Stéphane Legriel1,2,3, Benedicte Grigoresco4, Patricia Martel5, Matthieu Henry-Lagarrigue4, Virginie Lvovschi4, Gilles Troché4, Marlène Amara6, Gwenaelle Jacq4, Fabrice Bruneel4, Maguy Bernard7,8, Anne Marinier9, Jean-Pierre Bedos4. 1. Medical-Surgical Intensive Care Unit, Centre Hospitalier de Versailles - Site André Mignot, 177 rue de Versailles, 78150, Le Chesnay Cedex, France. slegriel@ch-versailles.fr. 2. Sorbonne Paris Cité-Medical School, Paris Descartes University, Paris, France. slegriel@ch-versailles.fr. 3. INSERM U970, Paris Cardiovascular Research Center, Paris, France. slegriel@ch-versailles.fr. 4. Medical-Surgical Intensive Care Unit, Centre Hospitalier de Versailles - Site André Mignot, 177 rue de Versailles, 78150, Le Chesnay Cedex, France. 5. Public Health Department, Centre Hospitalier Universitaire Ambroise Paré, Boulogne, France. 6. Microbiology Department, Centre Hospitalier de Versailles - Site André Mignot, 177 rue de Versailles, 78150, Le Chesnay Cedex, France. 7. Department of Metabolic Biochemistry, Pitié Salpêtrière-Charles Foix University Hospital (AP-HP), Paris, France. 8. Department of Oncology and Endocrine Biochemistry, Pitié Salpêtrière-Charles Foix University Hospital (AP-HP), Paris, France. 9. Biochemistry Department, Centre Hospitalier de Versailles - Site André Mignot, 177 rue de Versailles, 78150, Le Chesnay Cedex, France.
Abstract
BACKGROUND: Early diagnostic orientation for differentiating pneumonia from pneumonitis at the early stage after aspiration would be valuable to avoid unnecessary antibiotic therapy. We assessed the accuracy of procalcitonin (PCT) in diagnosing aspiration pneumonia (AP) in intensive care unit (ICU) patients requiring mechanical ventilation after out-of-hospital coma. METHODS: Prospective observational 2-year cohort study in a medical-surgical ICU. PCT, C-reactive protein (CRP) and white blood cell count (WBC) were measured at admission (H0) and 6 h (H), H12, H24, H48, H96, and H120 after inclusion. Lower respiratory tract microbiological investigations performed routinely in patients with aspiration syndrome were the reference standard for diagnosing AP. Performance of PCT, CRP, and WBC up to H48 in diagnosing AP was compared based on the areas under the ROC curves (AUC) and likelihood ratios (LR+ and LR-) computed for the best cutoff values. RESULTS: Of 103 patients with coma, 45 (44%) had AP. Repeated PCT assays demonstrated a significant increase in patients with AP versus without AP from H0 to H120. Among the three biomarkers, PCT showed the earliest change. ROC-AUC values were poor for all three biomarkers. Best ROC-AUC values for diagnosing AP were for CRP at H24 [0.73 (95%CI 0.61-0.84)] and PCT at H48 [0.73 (95%CI 0.61-0.84)]. LR+ was best for PCT at H24 (3.5) and LR- for CRP and WBC at H24 (0.4 and 0.4, respectively). CONCLUSIONS: Early and repeated assays of PCT, CRP, and WBC demonstrated significant increases in all three biomarkers in patients with versus without AP. All three biomarkers had poor diagnostic performance for ruling out AP. Whereas PCT had the fastest kinetics, PCT assays within 48 h after ICU admission do not help to diagnose AP in ICU patients with coma.
BACKGROUND: Early diagnostic orientation for differentiating pneumonia from pneumonitis at the early stage after aspiration would be valuable to avoid unnecessary antibiotic therapy. We assessed the accuracy of procalcitonin (PCT) in diagnosing aspiration pneumonia (AP) in intensive care unit (ICU) patients requiring mechanical ventilation after out-of-hospital coma. METHODS: Prospective observational 2-year cohort study in a medical-surgical ICU. PCT, C-reactive protein (CRP) and white blood cell count (WBC) were measured at admission (H0) and 6 h (H), H12, H24, H48, H96, and H120 after inclusion. Lower respiratory tract microbiological investigations performed routinely in patients with aspiration syndrome were the reference standard for diagnosing AP. Performance of PCT, CRP, and WBC up to H48 in diagnosing AP was compared based on the areas under the ROC curves (AUC) and likelihood ratios (LR+ and LR-) computed for the best cutoff values. RESULTS: Of 103 patients with coma, 45 (44%) had AP. Repeated PCT assays demonstrated a significant increase in patients with AP versus without AP from H0 to H120. Among the three biomarkers, PCT showed the earliest change. ROC-AUC values were poor for all three biomarkers. Best ROC-AUC values for diagnosing AP were for CRP at H24 [0.73 (95%CI 0.61-0.84)] and PCT at H48 [0.73 (95%CI 0.61-0.84)]. LR+ was best for PCT at H24 (3.5) and LR- for CRP and WBC at H24 (0.4 and 0.4, respectively). CONCLUSIONS: Early and repeated assays of PCT, CRP, and WBC demonstrated significant increases in all three biomarkers in patients with versus without AP. All three biomarkers had poor diagnostic performance for ruling out AP. Whereas PCT had the fastest kinetics, PCT assays within 48 h after ICU admission do not help to diagnose AP in ICU patients with coma.
Authors: Lorena M Schenk; Matthias Schneider; Christian Bode; Erdem Güresir; Christoph Junghanns; Marcus Müller; Christian Putensen; Hartmut Vatter; Julian Zimmermann; Patrick Schuss; Felix Lehmann Journal: Front Neurol Date: 2021-02-25 Impact factor: 4.003