The Effect of Iron Deficiency without Anemia on Cardiovascular Fitness: A Nhanes Study

Introduction:

The prevalence of iron deficiency (ID) without anemia is unknown. Iron is a vital constituent of hemoglobin, myoglobin, and some mitochondrial enzymes and its deficiency may result in reduced aerobic capacity. A few clinical studies have shown improvement in exercise performance after replacement of iron in subjects (subs) with ID and no anemia. As maximum oxygen consumption (Vo2 max) is the gold standard laboratory measure of cardiorespiratory fitness, in this study we investigated the prevalence of ID without anemia in general population and the effect of iron deficiency on Vo2 max in non-anemic subjects.

Methods

Data is obtained from continuous National Health and Nutrition Examination Survey (NHANES) 1999-2003, a nationally representative health survey combined with examination of non-institutionalized healthy subs. The data regarding demographics, age, race and smoking status, complete blood count, ferritin and estimated Vo2 max was obtained from questionnaires, laboratory and examination datasets respectively. The subs with normal hemoglobin were divided in to two groups based on ferritin level: Group A (GpA); ferritin ≤20 ng/mL, Group B (GpB); ferritin >20 ng/mL. Based on Vo2 max level subs were divided into: ‘below average’ (Vo2 max ≤ 30 ml/kg/min), and ‘average’ (Vo2 max >30 ml/kg/min).  The prevalence of ID without anemia was obtained using chi-square test. The correlation of Vo2 max and ferritin was obtained using linear regression. Odds ratio (OR) of ID with low Vo2 max was obtained using logistic regression after adjusting to age, race, gender and smoking. NHANES is a complex multistage probability sampling, and sampling weights were used in this analysis. P-value <0.05 was considered statistically significant. Analyses were performed using SAS 9.3 (SAS Institute, Cary NC).

Results:

Of 8628 subs with normal hemoglobin, 1074 subs had ID with weighted prevalence of 10.7% (95%CI 11.3-10). The prevalence of ID in males and in females was 0.86% (95%CI 1.15-0.57) and 9.8%(10.46-9.14), respectively. In individuals who had Vo2 max examination (n=2592), 408 subs had ID. The median age, hematocrit, platelet count and Vo2 max in GpA were 25, 39.2, 281, and 36.5 vs. GpB 29, 44.5, 262, and 40.8, respectively (table1). There was a weak correlation between Vo2 max and ferritin level with r-square=0.006. After adjusting for covariates, OR of below average Vo2 max in GpA was 0.93 (95%CI 0.64-1.34).  

Conclusion:

Weighted prevalence of iron deficiency in general adult population was 10.7% (95%CI  11.0-14.3). There was no significant difference between functional aerobic capacity indicating cardiovascular fitness between ID and non-ID subjects. Further studies are needed to validate our findings

 

Table1: Comparision of GpA and GpB
Variables	GpA (ferritin ≤ 20) (n=408)	GpB ( ferritin > 20) (n=2184)
Median age	25	29
Median Hematocit	39.2	44.5
Median RDW	12.5	12.2
Median Platelets	281	262
Median WBC	6.9	6.9
Median Neutrophils	4.1	4.0
Vo2 max	36.5	40.8