The present study was designed to analyze the dietary composition between categories of hemoglobin, TBI, and adherence to MMPs of PSC living at the highlands in the Southern Peruvian Andes. In the present study, 4.7% of PSC were diagnosed as anemic but when Hb was adjusted for altitude the prevalence of anemia increased to 65.6%.

One important finding is that PSC consuming low macronutrients consumes also low iron amounts in the diet. Therefore, PSC with anemia has fewer intakes of protein, fiber, beta carotene, niacin, AsA and sodium. These differences disappear when Hb is corrected by altitude. This happens because all PSC with erythrocytosis except one are re-classified as normal Hb and many PSC with normal uncorrected Hb are diagnosed as anemic after Hb correction for altitude. Several authors suggest that Hb should not be corrected by altitude [17, 19, 34, 35]. The arguments against the correction of Hb for altitude include the fact that the increase in Hb due to altitude is not universal and will depend on ethnicity and length of multigenerational seniority [36].

Population with longer data living at HA as native highlanders originating from the Tibetan and the Ethiopian plateaus present with a normal or only mildly elevated hemoglobin concentration [37]. People from ethnia Han in the Tibet with only 70 years living at HA have elevated levels of Hb compared with Tibetans living there for almost 25,000 years.

In addition, in Peru, the proportion of anemia attributable to ID was 22% of cases of anemia in children aged 6–59 months, and the proportion of anemia attributable to inflammation (27.8%). As other causes have not been identified is plausible to think that most of the cases of anemia at HA not due to ID or inflammation is due to the adjustment of Hb for altitude [10]. Moreover, populations living at HA have normal or higher iron status than those at low altitudes [38].

In Tibetan subjects with normal iron status and without deficiency of other vitamins, without hemoglobinopathies, anemia rates were very low but increase notably after Hb adjustment for altitude. Tibetan men had an apparent anemia (hemoglobin < 13 g/dL) prevalence of 1.4% (one male) and women had no anemia (hemoglobin < 12 g/dL). The WHO-recommended altitude adjustment, established using data on Andean highlanders, raised the prevalence of apparent anemia among Tibetan men to 77.8% (< 16.5 g/dL) and 86.5% (< 15.4 g/dL) among women [19].

According to our results, mean iron intake in PSC from Puno, Peru, is 10.78 ± 0.33 mg/day. This value is higher than the reported for Mexican children aged 12–50 months-old, residing also at high altitude. Mexican PSC has a mean iron intake of 6.2 ± 4.4 mg/day [39]. However, despite of this difference in iron intake, the prevalence of anemia after Hb adjustment is significantly higher in Puno (65.6%) than in the Mexican children (22.5%) [40]. Puno has in average more altitude (3800 m) than Mexico DF (2250 m).

The finding that only 7.44% of PSC has ID (TBI < 0 – ≥-4 mg/Kg) and 0.32% had IDA (TBI <-4 mg/Kg) suggest that high prevalence of anemia after Hb correction (65.6%) is not real.

Although mean total iron intake was lower in anemic PSC from Puno, the absence of significance seems to be due to the high deviation and to the low number of anemic PSC. This means that anemic PSC includes those with normal iron intake and others with low iron intake. A lack of difference in iron intake between adolescent with anemia and in those without anemia has been also reported [41]. This is in accordance to the fact that anemia has different causes and not only low iron intake [10]. In fact, children with IDA had significantly lower intakes of energy, protein, fat and various micronutrients, compared to those with normal iron status [42].

It is interesting to find that children with high TBI content has also more intake of AsA than children with ID and IDA (TBI < 0 mg/Kg). Several studies have demonstrated that AsA is an important enhancer of iron absorption not only supporting with an acid environment to the duodenum [43,44,45,46] but also regulating the of hypoxia inducible factor (HIF) which sense oxygen availability and iron homeostasis [47].

Low intake of niacin and beta carotene in PSC from Puno was associated with anemia as observed in other studies [48, 49].

Our study showed that only AsA was reduced in the anemic group. Supplementation with MMPs includes ascorbic acid (30 mg/sachet) but according to our analysis, this value seems to be insufficient to increase Hb concentration. Other authors recommend, 100–200 mg/day in other study [50]. These values are greater than that reported as average intake in our study.

The absorption of iron by the duodenum is essential to maintain its balance in the body, since, unlike most other essential nutrients; iron is not excreted in humans, this allows iron homeostasis. Also, this system has been developed to avoid iron overload that may have adverse effects in tissues [51].

A dietary intake contains heme and non-heme iron [52]. Heme iron has high bioavailability (15–35%) respect to non-heme iron (1–20%) [53]. Heme iron is composed of ferrous cation (Fe2+) and it is suggested to be absorbed as an intact metalloprotein via heme carrier protein 1 (HCP-1). In the enterocyte, ferrous iron is released from heme via heme oxygenase [54, 55]. Non-heme iron requires the conversion of the ferric (Fe3+) to ferrous cation which occurs in acid environments and depends on the divalent metal transporter 1 (DMT 1) to transport iron inside the enterocyte [56]. Thereafter, iron requires ferroportin to be exported to the systemic circulation. Ferroportin is down regulated by hepcidin [57].

Although fine regulation of iron absorption occurs to avoid iron overload in normal children, more recently, an alert has emerged from studies on gut microbiota [14, 58]. If a child receives an excess of iron it affects the gut microbiota by increasing the enteropathogenic count [59]. This in turn may result in systemic inflammation which could produce an anemic state [60].

Different studies suggest that modest response to iron interventions seems to be due mainly to a low adherence to the supplementation [61, 62]. However, none of these studies have reported data on dietary iron intake in children and how adherence to iron supplementation affects iron status. Then, it is necessary to know through nutritional surveys the daily dietary iron intake. Previous study showed that, both heminic and non-heminic iron were positively associated with serum ferritin [63].

Therefore, a contribution of this research is the evaluation of the nutritional composition, in particular the consumption of iron. An adequate consumption of iron results in normal TBI and, therefore, individuals are considered full of iron [63]. Likewise, the evaluation of the enhancing [64] and inhibitors [65] compounds of iron absorption will allow to know what is the best way to handle the type of diet used to fight against anemia.

In a population where iron is enough as observed in the present study in PSC from Puno, the supplementation with MMPs will not increase further TBI.

In populations where iron is deficient, home fortification with MMPs, compared with no intervention or placebo, reduced the risk of anemia in infants and increased hemoglobin concentrations and presented higher iron status [11]. The results of a meta-analysis in Latin America, evidenced that nutritional intervention reduced the prevalence of anemia from 45 to 25% [6].

The main strength of the study is the inclusion of nutritional and hematological evaluation which allow to know the real iron status of these children.

This is a cross-sectional study, which does not establish a causality between the variables studied. Likewise, the methodology of the 24-hour recall was only applied once; however, this technique has been widely used for nutritional evaluation, both by researchers and by international institutions and organizations.