Observational studies are a significant source of clinically relevant information. A recent expert interview with Professor Kamyar Kalantar-Zadeh highlights the value of observational studies in assessing intervention use and outcomes in a real-world setting. This article provides an overview of some key studies examining iron deficiency and IV iron use in haemodialysis patients.
Appropriate analysis of observational data – adjusting for confounders
Using observational data requires an understanding of the importance of the large number of variables involved. Maintenance haemodialysis patients experience fluctuations in markers such as serum ferritin and total iron binding capacity (TIBC) and their treatment is adjusted accordingly. Therefore, it is vital that the analysis of data for this patient group takes into account these time-varying confounders.
Using data from 58,058 maintenance haemodialysis patients in the United States of America, Professor Kalantar-Zadeh and colleagues determined whether the association between moderately high levels of serum ferritin, iron, iron saturation ratio, or IV iron and increased all-cause and cardiovascular mortality risk was predominately due to the time-varying confounding effect of malnutrition-inflammation-cachexia (MICS).
After applying time-dependent and multivariate adjustment for case mix, administered IV iron and erythropoietin doses, and available surrogates of malnutrition-inflammation-cachexia syndrome (MICS), the authors found that serum ferritin levels between 200 ng/ml and 1,200 ng/ml (reference 100 ng/ml to 199 ng/ml), serum iron levels between 60 ng/ml and 120 µg/ml (reference 50 µg/ml to 59 µg/ml), and iron saturation ratio between 30% and 50% (reference 45% to 50%) were associated with the lowest all-cause and cardiovascular death risks. They also concluded that the association between serum ferritin levels >800 ng/ml and mortality in maintenance dialysis patients is mostly due to the confounding effects of MICS.
To provide further analysis in the debate regarding iron supplementation and mortality in dialysis patients, 235 incident dialysis patients were enrolled in the INVOR study. Patients receiving iron (94% of whom were administered IV iron) had a significant reduction in all-cause mortality compared to those not receiving iron (HR (95% CI): 0.22 (0.08–0.58); p=0.002). These authors also adjusted for markers of inflammation and malnutrition and maintained a relationship between reduced all-cause mortality and iron supplementation. Dr Zitt and colleagues recognise the limited number of patients in their study, but also highlight the role of small observational studies in guiding future, larger studies.
Identifying the optimal dose of IV iron
Dr Kuragano and colleagues examined a possible association of ESA and iron doses with adverse events and prognosis in maintenance haemodialysis patients, using data from the Prospective Study of Treatment for Renal Anemia on Prognosis in haemodialysis patients (TRAP) cohort.
This sub-analysis included 1,095 patients and found that the risks of cerebrovascular and cardiovascular disease (CCVD), infection and hospitalization were higher in patients who did not maintain haemoglobin levels within the target range (10-11 g/dL) than those within the range. These risks were also significantly greater for patients treated with high doses of IV iron (≥50 mg/week) compared with no IV iron (CCVD HR 6.02, p=0.038; infection HR 5.22, p=0.001; hospitalization HR 2.77, p=0.015).
One dose does not fit all
The Dialysis Outcomes and Practice Patterns Study (DOPPS) is an important source for large scale observational studies. Dr Bailie and colleagues examined data from 32,435 patients receiving maintenance dialysis, to determine any relationship between IV iron dose and clinical outcomes.
Using the most common dose range (100–199 mg/month) as the reference, case-mix-adjusted mortality was similar for patients receiving 0, 1–99, and 200–299 mg/month, but significantly higher for the 300–399 mg/month (HR of 1.13, 95% CI of 1.00–1.27) and 400 mg/month or more (HR of 1.18, 95% CI of 1.07–1.30) groups. Although this result is caveated by further analysis which showed an increase in all-cause mortality in patients receiving ≥300 mg/month with a haemoglobin of ≥10 g/dL, but not <10 g/dL. This suggests that higher doses of IV iron may be suitable for patients with low haemoglobin concentrations. However, it also highlights one of the limitations of observational studies, in that not enough patients were in this category in DOPPS to make a strong conclusion.
IV iron use and hospitalizations
A recent analysis of the effect of IV iron on hospitalizations in patients on dialysis was performed as part of the DEcIDE-ESRD study. IV iron exposure for 1-, 3- and 6 months and incident hospitalizations during the follow-up period in 9,544 haemodialysis patients was examined. No consistent association was found for any dose level (0-900mg, 900-2100mg and >2100 mg/6 months) and risk for all-cause, cardiovascular or infectious hospitalizations.
- Kalantar-Zadeh K, Regidor DL, McAllister CJ, Michael B, Warnock DG. Time-Dependent Associations between Iron and Mortality in Hemodialysis Patients. J Am Soc Nephrol. 2005;16(10):3070-3080. doi:10.1681/ASN.2005040423.
- Zitt E, Sturm G, Kronenberg F, et al. Iron supplementation and mortality in incident dialysis patients: An observational study. PLoS One. 2014;9(12):1-19. doi:10.1371/journal.pone.0114144.
- Kuragano T, Matsumura O, Matsuda A, et al. Association between hemoglobin variability, serum ferritin levels, and adverse events/mortality in maintenance hemodialysis patients. Kidney Int. 2014;86(4):845-54. doi:10.1038/ki.2014.114.
- Bailie GR, Larkina M, Goodkin DA, et al. Data from the Dialysis Outcomes and Practice Patterns Study validate an association between high intravenous iron doses and mortality. Kidney Int. 2015;87(1):162-168. doi:10.1038/ki.2014.275.
- Tangri N, Miskulin DC, Zhou J, et al. Effect of intravenous iron use on hospitalizations in patients undergoing hemodialysis: A comparative effectiveness analysis from the DEcIDE-ESRD study. Nephrol Dial Transplant. 2015;30(4):667-675. doi:10.1093/ndt/gfu349.