Journal of Pediatric Gastroenterology and Nutrition
31:234237 September 2000 Lippincott Williams & Wilkins, Inc., Philadelphia
Vitamin A Supplementation in Acute Diarrhea
zmert, S. Songul Yalcn, and *Yahya Laleli
Kadriye Yurdakok, Elif O
Department of Social Pediatrics, Hacettepe University Institute of Child Health; *Duzen Laboratories, Ankara, Turkey
ABSTRACT
Background: Vitamin A supplementation reduces the severity
of subsequent diarrheal episodes. This study was conducted to
examine the effect of single oral high-dose vitamin A supple-
mentation on the duration of acute diarrhea in 6- to 12-month-
old infants who are not malnourished.
Method: In this double-blind, randomized, placebo-controlled
study, infants who were admitted to Hacettepe University Ihsan
Dogramaci Childrens Hospital Diarrheal Diseases Training
and Treatment Unit with acute diarrhea were randomly as-
signed either to a group receiving a single oral dose of 100,000
IU vitamin A or placebo. There were 60 infants in each group.
All infants were followed up until the diarrheal episode ended.
Serum vitamin A levels were determined both at admission and
2 weeks later.
Results: No effect of vitamin A supplementation could be
demonstrated on either the total duration of diarrhea (7.4 3.2
days in the treatment group vs. 7.8 3.1 days in the placebo
group) or on its duration after intervention (3.8 2.3 days in the
treatment group vs. 3.9 1.9 days in the placebo group; P >
Low serum vitamin A levels have been described in
many infections as well as during acute diarrhea (1,2).
Field trials have demonstrated a decrease in overall mor-
tality due to infectious diseases after prophylactic high-
dose oral vitamin A supplementation (35), but hospital-
based studies in patients with acute respiratory tract in-
fection and acute diarrhea have failed to show a dramatic
effect on clinical outcome (610) during infection. The
few clinical studies evaluating the effect of vitamin A
supplementation during acute diarrhea have not shown
the relationship between serum vitamin A levels of the
subjects and the clinical outcome (710). Therefore, we
conducted a double-blind, placebo-controlled trial to de-
termine the effect of administration of a single oral dose
of 100,000 IU vitamin A on the clinical outcome of
diarrhea and serum vitamin A levels in infants aged 6 to
12 months with acute diarrhea.
Received September 2, 1999; revised December 14, 2999; accepted
February 16, 2000.
Address correspondence and reprint requests to Kadriye Yurdakok,
Kuloglu sok. No 6/12, 06680 C
ankaya, Ankara, Turkey.
234
0.05 for both comparisons). Serum vitamin A levels were not
significantly different at admission (23.5 9.7 g/dL in the
treatment group vs. 24.1 9.7 g/dL in the placebo group; P
> 0.05) nor at the end of a follow-up period of 2 weeks (treat-
ment: 33.3 13.7 g/dL, placebo: 35.2 11.2 g/dL; P >
0.05). However, the increase in serum vitamin A levels at the
end of the 2-week follow-up interval for infants in both the
treatment and placebo groups were found to be significant com-
pared with levels at admission (P < 0.01). The mean weight
gain in both groups were similar by the end of the first month
(6.9 5.0% in the treatment group vs. 6.3 4.2% in the placebo
group; P > 0.05).
Conclusion: No effect of oral vitamin A supplementation on
serum vitamin A levels, duration of diarrhea, or weight gain
during an acute diarrheal episode could be demonstrated in our
study group of infants between 6 and 12 months of age who had
no malnutrition. JPGN 31:234237, 2000. Key Words: Chil-
drenDiarrheaSupplementationVitamin A. 2000 Lip-
pincott Williams & Wilkins, Inc.
MATERIALS AND METHODS
A double-blind, randomized, placebo-controlled clini-
cal trial was conducted at Hacettepe University Ihsan
Dogramaci Childrens Hospital Diarrheal Disease Train-
ing and Treatment Unit. Infants 6 to 12 months of age
with diarrhea of less than 5 days duration were included
in the study. Infants with chronic disease, malnutrition
(weight for age less than the 10th percentile according to
the National Center for Health Statistics) (11), associated
infectious disease, prior antibiotic use, or dysentery were
excluded. One hundred twenty infants were randomly
assigned to either treatment or placebo groups.
On admission, a standard history (12) was taken, and
a thorough physical examination was performed. Dehy-
dration was assessed and treated according to World
Health Organization (WHO) guidelines (12) by using
glucose in oral rehydration solution.
After obtaining verbal informed consent from the par-
ents, either a single dose of 100,000 IU water-soluble
vitamin A (Roche, Nutley, NJ, U.S.A.) or placebo
(starch) identically packaged was administered orally ac-
cording to systematic randomization by a nurse blinded
to the drug assignment. The packages were numbered
 VITAMIN A FOR DIARRHEA
011 or 022 by the pharmacist. Patients with a file number
ending with an odd number were administered the pack-
ages numbered 011, and those with an even number re-
ceived packages numbered 022. There was no dietary
intervention. Daily multivitamin administration was dis-
continued in patients who had been receiving it. Mothers
were given an observation chart and were instructed to
note the frequency of diarrhea every day until the termi-
nation of the episode. The progress of all infants was
monitored by telephone interview every other day by one
) until the infants recovery from di-
of the authors (EO
arrhea. Clinical recovery was defined as passage of
formed stool as described by the mother for at least 24
hours. Infants were evaluated at 2 weeks (second assess-
ment) and 1 month (third assessment) after study enroll-
ment.
Blood was collected on admission and after 2 weeks
for determination of vitamin A, protein, albumin, alka-
line phosphatase, calcium, and phosphorus levels. In ad-
dition, hemoglobin levels were measured. Serum was
stored at 20C until vitamin A level measurement by
high-performance liquid chromatography. Hemoglobin
was measured by a cell counter (Coulter, Hialeah, FL,
U.S.A.) and the other biochemical parameters by auto-
analyzer. Stool cultures (for Salmonella and Shigella)
were obtained, and stools were examined for parasite and
tested for pH, reducing substance, and steatocrit on ad-
mission. Anthropometric measurements of the infants
were repeated 1 month after the study enrollment.
Sample size was calculated according to the study by
Haffejee (13), assuming that there would be a 25% re-
duction in the duration of diarrhea in the vitamin A group
compared with the placebo group. Considering a signifi-
cance level of 5% and power of 80% and assuming a
withdrawal rate of 10%, a sample size of approximately
51 patients in each group was needed, and 60 patients
were assigned to each group. Statistical analysis was per-
formed by computer (EpiInfo version 5.0, CDC-WHO,
Stone Moutain, Georgia, 1990). Students and paired t-
tests and 2 tests were used when appropriate.
RESULTS
During the study period, 120 infants were assigned
either to treatment or placebo groups of 60 infants each.
The characteristics of each group on study enrollment
mean age, sex, anthropometric indices, duration and fre-
quency of diarrhea, dehydration status, breast-feeding
and low dose vitamin supplementation ratewere com-
parable (Table 1). Stool cultures obtained at hospital ad-
mission revealed only one instance of Shigella sonnei,
which was in the treatment group. All tested stools were
negative for parasite, reducing substance, and steatocrit.
The mean stool pH was 5.6 0.7 and 5.6 0.6 in the
vitamin A and placebo groups, respectively. Only a small
percentage of patients had dehydration, and thus most of
the patients may be considered to have had mild diarrhea.
235
TABLE 1. Characteristics at study enrollment of the
vitamin A and placebo groups
Vitamin A group Placebo group
(n 60) (n 60)
Age (Mo) 9.1 2.2 9.2 2.2
Male/female 40/20 39/21
Body weight (g) 8661 1693 8933 1335
Length (cm) 71.1 4.3 71.3 3.8
Duration of diarrhea (day) 3.6 1.9 3.7 1.9
Diarrhea frequency (per 24 hr) 6.2 2.9 6.1 3.5
Breast-feeding (n [%]) 37 (61.6) 31 (51.7)
Vitamin supplementation (n [%]) 44 (73.3) 39 (65.0)
Presence of dehydration (n [%] 7 (11.6) 9 (15)
P > 0.05 for all parameters between groups.
Blood samples for serum vitamin A level determina-
tion were obtained from all infants on admission. How-
ever, 57 samples in the treatment and 58 in the placebo
group were available for the study. The progress of all
patients (120) was monitored by telephone calls until
cessation of diarrhea. Forty-nine (82%) patients in the
treatment and 52 (87%) in the study group were admitted
for the second assessment. Parental consent was denied
for blood sampling in two infants in the treatment and
seven infants in the placebo group in this follow-up. At
the 1-month follow-up visit, 40 (67%) patients were
available in each group. Those who were available for
the second and third assessments did not differ from
those who were not available for assessment in charac-
teristics at study enrollment, duration of diarrhea, and
total duration of diarrhea after treatment (P < 0.05).
The mean total duration of diarrhea in the treatment
and placebo groups were 7.4 3.2 days and 7.8 3.1
days, respectively (P > 0.05). The mean duration of di-
arrhea in each group after treatment was 3.8 2.3 and
3.9 1.9 days, respectively. In each group, two patients
(6.6%) had persistent diarrhea (duration, 1518 days).
There were no children with a bulging fontanelle or other
findings of hypervitaminosis A.
Vitamin A levels of each group on study enrollment
and second assessment are given in Table 2. There was
no difference between the groups in vitamin A levels on
study enrollment. Vitamin A levels increased in each
group after 2 weeks, irrespective of supplementation, and
the increase was significant within each group (P <
0.01), but there continued to be no difference between
the groups (P > 0.05; Table 2).
The biochemical values of each group at study enroll-
ment and at second assessment are shown in Table 3. All
parameters were comparable between groups, both at en-
rollment and at the follow-up. The mean body weight
after 1 month and percentage of weight gain in both
groups were also comparable (9300 1380 g, 6.9 5.0%
in the treatment group and 9408 1323 g, 6.3 4.2% in
the placebo group; P > 0.05).
An additional statistical analysis using breast-fed in-
fants as the controlling variable also indicated that
J Pediatr Gastroenterol Nutr, Vol. 31, No. 3, September 2000
236 K ET AL.
K. YURDAKO

TABLE 2. Mean serum vitamin A levels in vitamin A and placebo groups

Vitamin A Placebo
group (n) group (n)
At enrollment
Vitamin A level (mean SD) 23.5 9.7 (57) 24.1 9.7 (58)
Vitamin A level <20 g/dL (%) 22/57 (38.6) 22/58 (37.9)
Vitamin A level <10 g/dL (%) 2 (3.5) 4 (6.9)
Second assessment
Vitamin A level (mean SD) 33.3 13.7 (47)* 35.2 11.2 (45)*
Vitamin A level <20 g/dL (%) 8/47 (17) 2/45 (4.4)
Vitamin A level <10 g/dL (%) 0 0

SD, standard deviation.

* P < 0.01 for serum vitamin A levels within groups at enrollment and at second assess-
ment.

breast-feeding status did not influence the effect of vita-
min A supplementation (P < 0.05).
DISCUSSION
In this study, no effect of oral single high-dose vitamin
A supplementation on shortening the duration of diarrhea
or on serum vitamin A levels could be demonstrated
during an episode of acute diarrhea in 6- to 12-month-old
infants who were not malnourished.
Duration of diarrhea and the weight gain in the month
after the acute diarrhea were similar in both the treatment
and placebo groups. In this study, the prevalence of sub-
clinical vitamin A deficiency (serum vitamin A level
below 20 g/dL) was found to be 39% in the treatment
group and 38% in the placebo group at hospital admis-
sion during the diarrheal episode, and was found to drop
in both groups 2 weeks later, decreasing to 17% in the
treatment group and, most remarkably, to 4.4% in the
placebo group. In fact, we believe that the prevalence
after the diarrheal episode reflects the real status of the
population, because the former reading may have been
confounded by the presence of infection. During the
course of acute diarrhea, no substantial effect of oral
vitamin A supplementation on vitamin A levels could be
demonstrated and in contrast to expectations, the propor-
tion of infants with subclinical deficiency of vitamin A in
the treatment group was higher than in the placebo
group, although the difference was not statistically sig-
nificant.
In our study, although the serum retinyl ester level,
which is a reflection of excess vitamin A intake, was not
measured, there was no infant with bulging fontanelle or
other signs of hypervitaminosis A. In fact, the dosage
was within the range recommended by WHO for 6- to
12-month-old infants (14).
The reports from previous studies regarding the effect
of prophylactic high-dose vitamin A supplementation on
the severity of subsequent diarrheal episodes are contro-
versial (15,16). In clinical settings, administration of
high doses of vitamin A during an acute episode has
produced different results (710) as well. Although no
desired effect could be shown for the acute episodic di-
arrhea, Henning et al. (7) nonetheless proposed that ad-
ministration of 200,000 IU oral vitamin A might at least
increase the hepatic reserves. Two recent reports (8,9)
have asserted that some effect occurs as a result of
supplementation, with patients in these studies being
more than 1 year of age. Hossain et al. (9) have shown
that a significantly higher proportion of children diag-
nosed specifically with shigellosis were clinically cured
when given 200,000 IU oral vitamin A supplementation
than those given placebo. Bhandari et al. (8) have shown
a positive effect of vitamin A supplementation only in
nonbreast-fed children. Dewan et al. (10) have demon-
strated an effect of vitamin A supplementation in chil-
dren who have malnutrition and clinical vitamin A defi-
ciency. In field trials it has been shown that 100,000 IU
of oral vitamin A supplementation decreases the preva-
lence of especially severe diarrhea (16).

TABLE 3. Biochemical values of the vitamin A and placebo groups

At enrollment Second assessment
Vitamin A group Placebo group Vitamin A group Placebo group
Protein (g/dL) 6.9 0.6 (44) 7.1 0.7 (44) 7.1 0.7 (30) 7.1 0.4 (27)
Albumin (g/dL) 4.8 0.4 (48) 4.9 0.5 (43) 4.7 0.4 (30) 4.7 0.3 (27)
Alkaline phosphatase (IU) 534.6 190.6 (46) 538.2 171.6 (43) 522.1 184.1 (28) 552.2 137.1 (23)
Calcium (mg/dL) 10.3 0.5 (35) 10.4 0.9 (38) 10.6 0.6 (28) 10.6 0.7 (26)
Phosphorus (mg/dL) 5.6 11.1 (35) 5.8 1.3 (38) 6.2 1.0 (23) 5.8 0.7 (17)
Hemoglobin (g/dL) 10.4 1.2 (49) 10.9 1.3 (50) 10.3 1.0 (32) 11.1 1.2 (37)

Data are means SD with number of subjects in parentheses.

J Pediatr Gastroenterol Nutr, Vol. 31, No. 3, September 2000

 VITAMIN A FOR DIARRHEA
Our study showed no effect of vitamin A supplemen-
tation in this patient group. That our results differed from
those in the previous studies may be explained by several
factors. Although there have been studies showing that
oral vitamin A is absorbed during acute diarrhea (17), the
oral dose may not have been adequately absorbed. Thus,
a parenteral dose may produce different results. Our
study population is different from the groups in the pre-
vious studies. The results of the present study and those
of a previous study conducted in infants in Ankara indi-
cate that clinical vitamin A deficiency is not as prevalent
in this area (18) as it may be in other areas studied. This
study included only infants between 6 and 12 months of
age who did not have either malnutrition or dysentery
(we had only one patient with S. sonnei), and the patients
mostly had mild diarrhea. Another study conducted in
patients with pneumonia treated with 200,000 or 400,000
IU oral supplementary vitamin A showed no beneficial
effect on recovery, especially in patients with no under-
lying malnutrition (6). Although the symptoms in our
supplemented group did not show deterioration, animal
studies suggest that either chronic excess or deficiency
may cause immune suppression (19). Breast-feeding sta-
tus did not influence the effect of vitamin A supplemen-
tation in our study. Therefore, in our study when infants
without dysentery and malnutrition were included, no
effect of oral supplementation could be demonstrated,
irrespective of the breast-feeding status.
Many investigators have attempted to show a relation
between diarrhea and low vitamin A levels and the
course or outcome of diarrhea. However, it may be pre-
mature to take the broad view that subclinical vitamin A
deficiency has a direct causal relationship in such cases.
In this study, the low levels of vitamin A observed in our
patients may not be a real indicator of the body stores
(and therefore of vitamin A status), but rather a defen-
sive, physiologically normal response to a pathologic
process in which serum vitamin A levels decrease, either
because of decreased mobilization or increased excre-
tion, or both, of vitamin A during infections (20). Most
of the children with apparent subclinical deficiency re-
covered after the infection, irrespective of supplementa-
tion.
Although the withdrawal rate seems to be high, we
still believe that the results can be justified, because there
was no difference in either group between the patients
who were withdrawn and those who attended all assess-
ments.
We conclude therefore that oral vitamin A supplemen-
tation during episodic acute diarrhea in infants between 6
and 12 months of age who have no signs of malnutrition,
has no effect on duration of diarrhea, weight gain, or
serum vitamin A levels.
237
Acknowledgment: The authors thank Mitch Halloran for his
help during the preparation of the manuscript.
This study was supported by grant SBAG-1647 from
The Scientific and Technical Research Council of Turkey
(TUBITAK).
REFERENCES
1. Lindo ES, Salazar M, Alvarez JO. Association of diarrhea and low
serum retinol in Peruvian children. Am J Clin Nutr 1993;58:1103.
2. Hussey GD, Klein M. A randomised, controlled trial of vitamin A
in children with severe measles. N Engl J Med 1990;323:1604.
3. Daulaire NM, Starbuck ES, Houston RM, et al. Childhood mor-
tality after a high dose of vitamin A in a high risk population. BMJ
1992;304:20710.
4. Sommer A, Tarwotjo I, Djunaedi E, et al. Impact of vitamin A
supplementation on mortality: a randomized controlled community
trial. Lancet 1986;1(8491):116973.
5. West KP Jr, Pokhrel RP, Katz J, et al. Efficiency of vitamin A in
reducing preschool mortality in Nepal. Lancet 1991;338:6971.
6. Si NV, Gryterr C, Vy NNT, et al. High dose vitamin A supple-
mentation in the course of pneumonia in Vietnamese children. Acta
Pediatr 1997;86:10525.
7. Henning B, Stewart K, Zaman K, et al. Lack of therapeutic efficacy
of vitamin A for non-cholera, watery diarrhoea in Bangladesh chil-
dren. Eur J Clin Nutr 1992;46:43743.
8. Bhandari N, Bahl R, Sazawal S, et al. Breast-feeding status alters
the effect of vitamin A treatment during acute diarrhea in children.
J Nutr 1997;127:5963.
9. Hossain S, Biswas R, Kabir I, et al. Single dose vitamin A treat-
ment in acute shigellosis in Bangladesh children: randomized
double blind controlled trial. BMJ 1998;316:4226.
10. Dewan V, Patwari AK, Jain M, et al. A randomized controlled trial
of vitamin A supplementation in acute diarrhea. Indian Pediatr
1995;32:215.
11. Hamill PV, Drizd TA, Johnson CL, et al. Physical growth: Na-
tional Center for Health Statistics percentiles. Am J Clin Nutr
1979;32:60729.
12. World Health Organization (WHO). A manual for the treatment of
acute diarrhea for use by physicians and other senior health work-
ers. Geneva: WHO; 1984. WHO/CDD/SER 80.2 REV 1.
13. Haffejee IE. Effect of oral folate on duration of acute infantile
diarrhoea. Lancet 1988;2(8606):3345.
14. Expanded Programme on Immunization: Using immunization con-
tacts as the gateway to eliminating vitamin A deficiency. A policy
document. Geneva: World Health Organization; 1994. WHO/EPI/
GEN/94.9.1994.
15. Stansfield SK, Pierre Louis M, Lerebours G, et al. Vitamin A
supplementation and increased prevalence of childhood diarrhoea
and acute respiratory infections. Lancet 1993;342:57882.
16. Barreto ML, Santos LMP, Asis AMO, et al. Effect of vitamin A
supplementation on diarrhoea and acute lower-respiratory-tract in-
fections in young children in Brazil. Lancet 1994;344:22831.
17. Molla A, Islam A, Molla AM. Change in serum vitamin A con-
centration after an oral dose in children with acute diarrhea. J
Pediatr 1983;103:10002.
18. Hatun E, Tezic T. Vitamin A status of healthy infants in Ankara,
Turkey. Turk J Pediatr 1995;37:18792.
19. Friedman A, Meidovsky A, Leitner G, et al. Decreased resistance
and immune response to Escherichia coli in chicks with low or
high intakes of vitamin A. J Nutr 1991;121:395400.
20. Erdman JW Jr, Poor CL, Dietz JM. Factors affecting the bioavail-
ability of vitamin A, carotenoids and vitamin E. Food Technol
1988;42:21421.
J Pediatr Gastroenterol Nutr, Vol. 31, No. 3, September 2000