📄 medicine

Impact of serial changes of D-dimer in patients with Kawasaki disease: Correlation between the peak D-dimer level and coronary arterial lesions

This study demonstrates that in patients with Kawasaki disease, higher peak D-dimer levels correlate with coronary artery lesions, suggesting that persistently elevated D-dimer levels after IVIG treatment may indicate ongoing vascular involvement and warrant further clinical monitoring.

Original authors: Kenichi Okumura, Kyosuke Izumi, Takuma Sugawara, Kento Itoi, Kunpei Sekine, Takuzou Wada, Motokazu Nishikado, Yuuko Amimoto, Hitoshi Awaguni, Ikumi Arai, Mayumi Nagasawa, Jun Shinozuka, Kenji Hamaoka

Published 2026-10-23

📖 5 min read🧠 Deep dive

CC BY 4.0

Original authors: Kenichi Okumura, Kyosuke Izumi, Takuma Sugawara, Kento Itoi, Kunpei Sekine, Takuzou Wada, Motokazu Nishikado, Yuuko Amimoto, Hitoshi Awaguni, Ikumi Arai, Mayumi Nagasawa, Jun Shinozuka, Kenji Hamaoka

Original paper licensed under CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/). ✨ This is an AI-generated explanation of the paper below. It is not written or endorsed by the authors. For technical accuracy, refer to the original paper. Read full disclaimer

Kawasaki disease is a condition that strikes young children, causing a sudden, intense inflammation that attacks the body's blood vessels. While the fever and rash are the most visible signs, the most dangerous part of the illness happens deep inside, where the inflammation can weaken the walls of the heart's own arteries. If these vessels become damaged, they can swell or form bulges, a complication that can lead to serious heart problems later in life. Doctors treat the disease quickly with a powerful mixture of antibodies to calm the immune system, but sometimes the inflammation does not stop as expected. To catch these stubborn cases early, medical teams rely on blood tests that act as signals for what is happening inside the body. One such signal is a substance called D-dimer, which the body produces when it is trying to repair damaged tissue or fight off clots. For years, researchers have known that children with this disease have high levels of D-dimer, but they did not know exactly how these levels changed after treatment or if the highest point the levels reached could predict whether the heart arteries would be hurt.

A team of doctors at Uji Tokusyukai Hospital in Japan set out to solve this puzzle by watching how D-dimer levels moved over time in sixty-two children with the disease. They compared these children to thirty-six other sick children who had fevers but did not have Kawasaki disease, ensuring that any differences they found were specific to the heart condition. The researchers took blood samples from the Kawasaki patients at several key moments: when they first arrived at the hospital, after they received their first dose of treatment, when they were sent home, and again one and four months later. They also measured the size of the children's coronary arteries using ultrasound, a painless imaging technique that allows doctors to see inside the blood vessels without surgery. The goal was to see if the highest level of D-dimer a child reached during their illness matched up with any swelling or damage found in the heart arteries.

The results showed a clear pattern that changes how doctors might monitor these patients. Before any treatment was given, the children with Kawasaki disease had much higher D-dimer levels than the sick children in the control group. However, something surprising happened after the standard treatment was administered. While other markers of inflammation, such as white blood cell counts, dropped quickly and returned to normal, the D-dimer levels did not fall as fast. In many cases, the levels remained high for days or even weeks after the fever broke. The researchers found that the highest point the D-dimer reached during the entire course of the illness was a strong predictor of heart trouble. Specifically, the size of the swelling in the right coronary artery, one of the main vessels feeding the heart, was directly linked to how high the D-dimer climbed.

The study identified a specific threshold that could help doctors decide when to be extra cautious. If a child's D-dimer level rose above 2.6 milligrams per deciliter at any point during their illness, there was a significant chance they would develop abnormalities in their coronary arteries. This cutoff point was able to correctly identify about seven out of ten children who would have heart issues, while also correctly ruling out the condition in a similar number of children who would remain healthy. The data suggested that when D-dimer levels fail to drop after the initial treatment, it is a sign that the inflammation in the blood vessels is still active, even if the child no longer has a fever. This delay in recovery of the blood marker suggests that the damage to the artery walls is still progressing, requiring the medical team to look closer and perhaps consider additional treatments.

The researchers also noted that the right coronary artery seemed to be the most sensitive to these changes, showing the strongest connection to the D-dimer levels. This aligns with previous observations that this specific artery is often the most frequently affected by the disease. The study did not find a similar strong link between the D-dimer levels and the other major arteries in the heart, such as the left main trunk or the left anterior descending artery, suggesting that the right side of the heart may have a unique vulnerability to this specific type of inflammatory response. While the study was limited to a single hospital and a relatively small number of patients, the findings offer a practical tool for clinicians. By tracking the peak of this blood marker rather than just the initial reading, doctors can get a clearer picture of the disease's true severity and the likelihood of long-term heart damage.

In the end, the work provides a reason to keep a close watch on these blood tests even after the acute phase of the illness seems to be over. If the D-dimer levels stay high, it is a signal that the body is still fighting a battle within the blood vessels. The authors suggest that in such cases, repeating the ultrasound scans of the heart and considering further medical intervention are necessary steps to protect the child's future heart health. This approach moves beyond simply treating the fever and looks at the deeper, lingering signs of vascular injury, offering a more complete strategy for managing a disease that can leave lasting marks on a child's heart.

Technical Summary: Impact of Serial Changes of D-dimer in Patients with Kawasaki Disease

Problem Statement
Kawasaki disease (KD) is the leading cause of acquired pediatric heart disease in developed countries, primarily due to the development of coronary artery lesions (CAL). While intravenous immunoglobulin (IVIG) is the standard treatment to suppress inflammation and prevent CAL, a subset of patients exhibits treatment resistance, leading to persistent or progressive coronary arteritis. Current clinical practice relies heavily on systemic inflammatory markers (e.g., C-reactive protein, white blood cell count) to monitor treatment response. However, these markers often normalize rapidly after IVIG administration, potentially masking ongoing local vascular inflammation. Previous research has established that D-dimer levels are elevated in KD, but existing studies have been limited to pre-IVIG measurements. There is a lack of data regarding the serial changes of D-dimer post-IVIG and the specific correlation between peak D-dimer levels and the severity of coronary artery abnormalities.

Methodology
This retrospective, single-center study conducted at Uji Tokusyukai Hospital analyzed 62 pediatric patients (under 14 years) diagnosed with KD between February 2020 and December 2024, alongside a control group of 36 febrile children without KD or rheumatic/immune disorders. All KD patients received standard IVIG therapy.

The study employed the following methodological approach:

Longitudinal Monitoring: D-dimer levels and other laboratory parameters (CRP, WBC, platelet count, albumin) were measured at five specific time points: pre-IVIG (admission), post-IVIG (after initial dose), at discharge, at 1 month (convalescent phase), and at 4 months (remission phase).
Peak Determination: The "peak" value for inflammatory markers was defined as the maximum level observed during the clinical course.
Coronary Assessment: Echocardiography was performed to measure the proximal right coronary artery (RCA), left main trunk (LMT), left anterior descending (LAD), and left circumflex (LCX) arteries. Z-scores were calculated based on Japanese multicenter reference data, with a Z-score >2.0 defining a coronary artery lesion (CAL).
Statistical Analysis: The study utilized unpaired t-tests for baseline comparisons, one-way ANOVA for serial changes, Pearson correlation coefficients to assess relationships between peak markers and Z-scores, and stepwise multivariate linear regression to identify independent predictors of CAL. Receiver Operating Characteristic (ROC) curve analysis was performed to determine the predictive performance of peak D-dimer levels.

Key Results

Baseline Differences: Pre-IVIG D-dimer levels were significantly higher in the KD group compared to the febrile control group ( $p < 0.001$ ).
Serial Changes: Unlike WBC and CRP, which showed significant improvement after IVIG, D-dimer levels did not normalize rapidly. The median day of peak D-dimer levels was day 7 (range 3–12). While levels decreased significantly by 1 and 4 months, they remained elevated immediately post-IVIG compared to controls.
Correlation with CAL:
- Pre-IVIG D-dimer levels showed no significant correlation with Z-scores of any coronary artery site.
- However, the peak D-dimer level demonstrated a significant positive correlation with the Z-score of the right coronary artery (RCA) ( $r = 0.44, p < 0.001$ ).
- Multivariate analysis confirmed peak D-dimer as the only significant independent predictor for the RCA Z-score.
- Peak albumin levels were identified as an independent predictor for the LAD Z-score.
Predictive Threshold: ROC analysis identified a peak D-dimer cutoff value of >2.6 mg/dL for predicting CAL. This threshold yielded a sensitivity of 70.0% and a specificity of 69.2%, with an Area Under the Curve (AUC) of 71.1%.
Clinical Outcomes: At the 4-month follow-up, 4 of the 62 patients had developed CAL. No moderate-to-giant aneurysms were observed in this cohort.

Significance and Claims
The authors claim that this study clarifies the utility of serial D-dimer monitoring in KD management. The primary significance lies in the finding that D-dimer levels reflect local vascular inflammation that may persist even after systemic markers (WBC, CRP) have normalized following IVIG treatment.

Specifically, the paper posits that:

Delayed Resolution: The failure of D-dimer levels to decrease rapidly after IVIG may indicate ongoing coronary artery involvement, suggesting a time lag in the resolution of vascular inflammation compared to systemic inflammation.
RCA Specificity: The strong correlation between peak D-dimer and the RCA Z-score supports the hypothesis that D-dimer is involved in the inflammatory process of the right coronary artery, which is the most frequently affected site in KD.
Clinical Recommendation: The authors suggest that if D-dimer levels fail to decrease or exceed the 2.6 mg/dL cutoff during the clinical course, clinicians should consider repeat echocardiography and evaluate the need for additional second-line therapies, rather than relying solely on the normalization of systemic inflammatory markers.

The study concludes that serial assessment of D-dimer, alongside albumin, provides essential information for detecting persistent coronary arteritis that might otherwise be missed by standard monitoring protocols. The authors acknowledge limitations regarding the study's retrospective nature, small sample size, and the absence of moderate-to-giant aneurysms in the cohort, noting that further accumulation of cases is required to validate these findings across different populations.

More like this