Background: The complexity of multiple-item criteria in acute respiratory distress syndrome (ARDS) often causes inconvenience for physicians in the management of patients with severe acute pancreatitis (SAP). We evaluated whether serum SP-A levels in the presence of diffuse alveolar damage (DAD) can be qualitatively assessed for diagnosis of SAP-induced ARDS.
Material and methods: Eighty rats were randomly divided into 2 groups (n=40 each) - the sham-operated (SO) group and the SAP group - and then randomly subdivided into 4 subgroups in a time-course manner. Furthermore, rats in the SAP group were subdivided into the SAP induced-ARDS group (ARDS group) and the SAP without ARDS group (non-ARDS group) according to the diagnostic standard of ARDS. The diagnostic cut-off values of SP-A for SAP-induced ARDS were determined by the receiver operating characteristic curve (ROC).
Results: Serum SP-A levels in Baseline, SO group, SAP group, ARDS group, and non-ARDS group were 43.15±14.29, 51.91±16.99, 193.4±35.37, 198.0+29.73, and 185.7±43.21 ug/ml, respectively. The best cut-off value for the serum SP-A level for the diagnosis of SAP-induced ARDS was 150 ug/ml and the area under the ROC curve of SP-A was 0.88. The sensitivity, specificity, positive predictive value, negative predictive value, and accuracy of SP-A in the diagnosis of SAP-induced ARDS were 100.0%, 81.8%, 71.4%, 100.0%, and 87.5%, respectively.
Conclusions: Serum SP-A levels may allow the detection of SAP-induced ARDS and may help to support the clinical diagnosis of ARDS. The optimal serum SP-A cut-off value to discriminate SAP-induced ARDS and other groups (SO group and non-ARDS group) is around 150 ug/ml.