ABSTRACT
Background
Limited real-world data are available in Europe, especially France, regarding the therapeutic management of anaemia in patients with dialysis-dependent chronic kidney disease (DD CKD).
Methods
This retrospective, longitudinal, observational study was based on medical records from the MEDIAL database of not-for-profit dialysis units in France. From January to December 2016, we included eligible patients (≥18 years), with a diagnosis of CKD and receiving maintenance dialysis. Patients with anaemia were followed up for 2 years after inclusion. Patient demographic data, anaemia status, CKD-related anaemia treatments, and treatment outcomes including laboratory test results were evaluated.
Results
Of 1632 DD CKD patients identified from the MEDIAL database, 1286 had anaemia; 98.2% of patients with anaemia were receiving haemodialysis at index date (ID). Of patients with anaemia, 29.9% had haemoglobin (Hb) levels of 10–11 g/dL and 36.2% had levels of 11–12 g/dL at ID. Furthermore, 21.3% had functional iron deficiency and 11.7% had absolute iron deficiency. The most commonly prescribed treatments at ID for patients with DD CKD–related anaemia were intravenous (IV) iron with erythropoietin-stimulating agents (ESAs) (65.1%). Among patients initiating ESA treatment at ID or during follow-up, 347 (95.3%) reached the Hb target of 10–13 g/dL and maintained response within the target Hb range for a median duration of 113 days.
Conclusions
Despite combined use of ESAs and IV iron, duration within the Hb target range was short, suggesting that anaemia management can be further improved.
Keywords: anaemia, CKD, ESA, France, haemodialysis
Graphical Abstract
Graphical Abstract.
Introduction
Chronic kidney disease (CKD) has an estimated prevalence of over half a billion patients worldwide [1]. In France, the prevalence of CKD in 2017 was estimated at approximately 6 million patients [1]. Anaemia is a frequent complication of CKD that contributes to morbidity and mortality in the CKD patient population [2], and the current standard of care for CKD-related anaemia is erythropoietin-stimulating agents (ESAs) combined with iron therapy [3]. Based on data from a Swedish study, CKD-related anaemia is present in up to 90% of dialysis-dependent (DD) patients, and approximately 80% of DD patients receive ESAs [4].
Iron storage deficiency leads not only to anaemia, but also to a reduced response to ESA treatment [5]. Several risk factors contribute to absolute and functional iron deficiency in CKD, including blood losses, impaired iron absorption and chronic inflammation [6]. Consequently, Kidney Disease: Improving Global Outcomes (KDIGO) and European Renal Best Practice (ERBP) guidelines recommend supplemental iron therapy to increase haemoglobin (Hb) levels prior to and during ESA treatment [3, 7]. Furthermore, maintenance of Hb levels in the range 10–11.5 g/dL (KDIGO [3]) or 10–12 g/dL (ERBP [7]) is recommended.
There are minimal real-world data available regarding the management of patients with DD CKD–related anaemia in Europe, and specifically in France. We investigated the real-world clinical management of patients with DD CKD–related anaemia in France.
MATERIALS AND METHODS
Study design
This was a retrospective, longitudinal, observational study of medical records from the MEDIAL database in France. The database included 71 not-for-profit dialysis units from the Pays de la Loire, Morbihan and Rhônes Alpes regions. Database records included patient demographic data, anaemia status, CKD-related anaemia treatments, concomitant treatments, and treatment outcomes including laboratory test results. The study inclusion period was from 1 January 2016 to 31 December 2016.
Patients
Eligible patients were aged ≥18 years, diagnosed with CKD and receiving maintenance dialysis (haemodialysis or peritoneal dialysis) during 2016, with or without a formal diagnosis of anaemia. Patients with at least one of the following criteria for anaemia were included in the DD CKD–related anaemia cohort: Hb <13 g/dL for males or <12 g/dL for females; a record of any anaemia treatment (ESAs, iron therapy or both); or prior anaemia diagnosis (International Classification of Diseases-10 code D63.8). Patients were required to have 1 year of medical history prior to index date (ID; date of first anaemic marker recorded during the study inclusion period), and up to 2 years of follow-up data after ID (follow-up period). Eligible patients with DD CKD–related anaemia must also have had data from at least one post-ID visit.
Patients were further categorized based on Hb level at ID (<10, ≥10–<11, ≥11–<12, ≥12 g/dL) for several pre-defined subgroups of interest: Hb level; inflammation status [C-reactive protein (CRP) ≤5, >5 mg/L]; diabetes status (yes, no); hypertension status (yes, no); ESA dosage range (<8000, ≥8000–≤16 000, and >16 000 units/week); ESA resistance status (ERS; yes, no) and ESA resistance index (ERI) for ESA-treated patients (<5, ≥5–<15, ≥15–≤30 and >30 IU/kg/week/g/dL) [8]; and dialysis vintage (length of time receiving dialysis before ID; <1, ≥1–≤5 and >5 years). ERS was defined as Hb level <11 g/dL over 4–6 months despite weekly doses of epoetin alfa (EPO) of >500 IU/kg or >30 000 IU/week (or ≥1.5 µg/kg per month for darbepoetin [9]) for each ESA treatment initiation. Dosage/unit conversion ratios used for ESA dosages ranges were 1 µg = 200 units for darbepoetin and 1 µg/month = 70–80 units/week for methoxy polyethylene glycol (PEG)-epoetin beta. ERI was calculated as weekly ESA dose per kilogram of body weight divided by Hb level over a 6-month period. Patient iron storage status was defined as adequate [transferrin saturation (TSAT) ≥20% and ferritin ≥100 ng/mL], functional iron deficiency (TSAT <20% and ferritin ≥100 ng/mL) or absolute iron deficiency (TSAT <20%, ferritin <100 ng/mL [5]).
Exclusion criteria included participation in a clinical trial at ID, objection to medical data collection and analysis, current or prior organ transplantation (unless explanted), active cancer or undergoing chemotherapy, or a history of cancer within the 5 years preceding 2016.
Endpoints
Key endpoints included an evaluation of the proportion of patients with DD CKD and DD CKD–related anaemia in 2016; a description of the DD CKD–related anaemia population, including demographic and clinical characteristics, proportions of patients in each subgroup of interest, and dialysis regimen at ID; clinical history and laboratory characteristics of patients with DD CKD–related anaemia at ID; anaemia management at ID and over the follow-up period, including annual frequency of red blood cell (RBC) transfusions and changes in anaemia treatment; Hb level evolution over the follow-up period for each treatment combination, and the duration of response to ESA treatment; and deaths in patients with DD CKD–related anaemia.
Therapeutic management and changes in Hb levels of patients with DD CKD–related anaemia (by Hb level at ID) were recorded at ID and every 2 months up to 2 years of follow-up from the ID.
Statistical analyses
Data were analysed descriptively for all variables in the overall DD CKD and DD CKD–related anaemia cohorts, and in the subgroups of interest with a minimum of 30 patients (sufficient for statistical estimation). Missing data were not imputed or replaced. Assuming 5% missing data, a minimum precision of 2.5% and 2.8% was calculated when describing DD CKD and DD CKD–related anaemia management, respectively, at ID, and modifications over the follow-up period (e.g. RBC transfusions, treatment changes).
Numbers of treatment changes during follow-up were analysed for all patients, whilst the sequence of treatment regimens during follow-up was analysed for all regimens with a duration of ≥45 days. Although KDIGO 2012 and ERBP guidelines recommend target Hb ranges for ESA therapy of 10–11.5 g/dL and 10–12 g/dL, respectively, to maximize improvements in patient quality of life [3, 7], our target range was 10–13 g/dL to capture patients for whom Hb briefly exceeds 12 g/dL without being considered as true ‘off target’, and to more accurately reflect clinical practice. However, this was applied only when analysing Hb levels following treatment initiation.
Duration of response to ESAs, for patients who initiated ESA treatment and had one or more value within the target Hb range, was defined as the time in days from the first date after ESA initiation when Hb was in the target range (start date) to the first post-start date when Hb was outside the target range (event of interest). Response duration was estimated using the Kaplan–Meier method accounting for censored outcomes (death, loss to follow-up, treatment response to end of follow-up, and treatment change). Loss to follow-up was defined as no data collected for ≥6 months.
Additional ad hoc analyses were conducted to assess whether inflammation and ESA resistance components (Hb and dosage) are correlated. A chi-squared test examined the association between Hb level (<10 or ≥10 g/dL) and CRP class at ID (≤5 or >5 mg/L), and a Pearson correlation test explored the correlation between darbepoetin dosage at ID (µg/kg/month) and CRP in the CKD anaemia population. The relationship between CRP class and darbepoetin dosage at ID in the CKD anaemia population was tested by Student's t-test with a threshold of error of 5% for equality of variances and equality of means, and equality of means calculated using the Satterthwaite method.
RESULTS
Patient demographics and baseline characteristics
From the MEDIAL database, 1729 medical records for DD patients were abstracted; 1632 met the criteria for CKD (DD CKD cohort) and 1286 (79% of the DD CKD cohort) met the definition for anaemia (DD CKD–related anaemia cohort) (Fig. 1). Patients with DD CKD–related anaemia had a mean [standard deviation (SD)] age of 70.4 (14.6) years and 57.2% (736/1286) were male (Table 1).
Figure 1:
Study population flow diagram of patient records extracted from the MEDIAL database.
Table 1:
Patient demographics and baseline characteristics in the DD CKD–related anaemia cohort.
| Hb levels | |||||
|---|---|---|---|---|---|
| <10 g/dL (N = 187) | ≥10–<11 g/dL (N = 359) | ≥11–<12 g/dL (N = 434) | ≥12 g/dL (N = 220) | Totala (N = 1286) | |
| Age at ID | |||||
| Mean (SD) | 70.42 (13.93) | 71.38 (15.03) | 70.23 (14.68) | 70.59 (14.04) | 70.43 (14.56) |
| Min–Max | 26.00–94.00 | 20.00–96.00 | 25.00–95.00 | 24.00–95.00 | 20.00–96.00 |
| Age at ID categories, n (%) | |||||
| 18–45 years | 12 (6.4) | 24 (6.7) | 32 (7.4) | 12 (5.5) | 87 (6.8) |
| 45–65 years | 49 (26.2) | 73 (20.3) | 96 (22.1) | 49 (22.3) | 296 (23.0) |
| 65–75 years | 37 (19.8) | 70 (19.5) | 112 (25.8) | 52 (23.6) | 295 (22.9) |
| ≥75 years | 89 (47.6) | 192 (53.5) | 194 (44.7) | 107 (48.6) | 608 (47.3) |
| Sex, n (%) | |||||
| Male | 94 (50.3) | 194 (54.0) | 257 (59.2) | 147 (66.8) | 736 (57.2) |
| BMI (kg/m²) | |||||
| N | 176 | 343 | 418 | 216 | 1167 |
| Mean (SD) | 26.05 (5.92) | 27.01 (5.81) | 27.24 (6.55) | 27.45 (5.77) | 26.98 (6.10) |
| Min–Max | 15.54–50.59 | 14.15–51.93 | 13.79–71.21 | 16.62–48.48 | 13.79–71.21 |
| Hb (g/dL) | |||||
| N | 187 | 359 | 434 | 220 | 1200 |
| Mean (SD) | 9.22 (0.67) | 10.49 (0.29) | 11.46 (0.32) | 12.76 (0.71) | 11.06 (1.20)b |
| Min–Max | 6.10–9.90 | 10.00–10.90 | 11.00–12.00 | 12.10–15.90 | 6.10–15.90 |
| Hypertension, n (%) | |||||
| N | 187 | 359 | 434 | 220 | 1286 |
| Yes | 59 (31.6) | 114 (31.8) | 122 (28.1) | 60 (27.3) | 375 (29.2) |
| CRP categories (mg/L), n (%)c | |||||
| N | 169 | 328 | 398 | 203 | 1099 |
| ≤5 mg/L | 73 (43.2) | 177 (54.0) | 229 (57.5) | 109 (53.7) | 588 (53.5)d |
| >5 mg/L | 96 (56.8) | 151 (46.0) | 169 (42.5) | 94 (46.3) | 511 (46.5)d |
| Diabetes, n (%) | |||||
| N | 187 | 359 | 434 | 220 | 1286 |
| Yes | 56 (29.9) | 102 (28.4) | 131 (30.2) | 71 (32.3) | 383 (29.8) |
| CCI | |||||
| N | 187 | 359 | 434 | 220 | 1286 |
| Mean (SD) | 10.41 (2.26) | 10.43 (2.46) | 10.25 (2.42) | 10.34 (2.27) | 10.31 (2.39) |
| CCI categories, n (%) | |||||
| N | 187 | 359 | 434 | 220 | 1286 |
| <8 | 12 (6.4) | 26 (7.2) | 34 (7.8) | 13 (5.9) | 92 (7.2) |
| 8–10 | 63 (33.7) | 115 (32.0) | 156 (35.9) | 79 (35.9) | 459 (35.7) |
| 11–12 | 97 (51.9) | 173 (48.2) | 204 (47.0) | 106 (48.2) | 606 (47.1) |
| >12 | 15 (8.0) | 45 (12.5) | 40 (9.2) | 22 (10.0) | 129 (10.0) |
Total number of patients from the DD CKD–related anaemia cohort, including 86 patients with missing data for Hb levels.
Mean (SD) calculated based on total number of patients with available data (excluding missing patients).
Defined by CRP in class (mg/L), n (%).
Percentages calculated based on total number of patients with available data (excluding missing patients).
BMI, body mass index.
Patient subgroups of interest
Most ESA-treated patients in the DD CKD–related anaemia cohort (69.5%, 743/1069) received dosages of <8000 units/week; only 7.3% (78/1069) received >16 000 units/week. Almost all patients (99.6%, 898/902) were considered non-resistant to ESAs. Only 11.4% (97/851) of patients had ERI of >15 IU/kg/week/g/dL, and 2.1% (18/851) had ERI >30 IU/kg/week/g/dL. Most patients had midrange (5–15 IU/kg/week/g/dL; 39.8%, 339/851) or lower (<5 IU/kg/week/g/dL; 48.8%, 415/851) ERIs.
Dialysis regimen and vintage
Most patients with DD CKD (99.3%, 1621/1632) were on maintenance haemodialysis and 0.8% (13/1632) received peritoneal dialysis. The mean (SD) duration of all dialysis sessions was 3.82 (0.45) h and session frequency was 2.52 (1.11) per week.
Almost half of patients with DD CKD (47.0%, 730/1552) had a dialysis vintage of ≥1–≤5 years, while 37.2% (577/1552) and 15.8% (245/1552) had received dialysis for <1 year and >5 years, respectively. Among patients with DD CKD–related anaemia, 58.0% (730/1259) had a dialysis vintage ≥1–≤5 years, while 22.6% (284/1259) and 19.5% (245/1259) had received dialysis for <1 year and >5 years, respectively.
Clinical history and laboratory characteristics at ID
Clinical history and laboratory characteristics of patients with DD CKD–related anaemia were categorized according to Hb levels at ID (Table 1). Of those with available Hb data at ID, patients predominantly had Hb levels of 10–11 g/dL (29.9%, 359/1200) and 11–12 g/dL (36.2%, 434/1200), with fewer patients having Hb levels of <10 g/dL (15.6%, 187/1200) and >12 g/dL (18.3%, 220/1200). A similar distribution by age group was observed for each Hb level category, but the proportion of males increased with increasing Hb level (Table 1). The majority of patients with DD CKD–related anaemia had a Charlson Comorbidity Index (CCI) of [11–12] (47.1%, 606/1286). Minimal differences were observed for CCI per Hb category level (Table 1).
Of patients with DD CKD–related anaemia, 35.5% (457/1286) received statins, 11.0% (141/1286) received angiotensin-converting enzyme inhibitors, 12.8% (165/1286) received angiotensin II receptor antagonists and 4.6% (59/1286) received immunosuppressive treatments. The proportion of patients receiving concomitant treatments was similar across Hb categories.
Overall, patient laboratory data at ID were heterogeneous between Hb level categories, with few clear trends observed (Table 2). However, mean CRP and ferritin levels were numerically higher in patients in the <10 g/dL Hb category versus those with higher Hb concentrations. Among patients with available data, 46.5% (511/1099) of patients had a CRP level outside the normal range (>5 mg/L), 33.1% (259/783) had a TSAT <20% and 14.6% (114/781) had a ferritin level <100 ng/mL. Furthermore, 21.3% (164/769) of patients had functional iron deficiency (combined TSAT <20% and ferritin ≥100 ng/mL), and 11.7% (90/769) had absolute iron deficiency (combined TSAT <20% and ferritin <100 ng/mL).
Table 2:
Laboratory characteristics at ID by Hb level.
| Hb levels | |||||
|---|---|---|---|---|---|
| <10 g/dL (N = 187) | ≥10–<11 g/dL (N = 359) | ≥11–<12 g/dL (N = 434) | ≥12 g/dL (N = 220) | Totala (N = 1286) | |
| Albumin (g/L) | |||||
| N | 129 | 265 | 314 | 164 | 872 |
| Mean (SD) | 35.9 (5.6) | 37.4 (4.5) | 38.5 (3.9) | 38.6 (3.7) | 37.8 (4.4) |
| CRP (mg/L) | |||||
| N | 169 | 328 | 398 | 203 | 1099 |
| Mean (SD) | 16.8 (29.0) | 10.2 (15.4) | 8.6 (15.1) | 9.9 (20.0) | 10.6 (19.1) |
| TSAT (%) | |||||
| N | 119 | 226 | 290 | 147 | 783 |
| Mean (SD) | 24.6 (13.3) | 24.3 (9.3) | 25.6 (9.8) | 24.5 (9.9) | 24.9 (10.3) |
| Ferritin (ng/mL) | |||||
| N | 115 | 226 | 289 | 150 | 781 |
| Mean (SD) | 503.4 (467.1) | 395.1 (335.3) | 366.9 (265.4) | 306.1 (271.1) | 384.6 (329.7) |
| TSAT categories (%), n (%) |
|||||
| N | 119 | 226 | 290 | 147 | 783 |
| <20% | 43 (36.1) | 81 (35.8) | 85 (29.3) | 50 (34.0) | 259 (33.1) |
| ≥20% | 76 (63.9) | 145 (64.2) | 205 (70.7) | 97 (66.0) | 524 (66.9) |
| Ferritin categories (ng/mL), n (%) | |||||
| N | 115 | 226 | 289 | 150 | 781 |
| <100 ng/mL | 16 (13.9) | 27 (11.9) | 36 (12.5) | 35 (23.3) | 114 (14.6) |
| ≥100 ng/mL | 99 (86.1) | 199 (88.1) | 253 (87.5) | 115 (76.7) | 667 (85.4) |
Total number of patients from the DD CKD–related anaemia cohort, including 86 patients with missing data for Hb levels.
Anaemia management at ID
At ID, 83.1% (1069/1286) of patients with DD CKD–related anaemia received at least one dose of ESA (with or without iron) and 77.4% (995/1286) received at least one dose of intravenous (IV) iron (with or without an ESA); 4.6% (59/1286) of patients did not receive anaemia treatment.
The most frequently reported combination therapy at ID was ESA plus IV iron (65.1%, 837/1286). Of patients receiving at least one dose of ESA, the most commonly used ESAs at ID were darbepoetin and PEG epoetin beta [57.9% (619/1069) and 27.5% (294/1069) of patients, respectively]. The mean (SD) dose of darbepoetin was 2.22 (1.97) µg/kg/month for patients on ESA plus IV iron combination therapy, and 1.84 (1.44) µg/kg/month for those on ESA monotherapy. Mean administration dose of darbepoetin was 0.63 µg/kg for patients on ESA monotherapy and ESA plus IV iron combination therapy (Supplementary data, Tables S1 and S2).
Treatment regimens at ID by Hb level
For patients with Hb levels of <10, ≥10–<11 and ≥11–<12 g/dL, the most common anaemia treatment received at ID was ESA plus IV iron, followed by ESA only. The most common treatment for patients with Hb level ≥12 g/dL was ESA plus IV iron, followed by IV iron monotherapy (Supplementary data, Table S3).
Anaemia management over the follow-up period
Use of RBC transfusions
During follow-up, 17.3% (222/1286) of patients with DD CKD–related anaemia received at least one RBC transfusion, ranging from 8.9% (14/157) treated only with IV iron to 20.3% (47/231) treated only with ESAs. Of the patients receiving RBC transfusions during follow-up, 59.9% (133/222) received exactly one RBC transfusion, with similar proportions (57.1%–66.0%) between patients receiving each therapy type. The proportion of patients receiving either one or multiple transfusions was 18.2% (152/837) for patients receiving IV iron plus ESA, 20.3% (47/231) for ESA only, 8.9% (14/157) for IV iron only and 15.3% (9/59) for no anaemia treatment at ID. However, the total annual number of RBC transfusions was heterogeneous between patients receiving each treatment scheme, with a high mean (SD) observed for IV iron-treated patients [2.5 (6.7)], and low mean (SD) observed for ESA-treated patients [0.9 (1.0)].
Treatment changes over the follow-up period
Of patients with DD CKD–related anaemia, 55.9% (719/1286) had one or more treatment change during follow-up. Most of these patients (85.8%, 617/719) only had a single treatment change, while 13.9% (100/719) had two changes. Initial treatment change after ID primarily consisted of either discontinuation (50.3%, 362/719) or treatment add-on (49.4%, 355/719).
Almost all patients who received ESAs only (96.1%, 222/231) or no anaemia treatment (98.3%, 58/59) at ID had one or more treatment change during follow-up. In contrast, 79.6% (125/157) of IV iron and 37.4% (313/837) of ESA plus IV iron patients at ID had one or more treatment change during follow-up. Of all patients with treatment changes, those receiving ESAs plus IV iron at ID exclusively had discontinuations (100%, 313/313), whereas a majority of those receiving ESAs only (90.5%, 201/222) or IV iron only (76%, 95/125) at ID had treatment add-ons.
Treatment regimen sequences
When treatment regimens lasting ≥45 days were considered, at ID, 37.8% (486/1286) of patients received darbepoetin plus IV iron treatment, 16.3% (209/1286) of patients received PEG epoetin beta plus IV iron, and 13.5% (173/1286) of patients received IV iron monotherapy (Fig. 2). Using the same criterion over the follow-up period, the most common treatment sequences were darbepoetin plus IV iron from ID to end of follow-up (36.9%, 475/1286); PEG epoetin beta plus IV iron at ID, switching to darbepoetin plus IV iron for the first treatment change and remaining on darbepoetin plus IV iron to end of follow-up (8.9%, 115/1286); and darbepoetin at ID, switching to darbepoetin plus IV iron for the first treatment change and remaining on darbepoetin plus IV iron until the end of follow-up (7.7%, 99/1286) (Fig. 2). Of the 62 patients with no treatment at ID, the majority went on to receive other treatments during the follow-up period—at the first treatment change, these treatments were predominantly IV iron only (2.2%, 28/1286) and darbepoetin only (1.2%, 15/1286) (Fig. 2).
Figure 2:
Significant† treatment sequences during the follow-up period. †Treatments administered for <45 days were not considered as a significant new regimen and were excluded from this diagram. The innermost sequence of the sunburst corresponds to the treatment administered during the ID period (−1/+1 month from ID). Only treatments or combinations received by more than one patient are displayed and only the first three sequences are presented (1% of patients had more than three sequences). Percentages are calculated based on the proportions of the overall patient population within each ring.
Changes in Hb level over the follow-up period
Over the follow-up period, Hb values were maintained at 10–12 g/dL, with a similar mean number of Hb level evaluations over the follow-up period, regardless of anaemia severity at ID (Table 3). For patients receiving either treatment add-ons or discontinuations, the coefficient of variation (CV) in mean Hb level was low (<10.3%) at each timepoint in the 3–6 months preceding the change (Fig. 3).
Table 3:
Hb evaluation frequency over the follow-up period.
| Hb levels | |||||
|---|---|---|---|---|---|
| <10 g/dL (N = 187) | ≥10–<11 g/dL (N = 359) | ≥11–<12 g/dL (N = 434) | ≥12 g/dL (N = 220) | Totala (N = 1286) | |
| Mean Hb level over the follow-up period, n | |||||
| N | 187 | 359 | 434 | 220 | 1249 |
| Missing | 0 | 0 | 0 | 0 | 37 |
| Mean (SD) | 10.57 (0.79) | 10.80 (0.54) | 11.04 (0.61) | 11.52 (0.95) | 10.99 (0.78) |
| Annual number of evaluationsb, n | |||||
| N | 187 | 359 | 434 | 220 | 1249 |
| Missing | 0 | 0 | 0 | 0 | 37 |
| Mean (SD) | 19.83 (9.59) | 19.35 (8.99) | 17.74 (7.58) | 17.61 (8.56) | 18.22 (8.75) |
Total number of patients from the DD CKD–related anaemia cohort, including 86 patients with missing data for Hb levels.
Total number of evaluations during the whole follow-up × 365/duration of follow-up in days.
Figure 3:
Hb levels prior to treatment change in the DD CKD–related anaemia patient population.
Duration of ESA treatment response
Of patients with DD CKD–related anaemia who initiated ESA treatment either at ID or during the follow-up period, 95.3% (347/364) had at least one Hb measure within the target range (10–13 g/dL) during follow-up (Supplementary data, Table S4). Overall, the median (95% confidence interval) response duration was 113.0 (98.0–125.0) days; a numerically shorter response duration [98.0 (98.0−125.0) days] was observed for patients with dialysis vintage of <1 year, compared with patients with dialysis vintage of ≥1–≤5 years [120.0 (92.0−169.0) days] and >5 years [106.0 (85.0−211.0) days] (Fig. 4). Additionally, the proportion of patients with Hb determinations outside the target range decreased with increasing dialysis vintage, with a corresponding increase in proportions of censored patients (Supplementary data, Table S4). Approximately one-third of patients were censored during follow-up (34.3%, 119/347), and the main reason for censoring was a change in treatment (71.4%, 85/119, of all censored events) (Supplementary data, Table S4).
Figure 4:
ESA treatment response duration over the follow-up period (a) overall and (b) by dialysis vintage.
Deaths
During follow-up, the mortality rate of patients with DD CKD–related anaemia was 12.1% (156/1286); 4% (51/1286) were lost to follow-up (primarily due to transfer to other dialysis centres). The primary single cause of patient death was cardiovascular events (23.9%, 33/138; Supplementary data, Fig. S1). Among patients receiving ESAs at ID and with more than one Hb value in the target range, there were two deaths (1.7%, 2/119); one due to aortic valve disease and the other to myocardial infarction.
Ad hoc analysis of CRP, Hb levels and ESA dosage
In total, 56.8% of patients with Hb level <10 g/dL had a CRP >5 mg/L compared with 44.6% of patients with Hb level ≥10 g/dL (Table 1). The mean (SD) CRP in all patients was 10.7 (19.1) mg/L compared with 16.8 (29.0) mg/L in patients with Hb level <10 g/dL (Table 2). This difference is statistically significant (P = .0033), indicating a relationship between Hb level and CRP class at ID in the CKD anaemia population.
CRP and darbepoetin dosage had a statistically significant linear relationship (P = .0373), with a correlation coefficient of r = 0.09. The average darbepoetin dosage at ID was 1.95 µg/kg/month for patients with CRP ≤5 mg/L and 2.20 µg/kg/month for patients with CRP >5 mg/L. For CRP class and darbepoetin dosage, there was a statistically significant difference in variances (P < .0001) but no significant difference in means (P = .1104).
DISCUSSION
This study investigated the real-world therapeutic management of CKD-related anaemia, in 1286 DD patients in France from 71 non-profit units, over a 2-year follow-up period. This is the first study of its kind for patients with DD CKD–related anaemia in France, and adds to a limited pool of data for this patient cohort in Europe. In particular, analysis of the median response duration for ESA therapy had not previously been investigated in this patient population. Our study also included analysis by Hb level to provide detailed insights into therapeutic anaemia management for patient subgroups.
We found that patients with DD CKD–related anaemia most frequently received a combination of IV iron and ESA therapy, with few receiving oral iron therapy, in line with Haute Autorité de Santé 2017 guidelines [10]. IV iron therapy has been shown to improve patient response to ESA treatment and reduce ESA dosage requirements in patients with DD CKD–related anaemia [11]. The KDIGO 2012 guidelines for patients with DD or non-dialysis-dependent CKD–related anaemia also recommend the use of supplemental IV iron in combination with ESAs to increase patient Hb levels or reduce ESA dosage [3].
Although more than two-thirds of patients with DD CKD–related anaemia had adequate iron stores at ID, 21.3% of patients had functional iron deficiency, which is usually indicative of an underlying inflammatory state [12], and 11.7% had absolute iron deficiency.
At ID, darbepoetin was the most common ESA treatment administered in this study population. For all ESA and IV iron treatment modalities at ID, higher mean and maximum doses were observed for ESA + iron combinations compared with ESA-only therapy. The mean dose of darbepoetin was also greater for patients on ESA plus IV iron combination therapy than for those on ESA monotherapy. Mean doses per administration observed in our study for ESA monotherapy and ESAs combined with IV iron were consistent with clinical trial data [13, 14], indicating that 0.45–0.75 µg/kg is the optimal weekly starting dose for darbepoetin treatment.
The median duration of ESA response was 113.0 days, and a numerically shorter response duration was observed for patients with dialysis vintage of <1 year versus ≥1 year. This is a relatively short duration given the chronic nature of CKD-related anaemia and likely contributes to the high proportion of treatment changes observed over the 2-year follow-up period. There are few studies focusing on ESA response duration in patients with DD CKD–related anaemia. However, it has previously been demonstrated that the initial increase in Hb level occurs within the first 4 weeks following initiation of ESA therapy in CKD-related anaemia patients [15], which is consistent with our findings. The rate of Hb increase in response to ESA therapy may vary depending on patient characteristics such as history of cardiovascular disease, or study methodology such as initial dose and dose frequency [16].
DD CKD anaemia patients with lower Hb levels experienced greater inflammation than those with higher levels; more patients with Hb level <10 g/dL had a CRP >5 mg/L compared with patients with Hb level ≥10 g/dL, and the mean CRP in the total patient population was lower compared with patients with Hb level <10 g/dL. Moreover, ad hoc analysis provided strong evidence of an association between Hb level (<10 or ≥10 g/dL) and CRP class at ID (≤5 or >5 mg/L). The findings of the present study also indicated a relationship between CRP and darbepoetin dosage, with higher levels of inflammation experienced by patients receiving higher doses of darbepoetin. The average darbepoetin dosage at ID was greater for patients with CRP >5 mg/L, and CRP and darbepoetin dosage had a statistically significant linear relationship. Moreover, for CRP class and darbepoetin dosage, there was a significant difference in variances. However, the correlation between darbepoetin dosage and CRP was weak with no significant difference in means for CRP class and darbepoetin dosage. As ESA molecules have different dosage systems and units, the correlation between CRP and ESA could only be assessed for patients taking darbepoetin as this represented a homogeneous subgroup that consisted of >50% of included patients.
Limitations of our study include those typical of observational studies. While MEDIAL is one of the largest French databases available for DD patients, only three geographical regions of France were covered, which may affect the generalizability of the findings. ERI data are inherently biased since ESA doses are not typically administered above 500 IU/kg for extended periods owing to deleterious effects on the patient [17]; instead, patients with long-term ESA resistance and substantial comorbidities are reverted to conventional dosages of ≤500 IU/kg. Additionally, not-for-profit health units in France typically do not treat patients with severe comorbidities, who are primarily referred to larger public hospitals, general hospitals and university hospitals with more extensive facilities; this would reduce the number of patients with elevated ERIs (>15 IU/kg/week/g/dL) in the MEDIAL database. A relatively low proportion (less than two-thirds) of patients with iron biomarker results was also reported in this study; this may be related to the lack of data sharing between external biological laboratories in some dialysis centres within the MEDIAL database. In general, patient records of scheduled RBC transfusions were not evenly distributed over the follow-up period. Annual numbers of RBC transfusions are derived from the total number of RBC transfusions received during the follow-up period and the duration of follow-up in days. Since all patients are included and have variable follow-up durations, this annualization calculation may lead to increased heterogeneity in results observed. Lastly, there were relatively few major cardiovascular events recorded from DD CKD anaemia patients over the follow-up period. As such, further statistical analysis on this small subset of data was considered unnecessary.
To conclude, we describe the real-life therapeutic management of patients with DD CKD–related anaemia in France. Results and outcomes from this study are consistent with the current literature and clinical guidelines on the use of ESA plus IV iron combination therapy in the DD CKD–related anaemia patient population. Despite this, duration within the Hb target range was short, indicating that management of anaemia can be further improved.
Supplementary Material
ACKNOWLEDGEMENTS
This study was initiated by Astellas Pharma France. Medical writing support was provided by Glen Dorrington, PhD, of Lumanity.
Contributor Information
Philippe Zaoui, CHU Grenoble-Alpes & AGDUC, Grenoble, Isère, France.
Cecile Courivaud, CHU Besançon, Besançon, Doubs, France.
Guy Rostoker, Hôpital Privé Claude Galien, Ramsay-Générale de Santé, Quincy sous Sénart, Essonne, France.
Gabriel Choukroun, Nephrology Dialysis Transplantation Department, CHU Amiens and MP3CV Research Unit, Université de Picardie Jules Verne, Amiens, France.
Laurence Dubel, Astellas Pharma France, Levallois-Perret, France.
Maria Lorenzo, Astellas Pharma Europe Ltd, Addlestone, Surrey, UK.
Sarah Harmand, IQVIA, La Défense, Hauts-de-Seine, France.
Victorio Menoyo Calonge, Association ECHO, Nantes, Loire-Atlantique, France.
FUNDING
This study was funded by Astellas Pharma France. Medical writing support was funded by Astellas Pharma Inc.
CONFLICT OF INTEREST STATEMENT
G.R. received funding from Astellas and consulting fees from Astellas-France. G.C. received honoraria from Astellas Pharma, Amgen, AstraZeneca, GSK, Takeda and Vifor Pharma, and support for attendance travel costs from Vifor Pharma and Sanofi. M.L. is an employee of Astellas Pharma Europe Ltd. L.D. is an employee of Astellas Pharma France. S.H. is an employee of IQVIA, HEOR vendors contracted to generate data on behalf of Astellas. P.Z. received honoraria and consulting fees from Vifor Pharma and Astellas. C.C. and V.M.C. have no conflicts of interest to disclose.
AUTHORS’ CONTRIBUTIONS
Study design and conduct was carried out by G.C., C.C., G.R., P.Z., L.D. and M.L. Data acquisition was done by V.M.C. and S.H. Analysis of study data was by G.C., C.C., G.R., P.Z., L.D., M.L. and S.H. Interpretation of study data was performed by G.C., C.C., G.R., P.Z., L.D., M.L. and V.M.C.
DATA AVAILABILITY STATEMENT
Researchers may request access to anonymized participant level data, trial level data, and protocols from Astellas-sponsored clinical trials at www.clinicalstudydatarequest.com. For the Astellas criteria on data sharing see: https://clinicalstudydatarequest.com/Study-Sponsors/Study-Sponsors-Astellas.aspx.
References
- 1. Bikbov B, Purcell CA, Levey AS.et al. Global, regional, and national burden of chronic kidney disease, 1990–2017: a systematic analysis for the Global Burden of Disease study 2017. Lancet 2020;395:709–33. 10.1016/S0140-6736(20)30045-3. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2. Stauffer ME, Fan T. Prevalence of anemia in chronic kidney disease in the United States. PLoS One 2014;9:e84943. 10.1371/journal.pone.0084943. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3. Kidney Disease: Improving Global Outcomes (KDIGO) Anemia Work Group . KDIGO clinical practice guideline for anemia in chronic kidney disease. Kidney Int Suppl 2012;2:279–335. https://kdigo.org/wp-content/uploads/2016/10/KDIGO-2012-Anemia-Guideline-English.pdf (27 November 2022, date last accessed). [Google Scholar]
- 4. Evans M, Bower H, Cockburn E.et al. Contemporary management of anaemia, erythropoietin resistance and cardiovascular risk in patients with advanced chronic kidney disease: a nationwide analysis. Clin Kidney J 2020;13:821–7. 10.1093/CKJ/SFAA054. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5. Cases A, Egocheaga MI, Tranche S.et al. Anemia of chronic kidney disease: protocol of study, management and referral to Nephrology. Nefrolgia (Engl Ed) 2018;38:8–12. 10.1016/j.nefroe.2018.01.007. [DOI] [PubMed] [Google Scholar]
- 6. Batchelor EK, Kapitsinou P, Pergola PE.et al. Iron deficiency in chronic kidney disease: updates on pathophysiology, diagnosis, and treatment. J Am Soc Nephrol 2020;31:456–68. 10.1681/ASN.2019020213. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7. Locatelli F, Bárány P, Covic A.et al. Kidney Disease: Improving Global Outcomes guidelines on anaemia management in chronic kidney disease: a European Renal Best Practice position statement. Nephrol Dial Transplant 2013;28:1346–59. 10.1093/ndt/gft033. [DOI] [PubMed] [Google Scholar]
- 8. López-Gómez JM, Portolés JM, Aljama P. Factors that condition the response to erythropoietin in patients on hemodialysis and their relation to mortality. Kidney Int Suppl 2008;111:S75–81. 10.1038/ki.2008.523. [DOI] [PubMed] [Google Scholar]
- 9. Bamgbola OF. Pattern of resistance to erythropoietin-stimulating agents in chronic kidney disease. Kidney Int 2011;80:464–74. 10.1038/ki.2011.179. [DOI] [PubMed] [Google Scholar]
- 10. Haute Authorité de Santé (HAS) . Résultats des indicateurs pour l'amélioration de la qualité et de la sécurité des soins. Prise en charge des patients hémodialysés chroniques. Campagne 2017 – Données 2017. 2017. https://www.has-sante.fr/upload/docs/application/pdf/2017-12/plaquette_dialyse_2017.pdf (27 November 2022, date last accessed). [Google Scholar]
- 11. Macdougall IC. Intravenous iron therapy in patients with chronic kidney disease: recent evidence and future directions. Clin Kidney J 2017;10:i16–24. 10.1093/ckj/sfx043. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12. Rambod M, Kovesdy CP, Kalantar-Zadeh K. Combined high serum ferritin and low iron saturation in hemodialysis patients: the role of inflammation. Clin J Am Soc Nephrol 2008;3:1691–701. 10.2215/CJN.01070308. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13. Macdougall IC. Darbepoetin alfa: a new therapeutic agent for renal anemia. Kidney Int Suppl 2002;80:55–61. 10.1046/j.1523-1755.61.s80.11.x. [DOI] [PubMed] [Google Scholar]
- 14. Macdougall IC, Matcham J, Gray SJ., on behalf of the NESP 960245/246 Study Group. Correction of anaemia with darbepoetin alfa in patients with chronic kidney disease receiving dialysis. Nephrol Dial Transplant 2003;18:576–81. 10.1093/ndt/18.3.576. [DOI] [PubMed] [Google Scholar]
- 15. Pfeffer MA, Burdmann EA, Chen C-Y.et al. A trial of darbepoetin alfa in type 2 diabetes and chronic kidney disease. N Engl J Med 2009;361:2019–32. 10.1056/NEJMoa0907845. [DOI] [PubMed] [Google Scholar]
- 16. Kidney Disease: Improving Global Outcomes (KDIGO) . KDIGO 2012 clinical practice guideline for the evaluation and management of chronic kidney disease. Kidney Int Suppl 2013;3:136–50. https://kdigo.org/wp-content/uploads/2017/02/KDIGO_2012_CKD_GL.pdf(27 November 2022, date last accessed). [DOI] [PubMed] [Google Scholar]
- 17. Koulouridis I, Alfayez M, Trikalinos TA.et al. Dose of erythropoiesis-stimulating agents and adverse outcomes in CKD: a metaregression analysis. Am J Kidney Dis 2013;61:44–56. 10.1053/j.ajkd.2012.07.014. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Availability Statement
Researchers may request access to anonymized participant level data, trial level data, and protocols from Astellas-sponsored clinical trials at www.clinicalstudydatarequest.com. For the Astellas criteria on data sharing see: https://clinicalstudydatarequest.com/Study-Sponsors/Study-Sponsors-Astellas.aspx.