Ashmead, et al.
2001 Latin
America(11) 

Ferrous 
bisglycinate

6 to 
36 mo.

40

RCT: Two groups:
FeSO4 vs. FBG @ 5 
mg/kg for 28 days

Hemoglobin 
and Ferritin

Both groups had significant hemoglobin ­ (p < 0.001);  
group treated with FBG had significant ­ (p < 0.005) in 
plasma ferritin. Iron bioavailabilities were 26.7% for FeSO4  
and 90.9%  for FBG.

Szarfarc, et al.
2001 Brazil(33)

Ferrous
bisglycinate
(FBG)

Pregnant
Women

145 (n=71
vs. n=74)

RCT: Two groups:
FeSO4  40 mg/d vs.
FBG 15 mg/day for a
min. of 13 weeks

Hemoglobin,
transferrin
saturation and
serum ferritin

Iron depletion was found in 30.8% of the women treated 
with  ferrous bisglycinate and in 54.5% of the women that
consumed ferrous sulfate. Higher non-compliance rates  
were seen with ferrous sulphate.

Aronstam,et al
1982
UK(12)

Ferrous
Glycine 
sulphate

adults

40

RCT-crossover: 
Ferrous fumarate or
FGS for 4 weeks
after which they
were switched to the
other drug

Hemoglobin, 
PCV and
MCV

Significant ­ in Hb, PCV and MCV were similar. In 3 
patients, side effects were severe with ferrous fumarate 
with interruption of treatment in 2 and withdrawal from
the study in the other. FGS group had no withdrawal of 
treatment. 

Sozmen, et al.
2003 
Turkey(18)

Iron
Polymaltose
complex
(IPC)

8-168 mo.

25 (n=14 
vs n=11)

RCT: Two groups:
FeSO4 vs. IPC @ 6
mg/kg for 3 mo. f/b 3
mg/kg for 3 mo. 

Hemoglobin,
Serum Iron,
Plasma Cu,
ceruloplasmin 
& Zinc

Hematological parameters showed equivalent rise in the 
two groups.  Copper  and ceruloplasmin metabolisms were 
affected  by ferrous iron supplementation.

Jacobs, et al.
1993 
South 
Africa(34)

Iron 
Polymaltose
complex
(IPC)

Healthy 
blood
donors

159
(Gp1:51;
Gp2: 53;
Gp3: 55)

RCT: Three groups: 
FeSO4 @ 60 mg  OD
vs. IPC @100 mg
OD vs. IPC @
100mg BD

Hemoglobin,
%age 
 transferrin 
saturation, 
serum ferritin 

80% of patients in Groups 1 & 3 reached normal Hb levels 
by 12 weeks, compared to 50% in Gp. 2. Similarly, the
proportion of patients improving their transferrin saturation
to normal was significantly better in Groups 1 & 3 than in
Gp. 2 (P < 0.01). However serum ferritin level, was  better in
Group 1 (p < 0.01); there was no difference in this respect 
between Groups 2 & 3

Devasthali,.
et al. 1991
India(26)

Carbonyl
 iron

Healthy
blood
donors

49

RCT: Two groups
FeSO4 vs Carbonyl
iron

Hb,MCV,
Retic. count, 
platelet count,
se iron, TIBC,
transferrin
saturation or
FEP

Mean values for hemoglobin concentration, mean
corpuscular volume, corrected reticulocyte count, platelet 
count, serum iron, total iron-binding capacity, transferrin
saturation or erythrocyte protoporphyrin did not differ
significantly between the two groups throughout the study.
After 16 weeks of therapy, the mean increase in hemoglobin
iron was similar in both groups (p = 0.2). Estimates of net
changes in total body iron suggested that the overall
bioavailability  of carbonyl iron was high, about 70% that
of  ferrous  sulfate.

Gordeuk VR,
et al. 1990
Ohio, 
USA(25)

Carbonyl 
iron

18-40
years-
Women

99

RCT: Two groups:
100 mg carbonyl
iron vs. placebo for
56 days

Hb, net iron
absorption,
rates of
deferral from
repeat
donation

At Day 56, estimated net iron absorption from therapy or 
diet, or both, was sufficient to replace iron in 85 percent of 
those receiving carbonyl iron but in only 29 percent of those
taking placebo (p less than 0.001). The rates of deferral
from  repeat donation were 8 percent in the carbonyl iron
group and 36 percent in the placebo group (p < 0.01).

Gordeuk VR, 
et al. 1987
Ohio, 
USA(29)

Carbonyl 
iron

Menstrua-
ting 
women

75

RCT: 3 groups: 1)
carbonyl iron 600 
mg; 2) FeSO4, 
300 mg (60 mg
Fe++); or 3)
placebo, each given
TDS *1wk 

Hb, MCV,
FEP, Se
ferritin, Se
iron, TIBC,
and %age
sat. of TIBC

The prevalence of gastrointestinal side effects was similar
in both groups taking iron. At the end of the study there was  
no laboratory evidence of change in iron status in women
who received carbonyl iron (n = 15). In those treated with  
ferrous sulfate (n = 17) the mean TIBC increased (p < 0.001),
and in the placebo group (n = 19) there were decreases in
mean MCV (p less than 0.01), serum ferritin (p <0.001),
and per cent saturation (p = 0.027) with an increase in 
mean TIBC (p = 0.004).