Acute Geriatric Ward (AGW) of the University
tional status in the elderly population, though the method Hospital of Bellvitge, Spain. The study included all
recommended by the European Society of Parenteral and patients over the age of 74 who were admitted to the
Enteral Nutrition (ESPEN) is the Mini Nutritional AGW. Exclusion criteria were: the presence of well-
Assessment (MNA).14-17 The MNA is the method most known liver disease, neoplasic disorders or terminal
commonly used for assessing the nutritional status of condition. At the time of admission to the AGW, each
older people. It was designed to evaluate and identify patient was evaluated for the presence of diseases asso-
those elderly people who are malnourished or at risk of ciated with nutritional status (dyslipidemia, diabetes,
same, in order to intervene as soon as possible and pressure ulcers and high blood pressure).
improve their prognosis.18 A short form of MNA exists Blood samples were obtained within 24-48 hours
(MNA-SF) which is used with malnutrition screening after admission for determination of serum proteins
tests. We should bear in mind that it is not applicable to (albumin, total proteins, C-reactive protein), renal
those patients diagnosed with dementia or other commu- function parameters (creatinine) and other biochemical
16
nication problems. However, the difficulty in achieving parameters (iron, ferritin, transferrin, hematocrit and
a regular size or weight in patients has resulted in the use haemoglobin).
of an index devised to investigate and predict complica- Experienced operators collected anthropometric
tions related to nutritional status in the elderly: the Geri- data: weight (to the nearest 0·1 kg using the same cali-
19.20
atric Nutritional Risk Index (GNRI). brated scale), standing height or knee-height (for
The GNRI index is a modification of the NRI (Nutri- stature prediction in the bedridden) and mid-upper arm
21
tional Risk Index) in which the value of “normal and calf circumferences (to the nearest 0.5 cm using a
weight patients” replaces the original formula of “ideal flexible tape). Estimated height (EH) was extrapolated
weight patients” (calculated from Lorentz’s formula) from knee-heel length according to the equations vali-
19 22
to be applied in the geriatric population. This index dated by Chumlea et al. Body mass index (BMI) was
takes into account two main parameters: serum calculated for all patients. Ideal body weight, necessary
albumin and the ratio between the current weight and for GNRI determination, was derived by using the
ideal weight of the individual. GNRI formula is struc- following equations of Lorentz:
Assessment of two methods Nutr Hosp. 2012;27(2):590-598 591
of nutritional screening
36. ASSESSING:01. Interacción 22/02/12 11:58 Página 592
*Ideal weight for men = height (cm) –100 [(height –150/4)] from wheelchair to bed and returning, doing one’s
personal toilet, getting on and off toilet, bathing self,
*Ideal weight for women = height (cm) –100 [(height –150/2,5)] walking on level surface, ascending and descending
stairs, dressing, controlling bowels and controlling
Weight loss in the previous three months was esti- bladder. Scoring ranges from 0 (completely dependent)
mated by interviewing patients and family members of to 100 (completely independent) and includes the cate-
each patient. gories of response between 2 and 4 alternatives, with
27
intervals of 5 points.
Mini Nutritional Assessment
Statistical analyses
The MNA is based on 18 items, including anthropo-
metric and dietary parameters. It is used to assess func- Data are presented as mean values and standard devia-
23-
tional status in elderly patients and to predict mortality. tions. We evaluated the relationship between the vari-
25 Baseline nutritional status was defined and graded ables and both the MNA and GNRI using Pearson’s
according to MNA and MNA-SF. This tool consists of simple correlation model, and we compared groups for
eighteen questions grouped in four rubrics addressing the quantitative variables using one-way ANOVA. Control
areas of anthropometry (BMI, weight loss, mid-upper for overall type I error was performed using the Bonfer-
arm and calf circumferences), general state (medication, roni post hoc comparison test. Patients were categorized
mobility, presence of pressure ulcers, lifestyle, and pres- and a severity score was assigned according to nutrition
ence of psychological stress or neuropsychological prob- state based on the MNA (MNA < 17 = 0; 17-23, 5 =1; ≥
lems), dietary assessment (autonomy of feeding, quality 24 = 2) and to nutrition risk as defined by the GNRI
and number of meals, fluid intake) and self-perception (GNRI < 92 = 0; 92-98 = 1, ≥ 98 = 2). We used the 2(Chi
regarding health and nutrition, respectively. A maximal squared test) or Fisher’s exact test (used when expected
score of thirty points is achievable on this questionnaire, values were < 5) to compare prevalence between nutri-
while threshold values are set as follows: adequately tional classes and Cohen’s kappa test to analyse the
nourished, MNA ≥ 24; at risk of malnutrition, MNA agreement between the assessment methods. To evaluate
between 17-23·5; and protein-energy malnourished, the association with the presence of disease related to
MNA < 17. nutritional status (bedsores) of both these tools, we calcu-
lated OR and 95% CI; for each calculation, the unex-
posed patients were those with a severity score = 2
Geriatric Nutritional Risk Index (GNRI ≥ 98 and MNA ≥ 24, respectively). In addition,
we carried out multiple nominal logistic regression
Nutritional risk of health complications was analyses to test independent associations. All statistical
assessed by the GNRI score through the equation of analyses were performed by SPSS 16.0 (2008, SPSS, Inc,
13
Bouillanne et al.: Chicago, IL). The level of significance was established as
a two-sided p-value = 0.05.
current weight (kg)
GRNI = 1,519×Albumin (g/l) + 41. 7× –––––––––––––––––––
ideal weight (kg)
Results
Categorization of the patients was performed
according to the following cut-offs: severe/moderate Baseline characteristics
risk, < 92; low risk, 92-98; no risk > 98. In the present
study we utilized the modification proposal devised by The sample comprised 29 (72.5%) female and 11
Cereda et al.26 The category of moderate risk (GNRI 92 (27.5%) men with a mean (± SD) age of 84.6 (± 5.59) and
to 98) and severe risk (GNRI < 92) have been included 83.45 (± 7.91) years, respectively. The major cause of
in one single category because these two categories hospitalization was acute heart failure (45% of cases) and
have been shown to present a similar increased risk exacerbation of chronic pulmonary disease (15%). The
(OR) of overall health complications and of those other most commonly associated comorbidity were: hyperten-
than mortality (bedsores or infections). Furthermore, sion (80%), pressure ulcers) (35%), dyslipidemia
this categorization enables us to obtain a three-cate- (32.5%), diabetes (25%) and depression (15%).
gory tool similar to the MNA.
Nutritional assessment scores
Barthel Index
The scores for each patient in the MNA and GNRI
The Barthel Index (BI) consists of 10 items that assess can be observed in figure 1. Statistical analyses showed
the patient’s ability to perform certain activities without differences in the scores of each group. The groups
help. It evaluates abilities such as feeding self, moving with the lowest scores were those with worse prognosis
592 Nutr Hosp. 2012;27(2):590-598 P. Durán Alert et al.
36. ASSESSING:01. Interacción 22/02/12 11:58 Página 593
200,00 30,00
150,00
20,00
100,00
GRNI score GRNI score
10,00
50,00
0,00 0,00
High risk Low risk No risk Malnutrition Risk malnutrition Well nourished
Geriatric nutritional risk index Mini nutritional assessment
Malnutrition (n = 17) Risk malnutrition (n = 13) Well nourished (n = 10) ANOVA Correlation
Mean SD Mean SD Mean SD p-value MNA vs GRNI
MNA 14.7647 1.99309 20.5769 1.80100 24.9500 1.23491 < 0.001** 0.673**
High risk (n = 13) Low risk (n = 8) No risk (n = 19) ANOVA Correlation
Mean SD Mean SD Mean SD p-value GRNI vs MNA
GRNI 80.0723 8.71111 95.8702 1.64769 111.1083 13.31489 < 0.001** 0.673**
*p < 0.05; **p < 0.001.
Fig. 1.—Scores for the two nutritional risk assessments. Geriatric Nutritional Risk Index (GNRI) and Mini Nutritional Assessment (MNA).
and risk of malnutrition in the MNA and GNRI (fig. 1). was approximately 39% (Kappa index = 0,393, p-value =
According to the MNA, 17 patients (42.5%) were < 0,001) (table I). However, the concordance between
malnourished, 13 patients (32.5%) were at risk of MNA short form and complete MNA was 81% (k =
malnutrition and 10 (25%) were well-nourished. 0,810, p-value = < 0,001) (table II).
According to the GNRI test, 13 patients (32.5%) had
high risk of complications related to nutrition, 8
patients (20%) had moderate risk of complications and Biochemical, anthropometric
19 patients (47.5%) were not at risk of nutritional and functional parameters
complications. Although both tests have good correla-
tion (r = 0.673, p = 0.002), discrepancies exist in the Results of a one-way analysis of variance and
classification of patients. The concordance of both tests analysis of linear correlation between anthropometric,
Table I
Distribution of the population among nutritional classes according to the Mini Nutritional Assessment (MNA) and the
Geriatric Nutritional Risk Index (GNRI)
MNA vs GRNI
MNA
Malnutrition Risk malnutrition Well nourished a,b
(MNA < 17) (MNA 17-23.5) (MNA > 24) Total
GRNI
High risk (< 92) n 11 (64.7%) 1 (7.7%) 1 (10.0%) 13 (32.5%)
Low risk (92-98) n 5 (29.4%) 3 (23.1%) 0 (0.0%) 8 (20.0%)
No risk ( > 98) n 1 (5.9%) 9 (69.2%) 9 (90.0%) 19 (47.5%)
Total n 17 (100.0%) 13 (100.0%) 10 (100.0%) 40 (100.0%)
a
Exact Fisher’s Chi square = 23.553, p-value = < 0.001.
bKappa index = 0.393, p-value = < 0.001.
Assessment of two methods Nutr Hosp. 2012;27(2):590-598 593
of nutritional screening
no reviews yet
Please Login to review.
