Blood sedimentation rate predicts death risk in dogs with chronic kidney disease

Researchers at Chonnam National University in Gwangju, South Korea, found that dogs whose erythrocyte sedimentation rate, or ESR, exceeded 16 millimeters per hour at their first evaluation faced more than three times the mortality risk of dogs below that threshold, even after accounting for conventional markers of kidney dysfunction such as blood urea nitrogen and serum creatinine.

The study included 272 dogs presented to a single referral animal hospital between October 2023 and January 2026. Of these, 145 were classified as clinically healthy controls, having shown no relevant abnormalities on history, physical examination, complete blood count, serum biochemistry, electrolyte analysis, urinalysis, and available imaging. In this study, the remaining 127 dogs carried a diagnosis of chronic kidney disease, established and staged according to the guidelines of the International Renal Interest Society using fasting serum creatinine and, where available, symmetric dimethylarginine. In this study, dogs with acute kidney injury, urinary tract infection, pyelonephritis, active systemic inflammation such as sepsis or acute pancreatitis, malignant neoplasia, or recent nephrotoxic drugs, transfusion, or immunosuppressive therapy were excluded from the CKD group to minimize potential confounding effects on ESR and systemic inflammatory biomarkers.

In this study, ESR was measured with an automated analyzer, the MINI-PET device, using anticoagulated whole blood, and results were recorded in millimeters per hour. Sedimentation is driven largely by erythrocyte aggregation into stacks called rouleaux, a process promoted by plasma proteins such as fibrinogen and immunoglobulins that reduce the negative surface charge on red cell membranes. Because these proteins rise during inflammation and chronic disease, ESR is nonspecific but responsive to the cumulative burden of systemic illness. In this cohort, the authors report that ESR was measured within minutes of venipuncture from the same EDTA sample used for the complete blood count, and no blood samples were stored before ESR analysis.

The results showed a clear gradient of ESR across disease severity. Median ESR was 9 millimeters per hour in healthy controls and 13 millimeters per hour in dogs with chronic kidney disease overall, a highly significant difference. Yet the separation was driven by advanced disease: dogs in IRIS stage 1 and stage 2 did not differ significantly from controls, with p values of 0.054 and 0.059 respectively, whereas stage 3 dogs had a median ESR of 15 and stage 4 dogs a striking median of 37 millimeters per hour, both significantly higher than controls and than earlier stages. Notably, stage 4 dogs were younger on average than stage 3 dogs, which the authors interpret as evidence that ESR elevation was not explained by age distribution alone.

Survival analysis sharpened the picture. Of the 127 dogs with chronic kidney disease, 53 were in stage 1, 46 in stage 2, 14 in stage 3, and 14 in stage 4; 75 dogs survived and 52 died during follow-up, which extended to a maximum of 786 days. Non-survivors had a median ESR of 21 millimeters per hour, with an interquartile range of 13 to 40, compared with 11 millimeters per hour, interquartile range 8 to 14, in survivors, a difference with a p value below 0.0001. Receiver operating characteristic analysis identified 16 millimeters per hour as the optimal cut-off, yielding a sensitivity of 64.58 percent and a specificity of 81.01 percent for all-cause mortality, with an area under the curve of 0.79.

That discrimination was comparable to the strongest conventional markers in the same cohort. Serum creatinine achieved an area under the curve of 0.78, the calcium-phosphorus product 0.77, and blood urea nitrogen 0.72, while the albumin-to-globulin ratio at 0.62 and white blood cell count at 0.61 showed limited discrimination. In univariable Cox proportional hazards regression, ESR of at least 16 millimeters per hour carried a hazard ratio of 4.10 for death. When the investigators built a multivariable model adjusting for blood urea nitrogen, creatinine, and cardiovascular comorbidity, ESR remained independently associated with mortality, with a hazard ratio of 3.42 and a 95 percent confidence interval of 1.78 to 6.64.

Correlation and regression analyses hinted at why a sedimentation marker might track outcomes in kidney disease. ESR correlated positively with serum creatinine, phosphate, potassium, the calcium-phosphorus product, systemic hypertension, proteinuria, and the presence of congestive heart failure due to myxomatous mitral valve disease at stage C or higher, and negatively with the albumin-to-globulin ratio. In multiple linear regression, creatinine, the calcium-phosphorus product, potassium, the albumin-to-globulin ratio, and advanced mitral valve disease were each independently associated with ESR levels. The authors caution that these are associations, not proof of mechanism: fibrinogen, immunoglobulin fractions, and other acute-phase proteins were not directly measured, and markers of mineral and bone disorder or vascular calcification were not evaluated, so ESR should not be read as a direct index of any single process.

The cardiovascular connection deserves emphasis. Dogs with mitral valve disease at stage C or higher were far more common among non-survivors, 42.3 percent versus 12.0 percent of survivors, and this comorbidity was associated with both higher ESR and mortality. The authors frame this in terms of the cardiovascular-renal axis, the bidirectional interplay between cardiac and renal dysfunction mediated by hemodynamic, neurohormonal, inflammatory, and metabolic pathways. Because chronic kidney disease in dogs is a systemic syndrome involving uremia, anemia, mineral imbalance, metabolic acidosis, proteinuria, hypertension, and nutritional deterioration, a nonspecific marker that integrates these downstream consequences may add information that static excretory markers such as creatinine cannot provide.

As an exploratory analysis, the team constructed two composite risk scores from multivariable Cox regression coefficients to examine whether ESR could provide additional prognostic information when interpreted together with conventional renal and clinical variables. The score without ESR, termed MCB, assigned 2 points for advanced mitral valve disease, 2.5 points for creatinine of at least 2.5 milligrams per deciliter, and 3 points for blood urea nitrogen of at least 45 milligrams per deciliter. The ESR-inclusive score, MCBE, weighted those components 2, 3, and 5 respectively and added 3 points for ESR of at least 16 millimeters per hour. In apparent performance within the same cohort, MCBE achieved an area under the curve of 0.87 with 71.15 percent sensitivity and 86.67 percent specificity at a cut-off of 6, while MCB reached 0.85 with 96.15 percent sensitivity but only 65.33 percent specificity at a cut-off of 2. The difference between the two areas was not statistically significant in a paired DeLong test, with p equal to 0.39.

Kaplan-Meier analysis reinforced the stratification pattern. When dogs were divided into four ordered groups by ESR, median survival time was not reached in the two lowest groups, fell to 245 days in the third group, and dropped to 35 days in the highest. The ESR-inclusive score produced the most orderly separation, with unreached median survival in the two lowest groups, 200 days in the third, and 31 days in the highest, whereas the score without ESR showed less consistent intermediate-risk ordering, with median survivals of 308 and 275 days in its second and third groups. The authors stress that both scores were derived and evaluated in the same cohort, so these figures represent apparent performance rather than validated predictive accuracy, and the study was not powered to detect small differences between the scores.

Several limitations temper the conclusions. The single-center retrospective design may limit generalizability, and healthy controls were not age-matched to the markedly older CKD group, although within the CKD cohort age was not significantly correlated with ESR and did not differ between survivors and non-survivors. The numbers of stage 3 and stage 4 dogs were small, survival was measured as all-cause mortality in a geriatric population with concurrent diseases, ESR was not corrected for hematocrit, and treatment was not standardized. Residual confounding by anemia, plasma protein composition, occult inflammation, or unmeasured inflammatory proteins cannot be excluded. The authors conclude that baseline ESR was associated with disease severity and all-cause mortality in dogs with chronic kidney disease and may offer complementary prognostic information alongside conventional renal and clinical variables, while the ESR-inclusive composite score remains exploratory and requires external validation, preferably in a prospective multicenter cohort, before any clinical use.

Subject of Research: Prognostic value of erythrocyte sedimentation rate in dogs with chronic kidney disease

Article Title: Baseline erythrocyte sedimentation rate is associated with mortality in dogs with chronic kidney disease

Article References: Moon, C.-H., Joo, J.-B., Ro, W.-B., & Lee, C.-M. (2026). Baseline erythrocyte sedimentation rate is associated with mortality in dogs with chronic kidney disease. Frontiers in Veterinary Science, 13, Article 1924110. https://doi.org/10.3389/fvets.2026.1924110

Image Credits: AI Generated

DOI: 10.3389/fvets.2026.1924110

Keywords: chronic kidney disease, dogs, erythrocyte sedimentation rate, prognosis, mortality, veterinary medicine, IRIS staging, inflammation, Cox regression, risk score, creatinine, biomarker