Disrupted coupling between ferritin and iron absorption in enteropathy-related anemia: Evidence for a dual-pathway model

Table 1 Characteristics of the study population, n (%)/mean ± SD

Parameter	Anemia		P value	Effect size
	Non-anemic group	Anemic group
Gender			0.683	N/A
Male	4 (17.4)	4 (13.3)
Female	19 (82.6)	26 (86.7)
Age	37.5 ± 15.8	39.3 ± 11.8	0.634	N/A
RBC	4.57 ± 0.47	4.46 ± 0.40	0.377	N/A
HB	13.03 ± 1.52	10.28 ± 1.15	< 0.001	2.01
MCV	87.06 ± 6.54	77.10 ± 6.44	< 0.001	1.54
MCH	28.67 ± 2.75	23.71 ± 2.93	< 0.001	1.74
Hematocrit	39.80 ± 3.91	33.99 ± 3.16	< 0.001	1.66
Iron	13.23 ± 6.80	5.21 ± 3.90	< 0.001	1.50
Ferritin	12.19 ± 8.90	5.51 ± 3.92	< 0.001	1.02
FACIT-F total	32.96 ± 9.48	29.30 ± 13.49	0.274	N/A
FACIT-F groups			0.078	N/A
1	2 (8.7)	7 (23.3)
2	5 (21.7)	10 (33.3)
3	10 (43.5)	4 (13.3)
4	6 (26.1)	9 (30.0)

Anemia was defined according to World Health Organization criteria: Male < 130 g/L, women < 120 g/L. Functional Assessment of Chronic Illness Therapy groups were categorized as follows: 1 = severe fatigue (0-20), 2 = moderate fatigue (21-30), 3 = mild fatigue (31-40), 4 = minimal or no fatigue (41-52), according to previously published Functional Assessment of Chronic Illness Therapy-Fatigue scoring guidelines. N/A: Not applicable; RBC: Red blood cells; HB: Hemoglobin; MCV: Mean corpuscular volume; MCH: Mean corpuscular hemoglobin; FACIT-F: Functional Assessment of Chronic Illness Therapy Fatigue scale.

Table 2 Summary of stepwise regression analyses for pathways linking enteropathy to erythrocyte status

Pathway tested	F-statistic	P value	R2	Outcome
Dual-pathway model
Enteropathy → iron absorption	F(4, 48) = 4.23	0.005	0.26	Supported
Enteropathy → ferritin stores	F(4, 48) = 5.09	0.002	0.30	Supported
Iron absorption → ferritin stores	F(1, 50) = 0.25	0.619	0.01	Not supported
Ferritin stores → RBC quality	F(1, 51) = 14.70	< 0.001	0.22	Supported
Ferritin stores → RBC quantity	F(1, 51) = 0.62	0.434	0.01	Not supported
Serial via serum iron model
Iron absorption → serum iron	F(1, 50) = 1.79	0.186	0.04	Not supported
Serial via ferritin model
Serum iron → RBC quality	F(1, 51) = 32.3	< 0.001	0.39	Supported
Serum iron → RBC quantity	F(1, 51) = 2.52	0.118	0.05	Not supported

The results indicate that simple serial mediation models (serial via serum iron model and serial via ferritin model) were not supported, owing to disruptions in key links of the proposed causal pathways. In contrast, the data most strongly support a dual-pathway model (dual-pathway model) in which enteropathy independently influences both intestinal iron absorption and serum ferritin levels. RBC: Red blood cell.