Protein-energy Kivu1’ 2 Pierre Fondu,3 Chantal Andr#{233}Van Steirteghem,3
malnutrition Hariga-Muller,4 and Israel
ABSTRACT
Nava Mozes,5 Mandelbaum7
M.
Protein-energy
norrnochrornic following
anemia. results.
As
and anemia
malnutrition
An attempt compared
in
to define
with
Jean
local
Neve,6
Kivu
is associated
with
the physiopathology controls,
in
both
a discrete
normocytic,
of this anemia
iron
and
total
iron
disclosed
binding
the
capacity
were low, but with siderophilin saturation and sideroblast counts either normal or elevated; serum and erythrocyte folate was normal, plasma vitamin B12 was normal or elevated, and serum ascorbic acid was normal or elevated. The riboflavin nutritional status was normal. During
refeeding,
iron
and
riboflavin
deficiencies
became
apparent.
Characteristic
findings
on
admission were the presence of giant erythroblasts and a diminished erythrocyte survival time implicated to an intracorpuscular hernolysis. Two resultsfrom the present study could contribute to explanation more
malnutrition,
the
anemia
aforementioned
low
iron of
in the
or
chronic
vitamin
selenium
in Kivu,
and
anemia.
Journal
of Clinical Nutrition
Downloaded from https://academic.oup.com/ajcn/article-abstract/31/1/46/4650537 by Washington University School of Medicine Library user on 05 June 2018
vitamin the
anemia
physiopathology
that
a selenium J. Clin.
E levels
anemia
of
that
cannot
is distinct deficiency
may
30: 46-56,
Nuir.
and,
perhaps
protein-energy
be explained
from play
that an
of the
important
1978.
ciated with PEM are less polymorphic there than in other regions. In particular, malaria, ancylostomiasis, sickle-cell anemia and megaloblastosis are exceptional (20-26). Here then was a priviliged opportunity to study the anemia of protein-energy deficiency in man. The purpose of this article is: 1 ) To demonstrate that the anemia of PEM, as observed in Kivu, is a classifiable, nonadaptive anemia that cannot be explained by isolated iron or vitamin deficiencies and ,
.
The American
plasma
conclusion,
nonadaptive
whose Am.
The causes of the anemia observed in protein-energy malnutrition (PEM) are still poorly understood Several hypothesis may be considered: 1) Some authors, relying mainly on data from animal experiments, believe it is a form of adaptation to a lowered oxygen requirement, the so-called “adaptive anemia” (1, 2). 2) Other authors think that protein deficiency in humans directly affects the erythrocytic survival time and the proliferation or the maturation of the erythroblasts; the mechanisms of these abnormalities, however, remain obscure (3-6). 3) Inasmuch as both iron and total iron binding capacity are generally decreased in PEM, PEM anemia might present certain similarities with the anemia of chronic disorders (7). 4) Many authors consider that the protein deficiency is only one component of a complex hematological picture in which vitamin and iron deficiencies play prominent roles (8-19). Earlier studies conducted in Kivu have shown that the hematological changes asso46
In
is a classifiable
It is suggested of this
low
levels.
deficiencies
disorders.
pathogenesis
abnormalities:
plasma
as observed
by isolated
role
for
importantly,
‘From the CEMUBAC Medical Mission to IRS, Lwiro, Republic of Zaire; the Departments of Pediatrics, Clinical Chemistry and Hematology of the H#{244}pital
Saint-Pierre and the Laboratory of Toxicology of the Institute of Pharmacy, Free University of Brussels, Belgium. 2
Address reprint requests to: Dr. P. Fondu, Service
de Pediatric, 1000 Bruxelles, Adjoints, of Brussels.
H#{244}pital Saint-Pierre, Belgique. Pediatric Department,
Rue
Haute
320,
Free University Inspecteur de Laboratoire, Department of Hematology, H#{244}pitalSaint-Pierre. Chercheur, Department of Pediatrics, Free University of Brussels. the Institute
partment
31:
JANUARY
“Assistant, of Pharmacy.
of Clinical
Laboratory
Chemistry,
of Toxicology of Ma#{238}trede Stages, De-
H#{244}pitalSaint-Pierre.
1978, pp. 46-56. Printed
in U.S.A.
PROTEIN-ENERGY
MALNUTRITION
whose physiopathology is distinct from that of the anemia of chronic disorders; and 2) To define the physiopathological mechanisms of this anemia Possibly important characteristics are the presence of giant erythroblasts, a moderately decreased red cell survival time, and low plasma levels of vitamin E and selenium. These findings, added to those already gathered, permit us to propose an overall model for the anemia of PEM in Kivu. .
and methods
Materials
The subjects studied were hospitalized at the CEMUBAC Pediatric Hospital at Lwiro (Kivu Province, Republic of Zaire, altitude 1,748 ± 20 rn). The investigations reported here concern 188 children ages 1-12 years with evident biological and clinical signs of edematous PEM (McLaren score not less than 4) (27); in 92% of the cases the minimum weight observed during refeeding was below the average weight of children of the same ethnic group, sex and height. This clinical picture corresponds to the description of rnarasmic kwashiorkor given by Vis (28). The investigations were started on the first morning after adrnission . There was no prior selection of cases on the basis of hemoglobin level. The treatment consisted in refeeding with skimmed milk and local foods (beans, meat, fish, bananas, and palm oil); no supplementary iron or vitamins were given . For certain investigations (ascorbic acid assays) a cohort of children was observed throughout 2 months’ refeeding. In other cases, the results obtained on admission were compared with results from children who had been renourished for at least 2 months and whose weight and serum proteins had been restored to normal. It was impossible to perform repeatedly all the determinations in the 188 patients presented here, but there was no apparent bias due to incomplete
follow-up or to seasonal nutritional fluctuations. The controls were 41 children of the same ethnic group and age range, whose McLaren score was less than 4, whose weight was above the 5th percentile for height, and who had no past history of PEM. For the iron and vitamin studies, healthy European children living in Brussels (altitude 50 m), ages 2 to 6 years, were also used as controls. The methods used to assess the serum proteins, the hernatological and parasitological examinations, the serum iron and total iron binding capacity and the sideroblast percentages have all been previously described (22). The weight deviation was calculated according to the equation:
w.dev.
%
weight
-
theoretical .
theoretical
weight .
-x
100
weight
For incoming patients, the weight considered was not the weight on admission but the minimum weight observed during refeeding. The theoretical the average weight of normal children of
weight is the same
Downloaded from https://academic.oup.com/ajcn/article-abstract/31/1/46/4650537 by Washington University School of Medicine Library user on 05 June 2018
AND
47
ANEMIA
ethnic group, sex, and height, which has been previously determined in a very broad sample. A microbiological method using Lactobacillus casei NCIB8O1O was used to assay the folate in serum (29) and erythrocytes (30). For the serum vitamin E assay, saponification and extraction were done by the method of Kayden et al . (3 1 ), and a subsequent correction was allowed for the carotene extracted at the same time (32). For the folate and vitamin E, samples were quick-frozen and the assays done later in Brussels; preliminary tests had shown that the storage conditions did not affect the results obtained. The serum ascorbic acid level was measured immediately by Roe and Kuether’s technique (33). The riboflavin nutritional status was assessed by the glutathione-reductase assay performed on freshly sampled erythrocytes (34). The erythrocyte and plasma selenium levels were measured by atomic absorption (35). The radioactive half-life per ml of red blood cells labeled with S 1-Chromium (TCr) was determined beforehand by autotransfusion in 10 children studied on admission (20). In this report these results were compared with those obtained: 1 ) by autotransfusion in 10 children who had been renourished for at least two months and whose packed cell volume (PCV) and reticulocyte count had been stable for at least 30 days before the start of the test; 2) by transfusing the red cells from 1 1 patients studied on the day of admission to compatible volunteers from the medical and paramedical staff. None of the 3 1 patients had malaria or sickle-cell anemia and the patients whose labeled cells were reinjected into healthy recipients had wholly normal hepatic test results. The labeling method used has already been described (20). The results of the analyses were also calculated according to the ICSH directions (36). In order to evaluate the mean value, at labeling time , of the erythrocyte destruction probability funclion (i (T)) taking into account the changes in the red cell volume (RCV) during treatment, the RCV value was determined twice in 20 patients using a technique previously described (20); after corrections for the physical decay of 51Cr activity and for an elution rate of 1% per day, i(T) was calculated using three different models of the red cell destruction probability function (37). All the analyses were done under fasting conditions. Student’s t test was used for the statistical analysis of most results; for the vitamin assays, however, the distribution was not always Gaussian and a free-distribution test (Mann and Whitney’s U-test) was used. The symbols are defined as: NS (nonsignificant) = P
>0.05;”=0.01