Serum Trimethylamine N-Oxide as a Diagnostic and Prognostic Biomarker in Dogs with Chronic Kidney Disease: A Pilot Study

Figures

• 

  Figure 1. Box plot comparing the median creatinine (A), SDMA (B), TMAO (C), and cystatin C (D) concentrations between control dogs (n = 30) and dogs with CKD (n = 32). Serum TMAO, creatinine, and SDMA levels in the CKD group were significantly higher than those in the control group (Mann–Whitney U test; p < 0.0001) (median, 79.64 ng/mL; IQR, 41.12–140.8 versus median, 28.66 ng/mL; IQR, 1.708–40.52). The cystatin C levels in the CKD group were higher than those in the control group (Mann–Whitney U test; p < 0.05). * p < 0.05 and **** p < 0.0001.

• 

  Figure 2. Scatter plots of serum TMAO concentration with creatinine (A), SDMA (B) and Cystatin C (C). Pearson’s correlation analysis displayed no correlation between serum TMAO and creatinine, SDMA and cystatin C (p = 0.2630, 0.7716, and 0.1006, respectively).

• 

  Figure 3. ROC curves of TMAO, creatinine, SDMA, and cystatin C concentrations to confirm their diagnostic accuracy in dogs with CKD. ROC curve for predicting CKD; TMAO (AUC = 0.8073), creatinine (AUC = 0.9558), SDMA (AUC = 0.9989) and cystatin C (AUC = 0.6844) compared with healthy dogs.

• 

  Figure 4. Box plot comparing the median TMAO concentration among control dogs (n = 30) and dogs with CKD stage 2 (n = 12), 3 (n = 11), and 4 (n = 9) dogs. TMAO levels were significantly higher in dogs with CKD stages 2, 3, and 4 than in healthy controls (p < 0.05 *). As presented in Table 1, the post hoc power analysis revealed that the minimum required sample size for adequate statistical power is 26 participants. Our current study utilized a total sample size of 62 participants, which substantially exceeds this requirement and ensures sufficient statistical power for the ANOVA analysis. This analysis demonstrates that the ANOVA results presented in Figure 4 are based on statistically adequate sample sizes with sufficient power to detect meaningful effects. We have incorporated these power analysis results into the manuscript to enhance the transparency and statistical rigor of our findings.

• 

  Figure 5. Box plot comparing the median HDL (A) and LDL concentrations (B) between control dogs (n = 30) and dogs with CKD (n = 30). Scatter plots of serum TMAO concentrations with HDL (C) and LDL concentrations (D). Serum LDL concentrations in the CKD group were significantly higher than those in the control group (Mann–Whitney U test; ** p < 0.01 in CKD) and serum HDL concentrations in the CKD group were not statistically significantly lower than those in the control group (Mann–Whitney U test; p > 0.05). Pearson’s correlation analysis for dogs with CKD showed that TMAO levels had a mild positive correlation with HDL (r = 0.4189, p = 0.0212) within the CKD group, and no correlation (p = 0.6228) was found between serum TMAO and LDL within the CKD group.

• 

  Figure 6. Box plot comparing the median concentrations of TMAO (A) between the young group (n = 12) and the old group (n = 20), (B) between the antibiotic-treated group (n = 17) and the untreated group (n = 15), (C) between the group with GI symptoms (n = 20) and the group without symptoms (n = 12), and (D) between the renal diet group (n = 16) and the normal diet group (n = 16). There was no statistically significant increase in TMAO concentrations based on age, antibiotic administration, the presence of GI symptoms, or renal diet status within the CKD group (Mann–Whitney U test; p > 0.05).

• 

  Figure 7. Scatter plots showing the TMAO concentration in relation to ALT ((A), n = 29), ALP ((B), n = 30), GGT ((C), n = 27), and TBIL levels ((D), n = 28) within the CKD group. Pearson’s correlation analysis revealed no significant correlations between TMAO and ALT, ALP, GGT, or TBIL within the CKD group.

• 

  Figure 8. Box plot comparing the median TMAO (A), creatinine (B), SDMA (C) and cystatin C (D) concentrations between the non-survivor group (n = 14) and survivor group (n = 18) at a 6-month follow-up in the CKD group. Serum TMAO, creatinine, and SDMA levels in non-survivors were significantly higher than in survivors in the CKD group (Mann–Whitney U test; * p < 0.05; 0.0142, ** p < 0.01 and *** p < 0.001 in non-survivors), (TMAO median, 137.0 ng/mL; IQR, 87.26– 165.2 versus median, 70.18 ng/mL; IQR, 27.86–84.50). Serum cystatin C in non-survivors was not statistically significantly higher than in survivors within the CKD group (Mann–Whitney U test; p > 0.05).

• 

  Figure 9. Box plot comparing the median HDL (A) and LDL (B) concentrations between the non-survivor group (n = 13) and survivor group (n = 17) at 6 months in the CKD group. The serum HDL and LDL concentrations in non-survivors were not statistically significantly higher than in survivors within the CKD group (Mann–Whitney U test; p > 0.05).

• 

  Figure 10. ROC curves of TMAO, creatinine, SDMA, and cystatin C concentrations to confirm their prognostic accuracy for mortality in dogs with CKD. ROC curve for predicting mortality in dogs with CKD; TMAO (AUC = 0.7563), creatinine (AUC = 0.8333), SDMA (AUC = 0.8393), and cystatin C (AUC = 0.5741) compared with survivor dogs.