Overruling of Procalcitonin-Guided Antibiotics for Lower Respiratory Tract Infections in Primary Care: Ancillary Study of a Randomized Controlled Trial
by
, , , , ,
José Knüsli
1,*,
Loïc Lhopitallier
2,
Andreas Kronenberg
3,4,
Jean-Yves Meuwly
5,
Onya Opota
6,
Marc-Antoine Perrenoud
6,
Marie-Anne Page
6,
Kevin C. Kain
7,8,
Aline Mamin
9,
Valérie D’Acremont
10,
Nicolas Senn
11,
Yolanda Mueller
11,
Isabella Locatelli
12 and
Noémie Boillat-Blanco
1 1
Infectious Diseases Service, Lausanne University Hospital, University of Lausanne, Rue du Bugnon 46, 1011 Lausanne, Switzerland
2
Gare10 Lausanne General Practice, Av. de la gare 10, 1003 Lausanne, Switzerland
3
Institute for Infectious Diseases, University of Bern, Friedbühlstrasse 51, 3001 Bern, Switzerland
4
Medix General Practice, Bubenbergplatz 8, 3011 Bern, Switzerland
5
Department of Radiology, Lausanne University Hospital, University of Lausanne, Rue du Bugnon 46, 1011 Lausanne, Switzerland
6
Microbiology Institute, Lausanne University Hospital, University of Lausanne, Rue du Bugnon 48, 1011 Lausanne, Switzerland
7
Tropical Disease Unit, Department of Medicine, University of Toronto, 200 Elizabeth Street, Toronto, ON M5G 2C4, Canada
8
Sandra Rotman Centre for Global Health, Toronto General Hospital, University Health Network, 105 St. George Street, Toronto, ON M5S 3E6, Canada
9
Division of Infectious Diseases and Centre for Emerging Viral Diseases, Faculty of Medicine, University of Geneva Hospitals, Rue Gabrielle-Perret-Gentil 4, 1211 Geneva, Switzerland
10
Digital Global Health Department, Centre for Primary Care and Public Health (Unisanté), University of Lausanne, Rue du Bugnon 44, 1011 Lausanne, Switzerland
11
Department of Family Medicine, Centre for Primary Care and Public Health (Unisanté), University of Lausanne, Rue du Bugnon 44, 1011 Lausanne, Switzerland
12
Department of Education, Research, and Innovation, Centre for Primary Care and Public Health (Unisanté), University of Lausanne, Rue du Bugnon 44, 1011 Lausanne, Switzerland
*
Author to whom correspondence should be addressed.
Antibiotics 2023, 12(2), 377; https://doi.org/10.3390/antibiotics12020377
Submission received: 24 January 2023
/
Revised: 9 February 2023
/
Accepted: 10 February 2023
/
Published: 12 February 2023
(This article belongs to the Special Issue Antimicrobial Treatment of Lower Respiratory Tract Infections)
Abstract
:Background: Lower respiratory tract infections (LRTIs) in primary care are a promising target for antibiotic stewardship. A clinical trial in Switzerland showed a large decrease in antibiotic prescriptions with procalcitonin guidance (cut-off < 0.25 µg/L) compared with usual care. However, one-third of patients with low procalcitonin at baseline received antibiotics by day 28. Aim: To explore the factors associated with the overruling of initial procalcitonin guidance. Design and Setting: Secondary analysis of a cluster randomized trial in which patients with an LRTI were included. Method: Using the characteristics of patients, their disease, and general practitioners (GPs), we conducted a multivariate logistic regression, adjusted for clustering. Results: Ninety-five out of 301 (32%) patients with low procalcitonin received antibiotics by day 28. Factors associated with an overruling of procalcitonin guidance were: a history of chest pain (adjusted OR [aOR] 1.81, 95% confidence interval 1.03–3.17); a prescription of chest X-ray by the GP (aOR 4.65, 2.32–9.34); a C-reactive protein measured retrospectively above 100 mg/L (aOR 7.48, 2.34–23.93, reference ≤ 20 mg/L); the location of the GP practice in an urban setting (aOR 2.27, 1.18–4.37); and the GP’s number of years of experience (aOR per year 1.05, 1.01–1.09). Conclusions: Overruling of procalcitonin guidance was associated with GPs’ socio-demographic characteristics, pointing to the general behavioral problem of overprescription by physicians. Continuous medical education and communication training might support the successful implementation of procalcitonin point-of-care tests aimed at antibiotic stewardship.
1. Introduction
Estimates put the global toll of antimicrobial resistance at 1.27 million deaths in 2019, which underlines the need for urgent real-life implementation of antibiotic stewardship [1]. More than 80% of human antimicrobial prescriptions take place in primary care, and acute respiratory infections are the most common reason for unnecessary prescriptions, making them promising targets for stewardship [2,3]. Indeed, the majority of patients with lower respiratory tract infections (LRTIs) receive antibiotics, although only 5 to 10% of them have pneumonia in primary care and will benefit from this treatment [4,5].
Procalcitonin is a host biomarker that can differentiate bacterial from viral infections [6]. The recent availability of point-of-care tests enables general practitioners (GPs) to rapidly measure procalcitonin from capillary blood [7]. Procalcitonin use in primary care is not yet common in Switzerland, although its reimbursement by the Swiss health insurance system is currently under consideration. This might help it become a standard tool for the management of LRTIs. Indeed, a randomized clinical trial conducted in Swiss primary care showed that procalcitonin at the point of care to guide initiation of antibiotics (using a cut-off of <0.25 µg/L) for patients with LRTIs led to a large decrease in antibiotic prescription compared with usual care (−26%, 95% confidence interval [95% CI] −41% to −10%) [8]. However, although only 8% of patients had elevated procalcitonin at baseline and a recommendation for antibiotic prescription, 40% received antibiotics within 28 days of the initial consultation, suggesting room for improvement of the guidance. We aimed to identify factors associated with an overruling of the initial procalcitonin guidance that could support the real-life implementation of the procalcitonin point-of-care test.
2. Results
2.1. Participants
Figure 1 presents the flow of the participants. Of the 469 patients originally included in the randomized controlled trial, 329/469 (70%) were in the UltraPro or procalcitonin groups, had their procalcitonin measured at the point of care during the initial consultation, and their antibiotic prescription status assessed at day 28. Of those, the majority (301/329, 91%) had a procalcitonin value < 0.25 µg/L and were included in this ancillary study. Patients were mostly female (61%), 29% were 65 or older, 28% had comorbidities, 53% abnormal vital signs, 39% a bacterial or mixed infection and 7% a retrospectively measured CRP above 100 mg/l (see Table 1). Among these 301 patients with a procalcitonin value < 0.25 µg/L, 95 (32%) were prescribed antibiotics by day 28 (overruling of initial procalcitonin guidance), 26/95 (27%) during the initial consultation, 34/95 (36%) between 1 and 7 days after the initial consultation and 35/95 (37%) between 8 and 28 days after the initial consultation. Procalcitonin measurement was repeated during follow-up visits for 9/301 (3%) patients; the newly measured values were still < 0.25 µg/L.
2.2. Univariate Analysis
Patients in whom procalcitonin guidance was overruled had more often a history of chest pain (odds ratio [OR] 1.74, 95% CI 1.03–2.93), a temperature ≥ 37.8 °C (OR 2.16, 95% CI 1.04–4.48), a focal abnormal finding upon lung auscultation (OR 1.83, 95% CI 1.06–3.16), a prescription of chest X-ray from the GP (OR 3.79, 95% CI 1.85–7.76), a measurement of CRP by the GP during the consultation (OR 2.77, 95% CI 1.58–4.86), and a CRP value measured retrospectively (result not communicated to the GPs) above 100 mg/l (OR 4.59, 95% CI 1.52–13.90, reference CRP ≤ 20 mg/L) (Table 1). Among patients with CRP measured by the GP during the initial consultation (97/301, 33%), the CRP value was not associated with an overruling of procalcitonin guidance. The detection of bacteria was also not associated with an overruling.
2.3. Multivariate Analysis
The backward selection determined a model with 5 independent variables (Figure 2), some factors being characteristics of the patients and others of the GPs: a history of chest pain (adjusted OR [aOR] 1.81, 95% CI 1.03–3.17); the prescription of a chest X-ray by the GP (aOR 4.65, 95% CI 2.32–9.34); a CRP value measured retrospectively above 100 mg/L (aOR 7.48, 95% CI 2.34–23.93, reference CRP ≤ 20 mg/L); the location of the GP practice in an urban setting (aOR 2.27, 95% CI 1.18–4.37); and the GP’s number of years of experience (aOR per year of experience 1.05, 95% CI 1.01–1.09). The presence of comorbidities, abnormal vital signs or a bacterial etiology of the LRTI were not associated with an overruling of the procalcitonin guidance.
The intraclass correlation coefficient was 0. The Hosmer–Lemeshow test had a p-value of 0.44, meaning that good calibration of the model should not be rejected. The C statistic was 0.71, showing a reasonably good predictive value for the model.
3. Discussion
3.1. Summary
Our study, conducted on a patient population with an overall low severity of disease at presentation, shows that overruling the initial procalcitonin-based guidance in patients with LRTIs in primary care was associated with the characteristics of the patient (retrospectively measured CRP > 100 mg/L, history of chest pain) as well as of the GP (location of the practice in an urban setting, number of years of experience). The prescription of a chest radiograph by the GP was also strongly associated with an overruling of procalcitonin guidance.
Regarding the characteristics of the patients, a high retrospectively measured CRP result, which was not communicated to the GP, was associated with an overruling of procalcitonin guidance. However, the GP’s willingness to measure CRP and its results (available to the GP) did not influence the decision to prescribe antibiotics. Indeed, GPs measured the CRP in only 2 of the 18 patients (11%) with a retrospectively measured CRP > 100 mg/L. It has been previously shown that a CRP > 100 mg/L is associated with pneumonia, or even bacterial pneumonia [9,10]. However, in our study, the clinical outcome of patients was not related to CRP value; we found the same proportion of patients with a persistence of dyspnea at day 7 (98/258 (38%) versus 7/18 (41%) patients with CRP ≤ 100 mg/L and CRP > 100 mg/L, respectively, p = 1.0) and with an admission to the hospital (3/258 (1%) versus 1/18 (6%), respectively, p = 0.6), and this irrespective of antibiotic prescription. This discrepancy between procalcitonin and CRP values was rare and occurred in only 18/301 (7%) patients. Another patients’ characteristic associated with an overruling of procalcitonin guidance was the presence of chest pain. As chest pain has not been proven to be a predictor of pneumonia in several studies [11,12,13], we suggest that this argument alone should not justify a prescription of antibiotics. Our finding is not in line with another study in which chest pain was a protective factor against prescription [14].
The characteristics of the GPs associated with an overruling of procalcitonin guidance (urban setting and length of experience) point to other psychosocial determinants of antibiotic prescriptions. The multivariate analysis included the demographics, comorbidities and disease severity of the patients, suggesting that practicing in an urban area is associated with an overruling of procalcitonin guidance beyond differences in patient characteristics between settings, possibly due to differences in clinical practices. A study conducted in a neighboring country also found higher rates of antibiotic prescription in urban areas [15], while another one did not find any association [16]. Regarding the number of years of experience, the more experienced GPs might distrust the procalcitonin level in favor of their own clinical impression or might have retained more liberal prescribing patterns, predating our knowledge of antibiotic resistance. This finding is in line with previous studies showing that GPs who have practiced longer are more likely to prescribe antibiotics inappropriately [17,18]. Interestingly, in our dataset, GPs practicing in urban areas tended to be less experienced than their counterparts in rural regions (median number of years of experience: 8 versus 13 years in the urban versus rural group, respectively, p = 0.04). This might explain why the location of practice and the number of years of experience were not associated with the overruling of procalcitonin guidance in the univariate regression (two opposite factors canceling out), while the association became apparent in the multivariate model.
The last factor associated with an overruling of procalcitonin guidance is the prescription of a chest X-ray by the GP. This result was shared by an earlier study that found an increase in antibiotic prescriptions when physicians requested X-rays [19]. Of note, 60% (18/30) of X-rays were negative (as read by the GPs) among patients who received an antibiotic prescription. Presumably, GPs could reserve X-rays for patients with the perceived highest pre-test probability of pneumonia. Nevertheless, X-rays are useful only when their results are expected to change the management of patients.
3.2. Strengths and Limitations
This study benefits from the strengths of the randomized controlled trial with a large number of practices included, the good quality and high completion of collected data, as well as the low rate of loss to follow-up. The biobank established during the trial made it possible to retrospectively analyze CRP and microbiology. We included several characteristics of the GPs in our variables, which proved to be associated with antibiotic prescriptions. Moreover, the calculated intraclass correlation coefficient was 0, suggesting that, conditional on the GP-dependent variables included in the model, no additional variation between GPs is estimated in the propensity to overrule the initial procalcitonin guidance. This lends another argument to the generalizability of our findings.
The main limitation of the study was the inability to differentiate between adequate and inadequate prescriptions. Consequently, no evidence-based improvement to the procalcitonin algorithm can be derived from our findings. This also means that our results neither support nor discourage the use of procalcitonin guidance beyond what was already proven in the original trial.
3.3. Comparison with Previous Studies
To the best of our knowledge, this is the first study to examine the factors associated with the overruling of procalcitonin-guided initiation of antibiotics for LRTIs in primary care. However, physicians explained the reasons for the overruling of procalcitonin guidance in two trials on procalcitonin-based management: one conducted in hospitals and one in primary care. Physicians justified overruling due to high clinical severity, respiratory or hemodynamic instability, signs of infection or when the patients requested antibiotics themselves. Physicians working in centers experienced with the use of procalcitonin were also more compliant with the recommendation [24,25].
A study on the associations between clinical presentation and antibiotic prescribing for acute exacerbation of chronic obstructive pulmonary disease in primary care, based on CRP measurement, showed that anomalies on lung auscultation were associated with higher odds of antibiotic prescription, while increased age and the presence of heart failure were associated with lower odds [26]. The auscultation and increased age did not appear in our multivariate model, although abnormal auscultation was indeed associated with higher odds of prescription in our univariate analysis; we had only a few patients (n = 0 in one group) with heart failure in our study, preventing us from using this variable in the analysis.
3.4. Implications for Research and Practice
Our results suggest that any future usage of procalcitonin guidance would probably benefit from new guidelines taking into account biomarker values for the management of LRTIs. This would help prevent uncertainty in the physicians’ assessment of patients when the procalcitonin values are at odds with X-ray results or CRP levels.
This study also demonstrated that the overruling of initial procalcitonin guidance was associated with GPs’ socio-demographic characteristics, pointing to the general behavioral problem of overprescription by physicians. A qualitative assessment of the drivers of antibiotic prescription might help identify the most adequate strategies and tools to support behavior change among GPs. In the meantime, continuous medical education on the use of procalcitonin and on “Less is More” approaches to prescribing, as well as communication training, might support the successful implementation of procalcitonin point-of-care tests aimed at antibiotic stewardship. This could lead to a further, clinically significant reduction in antibiotic prescriptions.
4. Materials & Methods
4.1. Study Design and Setting
This ancillary study is a secondary analysis of a three-group, open-label, cluster randomized trial in which 469 patients with LRTIs were included in Swiss primary care practices from September 2018 to March 2020 [8]. The detailed trial protocol has been published [27]. A total of 60 GPs were recruited, and their characteristics recorded. They were randomized to one of three study groups: antibiotics guided by sequential point-of-care procalcitonin and, in patients with elevated procalcitonin (≥0.25 µg/L), point-of-care lung ultrasonography (antibiotics recommended in patients with elevated procalcitonin and the presence of an infiltrate on lung ultrasound: UltraPro group); antibiotics guided by procalcitonin alone (antibiotics recommended in patients with elevated procalcitonin: procalcitonin group); or usual care. GPs could freely order additional diagnostic tests in all groups, as well as manage follow-up visits.
GPs included adult patients (aged 18 or older) presenting to their practice with an LRTI, defined as an acute cough and at least one of the following signs or symptoms: history of fever of more than 4 days, dyspnea, tachypnea (>22 cycles/min), or abnormal focal lung auscultation (i.e., definition of clinical pneumonia) [28]. Exclusion criteria were previously published [27]. Characteristics of the patients (demographics and comorbidities) and of their clinical presentation (symptoms and signs) as well as results of additional diagnostic tests ordered by the GPs outside the scope of the study protocol (chest X-ray, C-reactive protein [CRP]) were recorded by GPs in digitalized case-report forms using REDCap® (Research Electronic Data Capture, Vanderbilt University, Tennessee). Patients had standardized telephone interviews on day 7 and day 28 after the initial consultation to assess if antibiotics had been prescribed. All participants gave their written informed consent. The trial was approved by the Swiss Ethics Committees of the cantons of Vaud and Bern (2017-01246).
4.2. Biological Samples and Retrospective Laboratory Analyses
During the initial consultation of patients from the intervention groups, plasma and nasopharyngeal samples were taken, stored at −20 °C within 6 h, and then moved to −80 °C within 72 h. Plasma was retrospectively analyzed for CRP using enzyme-linked immunosorbent assays (R & D DuoSet®, R & D Systems, Minnesota). Nasopharyngeal swabs were retrospectively analyzed for microbiological pathogens using two multiplex polymerase chain reaction assays: BioFire® FilmArray® Torch, Respiratory Panel 2.1 plus (detection of adenovirus 2/3/6/7/8, coronavirus 229E/HKU1/NL63/OC43/MERS-CoV/SARS-CoV-2, human metapneumovirus, human rhinovirus/enterovirus, influenza A including H1/H3/H1-2009 and B, parainfluenza virus 1/2/3/4, respiratory syncytial virus, Bordetella parapertussis, Bordetella pertussis, Chlamydia pneumoniae, Mycoplasma pneumoniae; BioFire Diagnostics, Salt Lake City, UT, USA) and Bacterial pneumonia CAP FTD-29 (detection of Chlamydia pneumoniae, Haemophilus influenzae, Legionella pneumophila/longbeachae, Moraxella catarrhalis, Mycoplasma pneumoniae, Staphylococcus aureus, Streptococcus pneumoniae; Fast Track Diagnostics, Esch-sur-Alzette, Luxembourg).
We defined a “bacterial or mixed infection” as the identification of at least one respiratory bacterium (cycle threshold < 35 for Haemophilus influenzae, Moraxella catarrhalis, Streptococcus pneumoniae and/or any positive result for Chlamydia pneumoniae, Mycoplasma pneumoniae, Legionella pneumophila/longbeachae), regardless of the identification of viruses, which made the most clinical sense for antibiotic stewardship. We defined a “viral infection” as the identification of at least one respiratory virus but no identification of bacteria.
The GPs were not aware of the CRP value or of the microbiological results.
4.3. Study Population and Outcome
In this secondary analysis, we included all patients of the intervention groups with a low procalcitonin value (<0.25 µg/L) as measured by a point-of-care test during the initial consultation. The outcome was the overruling of the initial procalcitonin-based guidance, defined as an antibiotic prescription by day 28, despite a low procalcitonin value at baseline.
4.4. Statistical Analyses
Descriptive statistics were reported as numbers and percentages for categorical variables or median and interquartile range for continuous variables.
We investigated factors associated with an overruling of procalcitonin guidance by GPs with univariate and multivariate mixed-effects logistic regression models, corrected for clustering within practices with a random effect, using backward selection with the likelihood ratio criterion. Variables included the characteristics of the patients and their clinical presentation (including vital signs with previously defined cut-offs: heart rate ≥ 100/min, respiratory rate > 22/min, systolic blood pressure ≤ 100 mmHg, temperature ≥ 37.8 °C) [27,29], their retrospectively measured CRP and microbiological results, as well as the characteristics of the GPs. The intraclass correlation of the multivariate mixed-effects logistic model was calculated using the π2/3 estimator of the within-cluster variation, i.e., the level-1 variance component of a logistic regression model [30]. Goodness of fit was assessed using the Hosmer–Lemeshow test and the C statistic (area under the receiver operating characteristic curve).
All statistical analyses were performed using R Statistical Software version 4.1.1. p-values < 0.05 were considered statistically significant.
Author Contributions
Conceptualization, L.L., N.B.-B., A.K., V.D., N.S. and Y.M.; Methodology, N.B.-B. and J.K.; Software, J.K. and I.L.; Validation, N.B.-B., J.K. and I.L.; Formal Analysis, J.K. and I.L.; Investigation, L.L., J.-Y.M., O.O., M.-A.P. (Marc-Antoine Perrenoud), M.-A.P. (Marie-Anne Page), K.C.K. and A.M.; Resources, N.B.-B., O.O., M.-A.P. (Marc-Antoine Perrenoud), M.-A.P. (Marie-Anne Page), K.C.K. and A.M.; Data Curation, L.L. and J.K.; Writing—Original Draft Preparation, J.K. and N.B.-B.; Writing—Review & Editing, N.B.-B., L.L., A.K., J.-Y.M., V.D., N.S., Y.M. and I.L.; Visualization, J.K.; Supervision, N.B.-B.; Project Administration, N.B.-B. and L.L.; Funding Acquisition, N.B.-B., K.C.K. and J.K. All authors have read and agreed to the published version of the manuscript.
Funding
This study and the UltraPro trial were supported by: a Swiss National Science Foundation grant [grant number 407240_167133 to N.B.-B.]; a Eureka Eurostars grant [grant number E!113595 to N.B.-B.]; a Leenaards Foundation academic award to N.B.-B.; the Canadian Institutes of Health Foundation [grant number FDN-148439 to K.C.K.]; the Canada Research Chairs program to K.C.K.; a Vaud Academic Society thesis grant to J.K.; and an ECCMID 2022 travel grant to J.K. Brahms GmbH provided the BRAHMS procalcitonin point-of-care tests and readers and had no other role in the study. bioMérieux SA provided an unrestricted grant and had no other role in this study. The funders had no role in considering the study design or in the collection, analysis, interpretation of data, writing of the report, or decision to submit the article for publication.
Institutional Review Board Statement
The UltraPro trial was approved by the Swiss Ethics Committees of the cantons of Vaud and Bern (2017-01246).
Informed Consent Statement
Informed consent was obtained from all subjects involved in the study.
Data Availability Statement
The original UltraPro trial data are publicly available in the Zenodo repository at https://doi.org/10.5281/zenodo.4032527 (accessed on 20 January 2023). The microbiological data used in this study are available on request from the corresponding author; the data are not publicly available due to work still in progress on the microbiological samples database.
Acknowledgments
We thank all the patients who participated and made this study possible; all general practitioners (Allain-Pons Béatrice, Anex Frédéric, Antonini-Revaz Sylvie, Babic Ivana, Baraschi Constantin, Barbanneau-Sadoun Marie-Pierre, Barrois Dallière Olivia, Berdah Daniel, Bielinski Rainer, Blanc Louis, Blanc Müller Muriel, Brändle Carolina, Buff Evelyn, Collipal Tatiana, Cosandey-Tissot Marjorie, Darioli-Pignat Delphine, Dermigny Emilie, Draeyer Juerg, Fumeaux Christophe, Fuzat Laurent, Gennheimer Carl, Gropetti Gwenaël, Grunder Rolf, Hottinger Michaël, Hulliger Stéphane, Huynh Thanh-Liem, Jahns Maximilian, Jaunin-Stalder Nicole, Junod Michel, Keravec Erwan, Klay Michaël, Lombardo Patrick, Maire Ponci Céline, Marbouh Samira, Mayor Vladimir, Miéville Anne, Minnig Adrian, Montaru Rozain Corrine, Morel Abram, Mosimann Madeleine, Neumann Nicolas, Ochs Nicolas, Oppliger Séverine, Pasche Olivier, Paul Sophie, Persoz Christophe, Pierrisnard Marie-Amélie, Potin Mathieu, Rigamonti Véronique, Ronga Alexandre, Sagez Julien, Schwed François, Sibai Karim, Sierro Bard Fabienne, Steiner Anne-Sylvie, Vonnez Jean-Luc, Zahui Angenor, Zaugg Thomas, Zundel Rolf, Zundel-Maurhofer Doris) who enrolled patients in this study; all medical assistants who helped the general practices with the enrollment of patients and the measurement of procalcitonin; Cattaneo Nadia, Delaloye Martin, Ghidoni Céline, Legay Constance, and Vesin Timothée who helped us with patient follow-up; and Hodel Pascale who helped with training general practitioners in canton Bern. Chloé Fischer collected all X-rays in the GP practices.
Conflicts of Interest
The authors report no conflict of interest.
References
- Murray, C.J.; Ikuta, K.S.; Sharara, F.; Swetschinski, L.; Robles Aguilar, G.; Gray, A.; Han, C.; Bisignano, C.; Rao, P.; Wool, E.; et al. Global burden of bacterial antimicrobial resistance in 2019: A systematic analysis. Lancet 2022, 399, 629–655. [Google Scholar] [CrossRef] [PubMed]
- McDonagh, M.S.; Peterson, K.; Winthrop, K.; Cantor, A.; Lazur, B.H.; Buckley, D.I. Interventions to reduce inappropriate prescribing of antibiotics for acute respiratory tract infections: Summary and update of a systematic review. J. Int. Med. Res. 2018, 46, 3337–3357. [Google Scholar] [CrossRef] [PubMed]
- Federal Office of Public Health; Federal Food Safety and Veterinary Office. Swiss Antibiotic Resistance Report 2020. Usage of Antibiotics and Occurrence of Antibiotic Resistance in Switzerland; publication number: 2020-OEG-64; FOPH: Bern, Switzerland, 2020. [Google Scholar]
- Butler, C.C.; Hood, K.; Verheij, T.; Little, P.; Melbye, H.; Nuttall, J.; Kelly, M.J.; Mölstad, S.; Godycki-Cwirko, M.; Almirall, J.; et al. Variation in antibiotic prescribing and its impact on recovery in patients with acute cough in primary care: Prospective study in 13 countries. BMJ 2009, 338, b2242. [Google Scholar] [CrossRef] [PubMed]
- Ieven, M.; Coenen, S.; Loens, K.; Lammens, C.; Coenjaerts, F.; Vanderstraeten, A.; Henriques-Normark, B.; Crook, D.; Huygen, K.; Butler, C.C.; et al. Aetiology of lower respiratory tract infection in adults in primary care: A prospective study in 11 European countries. Clin. Microbiol. Infect. 2018, 24, 1158–1163. [Google Scholar] [CrossRef] [PubMed]
- Self, W.H.; Balk, R.A.; Grijalva, C.G.; Williams, D.J.; Zhu, Y.; Anderson, E.J.; Waterer, G.W.; Courtney, D.M.; Bramley, A.M.; Trabue, C.; et al. Procalcitonin as a Marker of Etiology in Adults Hospitalized With Community-Acquired Pneumonia. Clin. Infect. Dis. 2017, 65, 183–190. [Google Scholar] [CrossRef] [PubMed]
- Kutz, A.; Hausfater, P.; Oppert, M.; Alan, M.; Grolimund, E.; Gast, C.; Alonso, C.; Wissmann, C.; Kuehn, C.; Bernard, M.; et al. Comparison between B·R·A·H·M·S PCT direct, a new sensitive point-of-care testing device for rapid quantification of procalcitonin in emergency department patients and established reference methods—A prospective multinational trial. Clin. Chem. Lab. Med. 2016, 54, 577–584. [Google Scholar] [CrossRef]
- Lhopitallier, L.; Kronenberg, A.; Meuwly, J.-Y.; Locatelli, I.; Mueller, Y.; Senn, N.; D’Acremont, V.; Boillat-Blanco, N. Procalcitonin and lung ultrasonography point-of-care testing to determine antibiotic prescription in patients with lower respiratory tract infection in primary care: Pragmatic cluster randomised trial. BMJ 2021, 374, n2132. [Google Scholar] [CrossRef]
- Groeneveld, G.H.; van ’t Wout, J.W.; Aarts, N.J.; van Rooden, C.J.; Verheij, T.J.M.; Cobbaert, C.M.; Kuijper, E.J.; de Vries, J.J.C.; van Dissel, J.T. Prediction model for pneumonia in primary care patients with an acute respiratory tract infection: Role of symptoms, signs, and biomarkers. BMC Infect. Dis. 2019, 19, 976. [Google Scholar] [CrossRef]
- Siljan, W.W.; Holter, J.C.; Michelsen, A.E.; Nymo, S.H.; Lauritzen, T.; Oppen, K.; Husebye, E.; Ueland, T.; Mollnes, T.E.; Aukrust, P.; et al. Inflammatory biomarkers are associated with aetiology and predict outcomes in community-acquired pneumonia: Results of a 5-year follow-up cohort study. ERJ Open Res. 2019, 5, 00014–2019. [Google Scholar] [CrossRef]
- Hopstaken, R.M.; Muris, J.W.; Knottnerus, J.A.; Kester, A.D.; Rinkens, P.E.; Dinant, G.J. Contributions of symptoms, signs, erythrocyte sedimentation rate, and C-reactive protein to a diagnosis of pneumonia in acute lower respiratory tract infection. Br. J. Gen. Pract. 2003, 53, 358–364. [Google Scholar]
- Teepe, J.; Broekhuizen, B.D.L.; Loens, K.; Lammens, C.; Ieven, M.; Goossens, H.; Little, P.; Butler, C.C.; Coenen, S.; Godycki-Cwirko, M.; et al. Predicting the presence of bacterial pathogens in the airways of primary care patients with acute cough. CMAJ 2017, 189, E50–E55. [Google Scholar] [CrossRef] [PubMed] [Green Version]
- Moore, M.; Stuart, B.; Little, P.; Smith, S.; Thompson, M.J.; Knox, K.; van den Bruel, A.; Lown, M.; Mant, D. Predictors of pneumonia in lower respiratory tract infections: 3C prospective cough complication cohort study. Eur. Respir. J. 2017, 50, 1700434. [Google Scholar] [CrossRef] [PubMed]
- Hopstaken, R.M.; Butler, C.C.; Muris, J.W.M.; Knottnerus, J.A.; Kester, A.D.M.; Rinkens, P.E.L.M.; Dinant, G.-J. Do clinical findings in lower respiratory tract infection help general practitioners prescribe antibiotics appropriately? An observational cohort study in general practice. Fam. Pract. 2006, 23, 180–187. [Google Scholar] [CrossRef] [PubMed]
- Russo, V.; Monetti, V.M.; Guerriero, F.; Trama, U.; Guida, A.; Menditto, E.; Orlando, V. Prevalence of antibiotic prescription in southern Italian outpatients: Real-world data analysis of socioeconomic and sociodemographic variables at a municipality level. CEOR 2018, 10, 251–258. [Google Scholar] [CrossRef] [PubMed]
- Poss-Doering, R.; Kronsteiner, D.; Kamradt, M.; Andres, E.; Kaufmann-Kolle, P.; Wensing, M.; Szecsenyi, J.; Wensing, M.; Kamradt, M.; Poß-Doering, R.; et al. Antibiotic prescribing for acute, non-complicated infections in primary care in Germany: Baseline assessment in the cluster randomized trial ARena. BMC Infect. Dis. 2021, 21, 877. [Google Scholar] [CrossRef]
- Cadieux, G.; Tamblyn, R.; Dauphinee, D.; Libman, M. Predictors of inappropriate antibiotic prescribing among primary care physicians. CMAJ 2007, 177, 877–883. [Google Scholar] [CrossRef]
- Plate, A.; Kronenberg, A.; Risch, M.; Mueller, Y.; Di Gangi, S.; Rosemann, T.; Senn, O. Treatment of urinary tract infections in Swiss primary care: Quality and determinants of antibiotic prescribing. BMC Fam. Pract. 2020, 21, 125. [Google Scholar] [CrossRef]
- Llor, C.; Bjerrum, L.; Munck, A.; Hansen, M.P.; Córdoba, G.C.; Strandberg, E.L.; Ovhed, I.; Radzeviciene, R.; Cots, J.M.; Reutskiy, A.; et al. Predictors for antibiotic prescribing in patients with exacerbations of COPD in general practice. Ther. Adv. Respir. Dis. 2013, 7, 131–137. [Google Scholar] [CrossRef] [PubMed]
- Teepe, J.; Broekhuizen, B.D.L.; Loens, K.; Lammens, C.; Ieven, M.; Goossens, H.; Little, P.; Butler, C.C.; Coenen, S.; Godycki-Cwirko, M.; et al. Disease Course of Lower Respiratory Tract Infection With a Bacterial Cause. Ann. Fam. Med. 2016, 14, 534–539. [Google Scholar] [CrossRef]
- Verheij, T.J.; Cianci, D.; Velden, A.W.; van der Butler, C.C.; Bongard, E.; Coenen, S.; Colliers, A.; Francis, N.A.; Little, P.; Godycki-Cwirko, M.; et al. Clinical presentation, microbiological aetiology and disease course in patients with flu-like illness: A post hoc analysis of randomised controlled trial data. Br. J. Gen. Pract. 2022, 72, e217–e224. [Google Scholar] [CrossRef]
- Petursson, P. GPs’ reasons for “non-pharmacological” prescribing of antibiotics A phenomenological study. Scand. J. Prim. Health Care 2005, 23, 120–125. [Google Scholar] [CrossRef]
- Butler, C.C.; Rollnick, S.; Pill, R.; Maggs-Rapport, F.; Stott, N. Understanding the culture of prescribing: Qualitative study of general practitioners’ and patients’ perceptions of antibiotics for sore throats. BMJ 1998, 317, 637–642. [Google Scholar] [CrossRef] [PubMed]
- Albrich, W.C.; Dusemund, F.; Bucher, B.; Meyer, S.; Thomann, R.; Kuhn, F.; Bassetti, S.; Sprenger, M.; Bachli, E.; Sigrist, T.; et al. Effectiveness and safety of procalcitonin-guided antibiotic therapy in lower respiratory tract infections in “real life”: An international, multicenter poststudy survey (ProREAL). Arch. Intern. Med. 2012, 172, 715–722. [Google Scholar] [CrossRef] [PubMed]
- Burkhardt, O.; Ewig, S.; Haagen, U.; Giersdorf, S.; Hartmann, O.; Wegscheider, K.; Hummers-Pradier, E.; Welte, T. Procalcitonin guidance and reduction of antibiotic use in acute respiratory tract infection. Eur. Respir. J. 2010, 36, 601–607. [Google Scholar] [CrossRef] [PubMed]
- Gillespie, D.; Butler, C.C.; Bates, J.; Hood, K.; Melbye, H.; Phillips, R.; Stanton, H.; Alam, M.F.; Cals, J.W.; Cochrane, A.; et al. Associations with antibiotic prescribing for acute exacerbation of COPD in primary care: Secondary analysis of a randomised controlled trial. Br. J. Gen. Pract. 2021, 71, e266–e272. [Google Scholar] [CrossRef] [PubMed]
- Lhopitallier, L.; Kronenberg, A.; Meuwly, J.-Y.; Locatelli, I.; Dubois, J.; Marti, J.; Mueller, Y.; Senn, N.; D’Acremont, V.; Boillat-Blanco, N. Procalcitonin and lung ultrasonography point-of-care testing to decide on antibiotic prescription in patients with lower respiratory tract infection in primary care: Protocol of a pragmatic cluster randomized trial. BMC Pulm. Med. 2019, 19, 143. [Google Scholar] [CrossRef] [PubMed]
- Woodhead, M.; Blasi, F.; Ewig, S.; Garau, J.; Huchon, G.; Ieven, M.; Ortqvist, A.; Schaberg, T.; Torres, A.; van der Heijden, G.; et al. Guidelines for the management of adult lower respiratory tract infections—Full version. Clin. Microbiol. Infect. 2011, 17, E1–E59. [Google Scholar] [CrossRef]
- Marchello, C.S.; Ebell, M.H.; Dale, A.P.; Harvill, E.T.; Shen, Y.; Whalen, C.C. Signs and Symptoms That Rule out Community-Acquired Pneumonia in Outpatient Adults: A Systematic Review and Meta-Analysis. J. Am. Board Fam. Med. 2019, 32, 234–247. [Google Scholar] [CrossRef]
- Wu, S.; Crespi, C.M.; Wong, W.K. Comparison of methods for estimating the intraclass correlation coefficient for binary responses in cancer prevention cluster randomized trials. Contemp. Clin. Trials 2012, 33, 869–880. [Google Scholar] [CrossRef] [Green Version]
Figure 1.
Participants’ flowchart.
Figure 2.
Forest plot showing factors associated with an overruling of initial procalcitonin guidance: univariate and multivariate logistic regression models corrected for clustering within practices.
Figure 2.
Forest plot showing factors associated with an overruling of initial procalcitonin guidance: univariate and multivariate logistic regression models corrected for clustering within practices.
Table 1.
Patients’ characteristics, clinical presentation and general practitioners’ characteristics, according to the overruling of initial procalcitonin guidance: descriptive statistics and univariate logistic regression models. Values are n (% of the corresponding total number), unless stated otherwise. Odds ratios and p-values were calculated from univariate mixed-effects logistic regression and corrected for clustering.
Table 1.
Patients’ characteristics, clinical presentation and general practitioners’ characteristics, according to the overruling of initial procalcitonin guidance: descriptive statistics and univariate logistic regression models. Values are n (% of the corresponding total number), unless stated otherwise. Odds ratios and p-values were calculated from univariate mixed-effects logistic regression and corrected for clustering.
| Total Number | All Patients | No Overruling of Guidance | Overruling of Guidance | OR [95% CI] | p-Value |
|---|---|---|---|---|---|
| N = 301 | 206 (68%) | 95 (32%) | |||
| Demographics and comorbidities | |||||
| Female | 184 (61) | 129 (63) | 55 (58) | 0.84 [0.50, 1.42] | 0.52 |
| Age ≥ 65 years | 88 (29) | 58 (28) | 30 (32) | 1.31 [0.74, 2.32] | 0.35 |
| Active smoker | 62 (21) | 38 (18) | 24 (26) | 1.47 [0.79, 2.74] | 0.23 |
| Any comorbidity among the following | 84 (28) | 63 (32) | 21 (23) | 0.65 [0.35, 1.19] | 0.16 |
| Diabetes | 16 (5) | 11 (5) | 5 (5) | 0.99 [0.31, 3.14] | 0.99 |
| Chronic obstructive pulmonary disease | 17 (6) | 12 (6) | 5 (5) | 0.96 [0.30, 3.00] | 0.94 |
| Asthma | 52 (17) | 41 (20) | 11 (12) | 0.52 [0.25, 1.12] | 0.09 |
| Heart failure | 4 (1) | 4 (2) | 0 (0) | - | - |
| Other * | 4 (1) | 4 (2) | 0 (0) | - | - |
| Clinical presentation | |||||
| History of fever | 194 (65) | 128 (62) | 66 (70) | 1.37 [0.78, 2.40] | 0.27 |
| History of dyspnea | 205 (69) | 144 (70) | 61 (66) | 0.84 [0.47, 1.48] | 0.54 |
| History of chest pain | 140 (47) | 87 (43) | 53 (56) | 1.74 [1.03, 2.93] | 0.04 |
| History of sputum production | 209 (70) | 143 (70) | 66 (70) | 1.06 [0.60, 1.87] | 0.85 |
| Heart rate ≥ 100/min | 47 (16) | 30 (15) | 17 (18) | 1.29 [0.64, 2.58] | 0.48 |
| Respiratory rate > 22/min | 49 (16) | 36 (18) | 13 (14) | 0.87 [0.40, 1.87] | 0.72 |
| Systolic blood pressure ≤ 100 mmHg | 15 (5) | 7 (3) | 8 (9) | 2.21 [0.70, 6.99] | 0.18 |
| Temperature ≥ 37.8 °C | 44 (15) | 25 (12) | 19 (20) | 2.16 [1.04, 4.48] | 0.04 |
| Focal abnormal finding upon lung auscultation | 137 (46) | 86 (42) | 51 (54) | 1.83 [1.06, 3.16] | 0.03 |
| Management | |||||
| Prescription of chest X-ray by GP | 56 (19) | 26 (13) | 30 (32) | 3.79 [1.85, 7.76] | <0.001 |
| Biomarkers | |||||
| CRP measured by GP | 97 (33) | 51 (25) | 46 (49) | 2.77 [1.58, 4.86] | <0.001 |
| CRP value (measured retrospectively) | |||||
| ≤20 mg/L | 159 (58) | 116 (62) | 43 (48) | ref | ref |
| 20.01–100 mg/L | 99 (36) | 64 (34) | 35 (39) | 1.66 [0.92, 2.99] | 0.09 |
| >100 mg/L | 18 (7) | 7 (4) | 11 (12) | 4.59 [1.52, 13.90] | 0.007 |
| Microbiology | |||||
| Viral infection | 117 (41) | 79 (40) | 38 (43) | ref | ref |
| Bacterial or mixed infection | 112 (39) | 75 (38) | 37 (42) | 1.19 [0.65, 2.19] | 0.58 |
| No pathogen identified | 56 (20) | 42 (21) | 14 (16) | 0.75 [0.35, 1.60] | 0.45 |
| General practitioner’s characteristics | |||||
| Francophone | 274 (91) | 190 (92) | 84 (88) | 0.66 [0.22, 1.96] | 0.45 |
| >5 general practitioners in practice | 89 (30) | 57 (28) | 32 (34) | 1.37 [0.64, 2.94] | 0.42 |
| Location of practice in an urban setting | 187 (62) | 121 (59) | 66 (70) | 1.47 [0.72, 3.01] | 0.29 |
| Years of experience in practice (median [IQR]) | 9.0 [5.0, 19] | 8.0 [5.0, 19] | 10 [5.0, 19] | 1.02 [0.98, 1.07] | 0.35 |
OR: Odds ratio. 95% CI: 95% confidence interval. GP: General practitioner. CRP: C-reactive protein. IQR: Interquartile range. ref: Reference value for logistic regression. *: Active cancer, chronic kidney disease, or human immunodeficiency virus infection.
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content. |
© 2023 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https://creativecommons.org/licenses/by/4.0/).
Share and Cite
MDPI and ACS Style
Knüsli, J.; Lhopitallier, L.; Kronenberg, A.; Meuwly, J.-Y.; Opota, O.; Perrenoud, M.-A.; Page, M.-A.; Kain, K.C.; Mamin, A.; D’Acremont, V.; et al. Overruling of Procalcitonin-Guided Antibiotics for Lower Respiratory Tract Infections in Primary Care: Ancillary Study of a Randomized Controlled Trial. Antibiotics 2023, 12, 377. https://doi.org/10.3390/antibiotics12020377
AMA Style
Knüsli J, Lhopitallier L, Kronenberg A, Meuwly J-Y, Opota O, Perrenoud M-A, Page M-A, Kain KC, Mamin A, D’Acremont V, et al. Overruling of Procalcitonin-Guided Antibiotics for Lower Respiratory Tract Infections in Primary Care: Ancillary Study of a Randomized Controlled Trial. Antibiotics. 2023; 12(2):377. https://doi.org/10.3390/antibiotics12020377
Chicago/Turabian StyleKnüsli, José, Loïc Lhopitallier, Andreas Kronenberg, Jean-Yves Meuwly, Onya Opota, Marc-Antoine Perrenoud, Marie-Anne Page, Kevin C. Kain, Aline Mamin, Valérie D’Acremont, and et al. 2023. "Overruling of Procalcitonin-Guided Antibiotics for Lower Respiratory Tract Infections in Primary Care: Ancillary Study of a Randomized Controlled Trial" Antibiotics 12, no. 2: 377. https://doi.org/10.3390/antibiotics12020377
Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.
