A novel composite biomarker score for the identification of cognitive impairment in patients with heart failure: a pilot study

Figures

• 

  Figure 1.. Cognitive performance appears to decrease with worsened HF status. Cognitive values for heart failure (HF) participants were stratified by N-terminal pro-B-type natriuretic peptide (NT-proBNP) levels and compared to healthy control participants (A) cued name retrieval (CRN), (B) cued object retrieval (CRO), (C) 30-minute delay cued name retrieval (CRN30), (D) 30-minute delay cued occupation retrieval (CRO30), (E) Keep Track, and (F) Verbal Paired associated (VPA). CRN, CRO, CRN30, and CRO30 are all tasks part of FNAME. (G) A composite score (Actual Composite cognitive Score) was made using z-transformed values of CRN, CRO, CRN30, CRO30, keep track, and VPA. Groups were tested for normality via d’Agostino & Pearson test or Shapiro-Wilk test if n was too low. Multiple variables meeting normality were tested for significance via ordinary one-way ANOVA, and comparisons between groups were tested for significance Tukey’s multiple comparisons test. If values do not meet normality, multiple variables were tested for significance via the Kruskal-Wallis test, and comparison between groups was tested for significance through Dunn’s multiple comparisons test. p value indicated by *. p =…

• 

  Figure 2.. VCID neurodegenerative biomarkers appear to increase with worsened HF status, but not pTau217. Serum neurodegenerative values for heart failure (HF) participants were stratified by N-terminal pro-B-type natriuretic peptide (NT-proBNP) levels and compared to healthy control participants (A) neurofilament light protein (NfL), and (B) phosphorylated tau 181 (pTau181). (C) No significant differences were observed in plasma phosphorylated tau 217 (pTau217) levels in HF participants when stratified by NT-proBNP. Groups were tested for normality via d’Agostino & Pearson test or Shapiro-Wilk test if n was too low. Multiple variables were tested for significance via ordinary one-way ANOVA, and comparisons between groups were tested for significance using Tukey’s multiple comparisons test. (A, B) Significance was tested between the two variables using the Mann-Whitney test. (C) Correlation between NT-proBNP, a marker of heart failure severity, and (D) NfL, (E) pTau181, and (F) pTau217 were tested with a simple linear regression, interactions represented via R squared, and correlation is represented via Pearson r values. p value indicated by *. p=0.05–0.01 *, p = 0.009–0.001 **, p =…

• 

  Figure 3.. Serum cytokine values and the Composite Biomarker Cognitive Score increase with worsened HF status. Cytokine values for heart failure (HF) participants were stratified by N-terminal pro-B-type natriuretic peptide (NT-proBNP) levels and compared to healthy control participants (A) interleukin (IL)6, (B) IL-12p40, (C) IL-15, (D) and tumor necrosis factor-alpha (TNFα). (E) A Composite Biomarker Cognitive Score was made using transformed values of neurofilament light protein (NfL), IL-6, TNFα, and N-terminal pro-B-type natriuretic peptide (NT-proBNP). Groups were tested for normality via d’Agostino & Pearson test or Shapiro-Wilk test if n was too low. Multiple variables meeting normality were tested for significance via ordinary one-way ANOVA, and comparisons between groups were tested for significance Tukey’s multiple comparisons test. If values do not meet normality, multiple variables were tested for significance via the Kruskal-Wallis test, and comparison between groups was tested for significance through Dunn’s multiple comparisons test. p value indicated by *. p = 0.05–0.01 *, p = 0.009–0.001 **, p=0.0009–0.0001 ***, p < 0.0001 ****.

• 

  Figure 4.. The Composite Biomarker Cognitive Score correlated to the actual Composite cognitive Score. (A) Correlation and interaction between the Composite Biomarker Cognitive Score and the actual Composite cognitive Score were tested via simple linear regression, interactions represented via R-squared, and correlation is represented via Pearson r values. (B) Visualization of the Composite Biomarker Cognitive Score to predict the actual Composite cognitive Score is demonstrated in the dot line graph (measured in relative score). The Composite Biomarker Cognitive Score = NfL+NT-proBNP + IL-6 + TNFα. The actual Composite cognitive Score = CRN + CRN30 + CRO + CRO30 + VPA + KeepTrack.

• 

  Figure 5.. HF participants performed worse on cognitive and functional tests 12 months after their baseline visit. Values from baseline neuropsychological evaluations were compared to 12-month follow-up values and represented in the dot line graphs, showing the progression of each individual over time from baseline to 12-month follow-up for (A) the age and education normalized score from the Montreal cognitive assessment (MoCA), (B) the lure discrimination index from the mnemonic similarity task (MST), and both (C) the short delay free recall (MOAN standard Score) and D) the long delay free recall (MOAN standard Score) tasks from the rey auditory Verbal Learning test (AVLT). Individual participant’s colors are assigned for each person. Round, more saturated colors indicate individuals’ scores at baseline, and the less saturated square indicates 12-month follow-up values from the corresponding individuals. Normality was tested using the Shapiro-Wilk test. Significance between baseline and 12-month follow-up was tested for significance via Student’s paired test.

• 

  Figure 6.. HF participants had increased serum cytokine 12 months after their baseline visit. Values of baseline serum cytokine were compared to 12-month follow-up values and represented in the dot line graphs, showing the progression of cytokine values for each individual over time from baseline to 12-month follow-up for (A) Interferon-gamma (INFγ), (B) tumor necrosis factor-beta (TNFβ), (C) Eotaxin, (D) interleukin (IL)-6, and (E) tumor necrosis factor-alpha (TNFα). Individual participants’ colors are assigned for each person. Round, more saturated colors indicate individuals’ scores at baseline, and the less saturated square indicates 12-month follow-up values from the corresponding individuals. Normality was tested using the Shapiro-Wilk test. Significance between baseline and 12-month follow-up was tested via Student’s paired test if values met normality (Eotaxin, and IL-6), or Wilcoxon matched-pairs signed rank test if values did not meet normality (INFγ, TNFβ, and TNFα).