Aims Iron deficiency is a common finding among patients with heart failure (HF) and is associated with adverse outcomes, including decreased quality of life, increased risk of hospitalization, and decreased survival. Intravenous ferric carboxymaltose (FCM) has been shown to improve outcomes among patients with HF and concomitant iron deficiency, but FCM is associated with an increased risk of hypophosphataemia. We aimed to better characterize this risk among HF populations.
Methods and resultsThis pooled analysis examined data from 41 studies of adults with iron deficiency across disease states and therapeutic areas. Among the 7931 patients treated with FCM available for analysis, 14% made up the HF subgroup. Additional subgroups included women's health (36%), non-dialysis-dependent chronic kidney disease (NDD-CKD; 27%), haemodialysis-dependent chronic kidney disease (HD-CKD; 1%), gastrointestinal (10%), neurology (3%), and other (10%). The incidence of post-baseline moderate or severe hypophosphataemia (i.e. serum phosphate [PO 4 3À ] level <2.0 mg/dL) varied across the therapeutic areas, with the lowest incidences observed in the HD-CKD (0%), HF (8.1%), and NDD-CKD (12.8%) subgroups. The prevalence of moderate or severe hypophosphataemia among the women's health, other, gastrointestinal, and neurology subgroups was 30.1%, 40.6%, 51.0%, and 55.6%, respectively. In the HF subgroup, one patient (<0.1%) had a serum PO 4 3À of <1.0 mg/dL recorded, compared with 4.8% and 4.0% of the subjects in the neurology and gastrointestinal groups, respectively. With the exception of the HD-CKD subgroup, mean serum PO 4 3À levels decreased through weeks 2 to 4, and then returned toward baseline and plateaued by week 8. The strongest predictor of hypophosphataemia was preserved kidney function (estimated glomerular filtration rate: >60 mL/min/1.73 m 2 vs. <30 mL/min/1.73 m 2 ; odds ratio: 12.2). Among patients in the HF subgroup, the incidence of treatment-emergent adverse events potentially related to hypophosphataemia (e.g. cardiac failure, ventricular tachyarrhythmias, fatigue, muscle weakness, bone pain, neurological symptoms, and muscle pain) was lower among FCM-treated patients than among those receiving placebo, and lower among patients with a post-baseline PO 43À <2 mg/dL vs. those not meeting such criteria. Conclusions The risk of laboratory-assessed hypophosphataemia in HF patients treated with FCM was lower than that seen in patients in other therapeutic areas treated with FCM, and clinical events associated with hypophosphataemia are uncommon with FCM therapy in this population. Appropriate monitoring, particularly soon after administration, will allow for further refinement of management strategies.