Abstract
Rationale & Objective
The long-term outcomes associated with intravenous (IV) versus oral iron replacement therapy in patients with chronic kidney disease are unclear. We investigated the association of IV versus oral iron replacement therapy with incident end-stage kidney disease (ESKD) and with all-cause mortality in a national cohort of US veterans with an estimated glomerular filtration rate of < 60mL/min/1.73m2.
Study Design
A national Veterans Affairs (VA) historical cohort study.
Setting & Participants
Total of 17,428 incident new users of oral iron replacement therapy and a comparable group of 879 incident new users of IV iron replacement therapy during October 1, 2004, through September 30, 2006, with longitudinal follow-up until September 30, 2019.
Predictor
Incident new use of oral versus IV IRT.
Outcomes
ESKD and mortality.
Analytical Approach
We examined the association of oral versus IV IRT with outcomes using competing risk regression and Cox models. We used propensity score matching to account for differences in baseline characteristics.
Results
Event rates were overall higher in patients receiving IV iron replacement, but the risk was mitigated after propensity scores matching, with no significant association observed between the type of iron therapy and ESKD (subhazard ratio: 1.07; 95% CI, 0.82-1.39; P = 0.6), or all-cause mortality (hazard ratio: 1.07; 95% CI, 0.96-1.20; P = 0.2).
Limitations
Potential selection bias and residual confounding.
Conclusions
In this large national cohort of patients with chronic kidney disease, IV iron was not significantly associated with a higher risk of incident ESKD and all-cause mortality when compared to oral iron therapy. The comparative effectiveness of the different iron replacement modalities will need to be studied further in clinical trials.
Index Words: anemia, IV iron, oral iron, chronic kidney disease, end-stage kidney disease, mortality
Plain-Language Summary
Iron replacement therapy is commonly used in patients with advanced chronic kidney disease to treat iron deficiency anemia, but its effects on clinical outcomes (eg, survival or kidney failure) are unknown. We used a large nationwide cohort of US veterans to identify 18,307 patients with advanced chronic kidney disease, who were newly started on iron replacement therapy, including IV or oral iron replacement therapy. We compared associations of IV versus oral iron replacement with clinical outcomes. We found that there was no difference regarding survival and kidney failure with IV iron when compared with the oral iron therapy group.
Iron is a significant component of hemoglobin and a crucial element in red blood cell production. Furthermore, iron plays an important role in several other biologic pathways such as deoxyribonucleic acid synthesis and oxygen and electron transport.1 Iron deficiency is associated with adverse outcomes such as mortality and kidney failure in patients with chronic kidney disease (CKD).2 Furthermore, iron deficiency is also associated with higher risk of heart failure and mortality even in non-anemic patients (non-anemic absolute iron deficiency).3, 4, 5 Observational studies suggested that iron replacement therapy (IRT) is associated with lower mortality in anemic and iron deficient patients with CKD,6 although clinical trials examining the long-term benefits of IRT in patients with CKD are lacking.
Iron replacement therapy in CKD is used primarily as an ancillary therapy for anemia of CKD, and it can be implemented via oral and intravenous (IV) routes. The repletion route is determined by the severity of the iron deficiency anemia, the patient’s response (or lack thereof), the patient’s tolerance, or the availability (or lack thereof) of certain types of iron compounds. Oral iron is readily available, and it has a lower risk of severe adverse events, but it also has a slower and less efficient response, and its use is often marred by poor tolerance. IV iron has a faster response and it is more effective when compared to oral iron, but it is more cumbersome to administer, and it has been associated with acute reactions that may cause hypotension7 and concerns about causing oxidative stress, which could theoretically lead to a higher risk of cardiovascular disease, infection, kidney damage, and mortality.8,9 Despite the widely held belief that oral IRT is free of severe adverse effects, administration of oral iron has been reported to induce changes in the gut microbiome, which may predispose to gut inflammation and increased production and absorption of uremic toxins, potentially leading to poor renal and cardiovascular results.10
Studies comparing IV to oral iron in patients with CKD have been limited by relatively small sample sizes and short duration of follow-up, and hence they examined primarily laboratory end points (hemoglobin and iron parameters) or short-term safety outcomes.11,12 The long-term effects of IV versus oral iron therapy on clinical outcomes such as mortality or incidence of end-stage kidney disease (ESKD) in patients with CKD remain unclear. We hypothesized that the use of IV iron would be associated with higher risk of mortality and ESKD. To inform the field, we investigated the association of IV versus oral iron with all-cause mortality and with the incidence of ESKD in a large cohort of US veterans with nondialysis-dependent CKD.
Methods
Cohort Definition
We examined data from the Therapeutic Interventions in CKD (TRI-CKD) study. This is a retrospective cohort of 3,562,882 US veterans with estimated glomerular filtration rates (eGFRs) of ≥ 60 mL/min/1.73m2 recorded from October 1, 2004 to September 30, 2006, with longitudinal follow-up until September 30, 2019.13 We collected longitudinally patients’ subsequent eGFR values to identify those who developed incident eGFR <60 ml/min/1.73m2. We identified incident new use of IV or oral IRT following the development of incident CKD (Figure 1). We first excluded 2,992,362 patients who were not incident new iron users and 35,850 who started iron therapy after renal replacement therapy. We then excluded 466,774 patients with missing information about key characteristics at baseline (ie, at the time when they started IRT) and 49,589 patients with baseline eGFRs of ≥ 60 mL/min/1.73m2. Our final analytical sample thus consisted of 18,307 patients with eGFR < 60 mL/min/1.73m2, of whom 17,428 (95.2%) started oral iron and 879 (4.8%) started IV iron following the development of CKD. Among the 879 incident new users of IV iron, 254 patients (29%) did not receive any oral iron during the follow-up period. Of the 254 patients receiving only IV iron therapy, 176 patients (69%) received repeated doses of IV iron during follow-up.
Figure 1.
Flow chart of cohort creation.
Data Collection
We obtained information about baseline demographic characteristics, comorbid conditions, medications, vital signs, and laboratory variables from the Veterans Affairs (VA) Corporate Data Warehouse.14 Medication data were extracted from the Decision Support System National Data Extracts’ outpatient and inpatient pharmacy files and from Medicare Part D files.15 We also obtained information from the Corporate Data Warehouse about medications received from non-VA sources, including over the counter medications, herbal supplements, VA-prescribed medications filled at non-VA pharmacies, and medications prescribed by providers outside the VA. We defined baseline other medication use as the presence of at least one outpatient dispensation of ≥30 days during the 365 days before the baseline date. We identified comorbid conditions from the VA Inpatient and Outpatient Medical SAS Datasets based on the presence of International Classification of Disease-9 and International Classification of Disease -10 diagnostic and procedure codes and current procedural terminology codes, as well as from Centers for Medicare and Medicaid Services Data files. We used the presence of at least 1 inpatient or at least 2 outpatient codes recorded before the baseline date to define a comorbid condition. We calculated the Charlson Comorbidity Index using the Deyo modification for administrative data sets.16 Relevant laboratory characteristics were obtained from the VA LabChem files17 and eGFR was estimated from the 2009 Chronic Kidney Disease Epidemiology Collaboration equation.18 We collected information about urine protein–creatinine ratio, urine albumin–creatinine ratio, and urine dipstick protein from the Decision Support System National Data Extracts Laboratory Results file and the VA LabChem file in the Corporate Data Warehouse. We converted urine protein–creatinine ratio and urine dipstick protein to urine albumin–creatinine ratio using the conversion equations by Sumida et al19 and categorized the resulting urine albumin–creatinine ratio values as <30 mg/g; 30-<300 mg/g or ≥300 mg/g.
Exposure and Outcomes
We used an incident new user design to define exposure to IV or oral IRT. We considered patients as incident new users if they received a dispensation of ≥30 days of oral iron or any prescription for a course of IV iron, preceded by no dispensation of the same drugs during the 365 days before the first prescription, while having a record of VA pharmacy enrollment. We examined associations of IV versus oral IRT with outcomes using an intention-to-treat-like design where the treatment exposure at cohort entry is carried forward irrespective of future treatment status or cross-over in exposure. Because real-life exposure to IV iron in patients with nondialysis-dependent CKD is typically intermittent and often interspersed with PO iron administration, we conducted sensitivity analyses where we limited the IV iron group to patients who received exclusively IV iron, and (in a separate analysis) to patients who received repeated doses of exclusively IV iron. To describe overall exposure to iron therapy, we calculated the proportion of days covered (PDC).
Follow-up was started on the date of receiving the first iron prescription. Our co-primary outcomes were ESKD, defined as the initiation of dialysis or pre-emptive kidney transplant identified from the US Renal Data System,20 and all-cause mortality identified from the VA vital status files.21 We followed patients until the occurrence of any of the above outcomes, last recorded VA encounter date or end of follow-up (September 30, 2019 for mortality or June 30, 2018 for ESKD).
Statistical Analysis
We presented data as number (%) for categorical variables and mean with standard deviation (SD) or median with 25th-75th percentile for continuous variables, as appropriate. We compared characteristics between IV and oral iron users using standardized differences. We calculated cumulative incidence rates per 1,000 patient-years overall and stratified by treatment status. We examined the association of IV versus oral IRT with ESKD in competing risk regression models (with mortality as the competing event) using the Fine and Gray method22 and with all-cause mortality using Cox proportional hazard models. Propensity scores (PS) were calculated from baseline characteristics using logistic regression models, using as predictors the 180-day calendar window around the baseline date, patients’ baseline age, sex, race, ethnicity, marital status, insurance type, military service connectedness, income, baseline use of medications (erythropoiesis-stimulating agents, renin–angiotensin-aldosterone system inhibitors , potassium-sparing diuretics and mineralocorticoid receptor antagonists, thiazide diuretics, loop diuretics, other blood pressure–lowering agents, proton pump inhibitors, non-steroidal anti-inflammatory agents and opioid analgesics), comorbid conditions (diabetes mellitus, myocardial infarction, peripheral vascular and cerebrovascular disease, congestive heart failure, and the Charlson Comorbidity Index), body mass index, systolic and diastolic blood pressure, and baseline eGFR, urine albumin–creatinine ratio, hemoglobin, total iron saturation and serum ferritin. We used PS matching using 1-to-1 nearest-neighbor matching as our primary approach, with a PS-overlap weighting method as sensitivity analysis to account for differences in baseline characteristics between patients treated with oral or IV iron therapy. We examined heterogeneity of treatment effects by key baseline characteristics in subgroup analyses with the calculation of multiplicative interaction terms, using the same statistical approach as detailed above.
Analyses were conducted using Stata MP version 17.1 (StataCorp) and SAS 9.4 (SAS Institute Inc). The study was approved by the institutional review boards of the Memphis and Long Beach VA Medical Centers, with exemption from informed consent.
Results
Patients were overall 73 ± 10 years old, 97% were male, 75% were White, 17% were African American, and 62% had diabetes (Table 1). The mean ± SD baseline eGFR and blood hemoglobin and median (25th-75th percentile) iron saturation and serum ferritin levels were 43 ± 13 mL/min/1.73m2, 10.6 ± 1.8 g/dl, 13% (8-18) and 78 mcg/L (26-212), respectively. Patients starting IV iron had a higher prevalence of comorbid conditions, were less likely to be African American, and had lower eGFR, blood hemoglobin, total iron saturation and higher serum ferritin levels (Table 1). The baseline characteristics of patients receiving IV and oral iron were similar after PS matching (Table 1 and Table S1). The median PDC for oral iron users was 37%, with 25% of patients experiencing a PDC of >80%. The median PDC for exclusive IV iron users was 2.3%, with 6% of patients experiencing a PDC of >80%.
Table 1.
Baseline Characteristics of the Overall and the PS-Matched Cohort Receiving IV Iron Replacement and Oral Iron Replacement Therapies
| Overall Cohort |
PS-Matched Cohort |
||||||
|---|---|---|---|---|---|---|---|
| All |
Iron Treatment (IV) |
Iron Treatment (PO) |
All |
Iron Treatment (IV) |
Iron Treatment (PO) |
Standardized Difference | |
| (N = 18,307) | (n = 879) | (n = 17,428) | (n = 1,756) | (n = 878) | (n = 878) | ||
| Age (y) | 73 ± 10 | 71 ± 11 | 73 ± 10 | 71 ± 11 | 71 ± 11 | 71 ± 11 | 0.004 |
| Gender (Male) | 17,737 (97) | 853 (97) | 16,884 (97) | 1,709 (97) | 852 (97) | 857 (98) | −0.036 |
| Race | 0.000 | ||||||
| White | 13,795 (75) | 608 (69) | 13,187 (76) | 1,202 (68) | 608 (69) | 594 (68) | |
| African American | 3,019 (17) | 135 (15) | 2,884 (17) | 296 (17) | 134 (15) | 162 (18) | |
| Other | 1,493 (8.2) | 136 (16) | 1,357 (7.8) | 258 (15) | 136 (15) | 122 (14) | |
| Ethnicity (Hispanic) | 1,015 (5.5) | 46 (5.2) | 969 (5.6) | 90 (5.1) | 46 (5.2) | 44 (5) | 0.01 |
| Marital status | −0.029 | ||||||
| Single | 1,409 (7.7) | 72 (8.2) | 1,337 (7.7) | 167 (9.5) | 72 (8.2) | 95 (11) | |
| Married | 8,863 (48) | 401 (46) | 8,462 (49) | 803 (46) | 401 (46) | 402 (46) | |
| Divorced | 5,007 (27) | 290 (33) | 4,717 (27) | 543 (31) | 289 (33) | 254 (29) | |
| Widowed | 2,558 (14) | 103 (12) | 2,455 (14) | 204 (12) | 103 (12) | 101 (12) | |
| Unknown | 470 (2.6) | 13 (1.5) | 457 (2.6) | 39 (2.2) | 13 (1.5) | 26 (3) | |
| Service connected | 8,898 (49) | 447 (51) | 8,451 (49) | 891 (51) | 446 (51) | 445 (51) | −0.002 |
| Insurance type | −0.014 | ||||||
| None | 1,409 (7.7) | 72 (8.2) | 1,337 (7.7) | 433 (25) | 218 (25) | 215 (24) | |
| Medicare | 3,825 (21) | 218 (25) | 3,607 (21) | 1,191 (68) | 590 (67) | 601 (68) | |
| Medicaid | 113 (0.6) | 8 (0.9) | 105 (0.6) | 12 (0.7) | 8 (0.9) | 4 (0.5) | |
| Other | 1,294 (7.1) | 63 (7.2) | 1,231 (7.1) | 120 (6.8) | 62 (7) | 58 (6.6) | |
| Incomea | 16,228 (8,250-33,228) | 17,194 (8,527-33,876) | 16,200 (8,208-33,228) | 16,045 (8,323-33,876) | 17,250 (8,532-33,876) | 15,644 (7,927-33,792) | −0.017 |
| Erythropoiesis-stimulating agent | 319 (1.7) | 21 (2.4) | 298 (1.7) | 47 (2.7) | 21 (2.4) | 26 (3) | 0.035 |
| RAAS inhibitors | 12,249 (67) | 568 (65) | 11,681 (67) | 1,123 (64) | 567 (65) | 556 (63) | −0.026 |
| Thiazide diuretics | 5,954 (33) | 284 (32) | 5,670 (33) | 552 (31) | 284 (32) | 268 (31) | −0.039 |
| Other antihypertensives | 12,935 (71) | 643 (73) | 12,292 (71) | 1,283 (73) | 642 (73) | 641 (73) | −0.002 |
| Loop diuretics | 7,850 (43) | 461 (52) | 7,389 (42) | 905 (52) | 461 (53) | 444 (51) | −0.038 |
| Potassium-sparing diuretics | 2,314 (13) | 128 (15) | 2,186 (13) | 251 (14) | 128 (15) | 123 (14) | −0.016 |
| Mineralocorticoid receptor antagonists | 2,035 (11) | 119 (14) | 1,916 (11) | 231 (13) | 119 (14) | 112 (13) | −0.023 |
| Proton pump inhibitors | 8,992 (49) | 388 (44) | 8,604 (49) | 776 (44) | 388 (44) | 388 (44) | 0.000 |
| NSAIDs | 8,790 (48) | 412 (47) | 8,378 (48) | 819 (47) | 412 (47) | 407 (46) | −0.011 |
| Opioids | 6,162 (34) | 323 (37) | 5,839 (34) | 642 (37) | 322 (37) | 320 (36) | −0.004 |
| Diabetes mellitus | 11,298 (62) | 565 (64) | 10,733 (62) | 1,125 (64) | 564 (64) | 561 (64) | −0.007 |
| Myocardial infarction | 4,258 (23) | 228 (26) | 4,030 (23) | 448 (26) | 228 (26) | 220 (25) | −0.020 |
| Peripheral vascular disease | 5,602 (31) | 312 (36) | 5,290 (30) | 618 (35) | 312 (36) | 306 (35) | −0.014 |
| Cerebrovascular disease | 4,908 (27) | 250 (28) | 4,658 (27) | 494 (28) | 250 (28) | 244 (28) | −0.015 |
| Congestive heart failure | 7,327 (40) | 427 (49) | 6,900 (40) | 859 (49) | 427 (49) | 432 (49) | 0.011 |
| Charlson comorbidity index | 5.1 ± 2.9 | 5.7 ± 2.8 | 5 ± 2.9 | 5.7 ± 2.8 | 5.7 ± 2.8 | 5.7 ± 2.9 | 0.020 |
| Body mass index (kg/m2) | 29 ± 6.3 | 29 ± 6.3 | 29 ± 6.5 | 29 ± 6.5 | 29 ± 6.5 | 29 ± 6.5 | −0.004 |
| Systolic blood pressure (mm Hg) | 129 ± 21 | 131 ± 23 | 129 ± 21 | 131 ± 23 | 131 ± 23 | 131 ± 23 | 0.016 |
| Diastolic blood pressure (mm Hg) | 69 ± 12 | 69 ± 12 | 69 ± 12 | 69 ± 12 | 69 ± 12 | 70 ± 13 | 0.053 |
| Estimated GFR (mL/min/1.73m2) | 43 ± 13 | 34 ± 17 | 44 ± 13 | 34 ± 16 | 34 ± 17 | 34 ± 17 | −0.018 |
| UACR (mg/g) | 0.059 | ||||||
| <30 | 11,617 (64) | 439 (50) | 11,178 (64) | 840 (48) | 439 (50) | 401 (46) | |
| 30-300 | 3,751 (21) | 178 (20) | 3,573 (21) | 387 (22) | 178 (20) | 209 (24) | |
| >300 | 2,939 (16) | 262 (30) | 2,677 (15) | 529 (30) | 261 (30) | 268 (31) | |
| Ferritina (mcg/L) | 78 (26-212) | 107 (33-253) | 77 (26-210) | 105 (33-259) | 107 (33-253) | 103 (34-262) | 0.007 |
| Iron saturation∗ (%) | 13 (8-18) | 10 (6-16) | 13 (9-18) | 11 (6.7-16) | 10 (6.4-16) | 11 (7-16) | 0.010 |
| Hemoglobin (g/dL) | 10.6 ± 1.8 | 9.2 ± 1.6 | 10.6 ± 1.8 | 9.2 ± 1.6 | 9.2 ± 1.6 | 9.2 ± 1.6 | −0.002 |
Note: Data presented as mean ± standard deviation.
Abbreviations: GFR, glomerular filtration rate; NSAIDs, non-steroidal anti-inflammatory agents; IV, intravenous; PS, propensity scores; RAAS, renin–angiotensin-aldosterone system; UACR, urine albumin–creatinine ratio.
Median (25th-75th percentile) and number (percent).
Association of IV Versus Oral Iron Therapy With Risk of ESKD and Mortality
A total of 973 (5.31%) patients experienced incident ESKD (event rate, 23.06; 95% CI, 21.61-24.51 per 1,000 PY) and 12,497 patients (68%) died (event rate 282.91; 95% CI, 277.95-287.87 per 1,000 PY), during a median follow-up of 1.4 years. Crude event rates and unadjusted subhazard ratios for both outcomes were higher in patients receiving IV (vs oral) iron replacement in the overall cohort (Table S2), but the risk was mitigated after PS matching, with no significant association observed between the type of IRT and ESKD or mortality (Table 2). In the PS weighted analysis, IV (vs oral) IRT was not associated with a statistically significant risk of ESKD but showed a small but statistically significant association with all-cause mortality (Table S2). The association of IV (vs oral) IRT with ESKD and with mortality was similar in subgroup analyses, with none of interaction analyses showing statistical significance (Figure 2 and Figure S1). Results were similar when limiting the IV iron group to patients who used exclusively IV iron throughout their follow-up, and when limiting the latter group to those who received more than one dose of IV iron (Tables S3 and S4).
Table 2.
Event Rates, (sub)Hazard Ratios and 95% Confidence Intervals of Outcomes Associated with IV Iron Replacement Compared With Oral Iron Replacement in the PS-Matched Cohort
| ESKD |
All-Cause Death |
|||||
|---|---|---|---|---|---|---|
| Event Rate per 1,000 PY (95%CI) | Sub Hazard Ratio (95%CI) | P | Event Rate per 1,000 PY (95%CI) | Hazard Ratio (95%CI) | P | |
| Oral iron (n = 878) | 78.23 (64.84-94.38) | Referent | 0.62 | 385.5 (356.57-416.78) | Referent | 0.2 |
| IV iron (n = 878) | 86.15 (71.76-103.43) | 1.07 (0.82-1.39) | 416.82 (386.02-450.07) | 1.07 (0.96-1.2) | ||
Abbreviations: ESKD, end-stage kidney disease; IV, intravenous; PS, propensity score.
Figure 2.
Forest plot of (sub)hazard ratios and 95% confidence intervals for ESKD and all-cause mortality associated with IV iron replacement therapy in the PS-matched cohort. ESKD, end-stage kidney disease; IV, intravenous.
Discussion
We examined the association of IV versus oral iron therapy with the risk of incident ESKD and all-cause mortality in a large national cohort of US veterans with eGFR < 60 mL/min/1.73m2. We found that the risk of these long-term clinical outcomes was similar in patients treated with the different forms of IRT. We noted a small association of IV (vs oral) IRT with higher all-cause mortality in our sensitivity analysis. The validity and clinical relevance of this finding will require additional studies.
Anemia is commonly seen in CKD and its severity increases with advancement of CKD.23 Anemia in CKD is mostly due to reduced production of erythropoietin, but in addition to this, iron deficiency also contributes significantly/commonly to anemia of CKD.23 Iron deficiency could be due to true depletion of iron store or reduced mobilization from iron stores due to an inflammatory state. The most common indication for IRT in CKD is iron deficiency anemia, but the putative benefits of iron replacement may extend beyond the improvement of hemoglobin levels. Because of iron’s broader physiologic roles, iron deficiency has been noted to cause functional skeletal muscle impairment with improvement in muscle function with iron supplementation.24, 25, 26 Furthermore, IRT has also been shown to reduce the risk of hospitalization for heart failure or cardiovascular death in patients with congestive heart failure.27,28
The Kidney Disease: Improving Global Outcomes guidelines recommends a trial of PO or IV iron in patients with CKD who meet criteria for iron deficiency anemia before starting ESA,29 with iron deficiency anemia diagnosed using thresholds of transferrin saturation ≤30% or serum ferritin ≤500 μg/L. In clinical practice, the route of iron administration is most often determined by the severity of anemia and iron deficiency,30 in addition to the availability, affordability, and individual efficiency and tolerance of various agents. IV iron has been noted to have superior efficacy to oral iron therapy in the correction of hemoglobin levels of non-dialysis-dependent patients with CKD with anemia and low iron indices.29,31,32 Notwithstanding its superior efficacy on hemoglobin increase, the cumbersome nature of its administration, its higher cost, and potential concerns about its safety limit the widespread use of IV iron in this population. Our data shows that the vast majority (95%) of de-novo prescriptions for IRT consist of oral formulations in patients with nondialysis-dependent CKD. This observation suggests that despite its superior efficacy in correcting anemia and replacing iron stores, IV iron is rarely used as a first-line therapy and may primarily be reserved to patients with CKD who have an insufficient response or intolerance to oral iron products.
Besides cost, access and availability, another potential reason for why IV iron use is less common in patients with CKD is concern about its safety when compared to oral iron. Intravenous administration results in the bypassing of the natural regulatory mechanisms that prevent iron overload by limiting its intestinal absorption under conditions of iron excess. Iron overload can result in a host of potential adverse outcomes, including organ damage, increased risk of infection, and cancer risk.33 Experimental studies have also shown an effect of iron overload on atherosclerosis.34,35 Iron binds to circulating transferrin, but under conditions of excess iron there may be residual circulating free iron, unbound to transferrin, which can trigger iron-dependent oxidative stress and can accelerate disease progression by producing lipid oxidation, protein oxidation, inflammation, and endothelial dysfunction.36 Because of the nature of IV iron replacement (large amount of iron injected over a very short time directly into the venous circulation), these mechanisms are much more likely to be triggered with IV versus oral iron replacement. Additionally, hypersensitivity reactions have been another concerning complication of IV iron therapy, most likely related to IgE and IgG-mediated mechanisms in response to the dextran component, as well as complement activation.37,38 These concerns have been less prominent lately with the advent of newer IV iron formulations.38,39 Oral iron is devoid of most of the aforementioned concerns, and has therefore been generally regarded as a safer (albeit less effective) method of iron replacement. Nevertheless, a few recent studies have raised concerns about the impact of oral iron on gastrointestinal function and on the gut microbiome.40,41 By inducing constipation and by altering the composition of the gut microbiome, oral iron may facilitate intestinal permeability and the absorption of uremic toxins; as a result, it may lead to an increase in inflammation and oxidative stress and ultimately worsen clinical outcomes.40,42
Because of the aforementioned putative safety concerns of both IV and oral iron and the lack of long-term comparative effectiveness research regarding IV versus oral iron, there is currently equipoise regarding their long-term clinical impact. Our study showed that neither form of IRT has an advantage (or disadvantage) when examining their association with relevant long-term clinical outcomes such as all-cause mortality or the incidence of ESKD.
These results support those of short-term clinical trials reporting similar progression of CKD11 in patients treated with IV versus oral iron and also support clinical guidelines that do not favor one modality over the other and allow for individualized clinical decision-making.43,44 Further information from larger clinical trials powered to study long-term clinical outcomes would be welcome.
Our study is notable for the large patient base that was analyzed, being nationally representative and having detailed information on key confounders. Our study also has limitations that should be acknowledged. Our cohort consisted of mostly male US veterans, and hence it is unclear if the results apply to patients with CKD treated in different health care systems. Also, we did not consider the type and dosage of the agents used. Despite our efforts to account for major known confounders, the observational and retrospective design makes our results prone to unmeasured confounding. Even though we examined a very large base population, due to prevailing clinical practices we identified a relatively low number of incident new IV iron users and an even smaller number of patients who used exclusively IV iron throughout their follow-up, which limited the power of our analyses. Few patients in our cohort received iron exposure continuously (as indicated by the low PDC values, especially in the IV iron groups), which is indicative of prevailing treatment practices, and which makes it difficult to determine effectiveness or safety with continuous iron administration in this population.
Conclusion
In this large national cohort of patients with CKD, IV iron therapy was associated with a similar risk of ESKD and mortality when compared to oral iron therapy. The long-term clinical effects of IV versus oral iron therapy will need to be examined in clinical trials. Short of such trials, the results of our study support the current clinical practice of choosing the route of IRT based on considerations of short-term efficacy and safety in addition to individual patient characteristics and resource availability.
Article Information
Authors’ Full Names and Academic Degrees
Adedamola M. Adeboye, MD, Prabin Shrestha, PhD, Keiichi Sumida, MD, PhD, Fridtjof Thomas, PhD, Connie M. Rhee, MD, Kamyar Kalantar-Zadeh, MD, PhD, and Csaba P. Kovesdy, MD
Authors’ Contributions
Research idea and study design: AMA, PS, KS, KK-Z, CMR, and CPK; data acquisition: PS and CPK; data analysis/interpretation: AMA, PS, KS, KK-Z, and CPK; statistical analysis: AMA, FT, and CPK; supervision or mentorship: KK-Z, CMR, and CPK. Each author contributed important intellectual content during article drafting or revision and accepts accountability for the overall work by ensuring that questions pertaining to the accuracy or integrity of any portion of the work are appropriately investigated and resolved.
Support
The study is supported by grant I01HX002680 from the US Department of Veterans Affairs Health Services Research and Development Service (HSR&D) to Dr Kovesdy, and by resources from the Memphis and Long Beach VA Medical Centers. Data reported here have been supplied by the United States Renal Data System (USRDS). Support for VA/Centers for Medicare and Medicaid Services data is provided by the Department of Veterans Affairs, Veterans Health Administration, Office of Research and Development, Health Services Research and Development, VA Information Resource Center (Project Numbers SDR 02-237 and 98-004). The interpretation and reporting of these data are the responsibility of the authors and in no way should be seen as official policy or interpretation of the Department of Veterans Affairs or the US government.
Financial Disclosure
Dr Kovesdy has been a consultant for Ardelyx, Astra Zeneca, Bayer, Boehringer Ingelheim, GSK, and Takeda. Dr Rhee has previously received honoraria from Astra Zeneca, Dexcom, Fresenius, and Vifor. Dr Kalantar-Zadeh has received honoraria or support from the National Institutes of Health (NIH), Agency for Healthcare Research and Quality (AHRQ), UpToDate, CSL, and Kabi. Drs Kovesdy, Rhee and Kalantar-Zadeh are employees of the Department of Veterans Affairs. The remaining authors declare that they have no relevant financial interests.
Prior Presentation
Portions of these data were presented in abstract form at the 2022 American Society of Nephrology Kidney Week Meeting.
Peer Review
Received April 1, 2024, as a submission to the expedited consideration track with 2 external peer reviews. Direct editorial input from the Editor-in-Chief. Accepted in revised form June 4, 2025.
Footnotes
Complete author and article information provided before references.
Figure S1: Forest plot of (sub)hazard ratios and 95% confidence intervals for ESKD and all-cause mortality associated with IV iron replacement therapy in the overall cohort.
Table S1: Baseline Characteristics of the PS Matched Cohorts of Patients Receiving Oral Iron vs. Exclusively IV Iron Replacement, and Oral Iron vs. Exclusively IV Iron Replacement With >1 Dose of IV Iron Received.
Table S2: Associations of IV (vs. Oral) Iron Therapy With Incident ESKD and All-Cause Mortality in the Overall Cohort.
Table S3: Associations of Exclusively IV Iron (vs. Oral Iron) Therapy, and Exclusively IV Iron With >1 Dose Received (vs Oral Iron) Therapy with Incident ESKD and All-Cause Mortality in Propensity Score Matched Models.
Table S4: Associations of Exclusively IV Iron (vs. Oral Iron) Therapy, and Exclusively IV Iron With >1 Dose Received (vs Oral Iron) Therapy With Incident ESKD and All-Cause Mortality in Propensity Score Overlap Weighted Models.
Supplementary Materials
Figure S1; Tables S1-S4.
References
- 1.Abbaspour N., Hurrell R., Kelishadi R. Review on iron and its importance for human health. J Res Med Sci. 2014;19(2):164–174. [PMC free article] [PubMed] [Google Scholar]
- 2.Kovesdy C.P., Estrada W., Ahmadzadeh S., Kalantar-Zadeh K. Association of markers of iron stores with outcomes in patients with nondialysis-dependent chronic kidney disease. Clin J Am Soc Nephrol. 2009;4(2):435–441. doi: 10.2215/cjn.03980808. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Schrage B., Rübsamen N., Schulz A., et al. Iron deficiency is a common disorder in general population and independently predicts all-cause mortality: results from the Gutenberg Health Study. Clin Res Cardiol. 2020;109(11):1352–1357. doi: 10.1007/s00392-020-01631-y. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Chopra V.K., Anker S.D. Anaemia, iron deficiency and heart failure in 2020: facts and numbers. ESC Heart Fail. 2020;7(5):2007–2011. doi: 10.1002/ehf2.12797. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Philip K.E.J., Sadaka A.S., Polkey M.I., Hopkinson N.S., Steptoe A., Fancourt D. The prevalence and associated mortality of non-anaemic iron deficiency in older adults: a 14 years observational cohort study. Br J Haematol. 2020;189(3):566–572. doi: 10.1111/bjh.16409. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Paul S., Shrestha P., Sumida K., et al. Association of oral iron replacement therapy with kidney failure and mortality in CKD patients. Clin Kidney J. 2023;16(11):2082–2090. doi: 10.1093/ckj/sfad190. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Notebaert E., Chauny J.-M., Albert M., Fortier S., Leblanc N., Williamson D.R. Short-term benefits and risks of intravenous iron: a systematic review and meta-analysis. Transfusion. 2007;47(10):1905–1918. doi: 10.1111/j.1537-2995.2007.01415.x. [DOI] [PubMed] [Google Scholar]
- 8.Tiwari A.K.M., Mahdi A.A., Chandyan S., et al. Oral iron supplementation leads to oxidative imbalance in anemic women: a prospective study. Clin Nutr. 2011;30(2):188–193. doi: 10.1016/j.clnu.2010.08.001. [DOI] [PubMed] [Google Scholar]
- 9.Mansour W., Bissram M., Rosner M.H. Intravenous iron therapy and risk for progressive loss of kidney function in patients with chronic kidney disease. Nephron Clin Pract. 2011;118(2):c189–c194. doi: 10.1159/000320707. [DOI] [PubMed] [Google Scholar]
- 10.Rydzewska-Rosołowska A., Sroka N., Kakareko K., Rosołowski M., Zbroch E., Hryszko T. The links between microbiome and uremic toxins in acute kidney injury: beyond gut feeling-A systematic review. Toxins (Basel) 2020;12(12):788. doi: 10.3390/toxins12120788. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Agarwal R., Kusek J.W., Pappas M.K. A randomized trial of intravenous and oral iron in chronic kidney disease. Kidney Int. 2015;88(4):905–914. doi: 10.1038/ki.2015.163. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Macdougall I.C., Bock A.H., Carrera F., et al. FIND-CKD: a randomized trial of intravenous ferric carboxymaltose versus oral iron in patients with chronic kidney disease and iron deficiency anaemia. Nephrol Dial Transplant. 2014;29(11):2075–2084. doi: 10.1093/ndt/gfu201. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Hassan W., Shrestha P., Sumida K., et al. Association of uric acid-lowering therapy with incident chronic kidney disease. JAMA Netw Open. 2022;5(6) doi: 10.1001/jamanetworkopen.2022.15878. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.VA Information Center VIRec Resource Guide. 2012. United States Department of Veterans Affairs; Health Services Research and Development. Service. [Google Scholar]
- 15.VHA Pharmacy Prescription Data . 2008. VA Information Center (VIRec) [Google Scholar]
- 16.Deyo R.A., Cherkin D.C., Ciol M.A. Adapting a clinical comorbidity index for use with ICD-9-CM administrative databases. J Clin Epidemiol. 1992;45(6):613–619. doi: 10.1016/0895-4356(92)90133-8. [DOI] [PubMed] [Google Scholar]
- 17.Kovesdy C.P., Alrifai A., Gosmanova E.O., et al. Age and outcomes associated with BP in patients with incident CKD. Clin J Am Soc Nephrol. 2016;11(5):821–831. doi: 10.2215/cjn.08660815. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Levey A.S., Stevens L.A., Schmid C.H., et al. A new equation to estimate glomerular filtration rate. Ann Intern Med. 2009;150(9):604–612. doi: 10.7326/0003-4819-150-9-200905050-00006. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Sumida K., Nadkarni G.N., Grams M.E., et al. Conversion of urine protein-creatinine ratio or urine dipstick protein to urine albumin-creatinine ratio for use in chronic kidney disease screening and prognosis: an individual participant-based meta-analysis. Ann Intern Med. 2020;173(6):426–435. doi: 10.7326/m20-0529. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Saran R., Li Y., Robinson B., et al. US renal data system 2015 annual data report: epidemiology of kidney disease in the United States. Am J Kidney Dis. 2016;67(3 suppl 1) doi: 10.1053/j.ajkd.2015.12.014. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Sohn M.-W., Arnold N., Maynard C., Hynes D.M. Accuracy and completeness of mortality data in the Department of Veterans Affairs. Popul Health Metr. 2006;4:2. doi: 10.1186/1478-7954-4-2. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Fine J.P., Gray R.J. A proportional hazards model for the subdistribution of competing risk. J Am Stat Assoc. 1999;94(446):496–509. doi: 10.1080/01621459.1999.10474144. [DOI] [Google Scholar]
- 23.Shaikh H., Hashmi M.F., Aeddula N.R. StatPearls. StatPearls Publishing; 2023. Anemia of chronic renal disease. [Google Scholar]
- 24.Halon-Golabek M., Borkowska A., Herman-Antosiewicz A., Antosiewicz J. Iron metabolism of the skeletal muscle and neurodegeneration. Front Neurosci. 2019;13:165. doi: 10.3389/fnins.2019.00165. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Agarwal R. Nonhematological benefits of iron. Am J Nephrol. 2007;27(6):565–571. doi: 10.1159/000107927. [DOI] [PubMed] [Google Scholar]
- 26.Davies K.J., Donovan C.M., Refino C.J., Brooks G.A., Packer L., Dallman P.R. Distinguishing effects of anemia and muscle iron deficiency on exercise bioenergetics in the rat. Am J Physiol. 1984;246(6 Pt 1):E535–E543. doi: 10.1152/ajpendo.1984.246.6.E535. [DOI] [PubMed] [Google Scholar]
- 27.Anker S.D., Kirwan B.-A., van Veldhuisen D.J., et al. Effects of ferric carboxymaltose on hospitalisations and mortality rates in iron-deficient heart failure patients: an individual patient data meta-analysis. Eur J Heart Fail. 2018;20(1):125–133. doi: 10.1002/ejhf.823. [DOI] [PubMed] [Google Scholar]
- 28.Graham F.J., Pellicori P., Kalra P.R., Ford I., Bruzzese D., Cleland J.G.F. Intravenous iron in patients with heart failure and iron deficiency: an updated meta-analysis. Eur J Heart Fail. 2023;25(4):528–537. doi: 10.1002/ejhf.2810. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Drüeke T.B., Parfrey P.S. Summary of the KDIGO guideline on anemia and comment: reading between the (guide)line(s) Kidney Int. 2012;82(9):952–960. doi: 10.1038/ki.2012.270. [DOI] [PubMed] [Google Scholar]
- 30.Macdougall I.C. Intravenous iron therapy in patients with chronic kidney disease: recent evidence and future directions. Clin Kidney J. 2017;10(suppl 1):i16–i24. doi: 10.1093/ckj/sfx043. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Van Wyck D.B., Roppolo M., Martinez C.O., Mazey R.M., McMurray S., United States Iron Sucrose (Venofer) Clinical Trials Group A randomized, controlled trial comparing IV iron sucrose to oral iron in anemic patients with nondialysis-dependent CKD. Kidney Int. 2005;68(6):2846–2856. doi: 10.1111/j.1523-1755.2005.00758.x. [DOI] [PubMed] [Google Scholar]
- 32.Agarwal R., Rizkala A.R., Bastani B., Kaskas M.O., Leehey D.J., Besarab A. A randomized controlled trial of oral versus intravenous iron in chronic kidney disease. Am J Nephrol. 2006;26(5):445–454. doi: 10.1159/000096174. [DOI] [PubMed] [Google Scholar]
- 33.Malesza I.J., Bartkowiak-Wieczorek J., Winkler-Galicki J., et al. The dark side of iron: the relationship between iron, inflammation and gut microbiota in selected diseases associated with iron deficiency anaemia-A narrative review. Nutrients. 2022;14(17):3478. doi: 10.3390/nu14173478. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34.Morrison H.I., Semenciw R.M., Mao Y., Wigle D.T. Serum iron and risk of fatal acute myocardial infarction. Epidemiology. 1994;5(2):243–246. doi: 10.1097/00001648-199403000-00015. [DOI] [PubMed] [Google Scholar]
- 35.Ma J., Zhang H., Chen Y., Liu X., Tian J., Shen W. The role of macrophage iron overload and ferroptosis in atherosclerosis. Biomolecules. 2022;12(11):1702. doi: 10.3390/biom12111702. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36.Habib A., Finn A.V. The role of iron metabolism as a mediator of macrophage inflammation and lipid handling in atherosclerosis. Front Pharmacol. 2014;5:195. doi: 10.3389/fphar.2014.00195. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 37.Rampton D., Folkersen J., Fishbane S., et al. Hypersensitivity reactions to intravenous iron: guidance for risk minimization and management. Haematologica. 2014;99(11):1671–1676. doi: 10.3324/haematol.2014.111492. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 38.Li X., Kshirsagar A.V., Brookhart M.A. Safety of intravenous iron in hemodialysis patients. Hemodial Int. 2017;21(suppl 1):S93–S103. doi: 10.1111/hdi.12558. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 39.Hamstra R.D., Block M.H., Schocket A.L. Intravenous iron dextran in clinical medicine. JAMA. 1980;243(17):1726–1731. doi: 10.1001/jama.1980.03300430028018. [DOI] [PubMed] [Google Scholar]
- 40.Kortman G.A.M., Reijnders D., Swinkels D.W. Oral iron supplementation: potential implications for the gut microbiome and metabolome in patients with CKD. Hemodial Int. 2017;21(suppl 1):S28–S36. doi: 10.1111/hdi.12553. [DOI] [PubMed] [Google Scholar]
- 41.Rusu I.G., Suharoschi R., Vodnar D.C., et al. Iron supplementation influence on the gut microbiota and probiotic intake effect in iron deficiency-A literature-based review. Nutrients. 2020;12(7):1993. doi: 10.3390/nu12071993. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 42.Sumida K., Pierre J.F., Yuzefpolskaya M., Colombo P.C., Demmer R.T., Kovesdy C.P. Gut microbiota-targeted interventions in the management of chronic kidney disease. Semin Nephrol. 2023;43(2) doi: 10.1016/j.semnephrol.2023.151408. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 43.Bazeley J.W., Wish J.B. Recent and emerging therapies for iron deficiency in anemia of CKD: a review. Am J Kidney Dis. 2022;79(6):868–876. doi: 10.1053/j.ajkd.2021.09.017. [DOI] [PubMed] [Google Scholar]
- 44.Pellegrino A.M., Piemontese M., Grifa R., Aucella F. [Which is the role of the oral iron therapies for iron deficiency anemia in non-dialysis-dependent chronic kidney disease patients? Results from clinical experience] G Ital Nefrol. 2021;38(2) [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Figure S1; Tables S1-S4.