[Diagnostic Value of ROC Curve Evaluating Serum Related Indexes for Bloodstream Infection in Patients with Hematopathy]

Abstract

AbstractObjective: To evaluate the diagnostic value of serum PCT, CRP and SAA for bloodstream infection(BSI) in patients with hematopathy.

Methods: Sixty hematopathy patients with bloodstream infection from July 2016 to June 2018 were selected and enroued in bloodstream infection group. Sixty-five patients with negative blood culture during the same period were selected and enrolled in non-bloodstream infection group. The ROC curves were drawn and used to eualuate the diagnostic value of above montioned indexes.

Results: The levels of PCT, CRP and SAA in the bloodstream infection group were higher than those in the non-bloodstream infection group (P<0.05). ROC curve showed that AUC values of PCT, CRP, SAA and the combined test detection were 0.868, 0.746, 0.678 and 0.900, respectively, there was no significant difference in AUC between combined test and PCT test (P>0.05). AUC of combined test and PCT test were higher than those of CRP and SAA test, and the difference was statistically significant (P<0.05), but there was no significant difference in AUC between CRP and SAA (P>0.05). The optimal PCT detection threshold was 0.49 ng/ml, the sensitivity and specificity were 75.0% and 83.1%, respectively. The optimal critical value for CRP detection was 15.76 mg/L, the sensitivity and specificity were 60.0% and 80.0% respectively. The optimal SAA detection threshold was 35.66 mg/L, the sensitivity and specificity were 81.7% and 53.8%, respectively.

Conclusion: PCT, CRP and SAA detection have good diagnostic value for blood stream infection in patients with hematopathy. The diagnostic value of PCT is better than CRP and SAA, and there is no significant difference in diagnostic value between combined test and PCT test.