Objectives: Chronic heart failure may lead to worsening renal function. Cardiorenal syndrome (CRS) describes the interdependency of cardiac and renal function. The prognostic capabilities of both urinary biomarkers NT-proBNP and N-acetyl-ß-D-glucosaminidase (NAG) regarding mortality and chronic kidney disease (CKD) progression in a collective of ICD-patients were shown recently. Aim of the study was to assess the risk of poor outcome depending on biomarker levels and left ventricular ejection fraction (LVEF) in a mixed cohort of ICD patients.

Methods: 415 ICD patients were included in this study. Fresh spot urine and blood samples were used to assess NAG and urinary as well as plasmatic NT-proBNP. Urinary biomarkers were normalized to urinary creatinine. Follow-up was performed after 42 months. LVEF was estimated by echocardiography according to Simpsons’ method. Outcomes of interest were mortality, adequate ICD-shock-therapies and continuous CKD progression. We defined four subgroups regarding marker levels and LVEF. Subgroup 1: marker level > median combined with LVEF ≤ 40%. Subgroup 2: marker level > median combined with LVEF > 40%. Subgroup 3: marker level < median combined with LVEF > 40%. Subgroup 4: marker level < median combined with LVEF ≤ 40%.

Results: 177 patients showed a LVEF ≤ 40%. These patients had a significant higher rate of average age, male sex, primary prevention ICD indication and Diabetes compared to patients with an LVEF > 40%. Further patients with LVEF ≤ 40% had significant higher levels of NAG, urinary and plasmatic NT-proBNP as well as significant lower levels of eGFR compared to patients with LVEF > 40% (each p < 0.05). There was no difference regarding hypertension in these patients (p = n.s.). During follow-up 75 patients deceased, 56 patients received adequate ICD-shock-therapies and 27 patients suffered from continuous CKD progression.

In Kaplan-Meier-Analysis regarding NAG as well as urinary NT-proBNP as well as plasmatic NT-proBNP subgroup 1 showed a significant higher rate of mortality compared to all other subgroups respectively (each p < 0.05).  Further, in Kaplan-Meier-Analysis regarding NAG as well as plasmatic NT-proBNP subgroup 1 showed a significant higher rate of continuous CKD progression compared to all other subgroups (each p < 0.05). In Kaplan-Meier-Analysis regarding urinary NT-proBNP subgroup 1 showed a significant higher rate of continuous CKD progression compared to subgroup 3 and 4 (each p < 0.05), but there was no difference to subgroup 2 (p = n.s.).

There was a trend to higher AUCs of plasmatic NT-proBNP compared to urinary NT-proBNP without significant difference regarding prediction of mortality as well as continuous CKD progression in patients with LVEF ≤ 40% according to ROC analysis (each p = n.s.).

Regarding adequate ICD-shock-therapies there was a trend but no significant association between combination of higher markers levels as well as reduced LVEF and prediction of poor outcome in this cohort (each p = n.s.).

Conclusion: Patients with increased levels of NAG, urinary or plasmatic NT-proBNP and reduced LVEF represent highest risk-patients for poor outcome in a mixed cohort of ICD patients in the context of an advanced CRS. Particularly, the prediction regarding mortality and continuous CKD progression worsens essential if high levels of these biomarkers or reduced LVEF are not present.