Spanish Study of Hyperkalaemia Incidence and Prognosis in Patients With Heart Failure and Reduced Ejection Fraction

PARTICIPANT CENTRES AND RESEARCHERS: 20 National Public Health System hospitals. Cardiology specialist physicians

MAIN GOAL: To estimate prevalence and, in medium term basis (12 months), incidence of hyperkalaemia in heart failure (HF) outpatients with reduced ejection fraction (REF) and its relationship with non-optimal HF therapy and clinical outcomes (mortality and hospital admission).

SPECIFIC GOALS:

1. -Estimating Hyperkalaemia prevalence in these patients.
2. -Assessing the role of hyperkalaemia with the use of non-optimal therapy doses.
3. -Estimating 12 months hyperkalaemia incidence on these patients and describing the severity of the episodes.
4. -Estimating incidence of HF hospital admission and/or death in these patients on a medium term basis (12 months) and its association with existence of hyperkalaemia during follow up.
5. -Assessing hyperkalaemia risk during follow-up in relation with the existence of diabetes and/or renal failure (RF) at baseline visit.

STUDY DESIGN National multicentric prospective observational study that includes 12 months follow-up of consecutive cases of HF outpatients with REF. Inclusion baseline visit and follow -up visit at 12 months will be scheduled for collecting clinical and blood sample data of patients.

The study will be performed in ordinary conditions of clinical practice; no additional procedures or interventions will be performed. Patients will be selected in cardiology outpatient clinics of 20 Spanish centres, including the first 30 patients meeting inclusion and exclusion criteria. All patients will undergo the common studies according to usual clinical practice and at least both, potassium and renal creatinine clearance values will be collected.

To respond to the main objective, prevalence of hyperkalaemia at baseline will be determined as well as the appearance of new hyperkalaemia cases in the follow up of patients that had normal potassium levels at baseline visit; their relationship with outcomes of interest will also be determined

STUDY POPULATION:

The expected number of patients recruited in 20 Spanish research centres is 600.

CONSIDERATIONS ABOUT SIMPLE SIZE:

Assuming a 10% risk of outcomes (mortality or HF hospital admission) among the exposed (hyperkalaemia), 2.5% among the non-exposed and a 7:1 non-exposed/exposed ratio, 73 exposed and 511 non-exposed patients should be included to have a 80% power to detect these differences with a 95% confidence level. The recruitment goal is set at 600 patients.

This sample size allows a precision of ±2,4%, assuming a prevalence of hyperkalaemia at baseline of 10% and with a 95% confidence level. Furthermore, assuming 5% hyperkalaemia cumulative incidence through follow-up among hyperkalaemia-free patients at baseline, the proposed sample size, allows a precision of±1,8% with a 95% confidence level.

SELECTION OF CENTRES The number of participant centres throughout Spain will be 20 and a cardiologist will be the principal investigator at every centre. The centres have been selected either among those with excellence-certified HF units, included in CiberCV Consortium or with good performance in previous similar registries; this selection is not random, but based on interest and high performance criteria in the field of HF

VARIABLES AND CRF:

The main outcome variables are:

1. - Hyperkalaemia. Serum Potassium (K+) will be measured in baseline and follow-up visits, but also if there is any intermediate hospital admission. All available potassium measurements during follow-up will be reviewed and values of hyperkalaemia identified. Hyperkalaemia will be considered with (K+) Values > 5,4 mEq/L
2. - Proportion of patients using drugs with proven efficacy for HF with REF -but also linked with hyperkalaemia- (ACEIs/ARB-II/ARNI and MRAs) and proportion of patients with use of optimal doses (based on current ESC heart failure guidelines2) of the same drugs; reason for not using them (or not receiving optimal doses) will be registered. Proportion of patients in the following categories will also be computed:

• No ACEi/ARB/ARNI or at less than target dose and no MRA
• ACEi/ARB/ARNI at target dose and no MRA
• ACEi/ARB/ARNI at target dose and MRA at less than target dose
• ACEi/ARB/ARNI at target dose and MRA at target dose

1. - Hospital admissions during follow-up and main cause for them. Outcomes considered will include HF admissions caused by hyperkalaemia and related events. Outcome Serum (K+) values will be also registered.
2. - Mortality and cause. HF (and hyperkalaemia) hospital admissions and mortality will be independently considered as outcomes; joint outcome incidence of both will be also calculated.

STUDY DEVELOPMENT:

Every participating centre will include from inclusion date the first 30 heart failure outpatients who fulfil the inclusion criteria and none of the exclusion ones. The excluded patients will be also registered, pointing out the reason of exclusion. A follow-up and final visit will be scheduled after 12 months for collecting all the variables the study requires. Hospitalisation data since inclusion visit will be also collected, together with the information of blood sample test performed in such period for detecting hyperkalaemia episodes. Mortality data will include date and cause, where possible. If follow-up visit cannot be conducted, the reason for it will be registered. The envisaged length of the study is 24 months since approval date.

DATA COLLECTION:

Consecutively, every researcher will handle and explain the information sheet to the patients that might fulfil de inclusion criteria (and do not present any exclusion criteria) and will ask them to sign the informed consent. Researchers must warrant the accuracy and completion of the data collected for the study. Data registered at CRF should be consistent with source documents used for their collection.

DATA MANAGEMENT:

Data will be collected during initial and follow-up visits and they will be integrated into a unique data base of the web platform. Researchers are responsible of the information included in the database and will access by means of personal login and password. The online platform will include ranges and rules to minimize errors in data registering.

PLAN FOR DATA ANALYSIS Hyperkalaemia prevalence at basal visit and the rate of patients who do not receive and/or reach optimal doses of drugs of interest (ACIEs, ARB-II, MRAs) will be estimated and their corresponding 95% confidence intervals computed. The reason for not receiving and/or reaching optimal doses of drugs will be described, especially if it is due to hyperkalaemia.

For those patients who do not present hyperkalaemia at baseline, 12 month hyperkalaemia cumulative incidence (95% confidence interval) will be estimated; also it will be estimated the proportion of patients who have to modify their therapy due to hyperkalaemia.

The number of episodes of hyperkalaemia during follow-up in relation to the number of patients will be estimated and their severity and the therapy changes that induced described. Also 12 months cumulative incidence of clinical outcomes will be estimated: hospitalisation due to HF or to, hyperkalaemia (and related events) and mortality, considered as individual and composite outcomes. The association between hyperkalaemia and occurrence of outcomes will be analyses and also adjusted using potential confounding factors by means of logistic regression models

The quantitative variables will be generally described using either the mean and the standard deviation or the median and interquartile range according them following or not a normal distribution. When comparing groups, t-student test will be used for continuous variables and chi-share test for qualitative ones.

Both, intermediate and final analysis, are planned to be carried out after baseline-visit closure and at the end of follow up (12 months).