Kidney Week

Abstract: TH-PO076

Platelet Count and Serum Creatinine in the Differential Diagnosis of TMA: An Analysis from the CESAR Study

Session Information

• AKI: Biomarkers, Drugs, Onco-Nephrology
  October 25, 2018 | Location: Exhibit Hall, San Diego Convention Center
  Abstract Time: 10:00 AM - 12:00 PM

Category: Acute Kidney Injury

• 101 AKI: Epidemiology, Risk Factors, and Prevention

Authors

• Schoenermarck, Ulf, University Hospital Munich-Grosshadern, Muenchen, Germany

Ulf Schoenermarck,

• Schroppel, Bernd, University of Ulm, Ulm, Germany

Bernd Schroppel,

• Schmidbauer, Daniel, CRI - The Clinical Research Institute GmbH, Munich, Germany

Daniel Schmidbauer,

• Jeglitsch, Michael, Alexion, Munich, Germany

Michael Jeglitsch,

• Ries, Wolfgang, Diako Flensburg, Flensburg, Germany

Wolfgang Ries,

Background

Thrombotic thrombocytopenic purpura (TTP), Shiga toxin-producing E. coli hemolytic uremic syndrome (STEC-HUS), and atypical HUS (aHUS) require a fast differential diagnosis in order to initiate urgent therapy. TTP may be excluded by an ADAMTS13 activity assay result of >5–10%, however, this test is not immediately available at many centers. Data from the CESAR study¹ were used to assess whether routine laboratory parameters can be used to predict/rule out TTP.

Methods

CESAR was a cross-sectional, multicenter, epidemiologic study of the relative incidence of aHUS, TTP and STEC-HUS in Germany. We developed an algorithm to predict severe ADAMTS13 deficiency in adult patients at clinical presentation. All available variables were analyzed for correlation with final diagnosis. Thresholds were identified maximizing the sum of sensitivity and specificity (Youden Index: J=sensitivity+specificity-1) and then rounding to the nearest easy-to-recall value.

Results

A total of 219 adult patients with a mean (SD) age of 56.2 (18.9) years were enrolled, of whom 31 (14%) were diagnosed with TTP. Platelet (Plt) and serum creatinine (SCr) levels significantly correlated with TTP diagnosis (p<0.01). Multivariate analysis (Table 1) revealed that if Plt and SCr levels are above threshold values (Plt: 30 x 10⁹/L and SCr: 1.8 mg/dL), TTP is almost certainly ruled out (negative predictive value 98.1 and 92.3 for one or both values, respectively, above threshold).

Conclusion

This analysis shows that Plt and SCr are strong predictors to rule out a diagnosis of TTP. This finding confirms previous studies² and may help to accelerate differential diagnosis of TMA.

¹Schoenermarck U, et al. 2017; presented at ASN Kidney Week [SA-PO421]
²Coppo P, et al. PLoS One 2010;5(4):e10208.

Table. Multivariate analysis to predict TTP diagnosis

Values expressed as %. NPV, negative predictive value; Plt, platelets; PPV, positive predictive value; SCr, serum creatinine.

Funding

• Commercial Support –