We are grateful for the interest of Chaib et al., in our work on the estimated reduction of antibiotic prescribing if general practitioners (GPs) had used appropriate point-of-care tests (POCT) and had followed the current guidelines.1 As mentioned in the limitation section of the paper a hypothetical situation was considered, provided the following three assumptions were accomplished: having access to POCTs, these tests had been appropriately used by GPs, and GPs had followed the existing guidelines, as described in Fig. 1. The authors are concerned about the extrapolation of guidelines for pharyngitis and lower respiratory tract infections, but we want to remind the authors that these recommendations are based on studies carried out in the primary care setting, some of which were performed in our country.
They are completely right that C-reactive protein (CRP) does not reliably differentiate bacterial from viral infections in general practice. However, the myth that antibiotics should always be given in bacterial infections should be debunked as antibiotics should be withheld in viral infections as well as in self-limiting bacterial infections. Based on the evidence currently available from studies in general practice, CRP can assist clinicians in differentiating serious from self-limiting lower respiratory tract infections and has shown to be the best biomarker to assist and guide antibiotic prescribing decisions for these infectious diseases, such as pneumonia.2 Most lower respiratory tract infections are self-limiting and do not benefit from antibiotics (regardless of bacterial or viral aetiology). When CRP levels are lower than 20mg/l it is safe to withhold antibiotic treatment in all patients regardless of illness duration.
Regarding the use of procalcitonin, most of the studies have been carried out in settings other than primary care. For patients presenting to the emergency department with symptoms of a suspected infection, procalcitonin measurement is a useful diagnostic tool to identify bacterial infections, such as sepsis, allowing early initiation of the appropriate antibiotic treatment. In addition, in assessing the severity of sepsis, serum procalcitonin levels are an important diagnostic tool, especially in the early stages. Procalcitonin-guided management might reduce antibiotic prescribing in the emergency department, but studies carried out in primary care are scarce.2 In the largest study ever done including 2820 patients with lower respiratory tract infections in primary care, aimed at quantifying the diagnostic accuracy of inflammatory markers in addition to symptoms and signs for predicting pneumonia, the addition of CRP concentration to signs and symptoms improved the diagnostic information, but measurement of procalcitonin concentration did not add clinically relevant information.3 In another study Holm and colleagues showed a clear association between procalcitonin concentrations and radiographic pneumonia as well as bacterial infection, but the positive predictive value was too low to be useful in clinical practice.4 Furthermore, procalcitonin is not available as a POCT as it takes 20min to obtain a result and measurement costs are considerable making it undesirable for common high-incidence infections in primary care. CRP in contrast, is an office-based test with acceptable accuracy and costs that can be integrated into the ambulatory management of respiratory infections. POCTs should fulfil a list of criteria before considering their utilisation in clinical practice and for the time being there is insufficient evidence for considering procalcitonin in primary care.5
Conflicts of interestWe report receiving research grants from the European Commission (Sixth & Seventh Programme Frameworks and Horizon 2020), Catalan Society of Family Medicine, Instituto de Salud Carlos III and Alere.
