Anthropometric evaluation, risk factors for malnutrition, and nutritional therapy for children in teaching hospitals in Brazil

0021-7557/09/85-03/223
Jornal de Pediatria
Copyright © 2009 by Sociedade Brasileira de Pediatria
Original article
Anthropometric evaluation, risk factors for malnutrition, and
nutritional therapy for children in teaching hospitals in Brazil
Roseli O. S. Sarni,1 Maria de Fátima C. C. Carvalho,2
Cristina M. G. do Monte,3 Zuleica P. Albuquerque,4 Fabíola I. S. Souza5
Abstract
Objective: To evaluate risk factors for malnutrition, nutritional status and nutritional support provided in
hospitalized children.
Methods: This longitudinal study prospectively followed, for 3 consecutive months, all children under 5 years
of age (n = 907) hospitalized in general pediatric medical wards of 10 Brazilian university-based hospitals. For data
collection, a standard questionnaire was used and nutritional condition was evaluated at hospital admission and
discharge: weight-for-height, weight-for-age and height-for-age z score.
Results: Only 56.7% of the children had their nutritional classification documented in the medical record. At
hospital admission, 16.3 and 30.0% of the children had moderate/severe malnutrition and low stature, respectively.
Risk of malnutrition was associated with low birth weight and younger age. A high percentage of nutritional deficiencies
was observed in the children analyzed, although child’s nutritional condition and the adoption of appropriate nutritional
therapy were not documented in the medical records of the malnourished children.
Conclusion: These data underscore the importance of developing qualified hospital medical wards regarding
diagnosis and therapeutic approach to malnutrition, based on the conduct guidelines already available in Brazil.
J Pediatr (Rio J). 2009;85(3):223-228: Protein-energy malnutrition, hospitalized children and nutritional
therapy.
Introduction
Over recent decades, the prevalence of protein-energy
pediatric population still showed anthropometric indicators
malnutrition (PEM) has expressively decreased among
consistent with PEM, the height-for-age deficit being the
children aged < 5 years worldwide.1 The same trend has been
most common form of malnutrition in the country.2
observed in Brazil; however, data from the latest Brazilian
Childhood PEM is a multifactorial condition, which
National Demographic and Health Survey (Pesquisa Nacional
involves biological and social determinants, and may
de Demografia e Saúde, PNDS) revealed that 22% of the
occur due to insufficient supply of energy, macro- and
1. Pediatra. Doutora, Medicina, Universidade Federal de São Paulo – Escola Paulista de Medicina (UNIFESP-EPM), São Paulo, SP, Brazil. Médica assistente,
Departamento de Pediatria, UNIFESP-EPM, São Paulo, SP, Brazil. Professora assistente, Departamento de Pediatria, Faculdade de Medicina do ABC (FMABC),
Santo André, SP, Brazil. Membro, Departamento Científico de Nutrologia, Sociedade Brasileira de Pediatria.
2. Nutricionista. Mestre, Nutrição Humana, Universidade de Brasília (UnB), Brasília, DF, Brazil. Especialista, Saúde Coletiva, UnB, Brasília, DF, Brazil. Pesquisadora
associada, Observatório de Políticas de Segurança Alimentar e Nutrição (OPSAN), UnB, Brasília, DF, Brazil.
3. Pediatra. Doutora, Nutrição Humana, London School of Hygiene and Tropical Medicine (LSHTM), London, United Kingdom. Pesquisadora, Nutrição Materno-
Intanfil, Unidade de Pesquisas Clínicas, Faculdade de Medicina, Universidade Federal do Ceará (UFC), Fortaleza, CE, Brazil.
4. Pediatra. Mestre, Nutrição Humana, LSHTM, London, United Kingdom. Consultora nacional, Nutrição, Organização Pan-Americana da Saúde, 1993 a 2007.
5. Mestre, Ciências, UNIFESP-EPM, São Paulo, SP, Brazil. Professora colaboradora, Departamento de Pediatria, FMABC, Santo André, SP, Brazil.
This study was conducted in the Working Group on Child Malnutrition, Technical Area of Food and Nutrition, Health Policy Secretary (Grupo de Trabalho sobre
Desnutrição Infantil, Área Técnica de Alimentação e Nutrição, Secretaria de Políticas de Saúde), Brazilian Ministry of Health, and Brazilian Society of Pediatrics.
No conflicts of interest declared concerning the publication of this article.
Suggested citation: Sarni RO, Carvalho MF, do Monte CM, Albuquerque ZP, Souza FI. Anthropometric evaluation, risk factors for malnutrition, and nutritional
therapy for children in teaching hospitals in Brazil. J Pediatr (Rio J). 2009;85(3):223-228.
Manuscript received Dec 22 2008, accepted for publication Mar 03 2009.
doi:10.2223/JPED.1890
223

---

224 Jornal de Pediatria - Vol. 85, No. 3, 2009
Nutritional status and therapy in hospitalized children - Sarni RO et al.
micronutrients (a condition called primary nutrition) or as
Methods
a result of secondary malnutrition, which is characterized
This longitudinal study prospectively evaluated all
by an inadequate functional and biological use of the
children, aged between 28 days and 5 years, hospitalized
available nutrients or by increased energy expenditure in
in general pediatric medical wards of 10 university-based
the presence of associated diseases. Regardless of the
or teaching hospitals covering four regions of the country
cause, PEM is related to physiological alterations that
(Northeast, Southeast, Center-West, and South) and a total
have been reported to increase risk of complications and
of nine Brazilian capitals (Fortaleza, Natal, Recife, Salvador,
death in children.3
Rio de Janeiro, Belo Horizonte, Brasília, São Paulo, and
Worth mentioning that, despite the decreased prevalence
Porto Alegre). Data collection was performed simultaneously
of PEM, death rates due to severe malnutrition in hospital
in all the involved institutions for a 3-month period. Data
treatment remain high.4
were collected by health professionals, undergraduate or
graduate students of all institutions, who were previously
In order to reduce mortality rates, the World Health
trained by members of their institutions, also members
Organization (WHO) released, in 1999, the first version
of BMH Working Group on Child Malnutrition, on how to
of the protocol for in-hospital treatment of severely
correctly fill in the questionnaire and obtain anthropometric
malnourished children, which takes the pathophysiological
data. In order to ensure standardization of anthropometric
characteristics of this population into consideration. The
measurements, a manual with procedures was made
actual implementation of such protocol in hospital centers
available to the person in charge of each unit.
of several countries resulted in an expressive decrease in
death rates due to severe PEM.5-7
A standard and pre-encoded questionnaire was used for
data collection, which should be filled in within 48 hours of
Unfortunately, nutritional assessment and follow up of
hospital admission and, at most, 48 hours before hospital
children during hospitalization is not a routine in in-hospital
discharge, taking into account discharges during weekends.
treatment, which might hinder diagnosis and treatment of
This questionnaire included patient’s identification data
nutritional deficiencies.8,9 Another issue concerns the fact
(name, birth date, sex, and origin), history (birth
that the diagnosis of malnutrition is not always documented
conditions, birth weight, gestational age, breastfeeding,
in the child’s medical record at hospital admission or death
and immunization), hospital admission (main diagnosis
certificate, hindering a correct and comprehensive evaluation
and lag time, nutritional condition at hospital admission
of the problem.
and discharge, presence of edema or dehydration), and
As a consequence of the seriousness of this issue and
nutritional therapy employed (diet, administration route, use
little data available in the country, the Brazilian Ministry of
of multivitamin supplements, oligoelements and iron).
Health (BMH) informally established, in 2000, the Working
Anthropometric measurements (weight and height)
Group on Child Malnutrition. This group, under the shared
and classification of nutritional condition were performed
coordination of BMH, Pan American Health Organization
by health professionals of each institution, previously
and Brazilian Society of Pediatrics, worked together with
trained by the research coordinators, following WHO
the Technical Area of Food and Nutrition, of the BMH
recommendations.10 Weight-for-height (ZWH), weight-for-
Health Policy Secretary, in order to monitor and improve
age (ZW) and height-for-age (ZH) indicators were calculated
activities related to the implementation of WHO protocol,
as z scores, based on the reference curve proposed by
as well as to provide technical and scientific support to
WHO.11 Malnutrition and low stature were considered as ZWH
its implementation. However, the creation of this group,
< -2 and ZH < -2, respectively.12 Concerning nutritional
which was meant to be an advisory body, was not formally
support, health care centers followed their own standardized
approved by BMH and, thus, worked in an informal manner
conduct, and no previous training on nutritional therapy
throughout 1999-2000. The group comprised 13 hospital
was performed in the units involved.
institutions, among other partners.
The data collected were entered on databases, with
Within the strategies to guide and adjust WHO protocol
double typing and subsequent validation regarding internal
and its implementation in Brazil, this advisory group
consistency. The Statistical Package for the Social Sciences
proposed carrying out the present study. The purpose of
(SPSS) 13.0 was used for statistical analysis. During the
this study was to evaluate risk factors for malnutrition, to
study period, 959 questionnaires were returned; of these,
assess nutritional status and to describe nutritional support
52 were excluded: five because sex was not informed; four
provided in hospitalized children, at hospital admission
because child weight at admission was not informed; 16
and discharge, in university-based or teaching hospitals
because birth date was not informed; 19 because child was
of the Brazilian National Health System (Sistema Único
under 28 days of age; and eight because anthropometric
de Saúde, SUS), in addition to identify methods and
measurements were inaccurate and generated, for some
anthropometric indicators used for nutritional assessment
indicators, z score values < or > 6. Therefore, our final study
of children admitted to these institutions.
sample comprised data on 907 hospitalized children. For

---

Nutritional status and therapy in hospitalized children - Sarni RO et al.
Jornal de Pediatria - Vol. 85, No. 3, 2009 225
comparative analysis of anthropometric variables, at hospital
(1-59.7 months); and mean hospitalization time was 6.9
admission and discharge, we considered for ZWH, ZW and
days (2-58 days). Prevalence of low birth weight and
ZH data from 787, 867 and 785 children, respectively.
prematurity was 151/907 (16.6%) and 123/907 (13.6%),
To compare categorical variables, the chi-square test
respectively. Approximately 241/860 (28%) children were
was used, and risk of association between variables was
no longer on or had never been breastfed. The most
calculated using relative risk (RR). A stepwise forward
frequent diagnosis at admission was pneumonia, 429/907
logistic regression model was used in the multivariate
(42.3%), followed by diarrheic disease, 156/907 (17.2%)
analysis of risk factors for malnutrition, and Pearson’s
(Table 1). None of the children had malnutrition as the
coefficient was used to evaluate the correlation between
main cause of hospital admission.
time of hospitalization and nutritional condition at hospital
The presence of malnutrition at admission, among all
discharge. We adopted ? < 0.05. The study protocol
children analyzed, was associated with age under 12 months
was approved by the Research Ethics Committee of the
(RR = 2.3; 95%CI 1.63-3.27), low birth weight (RR = 1.66;
institutions participating in the study, and the children’s
95%CI 1.18-2.35), prematurity (RR = 1.69; 95%CI
parents or legal guardians were explained about all study
1.18-2.42), and incomplete immunization (RR = 1.47;
procedures and signed a free informed consent form.
95%CI 1.04-2.09). Prevalence of diarrheic disease, as a
diagnosis at admission, was also higher among malnourished
children than among those without malnutrition, who had
Results
pneumonia as the prevalent diagnosis (RR = 2.48; 95%CI
In the study population, there was male predominance,
1.74-3.53). The logistic regression model included age,
517 individuals/907 (56.2%); mean age was 10.5 months
sex, birth weight, gestational age, immunization, and
Table 1 -
Population characteristics

ZWH < -2
ZWH ? -2

(n = 142)
(n = 730)
RR
Adjusted RR
Variables
n (%)
n (%)
(95%CI)
(95%CI)*
Age (n = 907)


2.31 (1.63-3.27)†
0.96 (0.95-0.98)
? 6 months
66 (46.7)
209 (28.6)
6 to 12 months
38 (26.8)
160 (21.9)
12 to 24 months
24 (16.9)
170 (23.3)
24 to 36 months
8 (5.6)
87 (11.9)
> 36 months
6 (4.2)
104 (14.2)
Sex (n = 907)


1.04 (0.77-1.41)
Male
82 (57.7)
413 (56.6)
Female
60 (42.2)
317 (73.4)
Birth weight (n = 777)


1.66 (1.18-2.35)
1.77 (1.09-2.86)
< 2,500 g
35 (28)
107 (17.1)
? 2,500 g
90 (72)
517 (82.8)
Gestational age (n = 869)


1.69 (1.18-2.42)
Preterm
29 (21.2)
86 (12.3)
Term
108 (78.8)
614 (87.7)
Breastfeeding (n = 860)


1.17 (0.84-1.63)
No/never
42 (31.1)
189 (27.1)
Yes
93 (69.8)
507 (72.8)
Immunization (n = 809)


1.47 (1.04-2.09)
1.50 (0.95-2.37)
Incomplete
36 (29.5)
136 (20.7)
Complete
86 (70.5)
520 (79.3)
Diagnosis at admission (n = 907)


2.48 (1.74-3.53)‡
Diarrheic disease
44 (31)
100 (13.7)
Pneumonia
52 (36.6)
370 (50.7)
Sepsis
2 (1.4)
9 (12.3)
Meningitis
0
10 (13.7)
Cardiopathy
6 (4.2)
5 (0.7)
Other
38 (26.7)
236 (32.3)
95%CI = 95% confidence interval; RR = relative risk; ZWH = weight-for-height indicator.
* Stepwise forward logistic regression and adjusted relative risk taking into account age, sex, birth weight, gestational age, immunization, and diagnosis at admission.
† RR of children with and without malnutrition aged ? 12 months vs. > 12 months.
‡ RR of children with and without malnutrition in relation to the presence of pneumonia and diarrhea.

---

226 Jornal de Pediatria - Vol. 85, No. 3, 2009
Nutritional status and therapy in hospitalized children - Sarni RO et al.
diagnosis at admission. Children with low birth weight were
1.77 times as likely to have malnutrition, and each year
old represented a reduction of 4% (RR = 0.96) in the risk
for malnutrition, at admission. Regarding classification of
nutritional status, 515/907 (56.7%) children were diagnosed
with malnutrition, according to information abstracted from
their medical records.
Nutritional assessment revealed that 142/872 (16.3%)
and 264/880 (30%) children at admission, as well as 121/787
(15.4%) and 270/785 (34.4%) at discharge, were classified
as having moderate/severe malnutrition (ZWH < -2) and
low stature (ZH < -2), respectively (Figure 1).
Figure 2 - Correlation between time of hospitalization and weight-
for-height z score at hospital discharge of the study
children
simultaneously in 10 teaching hospitals in Brazil. A high
percentage of moderate/severe PEM (ZWH 16.3%) and
stature involvement (ZH 30%) was observed in the study
population. These percentages were higher than those
observed in a study conducted with children at the public
hospital of the city of Fortaleza, State of Ceará, northeastern
ZH = height-for-age z score; ZW = weight-for-age z score;
Brazil, using similar cutoff points: 6.9 and 18.2% for ZWH
ZWH = weight-for-height z score.
and ZH, respectively.13
* ZWHa, ZWa, ZHa = weight-for-height, weight-for-age and height-for-age z score
at hospital admission.
Unfortunately, despite the high percentage of PEM, only
† ZWHd, ZWd, ZHd = weight-for-height, weight-for-age, and height-for-age z
score at hospital discharge.
56.7% of the children had their nutritional classification
documented in the medical record, which indicates a lack
Figure 1 - Nutritional diagnosis at hospital admission and dis-
charge
of concern with the nutritional diagnosis of hospitalized
children. Similar evidence had already been identified in
a Brazilian multicenter study with adults. In that study,
81% of medical records did not exhibit information
An inversely proportional and statistically significant
about the patient’s nutritional condition, despite the high
correlation was observed between ZWH at discharge and
prevalence of PEM at admission (31%), with a twofold
hospitalization time (r = -0.132; p < 0.01). For malnourished
increase after 1-day hospitalization.14 The findings in the
children, r = -0.036 (p = 0.760); and for eutrophic children,
present study become even more relevant if we take
r = -0.085 (p = 0.070) (Figure 2).
into consideration that such information was collected in
Regarding nutritional therapy at admission, oral
university-based teaching hospitals – institutions which
administration was the most used route, 756/841 (89.9%).
shape what health professionals are taught. This aspect
Use of non-modified whole cow milk was observed in 77/214
indicates a need for developing strategies to educate
(36%) children under 6 months of age; between 6 and 12
and qualify professionals working in hospitals, in order
months, 97/143 (67.8%); and in malnourished children
to enhance the value of current techniques to assess
under 12 months of age, 28/77 (36.4%) (Table 2).
nutritional status in patients attending SUS hospitals,
Malnourished children usually received nutritional therapy
focusing on the pediatric population.
via probe (p = 0.014), multivitamin supplements (p < 0.001),
Studies show a consistent association between presence
oligoelements (p = 0.0007), and iron (p < 0.001).
of malnutrition in children < 5 years of age and risk
of death due to diarrheic disease and acute respiratory
infection.15 The greater the degree of malnutrition, the
Discussion
higher the risk of death. Since malnourished children
This is the first study on the prevalence of child
are more susceptible to infectious processes and present
malnutrition, in hospitalized children, conducted
a series of pathophysiological peculiarities, diagnosis,

---

Nutritional status and therapy in hospitalized children - Sarni RO et al.
Jornal de Pediatria - Vol. 85, No. 3, 2009 227
Table 2 - Characteristics of nutritional therapy adopted at hospital admission


< 12 months


? 12 months

ZWH < -2
ZWH ? -2

ZWH < -2
ZWH ? -2
Variables
n (%)
n (%)
p
n (%)
n (%)
p
Administration route (n = 841)


0.014*


< 0.001*
Oral
83 (87.7)
327 (93.7)

31 (83.8)
315 (97.5)
Probe
13 (13.3)
21 (6.0)

6 (16.2)
5 (1.5)
Parenteral
2 (2.0)
1 (0.3)

0
3 (0.9)
Diet (n = 703)


0.069†


9.024†
NMWCM
28 (36.4)
138 (48.8)

22 (75.9)
195 (85.5)
IF
44 (57.1)
133 (47.0)

3 (10.3)
25 (11)
NLPF
2 (2.6)
3 (1.1)

0
5 (2.2)
SEF
2 (2.6)
4 (1.4)

1 (3.4)
1 (0.4)
SF
1 (1.3)
5 (1.8)

3 (10.3)
2 (0.9)
Multivitamins (n = 550)


< 0.001


0.051
Yes
13 (39.4)
20 (8.8)

4 (15.4)
11 (4.7)
No
20 (98)
206 (91.1)

22 (84.6)
221 (95.2)
Vitamin A (n = 537)


0.498


0.100
Yes
1 (2)
2 (0.9)

1 (4)
0
No
48 (98)
223 (99.1)

24 (96)
225 (100)
Oligoelements (n = 544)


0.007


0.806
Yes
7 (14.3)
4 (1.8)

0
2 (0.9)
No
42 (85.7)
222 (98.2)

26 (100)
227 (99)
Iron (n = 534)


0.001


0.084
Yes
12 (23.5)
10 (4.6)

4 (15.4)
13 (5.8)
No
39 (76.5)
208 (95.4)

22 (84.6)
212 (94.2)
IF = modified infant polymeric formula; NLPF = no lactose polymeric formula; NMWCM = non-modified whole cow milk; SEF = semielement formula; SF = soy formula.
p = chi-square test.
* Probe vs. oral.
† NMWCM vs. IF.
follow up and care/attention are of utmost importance
Malnourished children are three times more likely to have
to reduce time of in-hospital treatment and morbidity
diarrheic disease at admission than eutrophic children. This
and mortality rates.12
finding is similar to that reported in a study with severely
Factors associated with malnutrition at hospital admission
malnourished children admitted to a reference center in
were: age < 1 year, prematurity, low birth weight, absence
the city of São Paulo,7 southeastern Brazil, and the city of
of breastfeeding, and incomplete immunization. These
Recife,4 northeastern Brazil.
results are consistent with those found by Lima et al.16
The results also demonstrated that child nutritional
at hospitals in the city of Recife, northeastern Brazil. PEM
condition was inversely proportional to time of hospitalization.
is considered a multifactorial condition, which is believed
To date, however, in contrast with the adult population, the
to occur not only due to lack of nutrients or presence of
percentage of weight loss in relation to time of hospitalization
associated disease; a statement that is also corroborated
or any association with clinical worsening or mortality
by the findings in the present study.
increase are yet to be defined for the pediatric population.20
Acute pulmonary infection was the main cause leading
Since mean time of hospitalization was around 1 week, ZWH
to admission to the hospitals under study, followed by
was used to evaluate the progression of nutritional condition
diarrheic disease, and is considered the most important
at the two time points; the short hospitalization period,
cause of hospitalization and death in individuals < 5 years
however, limited the use of the height-for-age indicator.
of age.17 In Brazil, it is estimated that 5.4 and 12.8% of
Despite the high percentage of nutritional deficiencies,
deaths in children under 1 year of age and aged 1 to 4
only a small number of children were submitted to
years, respectively, are due to pneumonia.18
an adequate nutritional therapy, i.e., according to
Diarrheic disease is still an important cause leading
recommendations by WHO and BMH.12,21 Our findings also
to mortality and hospital admission in Brazil, despite the
revealed a high percentage of use of non-modified whole
decrease observed in its prevalence over the past years.19
cow milk, including use in children with moderate/severe

---

228 Jornal de Pediatria - Vol. 85, No. 3, 2009
Nutritional status and therapy in hospitalized children - Sarni RO et al.
PEM. WHO and BMH recommend the use of cow milk-based
6. Falbo AR, Alves JG, Batista Filho M, Cabral-Filho JE. Implementação
do protocolo da Organização Mundial da Saúde para manejo da
foods, with proper dilution, addition of micronutrients and
desnutrição grave em Hospital do Nordeste do Brasil. Cad Saude
adjustment of macronutrients, in order to meet the specific
Publica. 2006;22:561-70.
7. Sarni RO, de Souza FI, Catherino P, Kochi C, Oliveira FL, Nobrega
needs of children with moderate/severe malnutrition during
FJ. Tratamento da desnutrição em crianças hospitalizadas em São
the nutritional stabilization and recuperation phases.12,21
Paulo. Rev Assoc Med Bras. 2005;51:106-12.
8. Ferreira HS, França AO. Evolução do estado nutricional de
Based on our results, we could detect a high percentage of
crianças submetidas à internação hospitalar. J Pediatr (Rio J).
2002;78:491-6.
nutritional deficiencies in the children evaluated. The results
9. do Monte CM, Ashworth A, Sa ML, Diniz RL. Effectiveness of
also demonstrated that health professionals do not routinely
nutrition centers in Ceara state, northeastern Brazil. Rev Panam
document nutritional status and the nutritional therapy
Salud Publica. 1998;4:375-82.
10. Physical status: the use and interpretation of anthropometry.
adopted in the medical records of malnourished children.
WHO Technical Report Series, 854. Geneva: WHO; 1995.
These data underscore the importance of developing qualified
11. World Health Organization. WHO child growth standards: length/
height-for-age, weight-for-age, weight-for-length, weight-for-
hospital medical facilities, including university-based and
height and body mass index-for-age: methods and development.
teaching hospitals, regarding diagnosis and therapeutic
Geneva: WHO; 2006.
12. World Health Organization. Management of severe malnutrition:
approach to malnutrition, based on the conduct guidelines
a manual for physicians and other senior health works. Geneva:
proposed by WHO and adapted by BMH and the Brazilian
WHO; 1999.
13. Rocha GA, Rocha EJ, Martins CV. The effects of hospitalization on the
Society of Pediatrics to the Brazilian reality.
nutritional status of children. J Pediatr (Rio J). 2006;82:70-4.
14. Waitzberg DL, Caiaffa WT, Correia MI. Hospital malnutrition: the
Brazilian national survey (IBRANUTRI): a study of 4000 patients.
Nutrition. 2001;17:573-80.
Acknowledgements
15. Rice Al, Sacco L, Hyder A, Black RE. Malnutrition as an
underlying cause of childhood deaths associated with infections
The authors are grateful to the professionals from the
diseases in developing countries. Bull World Health Organ.
Working Group on Child Malnutrition who helped to conduct
2005;78:1207-21.
16. Lima MC, Motta ME, Santos EC, Silva GA. Determinants of impaired
data collection: João Guilherme B. Alves (Recife, PE), Anna
growth among hospitalized children: a case-control study. Sao
Cleide Montarroyos (Recife, PE), Ana Paula Gomes Ribeiro
Paulo Med J. 2004;122:117-23.
17. Mulholland K. Magnitude of the problem of childhood pneumonia.
(Recife, PE), Ana Maria de Carvalho Melo (Recife, PE),
Lancet. 1999;354:590-2.
Hélcio de Sousa Maranhão (Natal, RN), Noélia Leal Lima
18. Brasil. Ministério da Saúde. DATASUS. Sistema de Informações
sobre mortalidade. Mortalidade - CID 10, 1998. Brasília: MS;
(Fortaleza, CE), Christiane Araújo Chaves Leite (Fortaleza,
2001.
CE), Maria Ceci Vale (Fortaleza, CE), Francisca Maria Paiva
19. Brasil. Ministério da Saúde. Atenção integrada às doenças
Lino (Fortaleza, CE), Hélio Fernandes da Rocha (Rio de
prevalentes na infância: curso de capacitação: introdução: módulo
1. 2ª ed. rev. Brasília, DF: MS; 2002.
Janeiro, RJ), Angela Peixoto de Mattos (Salvador, BA), Paulo
20. Merritt RJ, Blackburn GL. Nutritional assessment and metabolic
Pimenta Figueiredo Filho (Belo Horizonte, MG), Alfredo Floro
response to illness of the hospitalized child. In: Suskind RM, editor.
Textbook of pediatric nutrition. New York: Raven Press; 1981. p.
Cantalice (Porto Alegre, RS), Vera Lúcia V. Bezerra (Brasília,
285-307.
DF), and Margarete Arioza.
21. Brasil. Ministério da Saúde. Manual de atendimento da criança com
desnutrição grave em nível hospitalar. Brasília, DF: MS; 2005.
References
1. de Onis M, Frongillo EA, Blossner M. Is malnutrition declining?
An analysis of changes in levels of child malnutrition since 1980.
Bull World Health Organ. 2000;78:1222-33.
2. Sociedade Civil Bem Estar Familiar no Brasil (BEMFAM). Pesquisa
Nacional sobre Demografia e Saúde. Programa de Pesquisas de
Demografia e Saúde. Rio de Janeiro: BEMFAM; 1997.
3. Monte C. Desnutrição: um desafio secular à nutrição infantil. J
Pediatr (Rio J). 2000;76:S285-97.
Correspondence:
4. Schofield C, Ashworth A. Why have mortality rates for severe
Roseli Oselka Saccardo Sarni
malnutrition remained so high? Bull World Health Organ.
Rua René Zamlutti, 94/52
1996;74:223-9.
CEP 04116-260 - São Paulo, SP - Brazil
5. Sarni RO, de Souza FI, Catherino P, Kochi C, Oliveira FL, Nobrega
Tel.: +55 (11) 5571.9589
FJ. Treatment of severe malnourished children with WHO protocol:
Fax: +55 (11) 5571.9589
experience of a referral center in Sao Paulo, Brazil. Arch Latinoam
E-mail: rssarni@uol.com.br
Nutr. 2005;55:336-44.

---