BACKGROUND: The results obtained from various point-of-care (POC) test devices for estimating C-reactive protein (CRP) levels in a laboratory setting differ when compared to a laboratory reference test. We aimed to determine whether such differences meaningfully affect the accuracy and added diagnostic value in predicting radiographic pneumonia in adults presenting with acute cough in primary care. METHODS: A nested case control study of adult patients presenting with acute cough in 12 different European countries (the Genomics to combat Resistance against Antibiotics in Community-acquired LRTI in Europe [GRACE] Network). Venous blood samples from 100 patients with and 100 patients without pneumonia were tested with five different POC CRP tests and a laboratory analyzer. Single test accuracy values and the added value of CRP to symptoms and signs were calculated. RESULTS: Single test accuracy values showed similar results for all five POC CRP tests and the laboratory analyzer. The area under the curve of the different POC CRP tests and the laboratory analyzer (range 0.79-0.80) were all comparable and higher than the clinical model without CRP (0.70). Multivariable odds ratios were the same (1.2) for all CRP tests. CONCLUSIONS: Five POC CRP test devices and the laboratory analyzer performed with similar accuracy in detecting pneumonia both as single test, and when used in addition to clinical findings. Variability in results obtained from standard CRP laboratory and POC test devices do not translate into clinically relevant differences when used for prediction of pneumonia in patients with acute cough in primary care.
BACKGROUND: The results obtained from various point-of-care (POC) test devices for estimating C-reactive protein (CRP) levels in a laboratory setting differ when compared to a laboratory reference test. We aimed to determine whether such differences meaningfully affect the accuracy and added diagnostic value in predicting radiographic pneumonia in adults presenting with acute cough in primary care. METHODS: A nested case control study of adult patients presenting with acute cough in 12 different European countries (the Genomics to combat Resistance against Antibiotics in Community-acquired LRTI in Europe [GRACE] Network). Venous blood samples from 100 patients with and 100 patients without pneumonia were tested with five different POC CRP tests and a laboratory analyzer. Single test accuracy values and the added value of CRP to symptoms and signs were calculated. RESULTS: Single test accuracy values showed similar results for all five POC CRP tests and the laboratory analyzer. The area under the curve of the different POC CRP tests and the laboratory analyzer (range 0.79-0.80) were all comparable and higher than the clinical model without CRP (0.70). Multivariable odds ratios were the same (1.2) for all CRP tests. CONCLUSIONS: Five POC CRP test devices and the laboratory analyzer performed with similar accuracy in detecting pneumonia both as single test, and when used in addition to clinical findings. Variability in results obtained from standard CRP laboratory and POC test devices do not translate into clinically relevant differences when used for prediction of pneumonia in patients with acute cough in primary care.
Entities:
Keywords:
C-reactive protein; diagnostic tests; lower respiratory tract infections; point-of-care testing; primary health care
Authors: Leanne Chalmers; Petra Czarniak; Jeffery Hughes; Rebecca Iacob; Ya Ping Lee; Kiran Parsons; Richard Parsons; Bruce Sunderland; Tin Fei Sim Journal: Explor Res Clin Soc Pharm Date: 2022-06-02
Authors: Victoria Hardy; Matthew Thompson; Gina A Keppel; William Alto; M Ashworth Dirac; Jon Neher; Christopher Sanford; Jaime Hornecker; Allison Cole Journal: BMJ Open Date: 2017-01-25 Impact factor: 2.692
Authors: Yoel Lubell; Nga T T Do; Kinh V Nguyen; Ngan T D Ta; Ninh T H Tran; Hung M Than; Long B Hoang; Poojan Shrestha; Rogier H van Doorn; Behzad Nadjm; Heiman F L Wertheim Journal: Antimicrob Resist Infect Control Date: 2018-10-04 Impact factor: 4.887
Authors: Luca Arts; Endry Hartono Taslim Lim; Peter Marinus van de Ven; Leo Heunks; Pieter R Tuinman Journal: Sci Rep Date: 2020-04-30 Impact factor: 4.379
Authors: Petra Czarniak; Leanne Chalmers; Jeffery Hughes; Rebecca Iacob; Ya Ping Lee; Kiran Parsons; Richard Parsons; Bruce Sunderland; Tin Fei Sim Journal: Int J Clin Pharm Date: 2022-01-28