No Causal Association Between Red Cell Distribution Width and PAH

No causal association between elevated red cell distribution width and pulmonary arterial hypertension was established.

No causal association between elevated red cell distribution width (RDW) and pulmonary arterial hypertension (PAH) was established, according to Mendelian randomization study published in the European Respiratory Journal.

This multicenter case-control study included individual-level data that consisted of RDW laboratory measurements from 524 patients with PAH from the Imperial College PH Biobank and the UK PAH Cohort, and summary-level data from 118 patients with PAH and 15,889 common disease control individuals (women: 54%, mean age, 54.3 years; men: 46%, mean age, 58.1 years) from a hospital population-based study conducted at Vanderbilt University Medical Center in Nashville, Tennessee. The mean age of patients with PAH was 52.4 years for women (69%) and 58.2 years for men (31%).

A 2-sample Mendelian randomization procedure was used, in which effect estimates on the risk factor (RDW) and outcome (PAH) derived from dual non-overlapping datasets were evaluated to assess whether a genetic predisposition to higher RDW levels had an effect on the odds of developing PAH. The primary and secondary MR analyses used the inverse variance weighted (IVW) and weighted median (WM) methods, respectively, to evaluate causality. The primary analysis considered all genome-wide RDW variants (n=179) and the secondary analysis used genome-wide variants related to systemic iron status (n=5). Logistic regression was used to test the association between RDW and PAH occurrence after adjusting for age and sex.

There were 642 patients with PAH who were compared with 15,889 control individuals in the final cohort. An association was observed between RDW and PAH (odds ratio [OR], 1.90; 95% CI, 1.80-2.01). The primary Mendelian randomization analysis on 179 variants was found to be sufficiently powered to examine causation between RDW and PAH for ORs between 1.25 and 1.52 or higher. However, after multivariable adjustment, no causal association was established (ORcausal,1.07; 95% CI, 0.92-1.24; P =.57). No causal relationship between iron status and PAH was established in the secondary Mendelian randomization analysis on 5 variants linked to iron status (ORcausal, 1.09; 95% CI, 0.77-1.54; P =.91).

Study strengths include a large sample size; study limitations include the use of an unselected control group that was not at risk for PAH.

Related Articles

“The results suggest that at least some of the observed association of RDW with PAH is secondary to disease progression,” noted the authors. “Results of iron therapeutic trials in PAH should be interpreted with caution as any improvements observed may not be mechanistically linked to the development of PAH.” They recommended that future trials involve careful follow-up of at-risk populations and investigate underlying PAH mechanisms, causal relationships, and genetic susceptibility.


Ulrich A, Wharton J, Thayer TE, et al. Mendelian randomisation analysis of red cell distribution width in pulmonary arterial hypertension [published online November 19, 2019]. Eur Respir J. doi:10.1183/13993003.01486-2019