|
Overview :
Lempinen et al studied the ability of several tests to
identify patients with severe acute pancreatitis. Patients with severe
pancreatitis are managed better in the intensive care unit (ICU), while
patients with mild to moderate disease can be managed on a regular hospital ward.
The authors are from Helsinki
University Cenral
Hospital in Finland.
Tests used to assess a patient:
(1) urinary
trypsinogen-2 (serum trypsinogen-2 level is also elevated)
(2) urinary
trypsinogen activation peptide (TAP, amino-terminal peptide released on
activation of trypsinogen to trypsin)
(3) serum
C-reactive peptide (CRP)
(4) APACHE II
score
Timing of assessments:
(1) on admission
(2) at 24 hours after admission (vs 48 hours for Ranson
score)
|
Test
|
Mild to Moderate Pancreatitis
|
Severe Pancreatitis
|
|
urinary trypsinogen-2
|
<= 3,000 µg/L
|
> 3,000 ¨g/L
|
|
urinary TAP
|
<= 35 nmol/L
|
> 35 nmol/L
|
|
serum CRP
|
<= 150 mg/L
|
>= 150 mg/L
|
|
APACHE II
|
<= 8
|
> 8
|
The tests have only poor to fair sensitivity, while
specificity is fair to good (Table 1, page 269). This means that no single test
can be used to stratify a patient. For example, the admission urinary
trypsinogen-2 misses 28% of patients with severe pancreatitis and misclassifies
19% of patients who do not need to be admitted to the ICU.
|
Test
|
Youden Index on Admission
|
Youden Index at 24 Hours
|
|
urinary trypsinogen-2
|
0.53
|
0.60
|
|
urinary TAP
|
0.46
|
0.46
|
|
serum CRP
|
0.22
|
0.56
|
|
APACHE II
|
0.16
|
0.25
|
where:
• Youden index = S + E - 1, with 1.0 indicating a perfect
test and 0 a useless test.
|