Letter to the Editor
To the Editor:
In the previous month’s issue of the journal, Sibbel et al.1 report an intriguing possibility that the dialyzer design and membrane may affect the response to the erythropoiesis stimulating agents (ESAs). In their retrospective analysis, the investigators used propensity score-matching (PSM) in over 38,000 incident patients with end stage renal disease (ESRD) who started in-center hemodialysis (HD) between 2009 and 2013. They found that for the same target hemoglobin, patients initiating HD on Polyflux Revaclear (Baxter, Deerfield, IL) dialyzers with polyarylethersulfone-polyvinylpyrrolidone (PAS-PVP) membrane required significantly less Epogen at months 1–5, and 7 during the first year compared with those starting on Optiflux dialyzers (Fresenius Medical Care, Waltham, MA) with polysulfone (PS) membrane. Despite the large number of subjects in the analyses, some study methodological issues raise questions about the validity of the conclusions. Firstly, the PSM excluded variables such as inflammation, secondary hyperparathyroidism, residual kidney function, and HIV infection among others which are known to alter response to the ESAs.2 Secondly, despite the similar baseline hemoglobin levels, the PS group received significantly higher doses of Epogen at month 1. This could indicate different facility practice patterns. It is likely the facilities using PS dialyzers were more aggressive in titrating Epogen to achieve the target hemoglobin. A facility-based comparative analysis could have addressed this confounder. Furthermore, because the patients on PS dialyzers were on higher initial Epogen doses, a comparison in change in the Epogen dose from the baseline would have been more informative. Thirdly, during the study period the target hemoglobin level in the ESA-treated ESRD patients was lowered to 11 g/dl in 2011 and the Prospective Payment System for dialysis went into effect around the same time. Both affected the ESA utilization in ESRD patients. An analysis by the year when a patient was enrolled was not performed. Fourthly, use of venous pressure (VP) monitoring has not been validated to detect “microfiber clotting.” Due to postdialyzer placement of the VP monitor, dialyzer clotting does not increase the VP. Therefore, it is unlikely that the VP rise observed in the study reflected microfiber clotting. Furthermore, no difference in iron utilization between groups and a higher Kt/V in the PS group were observed indicating argue against increased clotting. We commend the investigators for their work. However, problems with PSM, limited analyses, and the retrospective nature of the study raise questions about the validity of the conclusions. A prospective, randomized controlled study is needed to define the role of dialyzer-related factors in responsiveness to the ESAs.
Shakil Aslam and Robert J. Kossmann are full-time employees of Fresenius Medical Care of North America.
Robert J. Kossmann
Fresenius Medical Care of North America
Renal Therapies Group
1. Sibbel S, Hunt A, Laplante S, Beck W, Gellens M, Brunelli SM. Comparative effectiveness of dialyzers: A longitudinal, propensity score-matched study of incident hemodialysis patients. ASAIO J. 2016.62: 613–622.
2. Elliott J, Mishler D, Agarwal R. Hyporesponsiveness to erythropoietin: Causes and management. Adv Chronic Kidney Dis. 2009.16: 94–100.