This study examined the advantages of adding iron to vitamin D vs. vitamin D alone on bone turnover, inflammatory, oxidative stress, and metabolic markers. The intervention was performed in women with low levels of hemoglobin and 25OHD. However, iron plus vitamin D was not superior to vitamin D alone.
The usefulness of iron therapy in the treatment of anemia has been proven, especially in those with deteriorated hematologic condition [23]. However, supplementation with high dose of ferrous sulfate in female athletes for eleven weeks did not result in an improvement in hematologic parameters and had only a preventive effect on reducing body iron which is in line with the present study [24]. In our study, minor decreases in hematologic indices were observed in the D-Fe group that is physiologically implausible. It has also shown that serum levels of transferrin saturation, ferritin and hemoglobin diminished by using a fortified food product containing calcium and vitamin D during 9 weeks [25]. Failure to see an increase in these variables in the present study could be related to menstruation [26] and insufficient iron dose used [27]. Further studies with different groups and interventions are required.
The salutary effect of vitamin D on bone health is well known. Vitamin D controls bone remodeling through inducing ligand receptor activator of NF-kB, regulation of phosphate homeostasis by increasing fibroblast growth factor 23, and increase bone response to mechanical stimulation via mitogen-activated protein kinase signaling pathway [28]. It has shown that consumption of 7000 IU/d cholecalciferol for 26 weeks in vitamin D deficient women can reduce PTH and CTX and increase arm bone mass density [29]. A recent study indicated that 1000 IU/day vitamin D supplementation for 9 months decreased bone turnover markers [30]. Lerchbaum et al., evaluated the effects of 20,000 IU/week vitamin D on bone turnover markers in 200 healthy men with 25OHD < 75 nmol/L in comparison to placebo. However, after 12 weeks of supplementation, there was no significant effect on bone metabolism or density [31]. In our study, it was assumed that vitamin D plus iron supplementation versus vitamin D alone would further increase 25OHD concentration, further improve bone formation, and further reduce bone resorption. We observed that both groups experienced approximately the same increase of 25OHD, bone turnover markers rose in both groups, and surprisingly, the bone formation marker, osteocalcin, was significantly higher in the D-P group. There are limited studies evaluating the effects of iron on 25OHD and bone turnover markers. Published results of the Safe-D study indicated that factors like age, body composition and iron status have impact on bone turnover markers [32]. Blanco-Rojo et al., reported no effects of iron-fortified juice consumption on bone turnover markers in women with ferritin < 40 ng/mL [33]. The 25OHD levels reduced compared to baseline in both groups, with no difference between groups. In rats fed a diet low in iron, a sharp reduction in the concentration of procollagen type I N-terminal propeptide (bone formation marker) and an increase in PTH and tartrate-resistant acid phosphatase 5b (bone resorption markers) have been observed [34]. However, the level of 25OHD in this study did not change. Toxqui et al., observed a decrease in bone turnover markers and an increase in 25OHD when a group of women with iron deficiency anemia received an iron and vitamin D-fortified skimmed milk for 16 weeks [35]. A recent study demonstrated no change in serum levels of PTH and vitamin D binding protein after simultaneous supplementation of vitamin D and iron [36]. Recovery from iron deficiency anemia has been associated with lower bone remodeling without significant change in 25OHD level [14]. There are several differences between our study and similar trials that may justify our findings. These differences include the presence of iron deficiency in all participants and longer study duration. Additionally, in all of these studies, procollagen type I N-terminal propeptide was measured that is the most accurate marker of the formation [28]. However, 1,25(OH)2 D3 induces mRNA, synthesis, and secretion of osteocalcin by human and rat bone cells in vitro which is sensitive and specific for the evaluation of formation in adults[37]. Unfortunately, none of these studies evaluated the effect of iron and vitamin D compared to vitamin D alone. Furthermore, we did not include a third group of iron-placebo in our trial. The increased bone turnover can be attributed to the coupling of formation and resorption. It has reported that serum osteocalcin is increased in patients with untreated osteomalacia [38]. So, it can be concluded that both groups experienced enhanced formation coupled with resorption. Additionally, calcium intake was less than recommended dietary intake in both groups. Low levels of calcium and phosphorus stimulate PTH secretion and consequent increased 1,25(OH)2 D3 concentration[39]. Both hormones increase bone resorption to provide calcium and phosphorus required for calcification and this could be another reason for the lack of decreased bone turnover in this trial. Four months supplementation with 20,000 IU/w vitamin D in 399 subjects with mean baseline 25OHD 34.0 nmol/L resulted in small, but significant reduction of procollagen of type 1 amino-terminal propeptide (P1NP), without any effect on CTX, Dickkopf-1, sclerostin, TNF-α, osteoprotegerin, and receptor activator of nuclear factor ĸB ligand [40]. Those with high baseline PTH who had a decrease in PTH after the intervention, experienced higher decrease of P1NP and also significant reduction of serum CTX and increased sclerostin. Authors concluded that vitamin D supplementation is not effective or may even exaggerate bone loss if not combined with sufficient calcium intake. It is also seen that the response of bone turnover markers is different to osteoporosis treatments [41]; anti-resorptive treatments, like bisphosphonates, cause an early bone resorption and a delayed bone formation decrease, while anabolic treatments such as teriparatide result in an initial increase of bone formation markers and subsequent increase of bone resorption. It seems that the action of vitamin D in our study was like an anabolic treatment.
Our study showed a significant increase of interleukin-6 in D-P group without any effect on in none of groups. Previous studies with vitamin D supplementation have shown conflicting results. A decline in CRP has been reported among patients in the intensive care unit after vitamin D supplementation [42], while supplementation with 100,000 IU/d vitamin D2 or D3 had no clinically meaningful effect on hsCRP level [43]. Monthly intake of 1.25 mg vitamin D for two years could not change inflammatory indexes like hs-CRP, IL-6, IL-8, and IL-10 in 200 patients with osteoarthritis and vitamin D deficiency [44]. A meta-analysis on diabetic patients showed that vitamin D supplementation could reduce hs-CRP but not TNF-α and IL-6 [45]. The difference in the results of studies could be due to supplementation dosage, treatment period, and the population studied. Higher serum concentration of hsCRP at baseline is effective on the results of vitamin D interventions [46]. Furthermore, supplemental vitamin D has increased interferon-γ and interleukin-10 in subjects that were vitamin D insufficient compared to sufficient healthy adults [47]. Perhaps lack of reduction of inflammatory markers in this study and similar studies is attributed to lower duration of intervention and recruiting healthy subjects; because most studies with positive results have been conducted in patients with inflammation [48]. Besides, observed increased interleukin-6 in the D-P group may be related to the lack of regulatory effect of vitamin D on some inflammatory markers, low concentrations of inflammatory markers at baseline, low dose of vitamin D, and a higher percentage of patients with vitamin D deficiency.
Antioxidant effects of vitamin D was first established by Wiseman in 1993 by the concept that vitamin D3 and the active form; 1,25(OH)2 D3 halted iron-dependent peroxidation of liposomal lipid [49]. Vitamin D3 mega dose of (200,000 IU) increased total antioxidant capacity in vitamin D sufficient elderly women, and did not affect malondialdehyde [50]. After 12 weeks of consumption of vitamin D3-fortified doogh compared to plain doogh, significant decrease of malondialdehyde and enhancement of glutathione and TAC were revealed [51]. Failing to show positive results in the present study may be related to differences in patients’ genotype for vitamin D receptor, low-dose supplements required for reduction of lipid peroxidation, and insufficient time to impact on oxidative stress markers.
According to some observational studies on type 2 diabetes patients, vitamin D may have positive effects on glucose homeostasis [52, 53]. However, a systematic review found a weak correlation between vitamin D supplementation and descending fasting blood glucose and addressing insulin resistance challenge in patients with type 2 diabetes or impaired glucose tolerance; while no effect was seen in subjects with normal glucose tolerance [54]. In a recent systematic review and meta-analysis on 20 randomized clinical trials including 1,464 patients with diabetic nephropathy, vitamin D supplementation had no impact on glycemic control indexes [55]. There are also some unpredictable findings. Supplementation with 50,000 IU/w vitamin D for eight weeks in healthy adults > 65 years resulted an increase in the number of people with insulin resistance from 13.2–36.8% [56]. The mean homeostatic model assessment of insulin resistance also increased from 1.39 ± 1.34 to 5.27 ± 3.72. Also, in Cox proportional hazard regression model, in contrast to multivariable binary logistic regression model, in order to predict the association between serum 25OHD level and the incidence of type 2 diabetes, a positive relationship was observed at the highest vitamin D quartile [57]. Here, an increase in FBS was revealed in our healthy young women that is inexplicable and should be evaluated in future studies. Short duration of the study and limited number of participants could cofound our results.
After the intervention a significant increase in TG and decrease in LDL was observed in the D-P group, while HDL increased significantly in the D-Fe group. In a meta-analysis (including 4 clinical trials, three of them with poor quality and high heterogeneity) to evaluate the effect of vitamin D supplementation on cardiometabolic risk factors in healthy adults, no significant effect was observed [58]. Some studies have reported a positive relationship between 25OHD and serum lipids [59, 60]. Ponda et al. observed that correcting vitamin D deficiency by administering 50,000 IU of vitamin D weekly during 8 weeks significantly increased LDL-cholesterol in vitamin D-deficient high cardiovascular-risk adults [61]. This controversy in results may be due to differences in study design, dose of vitamin, and participant’s conditions. Furthermore, sufficient 25OHD has been attributed to better physical health and healthy lifestyle [62]. Our participants' vitamin D insufficiency may be related to non-specific chronic diseases. Supplementation with 50 mg iron with 500 mg docosahexaenoic acid or placebo in 76 women suffering from iron deficiency anemia caused significant decrease of apoAI in the group receiving iron and placebo [63]. Because of no significant between-group differences and small number of studies evaluating iron effects on cardiovascular health, it seems too early for conclusion.
The lack of effect of iron on vitamin D activation can be related to one or more of the following reasons: (a): the length of follow-up in studies evaluating bone markers is usually longer than the duration of the present study; (b) the dose of iron was not sufficient for iron status improvement; and (c) as previously mentioned, cytochrome P450 contains iron; however, in an animal study conducted by Dhur et al., in 1989 to assess the effects of different degrees of iron deficiency on cytochrome P450 enzymes it was suggested that modification of iron-dependent enzymes may only happen after the third stage of iron deficiency [64]. A recent study in rats demonstrated negative effects of iron deficiency on renal 1α-hydroxylase activity and bone formation [65]. However, the authors concluded that the severity of iron deficiency anemia in their animal model is rare in human subjects. To the best of our knowledge, this is the first randomized, triple masked, clinical trial to assess the effect of vitamin D-iron co-supplementation, compared to vitamin D-placebo supplementation on bone health among healthy females. Strengths of the current study included the study methodology and adjustment of some confounding factors like physical activity level, sun exposure, age, and body composition. Further studies should be performed to explore in depth the effects of iron on vitamin D function and bone health.
This has some limitations which have to be pointed out. Firstly, all women were healthy and all parameters were within normal limits. It was assumed that most of participants with low hemoglobin levels are iron deficient; however, ferritin measurement did not prove our assumption and other reasons that can cause anemia may also affect the results of the present study. This study should have been performed in those with ferritin levels less than 30 ng/ml. Furthermore, we were not able to measure blood levels of 25OHD using liquid chromatography–mass spectrometry method. Another limitation is that we did no evaluate some demographic data such as socioeconomic status which could affect our results. We also did not assess factors like socioeconomic status which could affect our results. Finally, the present study was not statistically powerful enough to detect a difference between two groups in the primary outcome, 25OHD (~ 23%).