Age- and weight-based differences in haemodialysis prescription and delivery in children, adolescents and young adults

Age- and weight-based differences in haemodialysis prescription and delivery in children,... Abstract Background Limited systematic data are available on prescription and dosing of haemodialysis (HD) in children and adolescents compared with adults. We aimed to characterize age- and weight-based differences in HD delivery in children, adolescents and young adults. Methods This is a retrospective observational study including 1852 patients <30 years on chronic HD from childhood (53 903 HD sessions), receiving thrice weekly outpatient HD between 2004 and 2016 in the USA (6075 patient-years, of which 2535 were in patients aged 1–18 years; weight range 8.3–168 kg). Median individual prescriptions per year were calculated and overall 50% (IQR) and 90% distribution ranges over age and weight were derived. Repeated measurements analysis of variance assessed differences between age and weight groups. Results Prescriptions significantly differed among age and weight groups (P < 0.001). Lower weight patients (<75 kg) had higher (inter-quartile range, IQR) weight-normalized blood flow rate (highest in <25 kg: QB/kg = 6.5–9.1 mL/min/kg), urea dialytic clearance (KD/kg) and single pool Kt/V (spKt/V) (<25 kg: 1.43–1.78; 25–50 kg: 1.52–1.92; 50–75 kg: 1.43–1.74) than heavier patients (lowest in >100 kg: QB/kg = 3.1–4.0 mL/min/kg, spKt/V = 1.22–1.47, respectively). Adolescents had significantly lower QB/kg, KD/kg and spKt/V (1.34–1.71) compared with adults (1.45–1.79) and children <12 years (range of 25th percentiles: 1.37–1.44). Dialytic clearance derived from a mechanistic equation underpredicted KD in children but not in young adults. Significant growth retardation was observed, with the proportion of patients <3rd percentile (height for age) decreasing from 71% (1–2 years) to 15% (>18 years). Conclusion Delivered HD treatment varies with age and weight and is more intensified in children aged <12 years, compared with adolescents and overweight young adults, who appear to be at highest risk of receiving suboptimal treatment. Still, delivery of target or higher spKt/V values did not result in appropriate growth in these children, questioning the value of spKt/V as a measure of HD adequacy in children. Provided ranges of outpatient HD prescription can help clinicians and researchers in personalizing and optimizing delivery of dialysis treatment. chronic haemodialysis, clearance, dialysis dose, epidemiology, obesity, paediatrics INTRODUCTION Haemodialysis (HD) prescription and delivery is relatively standardized in adults [1] compared with children, where frequent adjustments in HD prescription are needed due to growth, with changes in nutrition needs, weight, body composition and volume of total body water (TBW). Potential differences in HD delivery between children, adolescents and adults is not well understood, and very limited systematic data are available regarding fine-tuning of HD prescription across various age and weight groups. In current clinical practice, reference ranges of key prescription parameters such blood flow (QB) are usually derived from body weight or body surface area (BSA) [2, 3], similar to dosing of therapeutics in paediatrics [4]. There are, however, no data on actual prescription and delivery of paediatric HD with ranges of tolerated treatment parameters routinely used in the outpatient setting. While HD prescription (QB: dialysate flow, QD: mass-transfer coefficient for urea KoA and duration of treatment) is expected to differ between children and adults, target values for HD delivery are adopted from adults, and similar dialysis dose would hence be expected to be delivered to paediatric and adult patients. Guidelines recommend measuring delivered dialysis dose in terms of single pool Kt/V (spKt/V; with K: urea dialytic clearance, t: dialysis time, V: urea distribution volume). Minimum spKt/V values of ≥1.2 (adolescents and adults [1]) and higher (children, because of increased nutritional requirements [5]) are recommended, with the target set by ∼0.2 units higher to account for overestimation due to urea rebound [1, 6]. The objective of this descriptive cohort analysis was to better characterize and understand age- and weight-based differences in HD prescription and delivery in children, adolescents and young adults, and to identify potential subpopulations at risk of receiving suboptimal treatment. MATERIALS AND METHODS Study design and participants This retrospective observational study was performed on a cohort of patients who received maintenance HD therapy between May 2004 and March 2016 in DaVita Kidney Care (DaVita Inc., Denver, CO, USA) dialysis centres. All patients that remained ≤18 years, and patients who reached adult age (>18 years) during the follow-up but remained <30 years, were extracted from standardized electronic medical records of the dialysis facilities. The scientific use of the de-identified data was approved by DaVita (DaVita Clinical Research®, Minneapolis, MN, USA), Institutional Review Board approval was not required since retrospective analysis was performed on the de-identified dataset. To be included, patients needed to have treatment records informing about both dialysis treatment (type of dialysis service, duration of treatment, KoA, pre- and post-dialysis weight) and associated demographic and laboratory parameters [age, height, QB, QD, pre- and post-dialysis blood urea nitrogen (BUN) concentration]. Treatment records were excluded if they represented a treatment other than thrice-weekly HD. Data was further screened for non-physiologic and impossible values, which were then eliminated (see section ‘Missing and erroneous data’). Variables and measurement Patient demographics Age (years), gender, weight (kg) and height (converted from inches to cm) were used as recorded. BSA (m2) and TBW (L/kg, according to Cheek et al. [7]) was derived from post-dialysis weight [8]. Dialysis prescription, urea clearance and dialysis dose The following prescription parameters were retrieved as recorded: QB, QD, KoA (mL/min) and treatment duration (min). Urea dialytic clearance (mL/min) was calculated using two methods: (i) theoretically expected clearance (CLD) according to a mechanistic equation accounting for QB, QD and KoA [9] and (ii) actually measured clearance (KD) back-calculated from spKt/V. Weight-normalized values of QD, QB, KoA, KD and CLD (mL/min/kg) were derived from post-dialysis weight and variable-volume spKt/V was calculated [10]. Efficiency of delivered clearance was assessed by indices KD/QB, QD/QB and KoA/QB [9]. The last available residual renal clearance of urea (CLR) from the previous 6 months was extracted. All BUN levels were measured in a central laboratory (DaVita, Inc., Deland, FL, USA) by photometric analysis (Beckman Coulter AU5400; dynamic range: 2.0–130 mg/dL, precision: 1.3–2.5%). Missing and erroneous data Since we used demographic data as explanatory variables for dialysis prescription and derived variables, records with missing data on age, gender, weight and/or height were excluded. Z-scores from the World Health Organization (WHO; absolute score >9 for weight for age or >5 for weight for length) and intradialytic weight changes of >+20%/−30% were used to screen and remove clearly erroneous unphysiologic records. To ensure that Kt/V is available for each treatment record and facilitate interpretation of prescription data, records with missing BUN and treatment duration were excluded. Further, the following variables/combinations indicating highly unlikely thrice-weekly conventional HD were excluded: QB > QD, treatment duration <90 min or >6 h, urea reduction ratio ≤0 or ≥100%. Statistical methods The median value per age was first calculated for each individual, in order to weigh subjects with different monitoring frequencies equally. Demographics were contrasted with WHO percentile curves for children up to 18 years (weight for age and height for age). Distribution ranges over age comprised 50% and 90% actual dialysis prescriptions (absolute and weight-normalized QB, QD, KoA, treatment time), and resulting absolute- and weight-normalized urea clearance and spKt/V were generated applying non-parametric statistics, by calculating at each year of age the 50th percentile (median), 25th and 75th percentiles (interquartile range, IQR), and 5th and 95th percentiles (90% interval). Distribution ranges over weight were generated in a similar way, by calculating percentiles for weight bands of 10 kg, up to a weight of 120 kg. All weights >120 kg were summarized at a weight of 150 kg, all other values were plotted at the upper limit of their weight band. Data were further summarized by median and IQR stratified over five age and weight groups (age groups: toddler: 1–2 years, early childhood: 3–5 years, school-age: 6–11 years, adolescents: 12–18 years, adults: 19–29 years; weight groups: <25 kg, 25–50 kg, 50–75 kg, 75–100 kg, >100 kg). Repeated measurements analysis of variance (ANOVA) was used to assess whether significant differences between groups exist, and to derive group means with 95% confidence intervals (95% CIs) and interpatient variability. A post hoc interaction analysis was performed to assess the combined relationship of age and weight on delivered spKt/V. The fraction of patients reaching spKt/V values of ≥1.2, ≥1.4 and ≥1.6 were calculated with 95% CI. Sensitivity analyses Since we neglected residual kidney function, its potential influence was investigated by exploring the correlation of CLR with age and weight, and by comparing generated spKt/V distribution ranges with total Kt/V (sum of dialysis spKt/V and residual renal Kt/V) ranges over age and weight. Because Cheek TBW estimates were used also for adult patients, the correlation and potential bias between Cheek and Watson TBW estimates [7, 11] was investigated for patients >18 years. Cheek TBW estimates were also compared with Morgenstern estimates [12]. Since HD prescription may also be based on BSA, BSA-normalized QB/m2 and spKtV distribution ranges over BSA were calculated and compared with QB/kg and spKtV ranges over weight. To assess potential bias in distribution ranges over weight introduced by repeated measurements, another dataset was created, summarizing the individual median for each weight band of 10 kg, and both percentiles over weight and percentage of patients reaching target spKt/V values were recalculated. RESULTS Study population The full extracted cohort comprised 66 343 recorded HD sessions with matching laboratory data from 1989 patients. A total of 8009 (12%) treatments did not meet the inclusion criteria of thrice-weekly (1–2 × /week: n = 2238, ≥4 × /week: n = 1944) conventional in-centre HD treatment (home HD: n = 1330, nocturnal dialysis: n = 728, backup dialysis: n = 345, additional day of dialysis: n = 103, ultrafiltration: n = 17, self in-centre HD: n = 7). This resulted in exclusion of 107 patients (5.4%). After cleaning from missing or erroneous data, a total of 53 903 HD sessions (92.4%) from 1852 patients (98.4%) of the included data were retained, comprising 6075 patient-years, of which 2535 patient-years (17 844 sessions) were from patients ≤18 years old (Figure 1; median follow-up duration of dialysis period was 2 years per patient, range: 1–14 years). Detailed numbers of patients per age group are given in Table 3. FIGURE 1: View largeDownload slide Flow chart: selection of study cohort. Horizontal line in histograms: count at least N = 20 patients. This number was achieved for all but the first group used for percentile calculations (1 year of age: N = 14 patients,  <10 kg: N = 11 patients). URR, urea reduction ratio. FIGURE 1: View largeDownload slide Flow chart: selection of study cohort. Horizontal line in histograms: count at least N = 20 patients. This number was achieved for all but the first group used for percentile calculations (1 year of age: N = 14 patients,  <10 kg: N = 11 patients). URR, urea reduction ratio. Median individual weight and height over age is contrasted with WHO growth curves in Figure 2. Most paediatric patients <12 years had low weight and height for age (Table 1). In contrast, large weight variability was observed in adolescents and adults (many patients exceeding 3rd and 97th percentiles, 20% and 11% in 12–18 years old, respectively). Weight range was 8.3–13.6 kg in toddlers (1–2 years), 9.5–22 kg in early childhood (3–5 years), 11.1–66.2 kg in school children (6–11 years), 20–163 kg in adolescents (12–18 years) and 23.3–168 kg in adults. Table 1. Demographics, HD prescription, resulting urea clearance and spKt/V in different age groups Group 1–2 years 3–5 years 6–11 years 12–18 years 19–29 years Number of individuals (number of patient-years) 24 (34) 49 (68) 129 (190) 1438 (2243) 1021 (3540) Number of sessions 175 436 1224 16 009 36 059 Median number of sessions (range) per individual year 4 (1–12) 5 (1–16) 5 (1–22) 6 (1–36) 12 (1–36) Demographics  Post-dialysis weight (kg) 10.3 (9.8–11.0) 14.7 (12.3–16.1) 24.7 (21.0–30.7) 55.4 (46.0–68.4) 60.2 (51.5–71.5)  Weight for age   <50th percentile (%) 97 81 78 65 61a   <3rd percentile (%) 21 37 21 20 15a  Height (cm) 79 (74–83) 93 (86–103) 125 (114–135) 160 (152–168) 163 (155–170)  Height for age   <50th percentile (%) 94 94 85 68 64a   <3rd percentile (%) 71 62 45 35 15a  BSA (m2) 0.48 (0.45–0.51) 0.61 (0.56–0.67) 0.93 (0.81–1.08) 1.58 (1.41–1.79) 1.66 (1.51–1.83) Absolute flows  QB (mL/min) 70 (60–100) 112 (86–150) 180 (150–217.62) 350 (300–400) 400 (350–450)  QD (mL/min) 500 (300–500) 500 (300–500) 500 (500–500) 600 (600–800) 600 (600–800)  KoA (mL/min) 231 (231–231) 309 (231–309) 556 (309–723.25) 1135 (1010–1170) 1140 (1030–1170)  KD (mL/min) 58 (52–64) 78 (70–91) 131 (108–154) 243 (208–272) 275 (247–300)  CLD (mL/min) 34 (23–44) 57 (42–75) 114 (84–137) 244 (208–270) 282 (257–307) Weight-normalized flows  QB/kg (mL/min/kg) 6.8 (5.5–9.9) 7.4 (6.3–9.4) 7.1 (6.0–8.3) 6.1 (4.9–7.2) 6.8 (5.6–7.8)  QD/kg (mL/min/kg) 42.4 (28.6–50.4) 28.7 (23.1–35.3) 19.2 (16.1–23.8) 11.9 (9.5–14.2) 11.0 (9.0–13.2)  KoA/kg (mL/min/kg) 22.7 (21.1–23.8) 19.6 (17.4–24.1) 22.4 (16.2–27.4) 19.1 (15.6–22.9) 18.3 (15.4–21.4)  KD/kg (mL/min/kg) 5.6 (5.0–6.2) 5.7 (5.0–6.3) 5.1 (4.5–5.7) 4.3 (3.6–5.0) 4.5 (3.9–5.2)  CLD/kg (mL/min/kg) 3.1 (2.4–3.8) 3.8 (2.9–4.8) 4.3 (3.4–5.2) 4.2 (3.5–5.1) 4.7 (3.9–5.4) Efficiency of dialysis  KD/QB 0.87 (0.66–0.96) 0.73 (0.62–0.85) 0.71 (0.63–0.81) 0.71 (0.64–0.79) 0.67 (0.62–0.74)  QD/QB 4.7 (3.4–8.0) 3.5 (2.9–4.7) 2.8 (2.3–3.3) 2.0 (1.7–2.3) 1.6 (1.5–1.9)  KoA/QB 3.3 (2.3–3.9) 2.7 (2.1–3.4) 3.1 (2.2–3.7) 3.2 (2.8–3.8) 2.8 (2.4–3.1) Dialyis dose calculation  TBW (L/kg, Cheek) 0.63 (0.61–0.65) 0.63 (0.59–0.66) 0.61 (0.57–0.63) 0.57 (0.53–0.6) 0.56 (0.52–0.6)  Duration (min) 178 (155–182) 180 (180–184) 183 (180–190) 204 (181–225) 210 (182–221)  spKt/V 1.43 (1.37–1.62) 1.56 (1.44–1.8) 1.56 (1.41–1.75) 1.53 (1.34–1.71) 1.61 (1.45–1.79) Group 1–2 years 3–5 years 6–11 years 12–18 years 19–29 years Number of individuals (number of patient-years) 24 (34) 49 (68) 129 (190) 1438 (2243) 1021 (3540) Number of sessions 175 436 1224 16 009 36 059 Median number of sessions (range) per individual year 4 (1–12) 5 (1–16) 5 (1–22) 6 (1–36) 12 (1–36) Demographics  Post-dialysis weight (kg) 10.3 (9.8–11.0) 14.7 (12.3–16.1) 24.7 (21.0–30.7) 55.4 (46.0–68.4) 60.2 (51.5–71.5)  Weight for age   <50th percentile (%) 97 81 78 65 61a   <3rd percentile (%) 21 37 21 20 15a  Height (cm) 79 (74–83) 93 (86–103) 125 (114–135) 160 (152–168) 163 (155–170)  Height for age   <50th percentile (%) 94 94 85 68 64a   <3rd percentile (%) 71 62 45 35 15a  BSA (m2) 0.48 (0.45–0.51) 0.61 (0.56–0.67) 0.93 (0.81–1.08) 1.58 (1.41–1.79) 1.66 (1.51–1.83) Absolute flows  QB (mL/min) 70 (60–100) 112 (86–150) 180 (150–217.62) 350 (300–400) 400 (350–450)  QD (mL/min) 500 (300–500) 500 (300–500) 500 (500–500) 600 (600–800) 600 (600–800)  KoA (mL/min) 231 (231–231) 309 (231–309) 556 (309–723.25) 1135 (1010–1170) 1140 (1030–1170)  KD (mL/min) 58 (52–64) 78 (70–91) 131 (108–154) 243 (208–272) 275 (247–300)  CLD (mL/min) 34 (23–44) 57 (42–75) 114 (84–137) 244 (208–270) 282 (257–307) Weight-normalized flows  QB/kg (mL/min/kg) 6.8 (5.5–9.9) 7.4 (6.3–9.4) 7.1 (6.0–8.3) 6.1 (4.9–7.2) 6.8 (5.6–7.8)  QD/kg (mL/min/kg) 42.4 (28.6–50.4) 28.7 (23.1–35.3) 19.2 (16.1–23.8) 11.9 (9.5–14.2) 11.0 (9.0–13.2)  KoA/kg (mL/min/kg) 22.7 (21.1–23.8) 19.6 (17.4–24.1) 22.4 (16.2–27.4) 19.1 (15.6–22.9) 18.3 (15.4–21.4)  KD/kg (mL/min/kg) 5.6 (5.0–6.2) 5.7 (5.0–6.3) 5.1 (4.5–5.7) 4.3 (3.6–5.0) 4.5 (3.9–5.2)  CLD/kg (mL/min/kg) 3.1 (2.4–3.8) 3.8 (2.9–4.8) 4.3 (3.4–5.2) 4.2 (3.5–5.1) 4.7 (3.9–5.4) Efficiency of dialysis  KD/QB 0.87 (0.66–0.96) 0.73 (0.62–0.85) 0.71 (0.63–0.81) 0.71 (0.64–0.79) 0.67 (0.62–0.74)  QD/QB 4.7 (3.4–8.0) 3.5 (2.9–4.7) 2.8 (2.3–3.3) 2.0 (1.7–2.3) 1.6 (1.5–1.9)  KoA/QB 3.3 (2.3–3.9) 2.7 (2.1–3.4) 3.1 (2.2–3.7) 3.2 (2.8–3.8) 2.8 (2.4–3.1) Dialyis dose calculation  TBW (L/kg, Cheek) 0.63 (0.61–0.65) 0.63 (0.59–0.66) 0.61 (0.57–0.63) 0.57 (0.53–0.6) 0.56 (0.52–0.6)  Duration (min) 178 (155–182) 180 (180–184) 183 (180–190) 204 (181–225) 210 (182–221)  spKt/V 1.43 (1.37–1.62) 1.56 (1.44–1.8) 1.56 (1.41–1.75) 1.53 (1.34–1.71) 1.61 (1.45–1.79) Variables are summarized as median (IQR) unless otherwise mentioned. For all variables listed, repeated measures ANOVA indicated significant differences between the mean of the five age groups, with all P-values  <1.33 × 10−25. Post hoc pairwise group comparisons are mentioned in the text. Respective estimated group means are given with 95% CIs in the Supplementary Data. a Compared with percentile at 18 years. CLD, theoretical urea clearance calculated from QB, QD and KoA [9]; KD, actual delivered clearance according to spKt/V. Table 1. Demographics, HD prescription, resulting urea clearance and spKt/V in different age groups Group 1–2 years 3–5 years 6–11 years 12–18 years 19–29 years Number of individuals (number of patient-years) 24 (34) 49 (68) 129 (190) 1438 (2243) 1021 (3540) Number of sessions 175 436 1224 16 009 36 059 Median number of sessions (range) per individual year 4 (1–12) 5 (1–16) 5 (1–22) 6 (1–36) 12 (1–36) Demographics  Post-dialysis weight (kg) 10.3 (9.8–11.0) 14.7 (12.3–16.1) 24.7 (21.0–30.7) 55.4 (46.0–68.4) 60.2 (51.5–71.5)  Weight for age   <50th percentile (%) 97 81 78 65 61a   <3rd percentile (%) 21 37 21 20 15a  Height (cm) 79 (74–83) 93 (86–103) 125 (114–135) 160 (152–168) 163 (155–170)  Height for age   <50th percentile (%) 94 94 85 68 64a   <3rd percentile (%) 71 62 45 35 15a  BSA (m2) 0.48 (0.45–0.51) 0.61 (0.56–0.67) 0.93 (0.81–1.08) 1.58 (1.41–1.79) 1.66 (1.51–1.83) Absolute flows  QB (mL/min) 70 (60–100) 112 (86–150) 180 (150–217.62) 350 (300–400) 400 (350–450)  QD (mL/min) 500 (300–500) 500 (300–500) 500 (500–500) 600 (600–800) 600 (600–800)  KoA (mL/min) 231 (231–231) 309 (231–309) 556 (309–723.25) 1135 (1010–1170) 1140 (1030–1170)  KD (mL/min) 58 (52–64) 78 (70–91) 131 (108–154) 243 (208–272) 275 (247–300)  CLD (mL/min) 34 (23–44) 57 (42–75) 114 (84–137) 244 (208–270) 282 (257–307) Weight-normalized flows  QB/kg (mL/min/kg) 6.8 (5.5–9.9) 7.4 (6.3–9.4) 7.1 (6.0–8.3) 6.1 (4.9–7.2) 6.8 (5.6–7.8)  QD/kg (mL/min/kg) 42.4 (28.6–50.4) 28.7 (23.1–35.3) 19.2 (16.1–23.8) 11.9 (9.5–14.2) 11.0 (9.0–13.2)  KoA/kg (mL/min/kg) 22.7 (21.1–23.8) 19.6 (17.4–24.1) 22.4 (16.2–27.4) 19.1 (15.6–22.9) 18.3 (15.4–21.4)  KD/kg (mL/min/kg) 5.6 (5.0–6.2) 5.7 (5.0–6.3) 5.1 (4.5–5.7) 4.3 (3.6–5.0) 4.5 (3.9–5.2)  CLD/kg (mL/min/kg) 3.1 (2.4–3.8) 3.8 (2.9–4.8) 4.3 (3.4–5.2) 4.2 (3.5–5.1) 4.7 (3.9–5.4) Efficiency of dialysis  KD/QB 0.87 (0.66–0.96) 0.73 (0.62–0.85) 0.71 (0.63–0.81) 0.71 (0.64–0.79) 0.67 (0.62–0.74)  QD/QB 4.7 (3.4–8.0) 3.5 (2.9–4.7) 2.8 (2.3–3.3) 2.0 (1.7–2.3) 1.6 (1.5–1.9)  KoA/QB 3.3 (2.3–3.9) 2.7 (2.1–3.4) 3.1 (2.2–3.7) 3.2 (2.8–3.8) 2.8 (2.4–3.1) Dialyis dose calculation  TBW (L/kg, Cheek) 0.63 (0.61–0.65) 0.63 (0.59–0.66) 0.61 (0.57–0.63) 0.57 (0.53–0.6) 0.56 (0.52–0.6)  Duration (min) 178 (155–182) 180 (180–184) 183 (180–190) 204 (181–225) 210 (182–221)  spKt/V 1.43 (1.37–1.62) 1.56 (1.44–1.8) 1.56 (1.41–1.75) 1.53 (1.34–1.71) 1.61 (1.45–1.79) Group 1–2 years 3–5 years 6–11 years 12–18 years 19–29 years Number of individuals (number of patient-years) 24 (34) 49 (68) 129 (190) 1438 (2243) 1021 (3540) Number of sessions 175 436 1224 16 009 36 059 Median number of sessions (range) per individual year 4 (1–12) 5 (1–16) 5 (1–22) 6 (1–36) 12 (1–36) Demographics  Post-dialysis weight (kg) 10.3 (9.8–11.0) 14.7 (12.3–16.1) 24.7 (21.0–30.7) 55.4 (46.0–68.4) 60.2 (51.5–71.5)  Weight for age   <50th percentile (%) 97 81 78 65 61a   <3rd percentile (%) 21 37 21 20 15a  Height (cm) 79 (74–83) 93 (86–103) 125 (114–135) 160 (152–168) 163 (155–170)  Height for age   <50th percentile (%) 94 94 85 68 64a   <3rd percentile (%) 71 62 45 35 15a  BSA (m2) 0.48 (0.45–0.51) 0.61 (0.56–0.67) 0.93 (0.81–1.08) 1.58 (1.41–1.79) 1.66 (1.51–1.83) Absolute flows  QB (mL/min) 70 (60–100) 112 (86–150) 180 (150–217.62) 350 (300–400) 400 (350–450)  QD (mL/min) 500 (300–500) 500 (300–500) 500 (500–500) 600 (600–800) 600 (600–800)  KoA (mL/min) 231 (231–231) 309 (231–309) 556 (309–723.25) 1135 (1010–1170) 1140 (1030–1170)  KD (mL/min) 58 (52–64) 78 (70–91) 131 (108–154) 243 (208–272) 275 (247–300)  CLD (mL/min) 34 (23–44) 57 (42–75) 114 (84–137) 244 (208–270) 282 (257–307) Weight-normalized flows  QB/kg (mL/min/kg) 6.8 (5.5–9.9) 7.4 (6.3–9.4) 7.1 (6.0–8.3) 6.1 (4.9–7.2) 6.8 (5.6–7.8)  QD/kg (mL/min/kg) 42.4 (28.6–50.4) 28.7 (23.1–35.3) 19.2 (16.1–23.8) 11.9 (9.5–14.2) 11.0 (9.0–13.2)  KoA/kg (mL/min/kg) 22.7 (21.1–23.8) 19.6 (17.4–24.1) 22.4 (16.2–27.4) 19.1 (15.6–22.9) 18.3 (15.4–21.4)  KD/kg (mL/min/kg) 5.6 (5.0–6.2) 5.7 (5.0–6.3) 5.1 (4.5–5.7) 4.3 (3.6–5.0) 4.5 (3.9–5.2)  CLD/kg (mL/min/kg) 3.1 (2.4–3.8) 3.8 (2.9–4.8) 4.3 (3.4–5.2) 4.2 (3.5–5.1) 4.7 (3.9–5.4) Efficiency of dialysis  KD/QB 0.87 (0.66–0.96) 0.73 (0.62–0.85) 0.71 (0.63–0.81) 0.71 (0.64–0.79) 0.67 (0.62–0.74)  QD/QB 4.7 (3.4–8.0) 3.5 (2.9–4.7) 2.8 (2.3–3.3) 2.0 (1.7–2.3) 1.6 (1.5–1.9)  KoA/QB 3.3 (2.3–3.9) 2.7 (2.1–3.4) 3.1 (2.2–3.7) 3.2 (2.8–3.8) 2.8 (2.4–3.1) Dialyis dose calculation  TBW (L/kg, Cheek) 0.63 (0.61–0.65) 0.63 (0.59–0.66) 0.61 (0.57–0.63) 0.57 (0.53–0.6) 0.56 (0.52–0.6)  Duration (min) 178 (155–182) 180 (180–184) 183 (180–190) 204 (181–225) 210 (182–221)  spKt/V 1.43 (1.37–1.62) 1.56 (1.44–1.8) 1.56 (1.41–1.75) 1.53 (1.34–1.71) 1.61 (1.45–1.79) Variables are summarized as median (IQR) unless otherwise mentioned. For all variables listed, repeated measures ANOVA indicated significant differences between the mean of the five age groups, with all P-values  <1.33 × 10−25. Post hoc pairwise group comparisons are mentioned in the text. Respective estimated group means are given with 95% CIs in the Supplementary Data. a Compared with percentile at 18 years. CLD, theoretical urea clearance calculated from QB, QD and KoA [9]; KD, actual delivered clearance according to spKt/V. FIGURE 2: View largeDownload slide Patient demographics. Illustration of post-dialysis weight (upper panel) and height (lower panel) over age for the cleaned dataset (median per age), and comparison with WHO growth curves (curved lines; solid line: 50th percentile, dashed lines: 3rd and 97th percentile). The individual median weight and height per age was calculated, individual changes over age are indicated by grey lines. FIGURE 2: View largeDownload slide Patient demographics. Illustration of post-dialysis weight (upper panel) and height (lower panel) over age for the cleaned dataset (median per age), and comparison with WHO growth curves (curved lines; solid line: 50th percentile, dashed lines: 3rd and 97th percentile). The individual median weight and height per age was calculated, individual changes over age are indicated by grey lines. Dialysis prescription, urea clearance and dialysis dose Distribution ranges are shown in Figure 3 (weight-normalized prescribed flows and duration of HD) and Figure 4 (resulting weight-normalized urea clearance and dialysis dose) over age and weight. Supplementary data, Figures S1 and S2 show distribution ranges for absolute prescribed and derived flows (QB, QD, KoA, KD, CLD) and for measures of dialysis efficiency (KD/QB, QD/QB, KoA/QB). FIGURE 3: View largeDownload slide Dialysis prescription over age (left) and over weight (right). Solid lines with dots: median. Grey area: IQR. Dashed lines: 5th and 95th percentiles. FIGURE 3: View largeDownload slide Dialysis prescription over age (left) and over weight (right). Solid lines with dots: median. Grey area: IQR. Dashed lines: 5th and 95th percentiles. FIGURE 4: View largeDownload slide Resulting urea clearance and dialysis dose over age (left) and weight (right). Solid lines and dots: median. Dashed lines: 5th and 95th percentiles. Grey shaded area: IQR. CLD: theoretical urea dialytic clearance calculated from QB, QD and KoA [9]. KD: urea dialytic clearance calculated from spKt/V. Horizontal solid line: spKt/V target of 1.4. Horizontal dotted lines: spKt/V of 1.2 (minimum target) and 1.6 (clinically frequently used target). FIGURE 4: View largeDownload slide Resulting urea clearance and dialysis dose over age (left) and weight (right). Solid lines and dots: median. Dashed lines: 5th and 95th percentiles. Grey shaded area: IQR. CLD: theoretical urea dialytic clearance calculated from QB, QD and KoA [9]. KD: urea dialytic clearance calculated from spKt/V. Horizontal solid line: spKt/V target of 1.4. Horizontal dotted lines: spKt/V of 1.2 (minimum target) and 1.6 (clinically frequently used target). Tables 1 and 2 summarize prescription parameters (median, IQR) stratified over five age and weight groups, respectively. Repeated measures ANOVA indicated that the mean of all listed variables is significantly influenced by age (Table 1, all P-values <1.33 × 10−25) and weight (Table 2, all P-values <2.06 × 10−50). Estimated group means are summarized with 95% CI in Supplementary data, Table S1. Weight generally better explained interpatient variability in weight-based prescription than age (mostly smaller interpatient variability estimated). Table 2. Demographics, HD prescription, resulting urea clearance and spKt/V in different weight groups Group <25 kg 25–50 kg 50–75 kg 75–100 kg >100 kg Number of individuals (number of patient-years) 129 (223) 621 (1615) 1025 (3140) 324 (808) 107 (289) [number of sessions] [1419] [13 245] [28 659] [7744] [2836] Demographics  Post-dialysis weight (kg) 18.6 (12.97–21.8) 43.7 (38.3–47.3) 60 (55.0–66.2) 83.1 (79.0–89.7) 112 (104.2–125.4)  Height (cm) 107 (88–117) 152 (142–157) 165 (157–170) 170 (163–175) 170 (165–175)  BSA (m2) 0.74 (0.57–0.84) 1.35 (1.24–1.44) 1.65 (1.57–1.76) 1.98 (1.91–2.07) 2.31 (2.22–2.45) Absolute flows (mL/min)  QB 150 (100–175) 350 (275–400) 400 (350–450) 400 (350–450) 400 (350–450)  QD 500 (319–500) 600 (581–800) 600 (600–800) 600 (600–800) 700 (600–800)  KoA 309 (231–556) 1030 (911–1170) 1140 (1030–1170) 1170 (1030–1183) 1170 (1140–1240)  KD 96 (74–118) 229 (194–254) 267 (242–292) 292 (266–319) 322(288–349)  CLD 75 (44–97) 236 (196–268) 276 (249–300) 287 (258–309) 284 (260–313) Weight-normalized flows (mL/min/kg)  QB/kg 7.4 (6.45–9.07) 7.8 (6.76–8.94) 6.5 (5.69–7.4) 4.8 (4.21–5.33) 3.5 (3.1–3.95)  QD/kg 25.6 (21.4–31.5) 14.5 (12.8–17.1) 11.1 (9.7–12.7) 7.8 (7–9.2) 5.9 (5.4–6.9)  KoA/kg 21.8 (16.8–25.7) 24.2 (21.3–27.2) 18.4 (16.4–20.4) 13.6 (12.4–14.8) 10.5 (9.5–11.4)  KD/kg 5.5 (4.9–6.3) 5.3 (4.7–5.9) 4.4 (3.9–4.9) 3.5 (3.1–3.8) 2.8 (2.4–3.1)  CLD/kg 3.9 (3.1–5.1) 5.5 (4.8–6.2) 4.5 (4.0–5.1) 3.4 (3.0–3.7) 2.5 (2.2–2.8) Efficiency of dialysis  KD/QB 0.72 (0.62–0.85) 0.69 (0.62–0.76) 0.67 (0.62–0.74) 0.73 (0.66–0.81) 0.78 (0.72–0.85)  QD/QB 3.33 (2.62–4.17) 2 (1.67–2.29) 1.71 (1.5–2) 1.67 (1.5–1.94) 1.72 (1.5–2)  KoA/QB 2.82 (2.06–3.71) 3.07 (2.64–3.71) 2.85 (2.5–3.29) 2.89 (2.5–3.3) 3.01 (2.58–3.31) Dialysis dose calculation  TBW (L/kg, Cheek) 0.63 (0.60–0.65) 0.58 (0.56–0.61) 0.57 (0.53–0.60) 0.52 (0.47–0.56) 0.49 (0.44–0.51)  Duration (min) 180 (180–186) 184 (180–210) 210 (183–219) 215 (201–240) 240 (212–245)   spKt/V 1.58 (1.43–1.78) 1.71 (1.52–1.92) 1.58 (1.43–1.74) 1.45 (1.31–1.58) 1.35 (1.22–1.47) Group <25 kg 25–50 kg 50–75 kg 75–100 kg >100 kg Number of individuals (number of patient-years) 129 (223) 621 (1615) 1025 (3140) 324 (808) 107 (289) [number of sessions] [1419] [13 245] [28 659] [7744] [2836] Demographics  Post-dialysis weight (kg) 18.6 (12.97–21.8) 43.7 (38.3–47.3) 60 (55.0–66.2) 83.1 (79.0–89.7) 112 (104.2–125.4)  Height (cm) 107 (88–117) 152 (142–157) 165 (157–170) 170 (163–175) 170 (165–175)  BSA (m2) 0.74 (0.57–0.84) 1.35 (1.24–1.44) 1.65 (1.57–1.76) 1.98 (1.91–2.07) 2.31 (2.22–2.45) Absolute flows (mL/min)  QB 150 (100–175) 350 (275–400) 400 (350–450) 400 (350–450) 400 (350–450)  QD 500 (319–500) 600 (581–800) 600 (600–800) 600 (600–800) 700 (600–800)  KoA 309 (231–556) 1030 (911–1170) 1140 (1030–1170) 1170 (1030–1183) 1170 (1140–1240)  KD 96 (74–118) 229 (194–254) 267 (242–292) 292 (266–319) 322(288–349)  CLD 75 (44–97) 236 (196–268) 276 (249–300) 287 (258–309) 284 (260–313) Weight-normalized flows (mL/min/kg)  QB/kg 7.4 (6.45–9.07) 7.8 (6.76–8.94) 6.5 (5.69–7.4) 4.8 (4.21–5.33) 3.5 (3.1–3.95)  QD/kg 25.6 (21.4–31.5) 14.5 (12.8–17.1) 11.1 (9.7–12.7) 7.8 (7–9.2) 5.9 (5.4–6.9)  KoA/kg 21.8 (16.8–25.7) 24.2 (21.3–27.2) 18.4 (16.4–20.4) 13.6 (12.4–14.8) 10.5 (9.5–11.4)  KD/kg 5.5 (4.9–6.3) 5.3 (4.7–5.9) 4.4 (3.9–4.9) 3.5 (3.1–3.8) 2.8 (2.4–3.1)  CLD/kg 3.9 (3.1–5.1) 5.5 (4.8–6.2) 4.5 (4.0–5.1) 3.4 (3.0–3.7) 2.5 (2.2–2.8) Efficiency of dialysis  KD/QB 0.72 (0.62–0.85) 0.69 (0.62–0.76) 0.67 (0.62–0.74) 0.73 (0.66–0.81) 0.78 (0.72–0.85)  QD/QB 3.33 (2.62–4.17) 2 (1.67–2.29) 1.71 (1.5–2) 1.67 (1.5–1.94) 1.72 (1.5–2)  KoA/QB 2.82 (2.06–3.71) 3.07 (2.64–3.71) 2.85 (2.5–3.29) 2.89 (2.5–3.3) 3.01 (2.58–3.31) Dialysis dose calculation  TBW (L/kg, Cheek) 0.63 (0.60–0.65) 0.58 (0.56–0.61) 0.57 (0.53–0.60) 0.52 (0.47–0.56) 0.49 (0.44–0.51)  Duration (min) 180 (180–186) 184 (180–210) 210 (183–219) 215 (201–240) 240 (212–245)   spKt/V 1.58 (1.43–1.78) 1.71 (1.52–1.92) 1.58 (1.43–1.74) 1.45 (1.31–1.58) 1.35 (1.22–1.47) Variables are summarized as median (IQR) unless otherwise mentioned. For all variables listed, repeated measures ANOVA indicated significant differences between the mean of the five age groups, with all P-values  <2.06 × 10−50. Respective estimated group means are given with 95% CIs in the Supplementary Data. CLD: theoretical urea clearance calculated from QB, QD and KoA [9]. KD, actual delivered clearance according to spKt/V. Table 2. Demographics, HD prescription, resulting urea clearance and spKt/V in different weight groups Group <25 kg 25–50 kg 50–75 kg 75–100 kg >100 kg Number of individuals (number of patient-years) 129 (223) 621 (1615) 1025 (3140) 324 (808) 107 (289) [number of sessions] [1419] [13 245] [28 659] [7744] [2836] Demographics  Post-dialysis weight (kg) 18.6 (12.97–21.8) 43.7 (38.3–47.3) 60 (55.0–66.2) 83.1 (79.0–89.7) 112 (104.2–125.4)  Height (cm) 107 (88–117) 152 (142–157) 165 (157–170) 170 (163–175) 170 (165–175)  BSA (m2) 0.74 (0.57–0.84) 1.35 (1.24–1.44) 1.65 (1.57–1.76) 1.98 (1.91–2.07) 2.31 (2.22–2.45) Absolute flows (mL/min)  QB 150 (100–175) 350 (275–400) 400 (350–450) 400 (350–450) 400 (350–450)  QD 500 (319–500) 600 (581–800) 600 (600–800) 600 (600–800) 700 (600–800)  KoA 309 (231–556) 1030 (911–1170) 1140 (1030–1170) 1170 (1030–1183) 1170 (1140–1240)  KD 96 (74–118) 229 (194–254) 267 (242–292) 292 (266–319) 322(288–349)  CLD 75 (44–97) 236 (196–268) 276 (249–300) 287 (258–309) 284 (260–313) Weight-normalized flows (mL/min/kg)  QB/kg 7.4 (6.45–9.07) 7.8 (6.76–8.94) 6.5 (5.69–7.4) 4.8 (4.21–5.33) 3.5 (3.1–3.95)  QD/kg 25.6 (21.4–31.5) 14.5 (12.8–17.1) 11.1 (9.7–12.7) 7.8 (7–9.2) 5.9 (5.4–6.9)  KoA/kg 21.8 (16.8–25.7) 24.2 (21.3–27.2) 18.4 (16.4–20.4) 13.6 (12.4–14.8) 10.5 (9.5–11.4)  KD/kg 5.5 (4.9–6.3) 5.3 (4.7–5.9) 4.4 (3.9–4.9) 3.5 (3.1–3.8) 2.8 (2.4–3.1)  CLD/kg 3.9 (3.1–5.1) 5.5 (4.8–6.2) 4.5 (4.0–5.1) 3.4 (3.0–3.7) 2.5 (2.2–2.8) Efficiency of dialysis  KD/QB 0.72 (0.62–0.85) 0.69 (0.62–0.76) 0.67 (0.62–0.74) 0.73 (0.66–0.81) 0.78 (0.72–0.85)  QD/QB 3.33 (2.62–4.17) 2 (1.67–2.29) 1.71 (1.5–2) 1.67 (1.5–1.94) 1.72 (1.5–2)  KoA/QB 2.82 (2.06–3.71) 3.07 (2.64–3.71) 2.85 (2.5–3.29) 2.89 (2.5–3.3) 3.01 (2.58–3.31) Dialysis dose calculation  TBW (L/kg, Cheek) 0.63 (0.60–0.65) 0.58 (0.56–0.61) 0.57 (0.53–0.60) 0.52 (0.47–0.56) 0.49 (0.44–0.51)  Duration (min) 180 (180–186) 184 (180–210) 210 (183–219) 215 (201–240) 240 (212–245)   spKt/V 1.58 (1.43–1.78) 1.71 (1.52–1.92) 1.58 (1.43–1.74) 1.45 (1.31–1.58) 1.35 (1.22–1.47) Group <25 kg 25–50 kg 50–75 kg 75–100 kg >100 kg Number of individuals (number of patient-years) 129 (223) 621 (1615) 1025 (3140) 324 (808) 107 (289) [number of sessions] [1419] [13 245] [28 659] [7744] [2836] Demographics  Post-dialysis weight (kg) 18.6 (12.97–21.8) 43.7 (38.3–47.3) 60 (55.0–66.2) 83.1 (79.0–89.7) 112 (104.2–125.4)  Height (cm) 107 (88–117) 152 (142–157) 165 (157–170) 170 (163–175) 170 (165–175)  BSA (m2) 0.74 (0.57–0.84) 1.35 (1.24–1.44) 1.65 (1.57–1.76) 1.98 (1.91–2.07) 2.31 (2.22–2.45) Absolute flows (mL/min)  QB 150 (100–175) 350 (275–400) 400 (350–450) 400 (350–450) 400 (350–450)  QD 500 (319–500) 600 (581–800) 600 (600–800) 600 (600–800) 700 (600–800)  KoA 309 (231–556) 1030 (911–1170) 1140 (1030–1170) 1170 (1030–1183) 1170 (1140–1240)  KD 96 (74–118) 229 (194–254) 267 (242–292) 292 (266–319) 322(288–349)  CLD 75 (44–97) 236 (196–268) 276 (249–300) 287 (258–309) 284 (260–313) Weight-normalized flows (mL/min/kg)  QB/kg 7.4 (6.45–9.07) 7.8 (6.76–8.94) 6.5 (5.69–7.4) 4.8 (4.21–5.33) 3.5 (3.1–3.95)  QD/kg 25.6 (21.4–31.5) 14.5 (12.8–17.1) 11.1 (9.7–12.7) 7.8 (7–9.2) 5.9 (5.4–6.9)  KoA/kg 21.8 (16.8–25.7) 24.2 (21.3–27.2) 18.4 (16.4–20.4) 13.6 (12.4–14.8) 10.5 (9.5–11.4)  KD/kg 5.5 (4.9–6.3) 5.3 (4.7–5.9) 4.4 (3.9–4.9) 3.5 (3.1–3.8) 2.8 (2.4–3.1)  CLD/kg 3.9 (3.1–5.1) 5.5 (4.8–6.2) 4.5 (4.0–5.1) 3.4 (3.0–3.7) 2.5 (2.2–2.8) Efficiency of dialysis  KD/QB 0.72 (0.62–0.85) 0.69 (0.62–0.76) 0.67 (0.62–0.74) 0.73 (0.66–0.81) 0.78 (0.72–0.85)  QD/QB 3.33 (2.62–4.17) 2 (1.67–2.29) 1.71 (1.5–2) 1.67 (1.5–1.94) 1.72 (1.5–2)  KoA/QB 2.82 (2.06–3.71) 3.07 (2.64–3.71) 2.85 (2.5–3.29) 2.89 (2.5–3.3) 3.01 (2.58–3.31) Dialysis dose calculation  TBW (L/kg, Cheek) 0.63 (0.60–0.65) 0.58 (0.56–0.61) 0.57 (0.53–0.60) 0.52 (0.47–0.56) 0.49 (0.44–0.51)  Duration (min) 180 (180–186) 184 (180–210) 210 (183–219) 215 (201–240) 240 (212–245)   spKt/V 1.58 (1.43–1.78) 1.71 (1.52–1.92) 1.58 (1.43–1.74) 1.45 (1.31–1.58) 1.35 (1.22–1.47) Variables are summarized as median (IQR) unless otherwise mentioned. For all variables listed, repeated measures ANOVA indicated significant differences between the mean of the five age groups, with all P-values  <2.06 × 10−50. Respective estimated group means are given with 95% CIs in the Supplementary Data. CLD: theoretical urea clearance calculated from QB, QD and KoA [9]. KD, actual delivered clearance according to spKt/V. Table 3. Achievement of different target spKt/V values per age and weight group Group 1–2 years 3–5 years 6–11 years 12–18 years 19–29 years n = 34 n = 68 n = 190 n = 2243 n = 3540 (N = 24) (N = 49) (N = 129) (N = 1438) (N = 1021) spKt/V achievement per age group spKt/V ≥1.2  n 31 65 173 1966 3401  % (95% CI) 91% (75–98%) 96% (87–99%); 91% (86–95%) 88% (86–89%) 96% (95–97%) spKt/V ≥1.4  n 20 58 145 1509 2887  % (95% CI) 59% (41–75%) 85% (74–92%) 76% (70–82%) 67% (65–69%) 82% (80–83%) spKt/V ≥1.6  n 10 31 85 870 1837  % (95% CI) 29% (16–48%) 46% (34–58%) 45% (38–52%) 39% (37–41%) 52% (50–54%) Group 1–2 years 3–5 years 6–11 years 12–18 years 19–29 years n = 34 n = 68 n = 190 n = 2243 n = 3540 (N = 24) (N = 49) (N = 129) (N = 1438) (N = 1021) spKt/V achievement per age group spKt/V ≥1.2  n 31 65 173 1966 3401  % (95% CI) 91% (75–98%) 96% (87–99%); 91% (86–95%) 88% (86–89%) 96% (95–97%) spKt/V ≥1.4  n 20 58 145 1509 2887  % (95% CI) 59% (41–75%) 85% (74–92%) 76% (70–82%) 67% (65–69%) 82% (80–83%) spKt/V ≥1.6  n 10 31 85 870 1837  % (95% CI) 29% (16–48%) 46% (34–58%) 45% (38–52%) 39% (37–41%) 52% (50–54%) Group <25 kg 25–50 kg 50–75 kg 75–100 kg >100 kg n = 223 n = 1615 n = 1025 n = 808 n = 289 (N = 129) (N = 621) (N = 3140) (N = 324) (N = 107) spK/V achievement per weight group spKt/V ≥1.2  n 213 1550 2946 704 223  % (95% CI) 96% (92–98%) 96% (95–97%) 94% (93–95%)a,1 87% (85–89%)a,2 77% (72–82%) spKt/V ≥1.4  n 180 1418 2438 479 104  % (95% CI) 81% (75–86%) 88% (86–89%)a,3 78% (76–79%)a,4 59% (56–63%) 36% (31–42%) spKt/V ≥1.6  n 109 1048 1475 176 25  % (95% CI) 49% (42–56%) 65% (62–67%) 47% (45–49%) 22% (19–25%) 9% (6–13%) Group <25 kg 25–50 kg 50–75 kg 75–100 kg >100 kg n = 223 n = 1615 n = 1025 n = 808 n = 289 (N = 129) (N = 621) (N = 3140) (N = 324) (N = 107) spK/V achievement per weight group spKt/V ≥1.2  n 213 1550 2946 704 223  % (95% CI) 96% (92–98%) 96% (95–97%) 94% (93–95%)a,1 87% (85–89%)a,2 77% (72–82%) spKt/V ≥1.4  n 180 1418 2438 479 104  % (95% CI) 81% (75–86%) 88% (86–89%)a,3 78% (76–79%)a,4 59% (56–63%) 36% (31–42%) spKt/V ≥1.6  n 109 1048 1475 176 25  % (95% CI) 49% (42–56%) 65% (62–67%) 47% (45–49%) 22% (19–25%) 9% (6–13%) n, number of patient-years (N, number of individual patients). a In the sensitivity analysis estimates were significantly 3–5% lower: 191% (89–92%), 281% (78–84%), 385% (83–88%), 473% (71–75%) for all other estimates 95% CIs were overlapping. Table 3. Achievement of different target spKt/V values per age and weight group Group 1–2 years 3–5 years 6–11 years 12–18 years 19–29 years n = 34 n = 68 n = 190 n = 2243 n = 3540 (N = 24) (N = 49) (N = 129) (N = 1438) (N = 1021) spKt/V achievement per age group spKt/V ≥1.2  n 31 65 173 1966 3401  % (95% CI) 91% (75–98%) 96% (87–99%); 91% (86–95%) 88% (86–89%) 96% (95–97%) spKt/V ≥1.4  n 20 58 145 1509 2887  % (95% CI) 59% (41–75%) 85% (74–92%) 76% (70–82%) 67% (65–69%) 82% (80–83%) spKt/V ≥1.6  n 10 31 85 870 1837  % (95% CI) 29% (16–48%) 46% (34–58%) 45% (38–52%) 39% (37–41%) 52% (50–54%) Group 1–2 years 3–5 years 6–11 years 12–18 years 19–29 years n = 34 n = 68 n = 190 n = 2243 n = 3540 (N = 24) (N = 49) (N = 129) (N = 1438) (N = 1021) spKt/V achievement per age group spKt/V ≥1.2  n 31 65 173 1966 3401  % (95% CI) 91% (75–98%) 96% (87–99%); 91% (86–95%) 88% (86–89%) 96% (95–97%) spKt/V ≥1.4  n 20 58 145 1509 2887  % (95% CI) 59% (41–75%) 85% (74–92%) 76% (70–82%) 67% (65–69%) 82% (80–83%) spKt/V ≥1.6  n 10 31 85 870 1837  % (95% CI) 29% (16–48%) 46% (34–58%) 45% (38–52%) 39% (37–41%) 52% (50–54%) Group <25 kg 25–50 kg 50–75 kg 75–100 kg >100 kg n = 223 n = 1615 n = 1025 n = 808 n = 289 (N = 129) (N = 621) (N = 3140) (N = 324) (N = 107) spK/V achievement per weight group spKt/V ≥1.2  n 213 1550 2946 704 223  % (95% CI) 96% (92–98%) 96% (95–97%) 94% (93–95%)a,1 87% (85–89%)a,2 77% (72–82%) spKt/V ≥1.4  n 180 1418 2438 479 104  % (95% CI) 81% (75–86%) 88% (86–89%)a,3 78% (76–79%)a,4 59% (56–63%) 36% (31–42%) spKt/V ≥1.6  n 109 1048 1475 176 25  % (95% CI) 49% (42–56%) 65% (62–67%) 47% (45–49%) 22% (19–25%) 9% (6–13%) Group <25 kg 25–50 kg 50–75 kg 75–100 kg >100 kg n = 223 n = 1615 n = 1025 n = 808 n = 289 (N = 129) (N = 621) (N = 3140) (N = 324) (N = 107) spK/V achievement per weight group spKt/V ≥1.2  n 213 1550 2946 704 223  % (95% CI) 96% (92–98%) 96% (95–97%) 94% (93–95%)a,1 87% (85–89%)a,2 77% (72–82%) spKt/V ≥1.4  n 180 1418 2438 479 104  % (95% CI) 81% (75–86%) 88% (86–89%)a,3 78% (76–79%)a,4 59% (56–63%) 36% (31–42%) spKt/V ≥1.6  n 109 1048 1475 176 25  % (95% CI) 49% (42–56%) 65% (62–67%) 47% (45–49%) 22% (19–25%) 9% (6–13%) n, number of patient-years (N, number of individual patients). a In the sensitivity analysis estimates were significantly 3–5% lower: 191% (89–92%), 281% (78–84%), 385% (83–88%), 473% (71–75%) for all other estimates 95% CIs were overlapping. Median prescriptions calculated for adults (QD of 600 mL/min, QB of 400 mL/min, KoA of 1140 mL/min, resulting KD of 275 mL/min, treatment duration of 210 min) were reached at 13 years/40 kg (QD), 19 years/60 kg (QB), 17 years/50 kg (KoA), 20 years/70 kg (KD) and 17 years/70 kg (duration). Patients >100 kg again had higher median QD (700 mL/min) and duration (240 min). KD also increased further in young adults (up to 300 mL/min at 29 years) and heavier patients (up to 320 mL/min >100 kg). Dialytic clearance (mean KD/QB ratio, 95% CI) was closer to QB in children (0.79, 0.74–0.84 in toddlers, 0.73, 0.72–7.30 in adolescents) compared with adults (0.70, 0.70–0.71). Conversely, patients >100 kg achieved significantly higher efficiency (0.81, 0.79–0.83), compared with those of 75–100 kg (0.76, 0.75–0.77) and lower weight (range of means: 0.70–0.72). QD exceeded QB in toddlers 5.4-fold (mean QD/QB ratio, 95% CI 5.2–5.6), and only 1.7 (1.70–1.76) in adults. Above 50 kg, mean QD/QB ratios were stable and independent of weight (range of mean ratios: 1.8–1.9; 90% interval: 1.3–2.3). Mean KoA/QB ratios varied between 3.0 and 3.4, highest values were observed in age group 6–11 years and weight group 25–50 kg. Normalized to weight, young children received ∼3–4 times higher QD/kg than adults. Weight-normalized QB/kg showed small differences over age, with significantly lower values (mean, 95% CI) in adolescents (6.1, 5.96–6.13 mL/min/kg) compared with adults (6.6, 6.53–6.70 mL/min/kg) and younger patients (e.g. 7.0, 6.7–7.2 mL/min/kg in 6- to 11-year-old patients). Larger differences were observed between weight groups, with lowest values in patients >100 kg (3.9, 3.7 4.1 mL/min/kg). More than 75% of patients >100 kg had QB/kg rates <5 mL/min/kg prescribed, with a median (IQR) of 3.5 (3.1–4.0) mL/min/kg. In the lowest weight groups QB/kg was almost double with 7.4 (6.5–9.1) mL/min/kg in patients <25 kg and 7.8 (6.8–8.9) mL/min/kg in patients 25–50 kg (i.e. >50% of patients exceeded flows of 7 mL/min/kg). The 95th percentile exceeded 10 mL/min/kg in children aged <10 years (Figure 3). Weight-normalized KoA used in infants and preschool children (range of means: 21.7–23.3 mL/min/kg) was only slightly higher than in adolescents and adults (19.4–19.3 mL/min/kg). Larger differences were observed over weight (mean, 95% CI), with almost 2-fold higher values in 25–50 kg patients (23.4, 23.1–23.6 mL/min/kg) compared with patients >100 kg (12.0, 11.4–12.6 mL/min/kg). Predicted CLD was similar to calculated KD in adults (mean, 95% CI: CLD = 4.7, 3.9–5.4 versus KD = 4.5, 3.9–5.2 mL/min/kg), but up to 45% lower in younger children (CLD = 3.1, 2.4–3.8 versus KD = 5.6, 5.0–6.2 mL/min/kg in 1–2 year olds). Similar to QB/kg, KD/kg (mean, 95% CI) was lower in adolescent patients (4.3, 4.2–4.3 mL/min/kg) compared with younger children (4.8, 4.7–5.0 mL/min/kg in 6–11 year olds) and adults (4.6, 4.5–4.6 mL/min/kg). KD/kg decreased over weight with lowest values of 3.1 (2.9–3.2 mL/min/kg) in patients >100 kg (Supplementary data, Figure S3). Observed HD prescriptions resulted in achieving target spKt/V of ≥1.4 (95% CI) (minimum ≥1.2) in 82% (80–83%) [96% (95–97%)] of young adults 19–29 years, but only 67% (65–69%) [88% (86–89%)] of adolescent patients aged 12–18 years, while estimated proportions in younger age groups (range: 59–85% [91–96%]) were not significantly different from adults. Target achievements were again lower in patients with high weight, with only 36% (31–42%) [77% (72–82%)] in patients weighing >100 kg. Interaction analysis indicated that weight >75 kg and adolescent age are independently associated with lower mean spKt/V (interaction terms non-significant with P > 0.05) as compared with other paediatric subpopulations. Sensitivity analyses Calculation of Cheek TBW resulted in physiologic plausible values across the entire age and weight range, including adults, while Watson TBW yielded estimates up to ≥1 L/kg in low-weight patients >16 years. Morgenstern estimated systematically lower TBW than Cheek (−13% in median), with the smallest difference around 10 years of age. This did, however, not affect the pattern of calculated KD distribution ranges. The pattern of QB/m2 and spKt/V ranges over BSA was similar to QB/kg and spKt/V ranges over weight (maximum ∼1.5 m2). Only a small percentage (4%) of patients had residual renal function (median: 1.8, IQR: 0.8–3.1 mL/min); neglecting CLR did not significantly influence calculated percentile curves. Comparing reported percentiles over weight (from median per year of age data) with percentiles from the sensitivity analysis dataset (individual median per weight band of 10 kg) did not suggest significant bias introduced by repeated measurements. DISCUSSION This study is the first to describe real-life HD prescription behaviour in a large cohort of children and young adults, having started dialysis at paediatric age. Distribution ranges comprising 50% (IQR) and 90% of actual HD prescriptions were derived for each year of age, and for weight bands of 10 kg, based on prescriptions of at least 20 patients (except ≤1 year/≤10 kg). Such data-driven distribution ranges may be useful to understand and fine-tune target prescription parameters of paediatric patients managed with standard HD. Upper limits are markedly higher than expected for a number of parameters, and given that ranges provided originate in the outpatient setting, it is assumed that these high values are well tolerated. Therefore, our results support the consideration of comfortably prescribing more intense HD treatments. Our results show that spKt/V ≥1.2 are routinely provided to young and lower weight patients, indicating that with respect to small solute clearance alone more intense HD regimens may not be necessary in this population. There seems, however, to be room for improvement in dialysis prescription and/or nutrition in all patients <19 years of age, and especially in children <12 years, given the large proportion of children with low weight and height for age. In addition, it has been suggested that spKt/V may underestimate dialysis dose in small patients [13]. In contrast, adolescent and young patients >75 kg are at risk of chronic undertreatment, especially given that dialysis dose may be overestimated in obese patients [14]. They may hence need more intense HD regimens (more frequent, prolonged) for best outcomes. This is particularly worrisome with recent reports on obesity in end-stage renal disease (ESRD) [15] recognizing limitations for kidney transplantation in this population, increasing their need for long-term HD. In adults, HD prescription is relatively standardized, compared with children [16], as the kidney disease outcomes quality initiative (KDOQI) guidelines recommend standard QB and QD prescription of ≥300 mL/min and ≥500 mL/min, respectively. Those minimum values were used indeed in almost all young adults, as shown by 5th percentiles exceeding those values in adults ≥19 years. QD rates of 300–800 mL/min were used in patients <10 years, consistent with recommended rates of 300–800 mL in paediatric patients [2], chosen independently from body size. QB rates are in contrast usually based on body weight (or BSA), with recommended rates of 5–7 mL/min/kg (150–200 mL/min/m2) [2], and consistent QB/kg values may thus be expected over weight and age (and BSA). QB/kg rates were lowest in adolescents (among age groups) and patients >100 kg (among weight groups), while highest rates were observed in the youngest and smallest patients, exceeding in >50% recommended rates of 5–7 mL/min/kg. This is in line with reported actual flow rates of ∼14 ± 6 mL/min/kg in 11 children [3], and not explained by use of the following equation [3] proposed for children: QB = (weight [kg] + 10) × 2.5, which would propose systematically lower QB than observed. In heavier patients median QB of 400 mL/min (90% interval: 300–500 mL/min) was probably considered maximized, resulting in QB/kg rates below 5 mL/min/kg in almost all paediatric patients >100 kg. The higher efficiency of dialytic clearance in toddlers, with KD closer to QB than in adults, may be partly explained by higher QD/QB ratios used. QD exceeded QB 4.7-fold in median (IQR 3.4–8.0) in this youngest age group, i.e. well exceeding the minimum factor of 1.5–2 recommended in children [17] and used in adults (IQR 1.5–1.9). Interestingly, in patients with high weight (≥100 kg) efficiency was not lower (IQR 1.5–2.0), and adolescents achieved slightly higher efficiency (1.7–2.3) than adults. This suggests that not only low efficiency but lower QB/kg in high-weight patients and adolescents may be the main limiting factor to achieve target dose (25th percentile <1.40 in adolescents and patients >90 kg). In adolescents, shorter dialysis duration and larger fraction of TBW (tendency observed in Figures 3 and 4, but not clearly in Table 1) may also contribute to lower delivered target spKt/V. Maximum spKt/V values were achieved in children >30–50 kg (median ≥1.70, Figure 4), while lowest values were seen in heavy patients. Also toddlers (1–2 years) had relatively low median spKt/V (Table 1) with only 59% achieving spKt/V of 1.4 (Table 3), possibly because of their higher fraction of TBW. However, a large proportion (91%) achieved minimal spKt/V >1.2, which may explain partly why no significant difference compared with older children could be demonstrated. Higher HD dose in paediatric (i.e. smaller) patients has been previously reported [18] and may be desired due to higher metabolic rate and nutritional requirements [19], while our data shows that spKt/V tends to decrease again in patients with weight decreasing below 25–30 kg. However, also a large proportion (96%) of patients <25 kg achieved minimal spKt/V of >1.2, compared with only ≤87% in patients >75 kg (Table 3). Following allometric principles (see illustration in Figure 5) [4, 19] a decrease of weight-normalized clearance and dialysis dose with weight could also be desired. Size-related differences in dialysis dose may, however, have implications for patient morbidity and mortality [18], and influence the extent of urea rebound and accuracy of dialysis dose calculations based on single-pool assumptions (with possible overestimation in paediatric patients [20–22] and underestimation in obese patients due to higher fat and lower body water fraction) [14]. Weight-dependent intensity of dialysis prescription may also have implications for drug dosing, since the amount of drugs removed during dialysis would be accordingly expected to be larger in low-weight patients (who receive higher spKt/V), which hence may require higher supplemental drug doses than proposed in adult dialysis patients [23]. FIGURE 5: View largeDownload slide Illustration of physiologically expected clearance change with weight based on allometric principles, using the mean clearance (KD) at a weight of 60 kg as reference (255 mL/min, corresponding to 4.25 mL/min/kg). Left: based on allometric principles, we would expect a linear relationship between the logarithm of weight and the logarithm of clearance, with a slope of 0.75 (solid line). Circles: actually observed KD. Right: weight-normalized clearance is accordingly expected to decrease with weight (solid line). In patients with increasing weight >60 kg, observed KD (circles) are, however, below this scaled ‘physiologic equivalent value’, while patients of ∼25–50 kg appear to receive more intense KD. Dashed lines: non-parametric regression line. FIGURE 5: View largeDownload slide Illustration of physiologically expected clearance change with weight based on allometric principles, using the mean clearance (KD) at a weight of 60 kg as reference (255 mL/min, corresponding to 4.25 mL/min/kg). Left: based on allometric principles, we would expect a linear relationship between the logarithm of weight and the logarithm of clearance, with a slope of 0.75 (solid line). Circles: actually observed KD. Right: weight-normalized clearance is accordingly expected to decrease with weight (solid line). In patients with increasing weight >60 kg, observed KD (circles) are, however, below this scaled ‘physiologic equivalent value’, while patients of ∼25–50 kg appear to receive more intense KD. Dashed lines: non-parametric regression line. This analysis has several limitations. Although a considerable number of paediatric patients was included given the low incidence of paediatric ESRD and dialysis dependency, patient numbers in the lower age and weight groups remain statistically small. This resulted in large 95% CI, and thus, our ability to detect significant differences between younger age groups was probably impeded. Given the retrospective nature of the study, we did not investigate the correlation of prescribed dialysis intensity with clinical outcomes beyond spKt/V. Haemodynamic tolerability (e.g. limiting filter size/KoA, QB and ultrafiltration rate), practical issues (e.g. patient comfort and accepted treatment duration to achieve target dry weight) and monitoring parameters other than spKt/V (e.g. normalized protein catabolic rate [24], interdialytic weight gain [25]), may be able to explain part of the interindividual variability in dialysis prescription, but were not investigated. We did not investigate additional measures of adequacy, such as fluid balance, middle-molecule clearance, mineral and bone disorders, growth and nutrition. Low prescription values may reflect an initial period of maintenance HD in which intensity is slowly up-titrated. The selection of thrice-weekly HD patients may have further excluded young patients, since for most children <10 kg more than three sessions per week may be needed to enable adapted nutrition [2]. Since we included only patients on conventional thrice-weekly in-centre HD, results may have limited utility for different dialysis schedules, such as home [26] or more frequent [27, 28] HD. Prescriptions may be different in countries, where fewer obese patients are treated, and change with transition from paediatric to adult dialysis units. Optimal dialysis prescription in children is still a subject of discussion. Given the growth retardation observed in our paediatric part of this cohort, despite good urea spKt/V delivered, the value of spKt/V as adequacy measure in children can be questioned. Observed age- and size-related differences in HD prescription and delivery may have implications not only for clinical outcomes, but also for urea and middle molecule kinetic assessment and modelling, as well as for drug removal and dose adjustment, deserving further investigations. Provided ranges of outpatient HD prescription can help clinicians and researchers in designing studies as well as personalizing and optimizing delivery of dialysis treatment. SUPPLEMENTARY DATA Supplementary data are available at ndt online. FUNDING This project has been supported by the Research Fund for Junior Researchers, University of Basel, Switzerland, and by the Eckenstein-Geigy Foundation, which is sponsoring research at Pediatric pharmacology at University Children’s Hospital Basel. AUTHORS’ CONTRIBUTIONS All authors equally contributed to research idea, study design, data analysis/interpretation. M.P. and O.M. were responsible for data acquisition, supervision and mentorship. V.G. performed statistical analysis. CONFLICT OF INTEREST STATEMENT M.P. is a consultant at Quantitative Solutions a Certara Company. V.G. and O.M. declare that they have no relevant financial interests. REFERENCES 1 National Kidney Foundation . KDOQI clinical practice guideline for hemodialysis adequacy: 2015 update . Am J Kidney Dis 2015 ; 66 : 884 – 930 CrossRef Search ADS PubMed 2 Fischbach M , Edefonti A , Schröder C et al. Hemodialysis in children: general practical guidelines . Pediatr Nephrol 2005 ; 20 : 1054 – 1066 Google Scholar CrossRef Search ADS PubMed 3 Cochat P , Lioux C. Maintenance hemodialysis during infancy. In: Warady BA , Schaefer FS , Fine RN , Alexander SR (eds). Pediatric dialysis. Dordrecht : Springer Science+Business Media , 2004 , p. 197 – 207 4 Bartelink IH , Rademaker CMA , Schobben AFAM et al. Guidelines on paediatric dosing on the basis of developmental physiology and pharmacokinetic considerations . Clin Pharmacokinet 2006 ; 45 : 1077 – 1097 Google Scholar CrossRef Search ADS PubMed 5 Hemodialysis Adequacy 2006 Work Group . Clinical practice guidelines for hemodialysis adequacy, update 2006 . Am J Kidney Dis 2006 ; 48 : S2 – S90 CrossRef Search ADS PubMed 6 NKF-DOQI clinical practice guidelines for hemodialysis adequacy . National Kidney Foundation . Am J Kidney Dis 1997 ; 30 (3 Suppl 2) : S15 – S66 7 Cheek DB , Mellits D , Elliott D. Body water, height, and weight during growth in normal children . Arch Pediatr Adolesc Med 1966 ; 112 : 312 – 317 Google Scholar CrossRef Search ADS 8 Mosteller RD. Simplified calculation of body-surface area . N Engl J Med 1987 ; 317 : 1098 Google Scholar PubMed 9 Michaels AS. Operating parameters and performance criteria for hemodialyzers and other membrane-separation devices . Trans Am Soc Artif Intern Organs 1966 ; 12 : 387 – 392 Google Scholar PubMed 10 Daugirdas JT. Second generation logarithmic estimates of single-pool variable volume Kt/V: an analysis of error . J Am Soc Nephrol 1993 ; 4 : 1205 – 1213 Google Scholar PubMed 11 Watson PE , Watson ID , Batt RD. Total body water volumes for adult males and females estimated from simple anthropometric measurements . Am J Clin Nutr 1980 ; 33 : 27 – 39 Google Scholar CrossRef Search ADS PubMed 12 Morgenstern BZ , Wühl E , Nair KS et al. Anthropometric prediction of total body water in children who are on pediatric peritoneal dialysis . J Am Soc Nephrol 2006 ; 17 : 285 – 293 Google Scholar CrossRef Search ADS PubMed 13 Spalding E , Chandna SM , Davenport A et al. Kt/V underestimates the hemodialysis dose in women and small men . Kidney Int 2008 ; 74 : 348 – 355 Google Scholar CrossRef Search ADS PubMed 14 Davenport A. Differences in prescribed Kt/V and delivered haemodialysis dose-why obesity makes a difference to survival for haemodialysis patients when using a “one size fits all” Kt/V target . Nephrol Dial Transplant 2013 ; 28 : iv219 – iv223 Google Scholar CrossRef Search ADS PubMed 15 Terrace JD , Oniscu GC. Paediatric obesity and renal transplantation: current challenges and solutions . Pediatr Nephrol 2016 ; 31 : 555 – 562 Google Scholar CrossRef Search ADS PubMed 16 Shroff R , Bonthius M , Fischbach M et al. SP865—Paediatric dialysis practice across the EU—a survey from the EPDWG/ERA-EDTA Registries . Nephrol Dial Transplant 2015 ; 30 : iii662 – iii662 Google Scholar CrossRef Search ADS 17 Müller D , Goldstein SL. Hemodialysis in children with end-stage renal disease . Nat Rev Nephrol 2011 ; 7 : 650 – 658 Google Scholar CrossRef Search ADS PubMed 18 Depner T , Daugirdas J , Greene T et al. Dialysis dose and the effect of gender and body size on outcome in the HEMO Study . Kidney Int 2004 ; 65 : 1386 – 1394 Google Scholar CrossRef Search ADS PubMed 19 Morton AR , Singer MA. The problem with Kt/V: dialysis dose should be normalized to metabolic rate not volume . Semin Dial 2007 ; 20 : 12 – 15 Google Scholar CrossRef Search ADS PubMed 20 Marsenić O , Pavlicić D , Bigović G et al. Effects of postdialysis urea rebound on the quantification of pediatric hemodialysis . Nephron 2000 ; 84 : 124 – 129 Google Scholar CrossRef Search ADS PubMed 21 Daugirdas JT , Greene T , Depner TA et al. Factors that affect postdialysis rebound in serum urea concentration, including the rate of dialysis: results from the HEMO Study . J Am Soc Nephrol 2004 ; 15 : 194 – 203 Google Scholar CrossRef Search ADS PubMed 22 Gotta V , Zhang L , Marsenic Couloures O et al. SP735 - body-weight dependency of urea kinetic parameters in adolescent hemodialysis patients . Nephrol Dial Transplant 2016 ; 31 : i339 – i340 Google Scholar CrossRef Search ADS 23 Gotta V , Dao K , Rodieux F et al. Guidance to develop individual dose recommendations for patients on chronic hemodialysis . Expert Rev Clin Pharmacol 2017 ; 10 : 737 – 752 Google Scholar CrossRef Search ADS PubMed 24 Marsenic O , Peco-Antić A , Jovanović O. Effect of dialysis dose on nutritional status of children on chronic hemodialysis . Nephron 2001 ; 88 : 273 – 275 Google Scholar CrossRef Search ADS PubMed 25 Fischbach M , Zaloszyc A , Shroff R. The interdialytic weight gain: a simple marker of left ventricular hypertrophy in children on chronic haemodialysis . Pediatr Nephrol 2015 ; 30 : 859 – 863 Google Scholar CrossRef Search ADS PubMed 26 Hothi DK , Stronach L , Harvey E. Home haemodialysis . Pediatr Nephrol 2013 ; 28 : 721 – 730 Google Scholar CrossRef Search ADS PubMed 27 Goldstein SL , Silverstein DM , Leung JC et al. Frequent hemodialysis with NxStageTM system in pediatric patients receiving maintenance hemodialysis . Pediatr Nephrol 2008 ; 23 : 129 – 135 Google Scholar CrossRef Search ADS PubMed 28 Laskin BL , Huang G , King E et al. Short, frequent, 5-days-per-week, in-center hemodialysis versus 3-days-per week treatment: a randomized crossover pilot trial through the Midwest Pediatric Nephrology Consortium . Pediatr Nephrol 2017 ; 32 : 1423 – 1432 Google Scholar CrossRef Search ADS PubMed © The Author(s) 2018. Published by Oxford University Press on behalf of ERA-EDTA. All rights reserved. This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/about_us/legal/notices) http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Nephrology Dialysis Transplantation Oxford University Press

Age- and weight-based differences in haemodialysis prescription and delivery in children, adolescents and young adults

Loading next page...
 
/lp/ou_press/age-and-weight-based-differences-in-haemodialysis-prescription-and-W5tWK5DtbB
Publisher
Oxford University Press
Copyright
© The Author(s) 2018. Published by Oxford University Press on behalf of ERA-EDTA. All rights reserved.
ISSN
0931-0509
eISSN
1460-2385
D.O.I.
10.1093/ndt/gfy067
Publisher site
See Article on Publisher Site

Abstract

Abstract Background Limited systematic data are available on prescription and dosing of haemodialysis (HD) in children and adolescents compared with adults. We aimed to characterize age- and weight-based differences in HD delivery in children, adolescents and young adults. Methods This is a retrospective observational study including 1852 patients <30 years on chronic HD from childhood (53 903 HD sessions), receiving thrice weekly outpatient HD between 2004 and 2016 in the USA (6075 patient-years, of which 2535 were in patients aged 1–18 years; weight range 8.3–168 kg). Median individual prescriptions per year were calculated and overall 50% (IQR) and 90% distribution ranges over age and weight were derived. Repeated measurements analysis of variance assessed differences between age and weight groups. Results Prescriptions significantly differed among age and weight groups (P < 0.001). Lower weight patients (<75 kg) had higher (inter-quartile range, IQR) weight-normalized blood flow rate (highest in <25 kg: QB/kg = 6.5–9.1 mL/min/kg), urea dialytic clearance (KD/kg) and single pool Kt/V (spKt/V) (<25 kg: 1.43–1.78; 25–50 kg: 1.52–1.92; 50–75 kg: 1.43–1.74) than heavier patients (lowest in >100 kg: QB/kg = 3.1–4.0 mL/min/kg, spKt/V = 1.22–1.47, respectively). Adolescents had significantly lower QB/kg, KD/kg and spKt/V (1.34–1.71) compared with adults (1.45–1.79) and children <12 years (range of 25th percentiles: 1.37–1.44). Dialytic clearance derived from a mechanistic equation underpredicted KD in children but not in young adults. Significant growth retardation was observed, with the proportion of patients <3rd percentile (height for age) decreasing from 71% (1–2 years) to 15% (>18 years). Conclusion Delivered HD treatment varies with age and weight and is more intensified in children aged <12 years, compared with adolescents and overweight young adults, who appear to be at highest risk of receiving suboptimal treatment. Still, delivery of target or higher spKt/V values did not result in appropriate growth in these children, questioning the value of spKt/V as a measure of HD adequacy in children. Provided ranges of outpatient HD prescription can help clinicians and researchers in personalizing and optimizing delivery of dialysis treatment. chronic haemodialysis, clearance, dialysis dose, epidemiology, obesity, paediatrics INTRODUCTION Haemodialysis (HD) prescription and delivery is relatively standardized in adults [1] compared with children, where frequent adjustments in HD prescription are needed due to growth, with changes in nutrition needs, weight, body composition and volume of total body water (TBW). Potential differences in HD delivery between children, adolescents and adults is not well understood, and very limited systematic data are available regarding fine-tuning of HD prescription across various age and weight groups. In current clinical practice, reference ranges of key prescription parameters such blood flow (QB) are usually derived from body weight or body surface area (BSA) [2, 3], similar to dosing of therapeutics in paediatrics [4]. There are, however, no data on actual prescription and delivery of paediatric HD with ranges of tolerated treatment parameters routinely used in the outpatient setting. While HD prescription (QB: dialysate flow, QD: mass-transfer coefficient for urea KoA and duration of treatment) is expected to differ between children and adults, target values for HD delivery are adopted from adults, and similar dialysis dose would hence be expected to be delivered to paediatric and adult patients. Guidelines recommend measuring delivered dialysis dose in terms of single pool Kt/V (spKt/V; with K: urea dialytic clearance, t: dialysis time, V: urea distribution volume). Minimum spKt/V values of ≥1.2 (adolescents and adults [1]) and higher (children, because of increased nutritional requirements [5]) are recommended, with the target set by ∼0.2 units higher to account for overestimation due to urea rebound [1, 6]. The objective of this descriptive cohort analysis was to better characterize and understand age- and weight-based differences in HD prescription and delivery in children, adolescents and young adults, and to identify potential subpopulations at risk of receiving suboptimal treatment. MATERIALS AND METHODS Study design and participants This retrospective observational study was performed on a cohort of patients who received maintenance HD therapy between May 2004 and March 2016 in DaVita Kidney Care (DaVita Inc., Denver, CO, USA) dialysis centres. All patients that remained ≤18 years, and patients who reached adult age (>18 years) during the follow-up but remained <30 years, were extracted from standardized electronic medical records of the dialysis facilities. The scientific use of the de-identified data was approved by DaVita (DaVita Clinical Research®, Minneapolis, MN, USA), Institutional Review Board approval was not required since retrospective analysis was performed on the de-identified dataset. To be included, patients needed to have treatment records informing about both dialysis treatment (type of dialysis service, duration of treatment, KoA, pre- and post-dialysis weight) and associated demographic and laboratory parameters [age, height, QB, QD, pre- and post-dialysis blood urea nitrogen (BUN) concentration]. Treatment records were excluded if they represented a treatment other than thrice-weekly HD. Data was further screened for non-physiologic and impossible values, which were then eliminated (see section ‘Missing and erroneous data’). Variables and measurement Patient demographics Age (years), gender, weight (kg) and height (converted from inches to cm) were used as recorded. BSA (m2) and TBW (L/kg, according to Cheek et al. [7]) was derived from post-dialysis weight [8]. Dialysis prescription, urea clearance and dialysis dose The following prescription parameters were retrieved as recorded: QB, QD, KoA (mL/min) and treatment duration (min). Urea dialytic clearance (mL/min) was calculated using two methods: (i) theoretically expected clearance (CLD) according to a mechanistic equation accounting for QB, QD and KoA [9] and (ii) actually measured clearance (KD) back-calculated from spKt/V. Weight-normalized values of QD, QB, KoA, KD and CLD (mL/min/kg) were derived from post-dialysis weight and variable-volume spKt/V was calculated [10]. Efficiency of delivered clearance was assessed by indices KD/QB, QD/QB and KoA/QB [9]. The last available residual renal clearance of urea (CLR) from the previous 6 months was extracted. All BUN levels were measured in a central laboratory (DaVita, Inc., Deland, FL, USA) by photometric analysis (Beckman Coulter AU5400; dynamic range: 2.0–130 mg/dL, precision: 1.3–2.5%). Missing and erroneous data Since we used demographic data as explanatory variables for dialysis prescription and derived variables, records with missing data on age, gender, weight and/or height were excluded. Z-scores from the World Health Organization (WHO; absolute score >9 for weight for age or >5 for weight for length) and intradialytic weight changes of >+20%/−30% were used to screen and remove clearly erroneous unphysiologic records. To ensure that Kt/V is available for each treatment record and facilitate interpretation of prescription data, records with missing BUN and treatment duration were excluded. Further, the following variables/combinations indicating highly unlikely thrice-weekly conventional HD were excluded: QB > QD, treatment duration <90 min or >6 h, urea reduction ratio ≤0 or ≥100%. Statistical methods The median value per age was first calculated for each individual, in order to weigh subjects with different monitoring frequencies equally. Demographics were contrasted with WHO percentile curves for children up to 18 years (weight for age and height for age). Distribution ranges over age comprised 50% and 90% actual dialysis prescriptions (absolute and weight-normalized QB, QD, KoA, treatment time), and resulting absolute- and weight-normalized urea clearance and spKt/V were generated applying non-parametric statistics, by calculating at each year of age the 50th percentile (median), 25th and 75th percentiles (interquartile range, IQR), and 5th and 95th percentiles (90% interval). Distribution ranges over weight were generated in a similar way, by calculating percentiles for weight bands of 10 kg, up to a weight of 120 kg. All weights >120 kg were summarized at a weight of 150 kg, all other values were plotted at the upper limit of their weight band. Data were further summarized by median and IQR stratified over five age and weight groups (age groups: toddler: 1–2 years, early childhood: 3–5 years, school-age: 6–11 years, adolescents: 12–18 years, adults: 19–29 years; weight groups: <25 kg, 25–50 kg, 50–75 kg, 75–100 kg, >100 kg). Repeated measurements analysis of variance (ANOVA) was used to assess whether significant differences between groups exist, and to derive group means with 95% confidence intervals (95% CIs) and interpatient variability. A post hoc interaction analysis was performed to assess the combined relationship of age and weight on delivered spKt/V. The fraction of patients reaching spKt/V values of ≥1.2, ≥1.4 and ≥1.6 were calculated with 95% CI. Sensitivity analyses Since we neglected residual kidney function, its potential influence was investigated by exploring the correlation of CLR with age and weight, and by comparing generated spKt/V distribution ranges with total Kt/V (sum of dialysis spKt/V and residual renal Kt/V) ranges over age and weight. Because Cheek TBW estimates were used also for adult patients, the correlation and potential bias between Cheek and Watson TBW estimates [7, 11] was investigated for patients >18 years. Cheek TBW estimates were also compared with Morgenstern estimates [12]. Since HD prescription may also be based on BSA, BSA-normalized QB/m2 and spKtV distribution ranges over BSA were calculated and compared with QB/kg and spKtV ranges over weight. To assess potential bias in distribution ranges over weight introduced by repeated measurements, another dataset was created, summarizing the individual median for each weight band of 10 kg, and both percentiles over weight and percentage of patients reaching target spKt/V values were recalculated. RESULTS Study population The full extracted cohort comprised 66 343 recorded HD sessions with matching laboratory data from 1989 patients. A total of 8009 (12%) treatments did not meet the inclusion criteria of thrice-weekly (1–2 × /week: n = 2238, ≥4 × /week: n = 1944) conventional in-centre HD treatment (home HD: n = 1330, nocturnal dialysis: n = 728, backup dialysis: n = 345, additional day of dialysis: n = 103, ultrafiltration: n = 17, self in-centre HD: n = 7). This resulted in exclusion of 107 patients (5.4%). After cleaning from missing or erroneous data, a total of 53 903 HD sessions (92.4%) from 1852 patients (98.4%) of the included data were retained, comprising 6075 patient-years, of which 2535 patient-years (17 844 sessions) were from patients ≤18 years old (Figure 1; median follow-up duration of dialysis period was 2 years per patient, range: 1–14 years). Detailed numbers of patients per age group are given in Table 3. FIGURE 1: View largeDownload slide Flow chart: selection of study cohort. Horizontal line in histograms: count at least N = 20 patients. This number was achieved for all but the first group used for percentile calculations (1 year of age: N = 14 patients,  <10 kg: N = 11 patients). URR, urea reduction ratio. FIGURE 1: View largeDownload slide Flow chart: selection of study cohort. Horizontal line in histograms: count at least N = 20 patients. This number was achieved for all but the first group used for percentile calculations (1 year of age: N = 14 patients,  <10 kg: N = 11 patients). URR, urea reduction ratio. Median individual weight and height over age is contrasted with WHO growth curves in Figure 2. Most paediatric patients <12 years had low weight and height for age (Table 1). In contrast, large weight variability was observed in adolescents and adults (many patients exceeding 3rd and 97th percentiles, 20% and 11% in 12–18 years old, respectively). Weight range was 8.3–13.6 kg in toddlers (1–2 years), 9.5–22 kg in early childhood (3–5 years), 11.1–66.2 kg in school children (6–11 years), 20–163 kg in adolescents (12–18 years) and 23.3–168 kg in adults. Table 1. Demographics, HD prescription, resulting urea clearance and spKt/V in different age groups Group 1–2 years 3–5 years 6–11 years 12–18 years 19–29 years Number of individuals (number of patient-years) 24 (34) 49 (68) 129 (190) 1438 (2243) 1021 (3540) Number of sessions 175 436 1224 16 009 36 059 Median number of sessions (range) per individual year 4 (1–12) 5 (1–16) 5 (1–22) 6 (1–36) 12 (1–36) Demographics  Post-dialysis weight (kg) 10.3 (9.8–11.0) 14.7 (12.3–16.1) 24.7 (21.0–30.7) 55.4 (46.0–68.4) 60.2 (51.5–71.5)  Weight for age   <50th percentile (%) 97 81 78 65 61a   <3rd percentile (%) 21 37 21 20 15a  Height (cm) 79 (74–83) 93 (86–103) 125 (114–135) 160 (152–168) 163 (155–170)  Height for age   <50th percentile (%) 94 94 85 68 64a   <3rd percentile (%) 71 62 45 35 15a  BSA (m2) 0.48 (0.45–0.51) 0.61 (0.56–0.67) 0.93 (0.81–1.08) 1.58 (1.41–1.79) 1.66 (1.51–1.83) Absolute flows  QB (mL/min) 70 (60–100) 112 (86–150) 180 (150–217.62) 350 (300–400) 400 (350–450)  QD (mL/min) 500 (300–500) 500 (300–500) 500 (500–500) 600 (600–800) 600 (600–800)  KoA (mL/min) 231 (231–231) 309 (231–309) 556 (309–723.25) 1135 (1010–1170) 1140 (1030–1170)  KD (mL/min) 58 (52–64) 78 (70–91) 131 (108–154) 243 (208–272) 275 (247–300)  CLD (mL/min) 34 (23–44) 57 (42–75) 114 (84–137) 244 (208–270) 282 (257–307) Weight-normalized flows  QB/kg (mL/min/kg) 6.8 (5.5–9.9) 7.4 (6.3–9.4) 7.1 (6.0–8.3) 6.1 (4.9–7.2) 6.8 (5.6–7.8)  QD/kg (mL/min/kg) 42.4 (28.6–50.4) 28.7 (23.1–35.3) 19.2 (16.1–23.8) 11.9 (9.5–14.2) 11.0 (9.0–13.2)  KoA/kg (mL/min/kg) 22.7 (21.1–23.8) 19.6 (17.4–24.1) 22.4 (16.2–27.4) 19.1 (15.6–22.9) 18.3 (15.4–21.4)  KD/kg (mL/min/kg) 5.6 (5.0–6.2) 5.7 (5.0–6.3) 5.1 (4.5–5.7) 4.3 (3.6–5.0) 4.5 (3.9–5.2)  CLD/kg (mL/min/kg) 3.1 (2.4–3.8) 3.8 (2.9–4.8) 4.3 (3.4–5.2) 4.2 (3.5–5.1) 4.7 (3.9–5.4) Efficiency of dialysis  KD/QB 0.87 (0.66–0.96) 0.73 (0.62–0.85) 0.71 (0.63–0.81) 0.71 (0.64–0.79) 0.67 (0.62–0.74)  QD/QB 4.7 (3.4–8.0) 3.5 (2.9–4.7) 2.8 (2.3–3.3) 2.0 (1.7–2.3) 1.6 (1.5–1.9)  KoA/QB 3.3 (2.3–3.9) 2.7 (2.1–3.4) 3.1 (2.2–3.7) 3.2 (2.8–3.8) 2.8 (2.4–3.1) Dialyis dose calculation  TBW (L/kg, Cheek) 0.63 (0.61–0.65) 0.63 (0.59–0.66) 0.61 (0.57–0.63) 0.57 (0.53–0.6) 0.56 (0.52–0.6)  Duration (min) 178 (155–182) 180 (180–184) 183 (180–190) 204 (181–225) 210 (182–221)  spKt/V 1.43 (1.37–1.62) 1.56 (1.44–1.8) 1.56 (1.41–1.75) 1.53 (1.34–1.71) 1.61 (1.45–1.79) Group 1–2 years 3–5 years 6–11 years 12–18 years 19–29 years Number of individuals (number of patient-years) 24 (34) 49 (68) 129 (190) 1438 (2243) 1021 (3540) Number of sessions 175 436 1224 16 009 36 059 Median number of sessions (range) per individual year 4 (1–12) 5 (1–16) 5 (1–22) 6 (1–36) 12 (1–36) Demographics  Post-dialysis weight (kg) 10.3 (9.8–11.0) 14.7 (12.3–16.1) 24.7 (21.0–30.7) 55.4 (46.0–68.4) 60.2 (51.5–71.5)  Weight for age   <50th percentile (%) 97 81 78 65 61a   <3rd percentile (%) 21 37 21 20 15a  Height (cm) 79 (74–83) 93 (86–103) 125 (114–135) 160 (152–168) 163 (155–170)  Height for age   <50th percentile (%) 94 94 85 68 64a   <3rd percentile (%) 71 62 45 35 15a  BSA (m2) 0.48 (0.45–0.51) 0.61 (0.56–0.67) 0.93 (0.81–1.08) 1.58 (1.41–1.79) 1.66 (1.51–1.83) Absolute flows  QB (mL/min) 70 (60–100) 112 (86–150) 180 (150–217.62) 350 (300–400) 400 (350–450)  QD (mL/min) 500 (300–500) 500 (300–500) 500 (500–500) 600 (600–800) 600 (600–800)  KoA (mL/min) 231 (231–231) 309 (231–309) 556 (309–723.25) 1135 (1010–1170) 1140 (1030–1170)  KD (mL/min) 58 (52–64) 78 (70–91) 131 (108–154) 243 (208–272) 275 (247–300)  CLD (mL/min) 34 (23–44) 57 (42–75) 114 (84–137) 244 (208–270) 282 (257–307) Weight-normalized flows  QB/kg (mL/min/kg) 6.8 (5.5–9.9) 7.4 (6.3–9.4) 7.1 (6.0–8.3) 6.1 (4.9–7.2) 6.8 (5.6–7.8)  QD/kg (mL/min/kg) 42.4 (28.6–50.4) 28.7 (23.1–35.3) 19.2 (16.1–23.8) 11.9 (9.5–14.2) 11.0 (9.0–13.2)  KoA/kg (mL/min/kg) 22.7 (21.1–23.8) 19.6 (17.4–24.1) 22.4 (16.2–27.4) 19.1 (15.6–22.9) 18.3 (15.4–21.4)  KD/kg (mL/min/kg) 5.6 (5.0–6.2) 5.7 (5.0–6.3) 5.1 (4.5–5.7) 4.3 (3.6–5.0) 4.5 (3.9–5.2)  CLD/kg (mL/min/kg) 3.1 (2.4–3.8) 3.8 (2.9–4.8) 4.3 (3.4–5.2) 4.2 (3.5–5.1) 4.7 (3.9–5.4) Efficiency of dialysis  KD/QB 0.87 (0.66–0.96) 0.73 (0.62–0.85) 0.71 (0.63–0.81) 0.71 (0.64–0.79) 0.67 (0.62–0.74)  QD/QB 4.7 (3.4–8.0) 3.5 (2.9–4.7) 2.8 (2.3–3.3) 2.0 (1.7–2.3) 1.6 (1.5–1.9)  KoA/QB 3.3 (2.3–3.9) 2.7 (2.1–3.4) 3.1 (2.2–3.7) 3.2 (2.8–3.8) 2.8 (2.4–3.1) Dialyis dose calculation  TBW (L/kg, Cheek) 0.63 (0.61–0.65) 0.63 (0.59–0.66) 0.61 (0.57–0.63) 0.57 (0.53–0.6) 0.56 (0.52–0.6)  Duration (min) 178 (155–182) 180 (180–184) 183 (180–190) 204 (181–225) 210 (182–221)  spKt/V 1.43 (1.37–1.62) 1.56 (1.44–1.8) 1.56 (1.41–1.75) 1.53 (1.34–1.71) 1.61 (1.45–1.79) Variables are summarized as median (IQR) unless otherwise mentioned. For all variables listed, repeated measures ANOVA indicated significant differences between the mean of the five age groups, with all P-values  <1.33 × 10−25. Post hoc pairwise group comparisons are mentioned in the text. Respective estimated group means are given with 95% CIs in the Supplementary Data. a Compared with percentile at 18 years. CLD, theoretical urea clearance calculated from QB, QD and KoA [9]; KD, actual delivered clearance according to spKt/V. Table 1. Demographics, HD prescription, resulting urea clearance and spKt/V in different age groups Group 1–2 years 3–5 years 6–11 years 12–18 years 19–29 years Number of individuals (number of patient-years) 24 (34) 49 (68) 129 (190) 1438 (2243) 1021 (3540) Number of sessions 175 436 1224 16 009 36 059 Median number of sessions (range) per individual year 4 (1–12) 5 (1–16) 5 (1–22) 6 (1–36) 12 (1–36) Demographics  Post-dialysis weight (kg) 10.3 (9.8–11.0) 14.7 (12.3–16.1) 24.7 (21.0–30.7) 55.4 (46.0–68.4) 60.2 (51.5–71.5)  Weight for age   <50th percentile (%) 97 81 78 65 61a   <3rd percentile (%) 21 37 21 20 15a  Height (cm) 79 (74–83) 93 (86–103) 125 (114–135) 160 (152–168) 163 (155–170)  Height for age   <50th percentile (%) 94 94 85 68 64a   <3rd percentile (%) 71 62 45 35 15a  BSA (m2) 0.48 (0.45–0.51) 0.61 (0.56–0.67) 0.93 (0.81–1.08) 1.58 (1.41–1.79) 1.66 (1.51–1.83) Absolute flows  QB (mL/min) 70 (60–100) 112 (86–150) 180 (150–217.62) 350 (300–400) 400 (350–450)  QD (mL/min) 500 (300–500) 500 (300–500) 500 (500–500) 600 (600–800) 600 (600–800)  KoA (mL/min) 231 (231–231) 309 (231–309) 556 (309–723.25) 1135 (1010–1170) 1140 (1030–1170)  KD (mL/min) 58 (52–64) 78 (70–91) 131 (108–154) 243 (208–272) 275 (247–300)  CLD (mL/min) 34 (23–44) 57 (42–75) 114 (84–137) 244 (208–270) 282 (257–307) Weight-normalized flows  QB/kg (mL/min/kg) 6.8 (5.5–9.9) 7.4 (6.3–9.4) 7.1 (6.0–8.3) 6.1 (4.9–7.2) 6.8 (5.6–7.8)  QD/kg (mL/min/kg) 42.4 (28.6–50.4) 28.7 (23.1–35.3) 19.2 (16.1–23.8) 11.9 (9.5–14.2) 11.0 (9.0–13.2)  KoA/kg (mL/min/kg) 22.7 (21.1–23.8) 19.6 (17.4–24.1) 22.4 (16.2–27.4) 19.1 (15.6–22.9) 18.3 (15.4–21.4)  KD/kg (mL/min/kg) 5.6 (5.0–6.2) 5.7 (5.0–6.3) 5.1 (4.5–5.7) 4.3 (3.6–5.0) 4.5 (3.9–5.2)  CLD/kg (mL/min/kg) 3.1 (2.4–3.8) 3.8 (2.9–4.8) 4.3 (3.4–5.2) 4.2 (3.5–5.1) 4.7 (3.9–5.4) Efficiency of dialysis  KD/QB 0.87 (0.66–0.96) 0.73 (0.62–0.85) 0.71 (0.63–0.81) 0.71 (0.64–0.79) 0.67 (0.62–0.74)  QD/QB 4.7 (3.4–8.0) 3.5 (2.9–4.7) 2.8 (2.3–3.3) 2.0 (1.7–2.3) 1.6 (1.5–1.9)  KoA/QB 3.3 (2.3–3.9) 2.7 (2.1–3.4) 3.1 (2.2–3.7) 3.2 (2.8–3.8) 2.8 (2.4–3.1) Dialyis dose calculation  TBW (L/kg, Cheek) 0.63 (0.61–0.65) 0.63 (0.59–0.66) 0.61 (0.57–0.63) 0.57 (0.53–0.6) 0.56 (0.52–0.6)  Duration (min) 178 (155–182) 180 (180–184) 183 (180–190) 204 (181–225) 210 (182–221)  spKt/V 1.43 (1.37–1.62) 1.56 (1.44–1.8) 1.56 (1.41–1.75) 1.53 (1.34–1.71) 1.61 (1.45–1.79) Group 1–2 years 3–5 years 6–11 years 12–18 years 19–29 years Number of individuals (number of patient-years) 24 (34) 49 (68) 129 (190) 1438 (2243) 1021 (3540) Number of sessions 175 436 1224 16 009 36 059 Median number of sessions (range) per individual year 4 (1–12) 5 (1–16) 5 (1–22) 6 (1–36) 12 (1–36) Demographics  Post-dialysis weight (kg) 10.3 (9.8–11.0) 14.7 (12.3–16.1) 24.7 (21.0–30.7) 55.4 (46.0–68.4) 60.2 (51.5–71.5)  Weight for age   <50th percentile (%) 97 81 78 65 61a   <3rd percentile (%) 21 37 21 20 15a  Height (cm) 79 (74–83) 93 (86–103) 125 (114–135) 160 (152–168) 163 (155–170)  Height for age   <50th percentile (%) 94 94 85 68 64a   <3rd percentile (%) 71 62 45 35 15a  BSA (m2) 0.48 (0.45–0.51) 0.61 (0.56–0.67) 0.93 (0.81–1.08) 1.58 (1.41–1.79) 1.66 (1.51–1.83) Absolute flows  QB (mL/min) 70 (60–100) 112 (86–150) 180 (150–217.62) 350 (300–400) 400 (350–450)  QD (mL/min) 500 (300–500) 500 (300–500) 500 (500–500) 600 (600–800) 600 (600–800)  KoA (mL/min) 231 (231–231) 309 (231–309) 556 (309–723.25) 1135 (1010–1170) 1140 (1030–1170)  KD (mL/min) 58 (52–64) 78 (70–91) 131 (108–154) 243 (208–272) 275 (247–300)  CLD (mL/min) 34 (23–44) 57 (42–75) 114 (84–137) 244 (208–270) 282 (257–307) Weight-normalized flows  QB/kg (mL/min/kg) 6.8 (5.5–9.9) 7.4 (6.3–9.4) 7.1 (6.0–8.3) 6.1 (4.9–7.2) 6.8 (5.6–7.8)  QD/kg (mL/min/kg) 42.4 (28.6–50.4) 28.7 (23.1–35.3) 19.2 (16.1–23.8) 11.9 (9.5–14.2) 11.0 (9.0–13.2)  KoA/kg (mL/min/kg) 22.7 (21.1–23.8) 19.6 (17.4–24.1) 22.4 (16.2–27.4) 19.1 (15.6–22.9) 18.3 (15.4–21.4)  KD/kg (mL/min/kg) 5.6 (5.0–6.2) 5.7 (5.0–6.3) 5.1 (4.5–5.7) 4.3 (3.6–5.0) 4.5 (3.9–5.2)  CLD/kg (mL/min/kg) 3.1 (2.4–3.8) 3.8 (2.9–4.8) 4.3 (3.4–5.2) 4.2 (3.5–5.1) 4.7 (3.9–5.4) Efficiency of dialysis  KD/QB 0.87 (0.66–0.96) 0.73 (0.62–0.85) 0.71 (0.63–0.81) 0.71 (0.64–0.79) 0.67 (0.62–0.74)  QD/QB 4.7 (3.4–8.0) 3.5 (2.9–4.7) 2.8 (2.3–3.3) 2.0 (1.7–2.3) 1.6 (1.5–1.9)  KoA/QB 3.3 (2.3–3.9) 2.7 (2.1–3.4) 3.1 (2.2–3.7) 3.2 (2.8–3.8) 2.8 (2.4–3.1) Dialyis dose calculation  TBW (L/kg, Cheek) 0.63 (0.61–0.65) 0.63 (0.59–0.66) 0.61 (0.57–0.63) 0.57 (0.53–0.6) 0.56 (0.52–0.6)  Duration (min) 178 (155–182) 180 (180–184) 183 (180–190) 204 (181–225) 210 (182–221)  spKt/V 1.43 (1.37–1.62) 1.56 (1.44–1.8) 1.56 (1.41–1.75) 1.53 (1.34–1.71) 1.61 (1.45–1.79) Variables are summarized as median (IQR) unless otherwise mentioned. For all variables listed, repeated measures ANOVA indicated significant differences between the mean of the five age groups, with all P-values  <1.33 × 10−25. Post hoc pairwise group comparisons are mentioned in the text. Respective estimated group means are given with 95% CIs in the Supplementary Data. a Compared with percentile at 18 years. CLD, theoretical urea clearance calculated from QB, QD and KoA [9]; KD, actual delivered clearance according to spKt/V. FIGURE 2: View largeDownload slide Patient demographics. Illustration of post-dialysis weight (upper panel) and height (lower panel) over age for the cleaned dataset (median per age), and comparison with WHO growth curves (curved lines; solid line: 50th percentile, dashed lines: 3rd and 97th percentile). The individual median weight and height per age was calculated, individual changes over age are indicated by grey lines. FIGURE 2: View largeDownload slide Patient demographics. Illustration of post-dialysis weight (upper panel) and height (lower panel) over age for the cleaned dataset (median per age), and comparison with WHO growth curves (curved lines; solid line: 50th percentile, dashed lines: 3rd and 97th percentile). The individual median weight and height per age was calculated, individual changes over age are indicated by grey lines. Dialysis prescription, urea clearance and dialysis dose Distribution ranges are shown in Figure 3 (weight-normalized prescribed flows and duration of HD) and Figure 4 (resulting weight-normalized urea clearance and dialysis dose) over age and weight. Supplementary data, Figures S1 and S2 show distribution ranges for absolute prescribed and derived flows (QB, QD, KoA, KD, CLD) and for measures of dialysis efficiency (KD/QB, QD/QB, KoA/QB). FIGURE 3: View largeDownload slide Dialysis prescription over age (left) and over weight (right). Solid lines with dots: median. Grey area: IQR. Dashed lines: 5th and 95th percentiles. FIGURE 3: View largeDownload slide Dialysis prescription over age (left) and over weight (right). Solid lines with dots: median. Grey area: IQR. Dashed lines: 5th and 95th percentiles. FIGURE 4: View largeDownload slide Resulting urea clearance and dialysis dose over age (left) and weight (right). Solid lines and dots: median. Dashed lines: 5th and 95th percentiles. Grey shaded area: IQR. CLD: theoretical urea dialytic clearance calculated from QB, QD and KoA [9]. KD: urea dialytic clearance calculated from spKt/V. Horizontal solid line: spKt/V target of 1.4. Horizontal dotted lines: spKt/V of 1.2 (minimum target) and 1.6 (clinically frequently used target). FIGURE 4: View largeDownload slide Resulting urea clearance and dialysis dose over age (left) and weight (right). Solid lines and dots: median. Dashed lines: 5th and 95th percentiles. Grey shaded area: IQR. CLD: theoretical urea dialytic clearance calculated from QB, QD and KoA [9]. KD: urea dialytic clearance calculated from spKt/V. Horizontal solid line: spKt/V target of 1.4. Horizontal dotted lines: spKt/V of 1.2 (minimum target) and 1.6 (clinically frequently used target). Tables 1 and 2 summarize prescription parameters (median, IQR) stratified over five age and weight groups, respectively. Repeated measures ANOVA indicated that the mean of all listed variables is significantly influenced by age (Table 1, all P-values <1.33 × 10−25) and weight (Table 2, all P-values <2.06 × 10−50). Estimated group means are summarized with 95% CI in Supplementary data, Table S1. Weight generally better explained interpatient variability in weight-based prescription than age (mostly smaller interpatient variability estimated). Table 2. Demographics, HD prescription, resulting urea clearance and spKt/V in different weight groups Group <25 kg 25–50 kg 50–75 kg 75–100 kg >100 kg Number of individuals (number of patient-years) 129 (223) 621 (1615) 1025 (3140) 324 (808) 107 (289) [number of sessions] [1419] [13 245] [28 659] [7744] [2836] Demographics  Post-dialysis weight (kg) 18.6 (12.97–21.8) 43.7 (38.3–47.3) 60 (55.0–66.2) 83.1 (79.0–89.7) 112 (104.2–125.4)  Height (cm) 107 (88–117) 152 (142–157) 165 (157–170) 170 (163–175) 170 (165–175)  BSA (m2) 0.74 (0.57–0.84) 1.35 (1.24–1.44) 1.65 (1.57–1.76) 1.98 (1.91–2.07) 2.31 (2.22–2.45) Absolute flows (mL/min)  QB 150 (100–175) 350 (275–400) 400 (350–450) 400 (350–450) 400 (350–450)  QD 500 (319–500) 600 (581–800) 600 (600–800) 600 (600–800) 700 (600–800)  KoA 309 (231–556) 1030 (911–1170) 1140 (1030–1170) 1170 (1030–1183) 1170 (1140–1240)  KD 96 (74–118) 229 (194–254) 267 (242–292) 292 (266–319) 322(288–349)  CLD 75 (44–97) 236 (196–268) 276 (249–300) 287 (258–309) 284 (260–313) Weight-normalized flows (mL/min/kg)  QB/kg 7.4 (6.45–9.07) 7.8 (6.76–8.94) 6.5 (5.69–7.4) 4.8 (4.21–5.33) 3.5 (3.1–3.95)  QD/kg 25.6 (21.4–31.5) 14.5 (12.8–17.1) 11.1 (9.7–12.7) 7.8 (7–9.2) 5.9 (5.4–6.9)  KoA/kg 21.8 (16.8–25.7) 24.2 (21.3–27.2) 18.4 (16.4–20.4) 13.6 (12.4–14.8) 10.5 (9.5–11.4)  KD/kg 5.5 (4.9–6.3) 5.3 (4.7–5.9) 4.4 (3.9–4.9) 3.5 (3.1–3.8) 2.8 (2.4–3.1)  CLD/kg 3.9 (3.1–5.1) 5.5 (4.8–6.2) 4.5 (4.0–5.1) 3.4 (3.0–3.7) 2.5 (2.2–2.8) Efficiency of dialysis  KD/QB 0.72 (0.62–0.85) 0.69 (0.62–0.76) 0.67 (0.62–0.74) 0.73 (0.66–0.81) 0.78 (0.72–0.85)  QD/QB 3.33 (2.62–4.17) 2 (1.67–2.29) 1.71 (1.5–2) 1.67 (1.5–1.94) 1.72 (1.5–2)  KoA/QB 2.82 (2.06–3.71) 3.07 (2.64–3.71) 2.85 (2.5–3.29) 2.89 (2.5–3.3) 3.01 (2.58–3.31) Dialysis dose calculation  TBW (L/kg, Cheek) 0.63 (0.60–0.65) 0.58 (0.56–0.61) 0.57 (0.53–0.60) 0.52 (0.47–0.56) 0.49 (0.44–0.51)  Duration (min) 180 (180–186) 184 (180–210) 210 (183–219) 215 (201–240) 240 (212–245)   spKt/V 1.58 (1.43–1.78) 1.71 (1.52–1.92) 1.58 (1.43–1.74) 1.45 (1.31–1.58) 1.35 (1.22–1.47) Group <25 kg 25–50 kg 50–75 kg 75–100 kg >100 kg Number of individuals (number of patient-years) 129 (223) 621 (1615) 1025 (3140) 324 (808) 107 (289) [number of sessions] [1419] [13 245] [28 659] [7744] [2836] Demographics  Post-dialysis weight (kg) 18.6 (12.97–21.8) 43.7 (38.3–47.3) 60 (55.0–66.2) 83.1 (79.0–89.7) 112 (104.2–125.4)  Height (cm) 107 (88–117) 152 (142–157) 165 (157–170) 170 (163–175) 170 (165–175)  BSA (m2) 0.74 (0.57–0.84) 1.35 (1.24–1.44) 1.65 (1.57–1.76) 1.98 (1.91–2.07) 2.31 (2.22–2.45) Absolute flows (mL/min)  QB 150 (100–175) 350 (275–400) 400 (350–450) 400 (350–450) 400 (350–450)  QD 500 (319–500) 600 (581–800) 600 (600–800) 600 (600–800) 700 (600–800)  KoA 309 (231–556) 1030 (911–1170) 1140 (1030–1170) 1170 (1030–1183) 1170 (1140–1240)  KD 96 (74–118) 229 (194–254) 267 (242–292) 292 (266–319) 322(288–349)  CLD 75 (44–97) 236 (196–268) 276 (249–300) 287 (258–309) 284 (260–313) Weight-normalized flows (mL/min/kg)  QB/kg 7.4 (6.45–9.07) 7.8 (6.76–8.94) 6.5 (5.69–7.4) 4.8 (4.21–5.33) 3.5 (3.1–3.95)  QD/kg 25.6 (21.4–31.5) 14.5 (12.8–17.1) 11.1 (9.7–12.7) 7.8 (7–9.2) 5.9 (5.4–6.9)  KoA/kg 21.8 (16.8–25.7) 24.2 (21.3–27.2) 18.4 (16.4–20.4) 13.6 (12.4–14.8) 10.5 (9.5–11.4)  KD/kg 5.5 (4.9–6.3) 5.3 (4.7–5.9) 4.4 (3.9–4.9) 3.5 (3.1–3.8) 2.8 (2.4–3.1)  CLD/kg 3.9 (3.1–5.1) 5.5 (4.8–6.2) 4.5 (4.0–5.1) 3.4 (3.0–3.7) 2.5 (2.2–2.8) Efficiency of dialysis  KD/QB 0.72 (0.62–0.85) 0.69 (0.62–0.76) 0.67 (0.62–0.74) 0.73 (0.66–0.81) 0.78 (0.72–0.85)  QD/QB 3.33 (2.62–4.17) 2 (1.67–2.29) 1.71 (1.5–2) 1.67 (1.5–1.94) 1.72 (1.5–2)  KoA/QB 2.82 (2.06–3.71) 3.07 (2.64–3.71) 2.85 (2.5–3.29) 2.89 (2.5–3.3) 3.01 (2.58–3.31) Dialysis dose calculation  TBW (L/kg, Cheek) 0.63 (0.60–0.65) 0.58 (0.56–0.61) 0.57 (0.53–0.60) 0.52 (0.47–0.56) 0.49 (0.44–0.51)  Duration (min) 180 (180–186) 184 (180–210) 210 (183–219) 215 (201–240) 240 (212–245)   spKt/V 1.58 (1.43–1.78) 1.71 (1.52–1.92) 1.58 (1.43–1.74) 1.45 (1.31–1.58) 1.35 (1.22–1.47) Variables are summarized as median (IQR) unless otherwise mentioned. For all variables listed, repeated measures ANOVA indicated significant differences between the mean of the five age groups, with all P-values  <2.06 × 10−50. Respective estimated group means are given with 95% CIs in the Supplementary Data. CLD: theoretical urea clearance calculated from QB, QD and KoA [9]. KD, actual delivered clearance according to spKt/V. Table 2. Demographics, HD prescription, resulting urea clearance and spKt/V in different weight groups Group <25 kg 25–50 kg 50–75 kg 75–100 kg >100 kg Number of individuals (number of patient-years) 129 (223) 621 (1615) 1025 (3140) 324 (808) 107 (289) [number of sessions] [1419] [13 245] [28 659] [7744] [2836] Demographics  Post-dialysis weight (kg) 18.6 (12.97–21.8) 43.7 (38.3–47.3) 60 (55.0–66.2) 83.1 (79.0–89.7) 112 (104.2–125.4)  Height (cm) 107 (88–117) 152 (142–157) 165 (157–170) 170 (163–175) 170 (165–175)  BSA (m2) 0.74 (0.57–0.84) 1.35 (1.24–1.44) 1.65 (1.57–1.76) 1.98 (1.91–2.07) 2.31 (2.22–2.45) Absolute flows (mL/min)  QB 150 (100–175) 350 (275–400) 400 (350–450) 400 (350–450) 400 (350–450)  QD 500 (319–500) 600 (581–800) 600 (600–800) 600 (600–800) 700 (600–800)  KoA 309 (231–556) 1030 (911–1170) 1140 (1030–1170) 1170 (1030–1183) 1170 (1140–1240)  KD 96 (74–118) 229 (194–254) 267 (242–292) 292 (266–319) 322(288–349)  CLD 75 (44–97) 236 (196–268) 276 (249–300) 287 (258–309) 284 (260–313) Weight-normalized flows (mL/min/kg)  QB/kg 7.4 (6.45–9.07) 7.8 (6.76–8.94) 6.5 (5.69–7.4) 4.8 (4.21–5.33) 3.5 (3.1–3.95)  QD/kg 25.6 (21.4–31.5) 14.5 (12.8–17.1) 11.1 (9.7–12.7) 7.8 (7–9.2) 5.9 (5.4–6.9)  KoA/kg 21.8 (16.8–25.7) 24.2 (21.3–27.2) 18.4 (16.4–20.4) 13.6 (12.4–14.8) 10.5 (9.5–11.4)  KD/kg 5.5 (4.9–6.3) 5.3 (4.7–5.9) 4.4 (3.9–4.9) 3.5 (3.1–3.8) 2.8 (2.4–3.1)  CLD/kg 3.9 (3.1–5.1) 5.5 (4.8–6.2) 4.5 (4.0–5.1) 3.4 (3.0–3.7) 2.5 (2.2–2.8) Efficiency of dialysis  KD/QB 0.72 (0.62–0.85) 0.69 (0.62–0.76) 0.67 (0.62–0.74) 0.73 (0.66–0.81) 0.78 (0.72–0.85)  QD/QB 3.33 (2.62–4.17) 2 (1.67–2.29) 1.71 (1.5–2) 1.67 (1.5–1.94) 1.72 (1.5–2)  KoA/QB 2.82 (2.06–3.71) 3.07 (2.64–3.71) 2.85 (2.5–3.29) 2.89 (2.5–3.3) 3.01 (2.58–3.31) Dialysis dose calculation  TBW (L/kg, Cheek) 0.63 (0.60–0.65) 0.58 (0.56–0.61) 0.57 (0.53–0.60) 0.52 (0.47–0.56) 0.49 (0.44–0.51)  Duration (min) 180 (180–186) 184 (180–210) 210 (183–219) 215 (201–240) 240 (212–245)   spKt/V 1.58 (1.43–1.78) 1.71 (1.52–1.92) 1.58 (1.43–1.74) 1.45 (1.31–1.58) 1.35 (1.22–1.47) Group <25 kg 25–50 kg 50–75 kg 75–100 kg >100 kg Number of individuals (number of patient-years) 129 (223) 621 (1615) 1025 (3140) 324 (808) 107 (289) [number of sessions] [1419] [13 245] [28 659] [7744] [2836] Demographics  Post-dialysis weight (kg) 18.6 (12.97–21.8) 43.7 (38.3–47.3) 60 (55.0–66.2) 83.1 (79.0–89.7) 112 (104.2–125.4)  Height (cm) 107 (88–117) 152 (142–157) 165 (157–170) 170 (163–175) 170 (165–175)  BSA (m2) 0.74 (0.57–0.84) 1.35 (1.24–1.44) 1.65 (1.57–1.76) 1.98 (1.91–2.07) 2.31 (2.22–2.45) Absolute flows (mL/min)  QB 150 (100–175) 350 (275–400) 400 (350–450) 400 (350–450) 400 (350–450)  QD 500 (319–500) 600 (581–800) 600 (600–800) 600 (600–800) 700 (600–800)  KoA 309 (231–556) 1030 (911–1170) 1140 (1030–1170) 1170 (1030–1183) 1170 (1140–1240)  KD 96 (74–118) 229 (194–254) 267 (242–292) 292 (266–319) 322(288–349)  CLD 75 (44–97) 236 (196–268) 276 (249–300) 287 (258–309) 284 (260–313) Weight-normalized flows (mL/min/kg)  QB/kg 7.4 (6.45–9.07) 7.8 (6.76–8.94) 6.5 (5.69–7.4) 4.8 (4.21–5.33) 3.5 (3.1–3.95)  QD/kg 25.6 (21.4–31.5) 14.5 (12.8–17.1) 11.1 (9.7–12.7) 7.8 (7–9.2) 5.9 (5.4–6.9)  KoA/kg 21.8 (16.8–25.7) 24.2 (21.3–27.2) 18.4 (16.4–20.4) 13.6 (12.4–14.8) 10.5 (9.5–11.4)  KD/kg 5.5 (4.9–6.3) 5.3 (4.7–5.9) 4.4 (3.9–4.9) 3.5 (3.1–3.8) 2.8 (2.4–3.1)  CLD/kg 3.9 (3.1–5.1) 5.5 (4.8–6.2) 4.5 (4.0–5.1) 3.4 (3.0–3.7) 2.5 (2.2–2.8) Efficiency of dialysis  KD/QB 0.72 (0.62–0.85) 0.69 (0.62–0.76) 0.67 (0.62–0.74) 0.73 (0.66–0.81) 0.78 (0.72–0.85)  QD/QB 3.33 (2.62–4.17) 2 (1.67–2.29) 1.71 (1.5–2) 1.67 (1.5–1.94) 1.72 (1.5–2)  KoA/QB 2.82 (2.06–3.71) 3.07 (2.64–3.71) 2.85 (2.5–3.29) 2.89 (2.5–3.3) 3.01 (2.58–3.31) Dialysis dose calculation  TBW (L/kg, Cheek) 0.63 (0.60–0.65) 0.58 (0.56–0.61) 0.57 (0.53–0.60) 0.52 (0.47–0.56) 0.49 (0.44–0.51)  Duration (min) 180 (180–186) 184 (180–210) 210 (183–219) 215 (201–240) 240 (212–245)   spKt/V 1.58 (1.43–1.78) 1.71 (1.52–1.92) 1.58 (1.43–1.74) 1.45 (1.31–1.58) 1.35 (1.22–1.47) Variables are summarized as median (IQR) unless otherwise mentioned. For all variables listed, repeated measures ANOVA indicated significant differences between the mean of the five age groups, with all P-values  <2.06 × 10−50. Respective estimated group means are given with 95% CIs in the Supplementary Data. CLD: theoretical urea clearance calculated from QB, QD and KoA [9]. KD, actual delivered clearance according to spKt/V. Table 3. Achievement of different target spKt/V values per age and weight group Group 1–2 years 3–5 years 6–11 years 12–18 years 19–29 years n = 34 n = 68 n = 190 n = 2243 n = 3540 (N = 24) (N = 49) (N = 129) (N = 1438) (N = 1021) spKt/V achievement per age group spKt/V ≥1.2  n 31 65 173 1966 3401  % (95% CI) 91% (75–98%) 96% (87–99%); 91% (86–95%) 88% (86–89%) 96% (95–97%) spKt/V ≥1.4  n 20 58 145 1509 2887  % (95% CI) 59% (41–75%) 85% (74–92%) 76% (70–82%) 67% (65–69%) 82% (80–83%) spKt/V ≥1.6  n 10 31 85 870 1837  % (95% CI) 29% (16–48%) 46% (34–58%) 45% (38–52%) 39% (37–41%) 52% (50–54%) Group 1–2 years 3–5 years 6–11 years 12–18 years 19–29 years n = 34 n = 68 n = 190 n = 2243 n = 3540 (N = 24) (N = 49) (N = 129) (N = 1438) (N = 1021) spKt/V achievement per age group spKt/V ≥1.2  n 31 65 173 1966 3401  % (95% CI) 91% (75–98%) 96% (87–99%); 91% (86–95%) 88% (86–89%) 96% (95–97%) spKt/V ≥1.4  n 20 58 145 1509 2887  % (95% CI) 59% (41–75%) 85% (74–92%) 76% (70–82%) 67% (65–69%) 82% (80–83%) spKt/V ≥1.6  n 10 31 85 870 1837  % (95% CI) 29% (16–48%) 46% (34–58%) 45% (38–52%) 39% (37–41%) 52% (50–54%) Group <25 kg 25–50 kg 50–75 kg 75–100 kg >100 kg n = 223 n = 1615 n = 1025 n = 808 n = 289 (N = 129) (N = 621) (N = 3140) (N = 324) (N = 107) spK/V achievement per weight group spKt/V ≥1.2  n 213 1550 2946 704 223  % (95% CI) 96% (92–98%) 96% (95–97%) 94% (93–95%)a,1 87% (85–89%)a,2 77% (72–82%) spKt/V ≥1.4  n 180 1418 2438 479 104  % (95% CI) 81% (75–86%) 88% (86–89%)a,3 78% (76–79%)a,4 59% (56–63%) 36% (31–42%) spKt/V ≥1.6  n 109 1048 1475 176 25  % (95% CI) 49% (42–56%) 65% (62–67%) 47% (45–49%) 22% (19–25%) 9% (6–13%) Group <25 kg 25–50 kg 50–75 kg 75–100 kg >100 kg n = 223 n = 1615 n = 1025 n = 808 n = 289 (N = 129) (N = 621) (N = 3140) (N = 324) (N = 107) spK/V achievement per weight group spKt/V ≥1.2  n 213 1550 2946 704 223  % (95% CI) 96% (92–98%) 96% (95–97%) 94% (93–95%)a,1 87% (85–89%)a,2 77% (72–82%) spKt/V ≥1.4  n 180 1418 2438 479 104  % (95% CI) 81% (75–86%) 88% (86–89%)a,3 78% (76–79%)a,4 59% (56–63%) 36% (31–42%) spKt/V ≥1.6  n 109 1048 1475 176 25  % (95% CI) 49% (42–56%) 65% (62–67%) 47% (45–49%) 22% (19–25%) 9% (6–13%) n, number of patient-years (N, number of individual patients). a In the sensitivity analysis estimates were significantly 3–5% lower: 191% (89–92%), 281% (78–84%), 385% (83–88%), 473% (71–75%) for all other estimates 95% CIs were overlapping. Table 3. Achievement of different target spKt/V values per age and weight group Group 1–2 years 3–5 years 6–11 years 12–18 years 19–29 years n = 34 n = 68 n = 190 n = 2243 n = 3540 (N = 24) (N = 49) (N = 129) (N = 1438) (N = 1021) spKt/V achievement per age group spKt/V ≥1.2  n 31 65 173 1966 3401  % (95% CI) 91% (75–98%) 96% (87–99%); 91% (86–95%) 88% (86–89%) 96% (95–97%) spKt/V ≥1.4  n 20 58 145 1509 2887  % (95% CI) 59% (41–75%) 85% (74–92%) 76% (70–82%) 67% (65–69%) 82% (80–83%) spKt/V ≥1.6  n 10 31 85 870 1837  % (95% CI) 29% (16–48%) 46% (34–58%) 45% (38–52%) 39% (37–41%) 52% (50–54%) Group 1–2 years 3–5 years 6–11 years 12–18 years 19–29 years n = 34 n = 68 n = 190 n = 2243 n = 3540 (N = 24) (N = 49) (N = 129) (N = 1438) (N = 1021) spKt/V achievement per age group spKt/V ≥1.2  n 31 65 173 1966 3401  % (95% CI) 91% (75–98%) 96% (87–99%); 91% (86–95%) 88% (86–89%) 96% (95–97%) spKt/V ≥1.4  n 20 58 145 1509 2887  % (95% CI) 59% (41–75%) 85% (74–92%) 76% (70–82%) 67% (65–69%) 82% (80–83%) spKt/V ≥1.6  n 10 31 85 870 1837  % (95% CI) 29% (16–48%) 46% (34–58%) 45% (38–52%) 39% (37–41%) 52% (50–54%) Group <25 kg 25–50 kg 50–75 kg 75–100 kg >100 kg n = 223 n = 1615 n = 1025 n = 808 n = 289 (N = 129) (N = 621) (N = 3140) (N = 324) (N = 107) spK/V achievement per weight group spKt/V ≥1.2  n 213 1550 2946 704 223  % (95% CI) 96% (92–98%) 96% (95–97%) 94% (93–95%)a,1 87% (85–89%)a,2 77% (72–82%) spKt/V ≥1.4  n 180 1418 2438 479 104  % (95% CI) 81% (75–86%) 88% (86–89%)a,3 78% (76–79%)a,4 59% (56–63%) 36% (31–42%) spKt/V ≥1.6  n 109 1048 1475 176 25  % (95% CI) 49% (42–56%) 65% (62–67%) 47% (45–49%) 22% (19–25%) 9% (6–13%) Group <25 kg 25–50 kg 50–75 kg 75–100 kg >100 kg n = 223 n = 1615 n = 1025 n = 808 n = 289 (N = 129) (N = 621) (N = 3140) (N = 324) (N = 107) spK/V achievement per weight group spKt/V ≥1.2  n 213 1550 2946 704 223  % (95% CI) 96% (92–98%) 96% (95–97%) 94% (93–95%)a,1 87% (85–89%)a,2 77% (72–82%) spKt/V ≥1.4  n 180 1418 2438 479 104  % (95% CI) 81% (75–86%) 88% (86–89%)a,3 78% (76–79%)a,4 59% (56–63%) 36% (31–42%) spKt/V ≥1.6  n 109 1048 1475 176 25  % (95% CI) 49% (42–56%) 65% (62–67%) 47% (45–49%) 22% (19–25%) 9% (6–13%) n, number of patient-years (N, number of individual patients). a In the sensitivity analysis estimates were significantly 3–5% lower: 191% (89–92%), 281% (78–84%), 385% (83–88%), 473% (71–75%) for all other estimates 95% CIs were overlapping. Median prescriptions calculated for adults (QD of 600 mL/min, QB of 400 mL/min, KoA of 1140 mL/min, resulting KD of 275 mL/min, treatment duration of 210 min) were reached at 13 years/40 kg (QD), 19 years/60 kg (QB), 17 years/50 kg (KoA), 20 years/70 kg (KD) and 17 years/70 kg (duration). Patients >100 kg again had higher median QD (700 mL/min) and duration (240 min). KD also increased further in young adults (up to 300 mL/min at 29 years) and heavier patients (up to 320 mL/min >100 kg). Dialytic clearance (mean KD/QB ratio, 95% CI) was closer to QB in children (0.79, 0.74–0.84 in toddlers, 0.73, 0.72–7.30 in adolescents) compared with adults (0.70, 0.70–0.71). Conversely, patients >100 kg achieved significantly higher efficiency (0.81, 0.79–0.83), compared with those of 75–100 kg (0.76, 0.75–0.77) and lower weight (range of means: 0.70–0.72). QD exceeded QB in toddlers 5.4-fold (mean QD/QB ratio, 95% CI 5.2–5.6), and only 1.7 (1.70–1.76) in adults. Above 50 kg, mean QD/QB ratios were stable and independent of weight (range of mean ratios: 1.8–1.9; 90% interval: 1.3–2.3). Mean KoA/QB ratios varied between 3.0 and 3.4, highest values were observed in age group 6–11 years and weight group 25–50 kg. Normalized to weight, young children received ∼3–4 times higher QD/kg than adults. Weight-normalized QB/kg showed small differences over age, with significantly lower values (mean, 95% CI) in adolescents (6.1, 5.96–6.13 mL/min/kg) compared with adults (6.6, 6.53–6.70 mL/min/kg) and younger patients (e.g. 7.0, 6.7–7.2 mL/min/kg in 6- to 11-year-old patients). Larger differences were observed between weight groups, with lowest values in patients >100 kg (3.9, 3.7 4.1 mL/min/kg). More than 75% of patients >100 kg had QB/kg rates <5 mL/min/kg prescribed, with a median (IQR) of 3.5 (3.1–4.0) mL/min/kg. In the lowest weight groups QB/kg was almost double with 7.4 (6.5–9.1) mL/min/kg in patients <25 kg and 7.8 (6.8–8.9) mL/min/kg in patients 25–50 kg (i.e. >50% of patients exceeded flows of 7 mL/min/kg). The 95th percentile exceeded 10 mL/min/kg in children aged <10 years (Figure 3). Weight-normalized KoA used in infants and preschool children (range of means: 21.7–23.3 mL/min/kg) was only slightly higher than in adolescents and adults (19.4–19.3 mL/min/kg). Larger differences were observed over weight (mean, 95% CI), with almost 2-fold higher values in 25–50 kg patients (23.4, 23.1–23.6 mL/min/kg) compared with patients >100 kg (12.0, 11.4–12.6 mL/min/kg). Predicted CLD was similar to calculated KD in adults (mean, 95% CI: CLD = 4.7, 3.9–5.4 versus KD = 4.5, 3.9–5.2 mL/min/kg), but up to 45% lower in younger children (CLD = 3.1, 2.4–3.8 versus KD = 5.6, 5.0–6.2 mL/min/kg in 1–2 year olds). Similar to QB/kg, KD/kg (mean, 95% CI) was lower in adolescent patients (4.3, 4.2–4.3 mL/min/kg) compared with younger children (4.8, 4.7–5.0 mL/min/kg in 6–11 year olds) and adults (4.6, 4.5–4.6 mL/min/kg). KD/kg decreased over weight with lowest values of 3.1 (2.9–3.2 mL/min/kg) in patients >100 kg (Supplementary data, Figure S3). Observed HD prescriptions resulted in achieving target spKt/V of ≥1.4 (95% CI) (minimum ≥1.2) in 82% (80–83%) [96% (95–97%)] of young adults 19–29 years, but only 67% (65–69%) [88% (86–89%)] of adolescent patients aged 12–18 years, while estimated proportions in younger age groups (range: 59–85% [91–96%]) were not significantly different from adults. Target achievements were again lower in patients with high weight, with only 36% (31–42%) [77% (72–82%)] in patients weighing >100 kg. Interaction analysis indicated that weight >75 kg and adolescent age are independently associated with lower mean spKt/V (interaction terms non-significant with P > 0.05) as compared with other paediatric subpopulations. Sensitivity analyses Calculation of Cheek TBW resulted in physiologic plausible values across the entire age and weight range, including adults, while Watson TBW yielded estimates up to ≥1 L/kg in low-weight patients >16 years. Morgenstern estimated systematically lower TBW than Cheek (−13% in median), with the smallest difference around 10 years of age. This did, however, not affect the pattern of calculated KD distribution ranges. The pattern of QB/m2 and spKt/V ranges over BSA was similar to QB/kg and spKt/V ranges over weight (maximum ∼1.5 m2). Only a small percentage (4%) of patients had residual renal function (median: 1.8, IQR: 0.8–3.1 mL/min); neglecting CLR did not significantly influence calculated percentile curves. Comparing reported percentiles over weight (from median per year of age data) with percentiles from the sensitivity analysis dataset (individual median per weight band of 10 kg) did not suggest significant bias introduced by repeated measurements. DISCUSSION This study is the first to describe real-life HD prescription behaviour in a large cohort of children and young adults, having started dialysis at paediatric age. Distribution ranges comprising 50% (IQR) and 90% of actual HD prescriptions were derived for each year of age, and for weight bands of 10 kg, based on prescriptions of at least 20 patients (except ≤1 year/≤10 kg). Such data-driven distribution ranges may be useful to understand and fine-tune target prescription parameters of paediatric patients managed with standard HD. Upper limits are markedly higher than expected for a number of parameters, and given that ranges provided originate in the outpatient setting, it is assumed that these high values are well tolerated. Therefore, our results support the consideration of comfortably prescribing more intense HD treatments. Our results show that spKt/V ≥1.2 are routinely provided to young and lower weight patients, indicating that with respect to small solute clearance alone more intense HD regimens may not be necessary in this population. There seems, however, to be room for improvement in dialysis prescription and/or nutrition in all patients <19 years of age, and especially in children <12 years, given the large proportion of children with low weight and height for age. In addition, it has been suggested that spKt/V may underestimate dialysis dose in small patients [13]. In contrast, adolescent and young patients >75 kg are at risk of chronic undertreatment, especially given that dialysis dose may be overestimated in obese patients [14]. They may hence need more intense HD regimens (more frequent, prolonged) for best outcomes. This is particularly worrisome with recent reports on obesity in end-stage renal disease (ESRD) [15] recognizing limitations for kidney transplantation in this population, increasing their need for long-term HD. In adults, HD prescription is relatively standardized, compared with children [16], as the kidney disease outcomes quality initiative (KDOQI) guidelines recommend standard QB and QD prescription of ≥300 mL/min and ≥500 mL/min, respectively. Those minimum values were used indeed in almost all young adults, as shown by 5th percentiles exceeding those values in adults ≥19 years. QD rates of 300–800 mL/min were used in patients <10 years, consistent with recommended rates of 300–800 mL in paediatric patients [2], chosen independently from body size. QB rates are in contrast usually based on body weight (or BSA), with recommended rates of 5–7 mL/min/kg (150–200 mL/min/m2) [2], and consistent QB/kg values may thus be expected over weight and age (and BSA). QB/kg rates were lowest in adolescents (among age groups) and patients >100 kg (among weight groups), while highest rates were observed in the youngest and smallest patients, exceeding in >50% recommended rates of 5–7 mL/min/kg. This is in line with reported actual flow rates of ∼14 ± 6 mL/min/kg in 11 children [3], and not explained by use of the following equation [3] proposed for children: QB = (weight [kg] + 10) × 2.5, which would propose systematically lower QB than observed. In heavier patients median QB of 400 mL/min (90% interval: 300–500 mL/min) was probably considered maximized, resulting in QB/kg rates below 5 mL/min/kg in almost all paediatric patients >100 kg. The higher efficiency of dialytic clearance in toddlers, with KD closer to QB than in adults, may be partly explained by higher QD/QB ratios used. QD exceeded QB 4.7-fold in median (IQR 3.4–8.0) in this youngest age group, i.e. well exceeding the minimum factor of 1.5–2 recommended in children [17] and used in adults (IQR 1.5–1.9). Interestingly, in patients with high weight (≥100 kg) efficiency was not lower (IQR 1.5–2.0), and adolescents achieved slightly higher efficiency (1.7–2.3) than adults. This suggests that not only low efficiency but lower QB/kg in high-weight patients and adolescents may be the main limiting factor to achieve target dose (25th percentile <1.40 in adolescents and patients >90 kg). In adolescents, shorter dialysis duration and larger fraction of TBW (tendency observed in Figures 3 and 4, but not clearly in Table 1) may also contribute to lower delivered target spKt/V. Maximum spKt/V values were achieved in children >30–50 kg (median ≥1.70, Figure 4), while lowest values were seen in heavy patients. Also toddlers (1–2 years) had relatively low median spKt/V (Table 1) with only 59% achieving spKt/V of 1.4 (Table 3), possibly because of their higher fraction of TBW. However, a large proportion (91%) achieved minimal spKt/V >1.2, which may explain partly why no significant difference compared with older children could be demonstrated. Higher HD dose in paediatric (i.e. smaller) patients has been previously reported [18] and may be desired due to higher metabolic rate and nutritional requirements [19], while our data shows that spKt/V tends to decrease again in patients with weight decreasing below 25–30 kg. However, also a large proportion (96%) of patients <25 kg achieved minimal spKt/V of >1.2, compared with only ≤87% in patients >75 kg (Table 3). Following allometric principles (see illustration in Figure 5) [4, 19] a decrease of weight-normalized clearance and dialysis dose with weight could also be desired. Size-related differences in dialysis dose may, however, have implications for patient morbidity and mortality [18], and influence the extent of urea rebound and accuracy of dialysis dose calculations based on single-pool assumptions (with possible overestimation in paediatric patients [20–22] and underestimation in obese patients due to higher fat and lower body water fraction) [14]. Weight-dependent intensity of dialysis prescription may also have implications for drug dosing, since the amount of drugs removed during dialysis would be accordingly expected to be larger in low-weight patients (who receive higher spKt/V), which hence may require higher supplemental drug doses than proposed in adult dialysis patients [23]. FIGURE 5: View largeDownload slide Illustration of physiologically expected clearance change with weight based on allometric principles, using the mean clearance (KD) at a weight of 60 kg as reference (255 mL/min, corresponding to 4.25 mL/min/kg). Left: based on allometric principles, we would expect a linear relationship between the logarithm of weight and the logarithm of clearance, with a slope of 0.75 (solid line). Circles: actually observed KD. Right: weight-normalized clearance is accordingly expected to decrease with weight (solid line). In patients with increasing weight >60 kg, observed KD (circles) are, however, below this scaled ‘physiologic equivalent value’, while patients of ∼25–50 kg appear to receive more intense KD. Dashed lines: non-parametric regression line. FIGURE 5: View largeDownload slide Illustration of physiologically expected clearance change with weight based on allometric principles, using the mean clearance (KD) at a weight of 60 kg as reference (255 mL/min, corresponding to 4.25 mL/min/kg). Left: based on allometric principles, we would expect a linear relationship between the logarithm of weight and the logarithm of clearance, with a slope of 0.75 (solid line). Circles: actually observed KD. Right: weight-normalized clearance is accordingly expected to decrease with weight (solid line). In patients with increasing weight >60 kg, observed KD (circles) are, however, below this scaled ‘physiologic equivalent value’, while patients of ∼25–50 kg appear to receive more intense KD. Dashed lines: non-parametric regression line. This analysis has several limitations. Although a considerable number of paediatric patients was included given the low incidence of paediatric ESRD and dialysis dependency, patient numbers in the lower age and weight groups remain statistically small. This resulted in large 95% CI, and thus, our ability to detect significant differences between younger age groups was probably impeded. Given the retrospective nature of the study, we did not investigate the correlation of prescribed dialysis intensity with clinical outcomes beyond spKt/V. Haemodynamic tolerability (e.g. limiting filter size/KoA, QB and ultrafiltration rate), practical issues (e.g. patient comfort and accepted treatment duration to achieve target dry weight) and monitoring parameters other than spKt/V (e.g. normalized protein catabolic rate [24], interdialytic weight gain [25]), may be able to explain part of the interindividual variability in dialysis prescription, but were not investigated. We did not investigate additional measures of adequacy, such as fluid balance, middle-molecule clearance, mineral and bone disorders, growth and nutrition. Low prescription values may reflect an initial period of maintenance HD in which intensity is slowly up-titrated. The selection of thrice-weekly HD patients may have further excluded young patients, since for most children <10 kg more than three sessions per week may be needed to enable adapted nutrition [2]. Since we included only patients on conventional thrice-weekly in-centre HD, results may have limited utility for different dialysis schedules, such as home [26] or more frequent [27, 28] HD. Prescriptions may be different in countries, where fewer obese patients are treated, and change with transition from paediatric to adult dialysis units. Optimal dialysis prescription in children is still a subject of discussion. Given the growth retardation observed in our paediatric part of this cohort, despite good urea spKt/V delivered, the value of spKt/V as adequacy measure in children can be questioned. Observed age- and size-related differences in HD prescription and delivery may have implications not only for clinical outcomes, but also for urea and middle molecule kinetic assessment and modelling, as well as for drug removal and dose adjustment, deserving further investigations. Provided ranges of outpatient HD prescription can help clinicians and researchers in designing studies as well as personalizing and optimizing delivery of dialysis treatment. SUPPLEMENTARY DATA Supplementary data are available at ndt online. FUNDING This project has been supported by the Research Fund for Junior Researchers, University of Basel, Switzerland, and by the Eckenstein-Geigy Foundation, which is sponsoring research at Pediatric pharmacology at University Children’s Hospital Basel. AUTHORS’ CONTRIBUTIONS All authors equally contributed to research idea, study design, data analysis/interpretation. M.P. and O.M. were responsible for data acquisition, supervision and mentorship. V.G. performed statistical analysis. CONFLICT OF INTEREST STATEMENT M.P. is a consultant at Quantitative Solutions a Certara Company. V.G. and O.M. declare that they have no relevant financial interests. REFERENCES 1 National Kidney Foundation . KDOQI clinical practice guideline for hemodialysis adequacy: 2015 update . Am J Kidney Dis 2015 ; 66 : 884 – 930 CrossRef Search ADS PubMed 2 Fischbach M , Edefonti A , Schröder C et al. Hemodialysis in children: general practical guidelines . Pediatr Nephrol 2005 ; 20 : 1054 – 1066 Google Scholar CrossRef Search ADS PubMed 3 Cochat P , Lioux C. Maintenance hemodialysis during infancy. In: Warady BA , Schaefer FS , Fine RN , Alexander SR (eds). Pediatric dialysis. Dordrecht : Springer Science+Business Media , 2004 , p. 197 – 207 4 Bartelink IH , Rademaker CMA , Schobben AFAM et al. Guidelines on paediatric dosing on the basis of developmental physiology and pharmacokinetic considerations . Clin Pharmacokinet 2006 ; 45 : 1077 – 1097 Google Scholar CrossRef Search ADS PubMed 5 Hemodialysis Adequacy 2006 Work Group . Clinical practice guidelines for hemodialysis adequacy, update 2006 . Am J Kidney Dis 2006 ; 48 : S2 – S90 CrossRef Search ADS PubMed 6 NKF-DOQI clinical practice guidelines for hemodialysis adequacy . National Kidney Foundation . Am J Kidney Dis 1997 ; 30 (3 Suppl 2) : S15 – S66 7 Cheek DB , Mellits D , Elliott D. Body water, height, and weight during growth in normal children . Arch Pediatr Adolesc Med 1966 ; 112 : 312 – 317 Google Scholar CrossRef Search ADS 8 Mosteller RD. Simplified calculation of body-surface area . N Engl J Med 1987 ; 317 : 1098 Google Scholar PubMed 9 Michaels AS. Operating parameters and performance criteria for hemodialyzers and other membrane-separation devices . Trans Am Soc Artif Intern Organs 1966 ; 12 : 387 – 392 Google Scholar PubMed 10 Daugirdas JT. Second generation logarithmic estimates of single-pool variable volume Kt/V: an analysis of error . J Am Soc Nephrol 1993 ; 4 : 1205 – 1213 Google Scholar PubMed 11 Watson PE , Watson ID , Batt RD. Total body water volumes for adult males and females estimated from simple anthropometric measurements . Am J Clin Nutr 1980 ; 33 : 27 – 39 Google Scholar CrossRef Search ADS PubMed 12 Morgenstern BZ , Wühl E , Nair KS et al. Anthropometric prediction of total body water in children who are on pediatric peritoneal dialysis . J Am Soc Nephrol 2006 ; 17 : 285 – 293 Google Scholar CrossRef Search ADS PubMed 13 Spalding E , Chandna SM , Davenport A et al. Kt/V underestimates the hemodialysis dose in women and small men . Kidney Int 2008 ; 74 : 348 – 355 Google Scholar CrossRef Search ADS PubMed 14 Davenport A. Differences in prescribed Kt/V and delivered haemodialysis dose-why obesity makes a difference to survival for haemodialysis patients when using a “one size fits all” Kt/V target . Nephrol Dial Transplant 2013 ; 28 : iv219 – iv223 Google Scholar CrossRef Search ADS PubMed 15 Terrace JD , Oniscu GC. Paediatric obesity and renal transplantation: current challenges and solutions . Pediatr Nephrol 2016 ; 31 : 555 – 562 Google Scholar CrossRef Search ADS PubMed 16 Shroff R , Bonthius M , Fischbach M et al. SP865—Paediatric dialysis practice across the EU—a survey from the EPDWG/ERA-EDTA Registries . Nephrol Dial Transplant 2015 ; 30 : iii662 – iii662 Google Scholar CrossRef Search ADS 17 Müller D , Goldstein SL. Hemodialysis in children with end-stage renal disease . Nat Rev Nephrol 2011 ; 7 : 650 – 658 Google Scholar CrossRef Search ADS PubMed 18 Depner T , Daugirdas J , Greene T et al. Dialysis dose and the effect of gender and body size on outcome in the HEMO Study . Kidney Int 2004 ; 65 : 1386 – 1394 Google Scholar CrossRef Search ADS PubMed 19 Morton AR , Singer MA. The problem with Kt/V: dialysis dose should be normalized to metabolic rate not volume . Semin Dial 2007 ; 20 : 12 – 15 Google Scholar CrossRef Search ADS PubMed 20 Marsenić O , Pavlicić D , Bigović G et al. Effects of postdialysis urea rebound on the quantification of pediatric hemodialysis . Nephron 2000 ; 84 : 124 – 129 Google Scholar CrossRef Search ADS PubMed 21 Daugirdas JT , Greene T , Depner TA et al. Factors that affect postdialysis rebound in serum urea concentration, including the rate of dialysis: results from the HEMO Study . J Am Soc Nephrol 2004 ; 15 : 194 – 203 Google Scholar CrossRef Search ADS PubMed 22 Gotta V , Zhang L , Marsenic Couloures O et al. SP735 - body-weight dependency of urea kinetic parameters in adolescent hemodialysis patients . Nephrol Dial Transplant 2016 ; 31 : i339 – i340 Google Scholar CrossRef Search ADS 23 Gotta V , Dao K , Rodieux F et al. Guidance to develop individual dose recommendations for patients on chronic hemodialysis . Expert Rev Clin Pharmacol 2017 ; 10 : 737 – 752 Google Scholar CrossRef Search ADS PubMed 24 Marsenic O , Peco-Antić A , Jovanović O. Effect of dialysis dose on nutritional status of children on chronic hemodialysis . Nephron 2001 ; 88 : 273 – 275 Google Scholar CrossRef Search ADS PubMed 25 Fischbach M , Zaloszyc A , Shroff R. The interdialytic weight gain: a simple marker of left ventricular hypertrophy in children on chronic haemodialysis . Pediatr Nephrol 2015 ; 30 : 859 – 863 Google Scholar CrossRef Search ADS PubMed 26 Hothi DK , Stronach L , Harvey E. Home haemodialysis . Pediatr Nephrol 2013 ; 28 : 721 – 730 Google Scholar CrossRef Search ADS PubMed 27 Goldstein SL , Silverstein DM , Leung JC et al. Frequent hemodialysis with NxStageTM system in pediatric patients receiving maintenance hemodialysis . Pediatr Nephrol 2008 ; 23 : 129 – 135 Google Scholar CrossRef Search ADS PubMed 28 Laskin BL , Huang G , King E et al. Short, frequent, 5-days-per-week, in-center hemodialysis versus 3-days-per week treatment: a randomized crossover pilot trial through the Midwest Pediatric Nephrology Consortium . Pediatr Nephrol 2017 ; 32 : 1423 – 1432 Google Scholar CrossRef Search ADS PubMed © The Author(s) 2018. Published by Oxford University Press on behalf of ERA-EDTA. All rights reserved. This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/about_us/legal/notices)

Journal

Nephrology Dialysis TransplantationOxford University Press

Published: Apr 18, 2018

There are no references for this article.

You’re reading a free preview. Subscribe to read the entire article.


DeepDyve is your
personal research library

It’s your single place to instantly
discover and read the research
that matters to you.

Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.

All for just $49/month

Explore the DeepDyve Library

Search

Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly

Organize

Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.

Access

Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.

Your journals are on DeepDyve

Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.

All the latest content is available, no embargo periods.

See the journals in your area

DeepDyve

Freelancer

DeepDyve

Pro

Price

FREE

$49/month
$360/year

Save searches from
Google Scholar,
PubMed

Create lists to
organize your research

Export lists, citations

Read DeepDyve articles

Abstract access only

Unlimited access to over
18 million full-text articles

Print

20 pages / month

PDF Discount

20% off