Chen et al developed symptom and clinical scores for diagnosing patients with the severe acute respiratory distress syndrome (SARS) during an epidemic. This can help identify patients who may benefit from more aggressive management and specific diagnostic tests. The authors are from the National Taiwan University.
Patient selection:
(1) potential exposure to SARS
(2) febrile (>= 38°C or >= 100.3°F)
Clinical parameters:
(1) myalgia
(2) diarrhea
(3) cough before or during fever
(4) rhinorrhea or sore throat
Laboratory parameters:
(5) absolute lymphocyte count
(6) platelet count
|
Parameter
|
Finding
|
Points
|
|
myalgia
|
absent
|
0
|
|
|
present
|
1
|
|
diarrhea
|
absent
|
0
|
|
|
present
|
1
|
|
cough before or after fever
|
absent
|
0
|
|
|
present
|
-2
|
|
rhinorrhea or sore throat
|
neither present
|
0
|
|
|
either or both present
|
-1
|
|
absolute lymphocyte count
|
>= 1,000 per µL
|
0
|
|
|
< 1,000 per µL
|
1
|
|
platelet count
|
>= 150,000 per µL
|
0
|
|
|
< 150,000 per µL
|
1
|
symptom score =
= SUM(points for clinical parameters #1 to #4)
clinical score =
= SUM(points for clinical and laboratory parameters)
Interpretation:
• minimum symptom score -3
• maximum symptom score: 2
• minimum clinical score -3
• maximum clinical score: 4
• The cutoff for the symptom score was >= 0.
• The cutoff for the clinical score was >= 1.
Criteria for placing the patient in isolation - one or more of the following:
(1) infiltrate on chest X-ray
(2) significantly abnormal laboratory finding (severe leukopenia)
(3) symptom score >= 0
(4) clinical score >= 1
(5) fever lasting more than 3 days
A patient not placed in isolation was sent to home quarantine and was evaluated for the next 3 days for signs of progression or recovery.
Performance in derivation set (Chen et al):
• The symptom score >= 0 was 100% sensitive and 76% specific.
• The clinical score >=1 was 100% sensitive and 86% specific.
• Performance is affected by age, comorbidity and site of exposure.
• Rhinorrhea and sore throat were reported in SARS patients from Canada and Hong Kong, which would affect performance.
Performance in validation set (Su et al):
• The sensitivity of the symptom score was 96% with a specificity of 52%.
• The sensitivity of the clinical score was 93% with a specificity of 71%.
• If the clinical score was only done on patients with a symptom score >=0, then the sensitivity was 90% and specificity was 80%.
