Kawasaki Disease Case: The 3-Year-Old With 7 Days of Fever and Incomplete Criteria
A toddler with prolonged fever meets only 3 of 5 classic criteria. Work through the diagnostic reasoning, the laboratory clues, and the critical treatment window that determines whether this child develops coronary aneurysms.
“The Fever Won’t Break” — A Kawasaki Disease Case Unfolds
Chief Complaint: A 3-year-old boy is brought to his pediatrician on day 5 of illness with persistent high fever. The parents report the fever has been continuous, ranging 39.0–40.2°C, and has not responded to alternating acetaminophen and ibuprofen. He was seen at an urgent care clinic on day 2 of illness and given a diagnosis of “viral syndrome.”
The parents are now concerned because his eyes have become red, his lips are cracked and bleeding, and he is increasingly irritable and refusing to eat or drink.
- Age: 3 years, 1 month
- Weight: 14.2 kg (45th percentile)
- Past Medical History: Unremarkable; no prior hospitalizations
- Immunizations: Up to date
- Contacts: Older sibling had a cold 2 weeks ago that resolved in 3 days
- Family History: No autoimmune or cardiac disease
Day 5 Examination: Counting the Criteria
The child is irritable and clinging to his mother. He is difficult to console and cries with minimal handling. The pediatrician performs a thorough examination:
Eyes: Bilateral non-exudative conjunctival injection, more prominent at the limbus. No discharge, no photophobia.
Oral cavity: Erythematous, dry, cracked lips with bleeding fissures. Pharynx diffusely erythematous without exudate. The tongue appears erythematous with prominent papillae (“strawberry tongue”).
Skin: A faint, polymorphous erythematous rash is noted on the trunk, fading. No desquamation. No vesicles or petechiae.
Extremities: No erythema or edema of the hands or feet. No periungual peeling.
Cervical lymph nodes: No cervical lymphadenopathy palpated (bilateral).
Cardiac: Tachycardic, regular rhythm, no murmur. Capillary refill 2 seconds.
Abdomen: Soft, mildly tender diffusely. The parents mention he has had some loose stools. No hepatosplenomegaly.
Criteria Scorecard at Day 5
| Classic Criterion | Present? | Finding |
|---|---|---|
| Fever ≥5 days | Yes | Continuous fever × 5 days, 39–40.2°C |
| Bilateral conjunctival injection (non-exudative) | Yes | Bilateral limbal-sparing injection, no discharge |
| Oral mucosal changes | Yes | Cracked lips, strawberry tongue, pharyngeal erythema |
| Extremity changes | No | No hand/foot erythema, edema, or peeling |
| Cervical lymphadenopathy (≥1.5 cm) | No | No palpable lymphadenopathy |
The child meets 3 of 5 classic criteria plus fever ≥5 days. Classic (complete) Kawasaki disease requires fever plus 4 of 5 criteria. The pediatrician refers the child to the emergency department with a note: “Possible Kawasaki disease — does not meet full criteria, may be viral. Needs lab workup.”
Before proceeding, consider: Does this child need to meet all classic criteria to warrant treatment? What is the concept of incomplete Kawasaki disease, and what additional data would help you decide?
Emergency Department Workup — Day 5
The ED attending orders a comprehensive laboratory workup. The results are available within 2 hours:
Additional results: sodium 133 mEq/L, urinalysis shows sterile pyuria (25 WBC/hpf, negative culture). GGT is 64 U/L (elevated). Total bilirubin 1.4 mg/dL. Blood culture is pending (will ultimately be negative).
When a child has fever ≥5 days and 2–3 classic criteria, the 2017 AHA Scientific Statement recommends checking CRP and ESR. If CRP ≥3.0 mg/dL or ESR ≥40 mm/hr, check supplemental labs. If ≥3 of 6 supplemental criteria are met, treat as Kawasaki disease OR obtain an echocardiogram — if the echo is abnormal, treat.
Supplemental Laboratory Criteria Scorecard
| Supplemental Criterion | Threshold | This Child | Met? |
|---|---|---|---|
| Albumin ≤3.0 g/dL | ≤3.0 | 3.2 g/dL | No (borderline) |
| Anemia for age | Hgb below normal | 11.8 g/dL | No |
| ALT elevation | Above normal | 86 U/L | Yes |
| Platelets ≥450,000 after day 7 | ≥450 | 524 (day 5) | Yes (already elevated) |
| WBC ≥15,000 | ≥15.0 | 18.4 | Yes |
| Urine WBC ≥10/hpf | ≥10 | 25/hpf | Yes |
This child meets 4 of 6 supplemental laboratory criteria (ALT elevation, thrombocytosis, leukocytosis, and sterile pyuria). Per the AHA algorithm, meeting ≥3 supplemental criteria in a child with fever ≥5 days and 2–3 classic criteria is sufficient to diagnose incomplete Kawasaki disease and initiate treatment.
The Critical Decision: Treat Now or Wait for More Criteria?
The ED attending discusses the case with the on-call pediatric hospitalist. The hospitalist reviews the criteria scorecard and notes the child has only 3 of 5 classic features. The hospitalist states: “He doesn’t meet full Kawasaki criteria. Let’s admit, observe, and repeat labs in 48 hours. If he develops a fourth criterion or the fever continues, we’ll consider treatment.”
On day 5, with 3 classic criteria, CRP 14.2, ESR 82, and 4 of 6 supplemental lab criteria met, what is the correct course of action?
What Happened: Days 5 Through 12
Day 5–7 (Observation): The child is admitted and observed. IV fluids are given for poor oral intake. The fever persists at 39–40°C. The rash fades. No new classic criteria develop. The hospitalist documents “likely viral illness, does not meet Kawasaki criteria” on daily progress notes. An echocardiogram is not ordered during this period.
Day 8: The child develops mild edema of the dorsum of both hands. The parents notice the irritability is worsening. The nurse documents a new finding: “swelling of hands bilaterally, redness of fingertips.” The team now counts 4 classic criteria (conjunctival injection, oral changes, rash history, and now extremity changes). The attending diagnoses Kawasaki disease and orders an echocardiogram and IVIG.
Day 9: IVIG (2 g/kg = 28.4 g) is administered over 12 hours. High-dose aspirin is started. The echocardiogram performed on day 9 shows mildly dilated left anterior descending coronary artery (LAD) with a Z-score of 2.8, just below the threshold for aneurysm definition (≥2.5 is abnormal, ≥5.0 is small aneurysm).
Day 10–11: The fever resolves within 36 hours of IVIG. The child’s irritability improves markedly. Repeat inflammatory markers show CRP trending down to 6.8 mg/dL.
Day 12: Follow-up echocardiogram shows progression: the LAD coronary artery now has a Z-score of 5.4 with a fusiform aneurysm measuring 4.2 mm. The right coronary artery (RCA) shows mild dilation with a Z-score of 3.1. The child is started on low-dose aspirin (3–5 mg/kg/day) indefinitely and referred to pediatric cardiology for long-term follow-up.
IVIG was not administered until day 9 of illness. The evidence is clear: the risk of coronary artery aneurysms increases significantly when IVIG is given after day 10. This child had a treatable condition on day 5 using the AHA incomplete Kawasaki algorithm, but treatment was withheld because the team required “complete” classic criteria. The coronary aneurysm detected on day 12 may have been prevented by earlier treatment.
The AHA Incomplete Kawasaki Disease Algorithm
The 2017 AHA Scientific Statement provides a structured algorithm for evaluating children with suspected incomplete Kawasaki disease. This algorithm is the essential diagnostic tool for cases exactly like this one — where the clinical features are suggestive but the “full” classic criteria are not met.
Incomplete Kawasaki disease carries the same risk of coronary artery aneurysms as classic Kawasaki disease — in fact, some studies suggest the risk may be higher because delayed diagnosis leads to delayed treatment. Children with incomplete Kawasaki disease are more likely to be treated after day 10, which is the primary driver of increased aneurysm risk. The terminology “incomplete” refers to the criteria, not the severity.
Treatment of Kawasaki Disease and IVIG-Resistant Cases
IVIG 2 g/kg as a single infusion over 10–12 hours, administered as early as possible and ideally within the first 10 days of illness. Combined with high-dose aspirin (30–50 mg/kg/day in divided doses, or 80–100 mg/kg/day in some protocols) until the child is afebrile for 48–72 hours, then transition to low-dose aspirin (3–5 mg/kg/day) for 6–8 weeks.
This child received IVIG on day 9 but the fever resolved only after 36 hours. What defines IVIG resistance, and what is the next step if the child remains febrile ≥36 hours after IVIG completion?
Before revealing the full outcome, consider: this child received IVIG on day 9 and already had coronary dilation on the day-9 echo. What would you expect to find at the 2-week and 6-week follow-up echocardiograms?
The fever resolved 36 hours after IVIG (borderline for resistance — the team elected to monitor closely rather than give a second dose). CRP normalized by day 14. Periungual desquamation of the fingers was noted on day 15, a late confirmatory feature that reinforced the Kawasaki diagnosis.
Day 12 echocardiogram: LAD coronary artery aneurysm, Z-score 5.4, fusiform, 4.2 mm. RCA mild dilation, Z-score 3.1.
6-week echocardiogram: LAD aneurysm persistent at Z-score 4.8 (slight improvement). RCA normalized. The child was continued on low-dose aspirin indefinitely and referred for ongoing pediatric cardiology surveillance.
6-month echocardiogram: LAD Z-score 3.2, showing regression. The aneurysm was reclassified as a small aneurysm with improving trend. The cardiologist recommended continued low-dose aspirin and annual follow-up echocardiography.
Had IVIG been administered on day 5 — when the child already met criteria for incomplete Kawasaki disease by the AHA algorithm — the 4-day treatment delay that likely contributed to the coronary aneurysm would have been avoided.
Echocardiogram Timing and Coronary Artery Classification
The echocardiogram is a critical tool in Kawasaki disease but its timing and interpretation require nuance. A normal early echo does not exclude the diagnosis, and coronary changes can evolve over weeks.
Recommended Echo Schedule
| Timing | Purpose | Key Findings to Assess |
|---|---|---|
| At diagnosis (baseline) | Establish baseline coronary dimensions | Coronary Z-scores, LV function, mitral regurgitation, pericardial effusion |
| 1–2 weeks | Detect early aneurysm formation | Coronary dilation or aneurysm development (this is when aneurysms are most commonly first detected) |
| 6–8 weeks | Assess for late changes and regression | Aneurysm size, regression or progression, thrombus formation |
Coronary Artery Z-Score Classification (AHA 2017)
| Classification | Z-Score | Clinical Significance |
|---|---|---|
| Normal | <2.0 | No coronary involvement |
| Dilation only | 2.0 to <2.5 | Borderline, warrants close follow-up |
| Small aneurysm | 2.5 to <5.0 | Low-dose aspirin; cardiology follow-up |
| Medium aneurysm | 5.0 to <10.0 | Aspirin + possible anticoagulation; long-term surveillance |
| Giant aneurysm | ≥10.0 or ≥8 mm | Aspirin + anticoagulation (warfarin or LMWH); highest risk for thrombosis and MI |
Coronary aneurysms typically develop during the second week of illness. The initial baseline echo may be completely normal. Treatment should never be delayed while waiting for echo confirmation — IVIG is given to prevent aneurysms from forming, not as a response to aneurysms already present.
What the Evidence Tells Us
The evidence for managing Kawasaki disease — including incomplete presentations — is anchored by the AHA 2017 Scientific Statement and supported by decades of observational data on IVIG timing, coronary outcomes, and the recognition of incomplete disease. This evidence base directly addresses the central failure in this case: treatment delay.
The American Heart Association defines incomplete Kawasaki disease as fever ≥5 days with 2–3 classic features plus supportive laboratory findings. The statement provides the definitive diagnostic algorithm incorporating supplemental lab criteria and echocardiographic findings. It recommends IVIG 2 g/kg within the first 10 days of illness and emphasizes that incomplete Kawasaki disease carries equal coronary artery risk to complete disease.
Study: Newburger et al. (1986) conducted the seminal randomized controlled trial establishing IVIG as the treatment for Kawasaki disease, comparing IVIG plus aspirin to aspirin alone in 168 children.
Key Finding: IVIG given within 10 days of fever onset reduced the incidence of coronary artery abnormalities from approximately 23% (aspirin alone) to 5% (IVIG + aspirin) at 7-week follow-up. Subsequent analyses confirmed that treatment before day 7 had the lowest aneurysm rates, while treatment after day 10 was associated with significantly diminished benefit.
Clinical Implication: The day-10 window is a true clinical threshold. In this case, IVIG on day 5 (when incomplete criteria were met) would have fallen within the optimal treatment window. Treatment on day 9 fell near the edge, and the coronary aneurysm detected on day 12 may represent the cost of the 4-day delay.
Study: Sudo et al. (2012) analyzed data from the Japanese nationwide Kawasaki disease survey, comparing coronary outcomes in children with complete versus incomplete presentations across over 23,000 cases.
Key Finding: Approximately 15–20% of children with Kawasaki disease present with incomplete criteria. Children with incomplete Kawasaki who were treated with IVIG had coronary artery aneurysm rates comparable to those with complete disease treated within the same timeframe. However, children with incomplete disease were significantly more likely to experience delayed treatment (beyond day 10), and delayed treatment was the primary driver of higher aneurysm rates in this group.
Clinical Implication: Incomplete Kawasaki disease is not a milder form — it is a diagnostic challenge. The worse outcomes associated with incomplete presentations are attributable to delayed recognition and treatment, not to inherently different disease biology. This case perfectly illustrates that failure mode.
Study: Yu et al. (2016) evaluated the diagnostic performance of the AHA supplemental laboratory criteria algorithm in a cohort of 195 children with incomplete Kawasaki disease, comparing outcomes in children treated based on the algorithm versus those in whom treatment was delayed.
Key Finding: Using the ≥3 of 6 supplemental criteria threshold, the algorithm had a sensitivity of 80% and specificity of 73% for identifying children who would develop coronary abnormalities if untreated. Children treated within 10 days using the algorithm had coronary aneurysm rates below 5%, comparable to complete Kawasaki disease treated within the same window.
Clinical Implication: The supplemental criteria algorithm is validated and effective. It provides a structured, evidence-based pathway for the exact clinical scenario in this case: a child with concerning features but fewer than 4 classic criteria. Clinicians who rely solely on the classic criteria will miss approximately 15–20% of Kawasaki disease cases.
Study: Tremoulet et al. (2008) identified risk factors for IVIG resistance in a prospective cohort of 362 children with Kawasaki disease, and tested the Kobayashi score for predicting treatment failure.
Key Finding: Approximately 10–20% of patients did not respond to initial IVIG, defined as persistent fever ≥36 hours after completion of the infusion. Risk factors included lower initial albumin, higher CRP, higher neutrophil count, lower platelet count, younger age, and higher ALT. IVIG-resistant patients had a significantly higher rate of coronary artery aneurysms (approximately 25% vs 5% in responders).
Clinical Implication: This child had several features associated with IVIG resistance (elevated ALT, high CRP, low-normal albumin). Early recognition of high-risk patients may justify closer monitoring for treatment failure and earlier consideration of adjunctive therapy. Borderline IVIG response (fever resolving at 36 hours) warrants vigilant follow-up.
Evidence Summary
| Study | Design | Key Finding | Clinical Impact |
|---|---|---|---|
| Newburger et al. (1986) | RCT | IVIG reduced coronary abnormalities from 23% to 5% | IVIG within 10 days is standard of care |
| Sudo et al. (2012) | Nationwide Survey | Incomplete KD has equal aneurysm risk; delay is the key variable | Incomplete = same urgency as complete |
| Yu et al. (2016) | Cohort Study | Supplemental criteria: 80% sensitivity, 73% specificity | AHA algorithm validation |
| Tremoulet et al. (2008) | Prospective Cohort | 10–20% IVIG resistance; 25% aneurysm rate in resistant cases | Early identification of high-risk patients |
The evidence is unambiguous: (1) IVIG within the first 10 days of illness reduces coronary aneurysm risk from approximately 25% to under 5%; (2) incomplete Kawasaki disease carries the same coronary risk as complete disease, but is more often treated late because clinicians wait for criteria that may never fully develop; (3) the AHA supplemental criteria algorithm is a validated tool for identifying children who need treatment before they meet classic criteria; and (4) every day of delay beyond day 7 incrementally increases coronary risk.
Clinical Takeaways
- Incomplete Kawasaki disease is not mild Kawasaki disease — it carries the same coronary aneurysm risk and demands the same treatment urgency
- The AHA algorithm provides a validated pathway: fever ≥5 days + 2–3 classic criteria + elevated CRP/ESR + ≥3 of 6 supplemental lab criteria = treat
- IVIG should be given within the first 10 days of illness — every day of delay beyond day 7 increases aneurysm risk. Do not wait for “complete” criteria
- A normal early echocardiogram does not exclude Kawasaki disease — aneurysms typically develop during the second week. The echo is for monitoring, not for gating treatment
- Sterile pyuria, elevated ALT, hypoalbuminemia, and thrombocytosis in a child with prolonged fever should immediately raise suspicion for Kawasaki disease
- IVIG resistance (~10–20%) is defined as persistent fever ≥36 hours post-infusion. Standard retreatment is a second dose of IVIG 2 g/kg
References
- McCrindle BW, Rowley AH, Newburger JW, et al. Diagnosis, treatment, and long-term management of Kawasaki disease: a scientific statement for health professionals from the American Heart Association. Circulation. 2017;135(17):e927–e999. https://doi.org/10.1161/CIR.0000000000000484
- Newburger JW, Takahashi M, Burns JC, et al. The treatment of Kawasaki syndrome with intravenous gamma globulin. New England Journal of Medicine. 1986;315(6):341–347. https://doi.org/10.1056/NEJM198608073150601
- Sudo D, Monobe Y, Yashiro M, et al. Coronary artery lesions of incomplete Kawasaki disease: a nationwide survey in Japan. European Journal of Pediatrics. 2012;171(4):651–656. https://doi.org/10.1007/s00431-011-1630-3
- Tremoulet AH, Best BM, Song S, et al. Resistance to intravenous immunoglobulin in children with Kawasaki disease. Journal of Pediatrics. 2008;153(1):117–121. https://doi.org/10.1016/j.jpeds.2007.12.021
- Yu JJ, Park YM, Yoon KL. Supplemental laboratory criteria in incomplete Kawasaki disease. Korean Journal of Pediatrics. 2016;59(2):68–73. https://doi.org/10.3345/kjp.2016.59.2.68