More than 10% of hemodialysis (HD) patients have prolonged intradialytic hypoxemia (PIH), a condition defined as arterial oxygen saturation below 90% for at least one-third of the treatment time.1Meyring-Wösten A. Zhang H. Ye X. et al.Intradialytic hypoxemia and clinical outcomes in patients on hemodialysis.Clin J Am Soc Nephrol. 2016; 11: 616-625Crossref PubMed Scopus (39) Google Scholar This is a matter of concern, as PIH is associated with significantly increased morbidity and mortality.1Meyring-Wösten A. Zhang H. Ye X. et al.Intradialytic hypoxemia and clinical outcomes in patients on hemodialysis.Clin J Am Soc Nephrol. 2016; 11: 616-625Crossref PubMed Scopus (39) Google Scholar In a large retrospective study in 983 HD patients, the risk of all-cause hospitalization was 2.2-fold higher in PIH patients compared to the nonhypoxemic group. Mortality risk was also significantly higher in HD patients with PIH, regardless from the presence of heart failure or chronic obstructive pulmonary disease (COPD).1Meyring-Wösten A. Zhang H. Ye X. et al.Intradialytic hypoxemia and clinical outcomes in patients on hemodialysis.Clin J Am Soc Nephrol. 2016; 11: 616-625Crossref PubMed Scopus (39) Google Scholar Intriguingly, patients with PIH also have higher requirement of erythropoietin (EPO) and other erythropoiesis-stimulating agents (ESAs), a finding that is counterintuitive, because hypoxia is the main stimulus for residual endogenous production of EPO.1Meyring-Wösten A. Zhang H. Ye X. et al.Intradialytic hypoxemia and clinical outcomes in patients on hemodialysis.Clin J Am Soc Nephrol. 2016; 11: 616-625Crossref PubMed Scopus (39) Google Scholar To start addressing this issue, we reasoned that hypoxemia has been shown to impair resilience against reactive oxygen species (ROS) and may increase erythrocyte apoptosis (eryptosis), characterized by reduced cell dimension, bleb formation, and phosphatidylserine exposure.2Lang F. Lang E. Foller M. Physiology and pathophysiology of eryptosis.Transfus Med Hemother. 2012; 39: 308-314Crossref PubMed Scopus (103) Google Scholar Therefore, we hypothesized that higher ESA requirement is the result of higher red blood cell (RBC) fragility and shorter half-life in PIH patients, due to an accumulation of intracellular reactive oxygen species (iROS). To test this idea, we designed a cross sectional study to compare RBC phenotype and function in HD patients with or without PIH and in healthy controls. In HD patients, online intradialytic arterial oxygen saturation (from arteriovenous fistula) was monitored using the Crit-Line Monitor3Balter P. Artemyev M. Zabetakis P. Methods and challenges for the practical application of Crit-Line™ monitor utilization in patients on hemodialysis.Blood Purif. 2015; 39: 21-24Crossref PubMed Scopus (9) Google Scholar (see Supplementary Methods). We collected blood samples and measured eryptosis and iROS production in RBCs from patients with or without PIH (see Supplementary Methods). Finally, we measured reticulocytes (RCs) and levels of the different RC subsets. From March 2019 to October 2020, we included 56 patients on chronic HD and 18 healthy controls. Of the 56 HD patients, 12 had PIH. Although PIH has been previously associated with older age and longer dialysis vintage,1Meyring-Wösten A. Zhang H. Ye X. et al.Intradialytic hypoxemia and clinical outcomes in patients on hemodialysis.Clin J Am Soc Nephrol. 2016; 11: 616-625Crossref PubMed Scopus (39) Google Scholar in our cohort demographic characteristics did not significantly differ between HD patients with or without PIH. Kt/V was also similar between the 2 groups. Consistent with prior reports,1Meyring-Wösten A. Zhang H. Ye X. et al.Intradialytic hypoxemia and clinical outcomes in patients on hemodialysis.Clin J Am Soc Nephrol. 2016; 11: 616-625Crossref PubMed Scopus (39) Google Scholar PIH patients had lower hematocrit levels, despite significantly higher use of ESA (Table 14Daugirdas J.T. Second generation logarithmic estimates of single-pool variable volume Kt/V: an analysis of error.J Am Soc Nephrol. 1993; 4: 1205-1213Abstract Full Text Full Text PDF PubMed Scopus (1165) Google Scholar). Healthy controls were significantly younger than HD patients (Table 1).Table 1Patient characteristics and clinical parameters at sample collectionCharacteristic or parameterControls n = 18HD patients, total n = 58HD hypoxemic patients n = 12HD nonhypoxemic patients n = 46Age, yr32.3 ± 7.064.4 ± 14.468.2 ± 14.663.4 ± 14.3Men, %6 (33.3%)35 (59.3%)7 (58.3%)28 (60.9%)Race, %Asian2 (11.1%)3 (5.2%)1 (8.3%)2 (4.3%)Black0 (0%)29 (50.0%)5 (41.7%)24 (52.2%)White16 (88.9%)26 (44.8%)6 (50.0%)20 (43.5%)Ethnicity, %Hispanic or Latino11 (19.0%)1 (8.3%)10 (21.7%)Dialysis vintage, yr4 (1.75-8)1 (1-3)5 (2-9)Dialysis adequacy, Kt/VaKt/V was calculated based on the Daugirdas formula.41.60 ± 0.311.58 ± 0.291.60 ± 0.31Comorbidities, %Diabetes20 (34.4%)2 (16.7%)18 (39.1%)Hypertension45 (77.6%)8 (66.7%)37 (80.4%)COPD6 (10.3%)3 (25.0%)3 (6.5%)CHF19 (32.8%)3 (25%)16 (34.8%)Systolic BP145.1 ± 22.0140.5 ± 22.4146.4 ± 21.9Diastolic BP75.3 ± 14.568.8 ± 12.177.2 ± 14.7Laboratory testsWBC count, 1000/μl6.9 ± 2.17.36 ± 2.56.77 ± 2.0RBC count, 1000/μl3.8 ± 0.43.6 ± 0.43.79 ± 0.5sAlbumin, g/dl3.9 ± 0.33.8 ± 0.53.9 ± 0.3HGB, g/dl11.1 ± 1.210.5 ± 0.8111.2 ± 1.3HCT, %35.3 ± 3.833.70 ± 2.8135.69 + 3.95MCV, fl94.5 ± 6.794.4 ± 7.794.6 + 6.5MCH, pg/cell30.0 ± 2.430.0 ± 3.230.1 + 2.4RDW, %16.3 ± 2.017.12 ± 2.216.1 + 1.9Ferritin, ng/ml1002 (584.9-1256.5)625 (583.3-1069)1005 (594.5-1334)PTH, pg/ml596 (379-996.3)485 (383-793)632 (377-1048)Calcium8.9 ± 0.69.0 ± 0.48.8 ± 0.6Potassium5.0 ± 0.75.1 ± 0.55.0 ± 0.7TherapyESA, %26 (44.8%)9 (75.0%)17 (37.0%)Iron, %18 (31.0%)4 (33.3%)14 (30.4%)Antihypertensive, %11 (19.0%)2 (16.7%)9 (19.6%)ACEi, %4 (6.9%)0 (0%)4 (8.7%)Vitamin D, %54 (93.1%)10 (83.3%)44 (95.7%)Data are represented as mean ± standard deviation or as median (interquartile range). Numbers in boldface type are significantly different between groups. Significance set at P < 0.05.ACEi, angiotensin converting enzyme inhibitor; BP, blood pressure; CHF, congestive heart failure; COPD, chronic obstructive pulmonary disease; ESA, erythropoiesis stimulating agents; HCT, hematocrit; HD, hemodialysis; HGB, hemoglobin; MCH, mean corpuscular hemoglobin; MCV, mean corpuscular volume; PTH, parathyroid hormone; RBC, red blood cell; RDW, red cell distribution width; WBC, white blood cell.a Kt/V was calculated based on the Daugirdas formula.4Daugirdas J.T. Second generation logarithmic estimates of single-pool variable volume Kt/V: an analysis of error.J Am Soc Nephrol. 1993; 4: 1205-1213Abstract Full Text Full Text PDF PubMed Scopus (1165) Google Scholar Open table in a new tab Data are represented as mean ± standard deviation or as median (interquartile range). Numbers in boldface type are significantly different between groups. Significance set at P < 0.05. ACEi, angiotensin converting enzyme inhibitor; BP, blood pressure; CHF, congestive heart failure; COPD, chronic obstructive pulmonary disease; ESA, erythropoiesis stimulating agents; HCT, hematocrit; HD, hemodialysis; HGB, hemoglobin; MCH, mean corpuscular hemoglobin; MCV, mean corpuscular volume; PTH, parathyroid hormone; RBC, red blood cell; RDW, red cell distribution width; WBC, white blood cell. We started our analyses by measuring eryptosis and found that HD individuals with PIH had significantly higher levels of annexin-V−positive RBCs than healthy controls. In contrast, annexin-V−positive RBCs tended to be higher in HD patients without PIH compared to healthy controls, but this difference did not reach statistical significance (Figure 1a, b). We next tested the hypothesis that increased eryptosis in HD patients with PIH is due to increased intracellular ROS production. The analyses were performed within 2 hours from withdrawal to prevent altered iROS production. As shown in Figure 1c, d, HD patients with PIH had significantly higher iROS levels than nonhypoxemic patients. Consistent with previously published results,S1 there was no significant difference in the iROS levels between healthy controls and nonhypoxemic HD patients, regardless of hypoxic status. Altogether, these data support the hypothesis that PIH increases iROS production in RBCs, leading to accelerated RBC eryptosis and loss. This mechanism explains, at least in part, the increased requirement for ESA in HD patients with PIH. An alternative, additional mechanism responsible for higher ESA requirement in PIH patients could be an impaired bone marrow response to ESA, leading to lower reticulocyte production. One of the main reasons for resistance to ESA therapy is chronic inflammation, a condition that often associates with PIH.1Meyring-Wösten A. Zhang H. Ye X. et al.Intradialytic hypoxemia and clinical outcomes in patients on hemodialysis.Clin J Am Soc Nephrol. 2016; 11: 616-625Crossref PubMed Scopus (39) Google Scholar We analyzed the levels of serum ferritin and albumin, markers of systemic inflammation.5Kalantar-Zadeh K. Kalantar-Zadeh K. Lee G.H. The fascinating but deceptive ferritin: to measure it or not to measure it in chronic kidney disease?.Clin J Am Soc Nephrol. 2006; 1: S9-S18Crossref PubMed Scopus (141) Google Scholar Ferritin levels in HD patients were high overall, but did not significantly differ between patients with or without PIH. Similarly, we found no significant difference in serum albumin levels between the 2 HD groups (Table 1). Next, we measured percentages of total reticulocytes and of those with high, intermediate, and low expression of CD71 (transferrin receptor), the expression of which is lost during reticulocytes maturation6Borrione P. Spaccamiglio A. Parisi A. et al.A biparametric flow cytometry analysis for the study of reticulocyte patterns of maturation.Int J Lab Hematol. 2010; 32: 65-73Crossref PubMed Scopus (6) Google Scholar (see Supplementary Methods). We found no significant differences in the total reticulocyte percentages (Figure 1e, f) or in any of the 3 subpopulations (Figure 1g), suggesting no impaired response to EPO in HD patients with PIH. Multiple causes have been implicated in the pathogenesis of PIH, including, among others, intradialytic hypotension, sleep apnea, congestive heart failure, and COPD.7Campos I. Chan L. Zhang H. et al.Intradialytic hypoxemia in chronic hemodialysis patients.Blood Purif. 2016; 41: 177-187Crossref PubMed Scopus (32) Google Scholar Regardless of its etiology, PIH is a serious medical condition associated with intradialytic complications, such as hypotension and cramps,7Campos I. Chan L. Zhang H. et al.Intradialytic hypoxemia in chronic hemodialysis patients.Blood Purif. 2016; 41: 177-187Crossref PubMed Scopus (32) Google Scholar but also with poor clinical outcomes in the long term.1Meyring-Wösten A. Zhang H. Ye X. et al.Intradialytic hypoxemia and clinical outcomes in patients on hemodialysis.Clin J Am Soc Nephrol. 2016; 11: 616-625Crossref PubMed Scopus (39) Google Scholar Hemodialysis patients with PIH have increased ESA requirements, and our data support the concept that this phenomenon is due to increased RBC eryptosis and turnover, induced by iROS accumulation. The levels of total reticulocytes and the percentages of reticulocytes in different stages of maturation were not significantly different between PIH patients and non-PIH controls, suggesting that their response to EPO is not impaired.1Meyring-Wösten A. Zhang H. Ye X. et al.Intradialytic hypoxemia and clinical outcomes in patients on hemodialysis.Clin J Am Soc Nephrol. 2016; 11: 616-625Crossref PubMed Scopus (39) Google Scholar Although EPO deficiency represents a central mechanism in the pathogenesis of anemia in HD patients, the process is indeed multifactorial, and involves systemic inflammation that reduces the response to EPO by erythroid cells in the bone marrow. Although HD patients in our study showed signs of inflammation compared to healthy controls, we did not detect significant differences in ferritin or serum albumin levels between PIH and non-PIH patients, suggesting that increased ESA requirement in PIH patients was not due to more severe chronic inflammation. Reduced RBC survival in HD patients can be promoted by different stimuli, including oxidative stress, various diseases (including diabetes, heart failure, and sepsis) and also uremic toxins8Lang E. Lang F. Triggers, inhibitors, mechanisms, and significance of eryptosis: the suicidal erythrocyte death.Biomed Res Int. 2015; 2015: 513518Crossref PubMed Scopus (92) Google Scholar that promote eryptosis by increasing iROS production.9Dias G.F. Bonan N.B. Steiner T.M. et al.Indoxyl sulfate, a uremic toxin, stimulates reactive oxygen species production and erythrocyte cell death supposedly by an organic anion transporter 2 (OAT2) and NADPH oxidase activity-dependent pathways.Toxins (Basel). 2018; 10Crossref PubMed Scopus (24) Google Scholar A recent study showed that uremic toxin indoxyl sulfate increases iROS production in a dose-dependent manner, in both hypoxic and nonhypoxic conditions, suggesting that hypoxia and uremic toxins promote eryptosis in an independent manner.S2 We did not have information on uremic toxins in our cohort, but this is a variable that will be important to consider in future studies. The major limitation of our report is the relatively small sample size. However, the results of our comprehensive RBC phenotypic analyses will form the basis for the sample size estimation of larger investigations. The pathophysiology of PIH is only partially understood and our findings are important, as they support the concept that increased ESA requirement in PIH patients is due, at least in part, to increased RBC fragility. Our study provides a rationale for future studies testing the hypothesis that therapeutic strategies for anemia in PIH should aim at prolonging RBC survival (possibly by reducing ROS production), together with increasing erythrocyte production. All the authors declared no competing interests. This project was supported by a grant from Renal Research Institute (RRI), New York, NY. Download .pdf (.09 MB) Help with pdf files Supplementary File (PDF) Supplementary Methods Supplementary References