Tomislav Cikara
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Tomislav Cikara has a total of 10 co-authors. Here are the top co-authors by number of shared publications:
- Marko Lucijanic
Department of Hematology, University Hospital Dubrava, 10000 Zagreb, Croatia
2 shared publications - Irzal Hadžibegović
Department of Cardiovascular Diseases, University Hospital Dubrava, 10000 Zagreb, Croatia
1 shared publication - Ivan Papic
Pharmacy Department, University Hospital Dubrava, 10000 Zagreb, Croatia
1 shared publication - Ivana Jurin
Department of Cardiovascular Diseases, University Hospital Dubrava, Zagreb, Croatia
1 shared publication - Maja Ortner Hadžiabdić
Faculty of Pharmacy and Biochemistry, University of Zagreb, 10000 Zagreb, Croatia
1 shared publication - Marin Pavlov
Department of Cardiovascular Diseases, University Hospital Dubrava, 10000 Zagreb, Croatia
1 shared publication
Latest Publications
- Open Access
- 13 July 2026
- Published by MDPI AG in Journal of Clinical Medicine
- Vol. 15 (14) , 5465
Hemoglobin Trajectories on SGLT2 Inhibitor Therapy in Heart Failure: Anemia Marks Adverse Prognosis While Erythrocytosis Is Transient and Not Associated with Adverse Outcomes
Background/Objectives: Sodium-glucose cotransporter 2 inhibitors (SGLT2is) raise hemoglobin in patients with heart failure (HF). We characterized the prevalence, dynamics, predictors and prognostic implications of three hemoglobin states (anemia, normal, and erythrocytosis) at baseline and during the first year on SGLT2i therapy. Methods: Prospective single-center registry of 1244 HF patients with a baseline hemoglobin who initiated dapagliflozin or empagliflozin (March 2022–October 2025). Hemoglobin was categorized at baseline, 6 and 12 months using WHO sex-specific thresholds. We analyzed inter-category transitions, predictors of each hemoglobin category at each timepoint, time-to-event outcomes (all-cause death, MACE, thrombosis) in landmark analyses, and the correlation of hemoglobin change with concurrent changes in left ventricular ejection fraction and NT-proBNP. Results: At baseline, 23.6% had anemia, 71.1% were normal, and 5.3% had erythrocytosis. The prevalence of erythrocytosis remained low throughout, with a modest, transient rise to 8.0% at 6 months before returning toward baseline (5.9% at 12 months), even though the underlying composition of patients was highly dynamic. Among baseline-normal patients, downward transition to anemia (7.8% at 6 months, 9.2% at 12 months) occurred at rates equal to or greater than upward transition to erythrocytosis (7.7% and 6.0%, respectively). The prevalence of anemia decreased significantly between baseline and the follow-up timepoints. Across timepoints, the patient’s own baseline hemoglobin was the dominant independent predictor of both subsequent erythrocytosis and subsequent anemia. Other independent predictors of erythrocytosis were higher adherence to SGLT2i, male sex, current smoking, COPD or asthma, higher heart rate and lower serum albumin. Other independent predictors of anemia were non-adherence to or discontinuation of SGLT2i, peripheral arterial disease, advanced age, diabetes, lower eGFR, higher NT-proBNP and lower serum albumin. Increases in hemoglobin from baseline were associated with greater concurrent reductions in NT-proBNP at 6 and 12 months but were not associated with changes in LVEF at either timepoint. Anemia was independently associated with all-cause death at the baseline and 6-month landmarks (adjusted HRs are 2.23 and 2.45, respectively; time-varying HR is 2.55), with a consistent but non-significant point estimate at 12 months (aHR of 1.90), and with significant 6-month MACE estimate (aHR of 3.07). Erythrocytosis was not associated with mortality, MACE or thrombosis at any timepoint, although the small number of patients and thrombotic events limits interpretation. Dapagliflozin and empagliflozin were associated with indistinguishable hemoglobin trajectories. Conclusions: On SGLT2i therapy, anemia carries the dominant prognostic signal and identifies a high-risk subgroup; erythrocytosis is uncommon, transient and showed no detectable association with adverse outcomes, though thrombotic events were too few for firm safety conclusions. Routine monitoring of hemoglobin for anemia is of higher clinical importance compared to erythrocytosis; these single-center findings require external validation.
- 13 July 2026
- Published by MDPI AG in Journal of Clinical Medicine
- Vol. 15 (14) , 5465
Open AccessHemoglobin Trajectories on SGLT2 Inhibitor Therapy in Heart Failure: Anemia Marks Adverse Prognosis While Erythrocytosis Is Transient and Not Associated with Adverse Outcomes
Background/Objectives: Sodium-glucose cotransporter 2 inhibitors (SGLT2is) raise hemoglobin in patients with heart failure (HF). We characterized the prevalence, dynamics, predictors and prognostic implications of three hemoglobin states (anemia, normal, and erythrocytosis) at baseline and during the first year on SGLT2i therapy. Methods: Prospective single-center registry of 1244 HF patients with a baseline hemoglobin who initiated dapagliflozin or empagliflozin (March 2022–October 2025). Hemoglobin was categorized at baseline, 6 and 12 months using WHO sex-specific thresholds. We analyzed inter-category transitions, predictors of each hemoglobin category at each timepoint, time-to-event outcomes (all-cause death, MACE, thrombosis) in landmark analyses, and the correlation of hemoglobin change with concurrent changes in left ventricular ejection fraction and NT-proBNP. Results: At baseline, 23.6% had anemia, 71.1% were normal, and 5.3% had erythrocytosis. The prevalence of erythrocytosis remained low throughout, with a modest, transient rise to 8.0% at 6 months before returning toward baseline (5.9% at 12 months), even though the underlying composition of patients was highly dynamic. Among baseline-normal patients, downward transition to anemia (7.8% at 6 months, 9.2% at 12 months) occurred at rates equal to or greater than upward transition to erythrocytosis (7.7% and 6.0%, respectively). The prevalence of anemia decreased significantly between baseline and the follow-up timepoints. Across timepoints, the patient’s own baseline hemoglobin was the dominant independent predictor of both subsequent erythrocytosis and subsequent anemia. Other independent predictors of erythrocytosis were higher adherence to SGLT2i, male sex, current smoking, COPD or asthma, higher heart rate and lower serum albumin. Other independent predictors of anemia were non-adherence to or discontinuation of SGLT2i, peripheral arterial disease, advanced age, diabetes, lower eGFR, higher NT-proBNP and lower serum albumin. Increases in hemoglobin from baseline were associated with greater concurrent reductions in NT-proBNP at 6 and 12 months but were not associated with changes in LVEF at either timepoint. Anemia was independently associated with all-cause death at the baseline and 6-month landmarks (adjusted HRs are 2.23 and 2.45, respectively; time-varying HR is 2.55), with a consistent but non-significant point estimate at 12 months (aHR of 1.90), and with significant 6-month MACE estimate (aHR of 3.07). Erythrocytosis was not associated with mortality, MACE or thrombosis at any timepoint, although the small number of patients and thrombotic events limits interpretation. Dapagliflozin and empagliflozin were associated with indistinguishable hemoglobin trajectories. Conclusions: On SGLT2i therapy, anemia carries the dominant prognostic signal and identifies a high-risk subgroup; erythrocytosis is uncommon, transient and showed no detectable association with adverse outcomes, though thrombotic events were too few for firm safety conclusions. Routine monitoring of hemoglobin for anemia is of higher clinical importance compared to erythrocytosis; these single-center findings require external validation.
- Open Access
Corticosteroid Dosing Level, Incidence and Profile of Bacterial Blood Stream Infections in Hospitalized COVID-19 Patients
COVID-19 patients with severe or critical symptoms are often treated with corticosteroids, per contemporary guidelines. Due to their immunosuppressive and immunomodulatory properties, corticosteroids are associated with the development of superinfections. We aimed to retrospectively assess patterns of corticosteroid use and the profiles of bacterial blood stream infections associated with exposure to different dosing levels, in a cohort of 1558 real-life adult COVID-19 patients. A total of 1391 (89.3%) patients were treated with corticosteroids, with 710 (45.6%) patients receiving low, 539 (34.6%) high and 142 (9.1%) very high corticosteroid doses. Bacteremia developed in a total of 178 (11.4%) patients. The risk of bacteremia was of similar magnitude between the no and low-dose corticosteroid treatments (p = 0.352), whereas it progressively increased with high (OR 6.18, 95% CI (2.66–14.38), p < 0.001) and very high corticosteroid doses (OR 8.12, 95% CI (3.29–20.05), p < 0.001), compared to no corticosteroid treatment. These associations persisted after multivariate adjustments and were present independently of sex, comorbidity burden, and mechanical ventilation. The profiles of individual bacterial pathogens differed depending on the used corticosteroid doses. High and very high corticosteroid doses are frequently used for real-life COVID-19 patients with severe and critical clinical presentations and are associated with a higher risk of bacteremia independently of sex, comorbidity burden, and mechanical ventilation use.
Open AccessCorticosteroid Dosing Level, Incidence and Profile of Bacterial Blood Stream Infections in Hospitalized COVID-19 Patients
COVID-19 patients with severe or critical symptoms are often treated with corticosteroids, per contemporary guidelines. Due to their immunosuppressive and immunomodulatory properties, corticosteroids are associated with the development of superinfections. We aimed to retrospectively assess patterns of corticosteroid use and the profiles of bacterial blood stream infections associated with exposure to different dosing levels, in a cohort of 1558 real-life adult COVID-19 patients. A total of 1391 (89.3%) patients were treated with corticosteroids, with 710 (45.6%) patients receiving low, 539 (34.6%) high and 142 (9.1%) very high corticosteroid doses. Bacteremia developed in a total of 178 (11.4%) patients. The risk of bacteremia was of similar magnitude between the no and low-dose corticosteroid treatments (p = 0.352), whereas it progressively increased with high (OR 6.18, 95% CI (2.66–14.38), p < 0.001) and very high corticosteroid doses (OR 8.12, 95% CI (3.29–20.05), p < 0.001), compared to no corticosteroid treatment. These associations persisted after multivariate adjustments and were present independently of sex, comorbidity burden, and mechanical ventilation. The profiles of individual bacterial pathogens differed depending on the used corticosteroid doses. High and very high corticosteroid doses are frequently used for real-life COVID-19 patients with severe and critical clinical presentations and are associated with a higher risk of bacteremia independently of sex, comorbidity burden, and mechanical ventilation use.