ULtrafiltration-Rate Induced CArdiac Strain (ULRICA) - Study

Research plan

ULtrafiltration-Rate Induced CArdiac strain (ULRICA) - study

Survey of the field:

Patients on chronic haemodialysis suffer from an increased risk for morbidity and mortality with the main cause of death being cardiovascular disease. Besides retention of uremic solutes, dialysis patients with poor urine output suffer from water retention. Excessive water retention results in inter dialytic weight gain (IDWG). IDWG is one prominent risk factor for cardiovascular events.

With time, the effects of uraemia and dialysis treatment itself cause a cardiac strain leading to an impact on cardiac function. In the course of chronic haemodialysis, a majority of the patients develop heart failure with preserved ejection fraction. Biomarkers for such clinical involvement are i.e., B-type natriuretic peptide (BNP), and Troponin.

BNP is one of two natriuretic peptides synthesised in the heart. It is released as a response to stretching of the myocardial wall. BNP is produced as a 108 amino acid long peptide. It is cleaved into two parts, the active 32 amino acid BNP with a molecular weight of 3.5 kDa and the inactive NT-proBNP with a molecular weight of 8.5 kDa.

The main effect of BNP in healthy subjects is vasodilatation, natriuresis and diuresis. In addition to their roles in water and electrolyte balance, natriuretic peptides participate in seemingly unrelated processes, such as immune response and lipid metabolism. BNP is cleared by receptor mediated endocytosis and lysosomal degradation and by neutral endopeptidases present in renal tubular cells and cells in the vascular walls. NT-proBNP is thought to be excreted by the kidney and has a half-life of approximately 120 minutes compared to 20 minutes for BNP. The different half-lives and clearance mechanisms explain why the levels in serum of patients are much higher for NT-proBNP than BNP although it is excreted in a 1:1 ratio.

Patients with chronic kidney disease (CKD) and especially those on dialysis have increased levels of natriuretic peptides. Both BNP and proBNP have been shown to correlate with increased risk of mortality. In patients with CKD, BNP and proBNP levels have been shown to correlate well with echocardiographic findings such as left ventricular filling pressure, left atrium size and left ventricular mass index, but not with left ventricular ejection fraction. Other studies found no such correlations. The elevated levels of proBNP and BNP in patients on dialysis have been explained not only by cardiac dysfunction/hypertrophy but also with volume overload.

To perform haemodialysis, the investigator is dependent on a functioning vascular access. The first choice has for a long time been to create an arteriovenous fistula between A. radialis and V. cephalica in the forearm. According to data from DOPPS (Dialysis Outcomes and Practice Pattern Study) the use of upper arm fistulas is increasing. The increased use of upper arm fistulas is also seen in the Swedish register. The advantage of creating an upper arm fistula comes mainly from the surgical point of view. However, there is a risk of creating a high-output fistula where a large proportion of arterial blood is shunted from the left-sided circulation to the right-sided circulation. The increase in preload can lead to increased cardiac output that over time may lead to cardiac hypertrophy and eventually heart failure. A recent study showed marked changes in left ventricular mass six weeks after fistula creation, especially in those patients with fistula flow >600 ml/min.

Purpose and aims:

Our aim is to study the effects of haemodialysis on cardiac biomarkers and see if negative effects that haemodialysis can have on the heart can be reduced by changes in haemodialysis prescription.

Specifically, in this study, the investigator investigates if cardiac strain caused by dialysis is reduced by lowering the ultrafiltration rate.

Method:

Study design: prospective multicentre study, cross-over design.

Study population: Prevalent hemodialysis (HD) patients. Study sites are dialysis centres at Umeå, Skellefteå, Östersund, Uppsala, Linköping Skövde, Stockholm (Diaverum), Stockholm (Danderyd), Falköping (Diaverum), Malmö, Jönköping, and Örebro.

Inclusion criteria: Adult HD patients with an intra-dialytic weight gain (IDWG) ≥2.5% of target weight over the long interval (3 days)26 during the four weeks prior to inclusion. Patients need to be able to understand the study information and be able to give consent.

Exclusion criteria: Active systemic inflammatory state such as extensive malignancy or acute septic infection. Uncontrolled hyperphosphatemia decided by the local routine of the physician (due to somewhat decreased efficacy of the LF filter to eliminate phosphate)

Study protocol: For each patient, two dialysis sessions are performed in two consecutive weeks. Blood samples, blood pressure and heart rate are taken before HD, at 180 min and after HD. For the dialysis sessions low-flux dialyzers are used. Dialysis 1 is done at usual ultrafiltration rate (UR). Dialysis 2 is done at a reduced or increased UR. Dialysis time is calculated as follows: [hours]=(100 x ultrafiltration need [L])/(0.63 x target weight [kg])28 Treatment with antihypertensives, sodium concentration in dialysate, type of dialysate and target weight should remain unchanged between both dialysis sessions. Such as: If standard dialysis is performed with an increased UR a reduced UR is performed for Dialysis 2; If standard dialysis is performed with a reduced UR Dialysis 2 can be done with an increased UR (while dialysis time can be maintained, if preferred). Blood samples are taken after finalizing of the UR period.

Outcome parameters: Difference in pre- and post-dialytic serum troponin and pro-BNP concentrations drawn at the end of the UF-period.

Secondary outcomes: Adverse effects related to dialysis (e.g. art. hypotension, muscle cramps), recovery time, filter clotting.

Variables: Age, sex, access type (incl. fistula flow), diurnal diuresis, target weight, pre- and post-dialysis weight, height, IDWG, ultrafiltration volume, dialysate flow, blood flow, dialysis time, anticoagulation and blood pressure, heart rate, blood count, urea, CRP, treatment with RAAS-blockade, antihypertensive agents, comorbidities, time on dialysis, ordinary HD modality and dialysis schedule.

When available: QRS-time in last ECG, prevalence of atrial fibrillation, volume state in body-composition monitoring.

Laboratory samples are drawn predialysis, at 180 minutes and at the end of the ultrafiltration (UR) period.

Correction for fluid shift in blood or plasma samples: Results of large molecules, cells and platelets are adjusted for the change in blood volume caused by ultrafiltration according to Schneditz et al 2012.

Protection of data privacy: Coding of patient identity with local lists. Analysis and publication of results in anonymized, aggregated form.

Power analysis: Assuming an effect size of 0.6, the study would require a sample size of 40 pairs (two consecutive dialysis sessions in a patient) to achieve a power of 95% at a level of significance of 5% (two sided).29

Statistical analysis: paired sampled T-test for the difference of the means.