Biomedicines (Mar 2023)

Mid-Infrared Spectroscopy as a New Tool for Ruling Out Spontaneous Bacterial Peritonitis: A Proof-of-Concept Study

  • Marwin A. Farrugia,
  • Maëna Le Corvec,
  • Christophe Renou,
  • Jean-Baptiste Nousbaum,
  • Dann J. Ouizeman,
  • Olivier Sire,
  • Olivier Loréal,
  • Hugues Tariel,
  • Jérôme Bernard,
  • Thierry Piche,
  • Albert Tran,
  • Hafid Ait-Oufella,
  • Luce Landraud,
  • Philippe Gual,
  • Rodolphe Anty,
  • the Cytokine Ascites Group

DOI
https://doi.org/10.3390/biomedicines11030838
Journal volume & issue
Vol. 11, no. 3
p. 838

Abstract

Read online

Background and aims: A highly sensitive and specific point-of-care method for diagnosing spontaneous bacterial peritonitis (SBP) is currently lacking. The objective of the present study is to evaluate the diagnostic value of a rapid, easy-to-use, mid-infrared fiber evanescent wave spectroscopy (MIR-FEWS) method for ruling out SBP. Patients and Methods: Cirrhotic patients (n = 256) at five centers in France were included for suspected SBP or for the scheduled evacuation of ascites fluid. The mid-infrared spectrum of 7 µL of an ascites fluid sample was recorded using a MIR-FEWS system. To define a model for the diagnosis of SBP, the patients were divided into a calibration group (n = 170) and a validation group (n = 86). Results: Most of the patients were male (71%). The mean age was 60.25 years. Alcohol-related liver disease was the most common cause of cirrhosis. SBP was observed in 18% of the patients. For the diagnosis of SBP in the calibration and validation groups, respectively, the model gave areas under the receiver operating characteristic curves of 0.87 and 0.89, sensitivities of 90% and 87%, specificities of 78% and 80%, positive predictive values of 48% and 50%, negative predictive values of 97% and 96%, positive likelihood ratio of 4.09 and 4.35, negative likelihood ratio of 0.13 and 0.16, Youden index of 0.68 and 0.67, and correct classification rates of 80% and 81%. Conclusion: The results of this proof-of-concept study show that MIR-FEWS is a highly sensitive diagnostic method for ruling out SBP. The method warrants further investigation.

Keywords