| Study characteristics |
| Methods |
Study design
Study dates
|
| Participants |
Study characteristics
Setting: multicentre (2 clinical core consortiums headquartered at the Renal Research Institute in New York City and University of California San Francisco (later transferred to Stanford University; 10 clinical centres in the United States and Canada) Country: USA Inclusion criteria: patients undergoing HD 3 times/week; concomitant medical conditions; the ability to complete cardiac MRI; verbal communication ability in English or Spanish to permit completion of the quality of life interview; residential proximity to the dialysis units; achieved mean eKt/V at least 1.1 on at least two consecutive baseline sessions; patients aged 13 to 17 years were permitted Exclusion criteria: unable or unwilling to follow the study protocol for any reason (including mental incompetence); unable or unwilling to provide informed consent or sign the Institutional Review Board‐approved consent form; requires HD > 3 times/week owing to medical comorbidity; currently pregnant, or planning to become pregnant within the duration of follow‐up; currently on home HD; history of poor adherence to 3 times/week HD or PD; inability or unwillingness to come for in‐centre HD 6 days/week, including inability to arrange adequate transportation;expected geographic unavailability at a participating HD unit for 42 consecutive weeks or 44 weeks total during follow‐up; currently in an acute or chronic care hospital; contraindication to heparin, including allergy‐ or heparin‐induced thrombocytopenia; expectation that native kidneys will recover; currently on daily or nocturnal HD or < 3 months as the patient discontinued daily or nocturnal HD; < 3 months as patient returned to HD after kidney transplantation or use of an alternative dialysis modality (such as PD); current use of investigational drugs or participation in another clinical trial; scheduled for living donor kidney transplant, change to PD, or plans to relocate to a non‐study centre within the follow‐up period; life expectancy < 6 months; medical history that might limit the patient’s ability to take the trial treatments and complete the full duration of follow‐up (including currently receiving chemo or radiotherapy for a malignant neoplastic disease other than localized non‐melanoma skin cancer, active systemic infection (including tuberculosis, disseminated fungal infection, active AIDS), and cirrhosis with encephalopathy); medical conditions that would prevent the patient from receiving the cardiac MRI procedure (e.g. inability to remain still for the procedure, a metallic object in the body that is a contraindication to MRI such as cardiac pacemaker, cochlear implant, brain aneurysm clips, mechanical heart valves, recently placed artificial joints, and older vascular stents); inability to communicate verbally in English or Spanish; vascular access being used for HD is a non‐tunnelled catheter; residual kidney function was urea clearance 3 mL/min/35 L urea volume
Baseline characteristics
Number (analysed/randomised): intervention group (99/125); control group (84/120); however number of participants analysed varied based on the outcome (here is reported the lowest number of participants analysed, considering the outcomes of interest of the review) Mean age ± SD (years): intervention group (48.9 ± 13.6); control group (52.0 ± 14.1) Sex (M/F): intervention group (78/47); control group (73/47) Dialysis type: HD Median dialysis vintage. IQR (years): intervention group (3.85, 0.69 to 17.31)); control group (3.40, 0.58 to 12.94) -
Comorbidities
CVD: not reported Diabetes: intervention group (50/125); control group (50/120) Hypertension: intervention group (117/125); control group (111/120) Depression (clinician diagnosis): not reported
|
| Interventions |
Intervention classification
Intervention group
Control group
Co‐interventions
|
| Outcomes |
Outcomes reported
Fatigue outcome measures used: validation data available -
Change during 12 months in LV mass
Death: assessed until the end of treatment -
Change in 12 months of the self‐reported physical health
-
SF‐36: assessed at baseline, 4 and 12 months
RAND Physical Health Composite: assessed at baseline, 4 and 12 months
Short Physical Performance Battery (range from 1 to 12; higher values represent better physical function): assessed until 12 months HUI‐3 Multi‐attribute utility scale (range from 0 to 1; higher scores represent better health): assessed until 12 months Feeling Thermometer Scores (range from 0 to 100, with 100 representing best imaginable health state): assessed until 12 months Short Physical Performance Battery (range from 1 to 12; higher values represent better physical function): assessed until 12 months -
Depression
-
Cognitive function
-
Executive function
Trial Making B score (5‐minute limit) (ranges from 0‐300 seconds; less time represents better executive control and less cognitive impairment): assessed at baseline, 4 and 12 months -
Attention
Digit Symbol Substitution Test: assessed until 12 months Trail‐Making Test, Form A: assessed until 12 months
-
Psychomotor speed
-
Memory
Rey Auditory Verbal Learning Test, immediate and delayed recall: assessed until 12 months Letter‐Number Sequencing: assessed until 12 months
-
Verbal fluency
Sleep and hour slept each night Sleep Problems Index (ranges from 0 to 100; higher values represent more problems): assessed at baseline, 4 and 12 months Caregiver burden Cousineau Caregiver Burden (ranges from 0 to 100; higher scores represent greater anxiety. Scores were calculated for only participants with unpaid caregivers): assessed at baseline, 4 and 12 months Laboratory results (pre‐dialysis SCr, phosphate, urea nitrogen, albumin, interdialytic weight gain, extracellular fluid load, normalized protein catabolic rate, BMI, lean body mass by single frequency bioimpedance analysis, calcium, calcium‐phosphate product, PTH, pre‐dialysis Hb, iron, transferrin, ferritin): assessed at baseline 4 and 12 months Dialysis outcomes (target dry weight, prescribed treatment time): assessed at 12 months Safety (vascular access complication, iron losses, metabolic complication): assessed at 12 months Weekly average BP: assessed until 12 months Weekly average pre‐dialysis pulse pressure: assessed until 12 months Proportion of patients with weekly average pre‐dialysis SBP < 110 mm Hg: until 12 months Number of prescribed antihypertensive agents: assessed until 12 months Eritropoiesys: assessed at 12 months Hospitalisation, cardiovascular hospitalisation and total hospital days: assessed at 12 months Cost‐effectiveness: assessed at 12 months End‐diastolic, end‐systolic, and stroke volumes; ejection fraction; cardiac output: assessed until 12 months Heart rate variability measures: assessed until 12 months Rate of intradialytic hypotension episodes: assessed until 12 months Phosphate binder dose, Vitamin D analogue dose: assessed until 12 months Hypertension: assessed until 12 months
|
| Notes |
Additional information
Funding: NIDDK, CMS, National Institutes of Health (NIH) Research Foundation, Fresenius Medical Care, Renal Research Institute, and Satellite Health Care. These trials were supported by NIDDK grantsU01DK066597 (Data Coordinating Center), 2U01DK066579 (Dr Levin), 3U01DK066481 (Dr Chertow), and 3U01DK066480 (Dr Rocco) Conflicts of interest/disclosures: none Trial registration identification number: NCT00264758 A priori published protocol: accessible at https://clinicalresearch.ccf.org/fhn/index.html
|
| Risk of bias |
| Bias |
Authors' judgement |
Support for judgement |
| Random sequence generation (selection bias) |
Low risk |
Quote from Suri 2007: "Eligible subjects are then randomly assigned 1:1 to the frequent haemodialysis intervention or control arms, by a central, web‐based program. Randomization is stratified by clinical centre and diabetic status, using permuted blocks."
Comment: A web‐based program is considered as low risk of bias. No imbalance between intervention groups was apparent |
| Allocation concealment (selection bias) |
Unclear risk |
Quote from FHN Trial Group 2010: "Randomization was stratified according to clinical centre and diabetes status, with the use of randomly permuted blocks. Although treatment assignments could not be concealed, between group comparisons of the outcomes were concealed from the investigators throughout the course of the trial." |
| Blinding of participants and personnel (performance bias)
All outcomes |
High risk |
Quote from Kuella 2013: "Unblinded intervention."
Comment: An open‐label study is considered as high risk of bias. Possible deviations from the intended intervention that arose from the trial context were not reported |
| Blinding of outcome assessment (detection bias)
All outcomes |
Low risk |
Quote from Suri 2013: "A vascular access outcomes committee blinded to group allocation reviewed all access events to determine whether the event met the definition of repair or loss. [...] An independent outcomes committee blinded to group allocation reviewed these forms, discharge summaries, and supplementary chart information to determine whether each death or hospitalisation was access related or non–access related."
Quote from Suri 2014: "Patients also completed several questionnaires that were centrally administered by telephone before randomisation and 4 (F4) and 12 months (F12) after randomisation."
Quote from Ornt 2013: "An independent data and Safety Monitoring Board reviewed safety data and interim results."
Comment: An independent Data Safety Monitoring Board assessed the outcomes.The outcomes were assessed with an appropriate measure, without differences between groups. However, subjective measures were used, it was stated that outcomes were assessed without knowledge of treatment allocation, and knowledge of treatment assignment may not have influenced reporting. Participant beliefs about the superiority/inferiority of either intervention could have influenced their assessment of the outcome, but there was no evidence that this was likely |
| Incomplete outcome data (attrition bias)
All outcomes |
High risk |
Overall, attrition seemed to be > 5% of loss to follow‐up with some difference between groups. Attrition was not reported in sufficient detail to permit judgment in the first phase of the study. Tamura 2010 reported that 239 participants were randomised but there were no data on the missing participants |
| Selective reporting (reporting bias) |
Low risk |
Protocol was published. Fatigue was reported (Unruh 2013) in accordance with a pre‐specified analysis plan, using multiple eligible outcome measurements (scales, time points). Fatigue was reported in a format that was not extractable for meta‐analysis. All outcomes that should be addressed (fatigue, cardiovascular disease, and death) were reported |
| Other bias |
Low risk |
There was no evidence of different baseline characteristics, or different non‐randomised co‐interventions between groups. Founding were unlikely to influence the data analysis and authors had no conflicts of interest |
