Encyclopedia
Please note this is an old version of this entry, which may differ significantly from the current revision.
Subjects:
Infectious Diseases
Pneumonia is a common and severe illness that requires prompt and effective management. Advanced, rapid, and accurate tools are needed to diagnose patients with severe bacterial pneumonia, and to rapidly select appropriate antimicrobial therapy, which must be initiated within the first few hours of care. Two multiplex molecular tests, Unyvero HPN and FilmArray Pneumonia+ Panel, have been developed using the multiplex polymerase chain reaction (mPCR) technique to rapidly identify pathogens and their main antibiotic resistance mechanisms from patient respiratory specimens. Performance evaluation of these tests showed strong correlations with reference techniques.
- rapid tests
- sepsis
- nosocomial pneumonia
- diagnosis
1. Introduction
Severe pneumonia remains a major cause of morbidity and mortality worldwide, and its therapeutic management is a public health issue. The severity of pneumonia is generally defined by clinical criteria leading to admission to an intensive care unit (ICU) [1]. One of them is the need for mechanical ventilation (invasive or non-invasive), or severe hypoxemia defined by a PaO2/FiO2 ratio of less than 300 mmHg, requiring oxygen administration through a high-flow nasal cannula or a non-rebreathing mask. Among hospital-acquired pneumonia (HAP), the most severe and frequent is the ventilator-associated pneumonia (VAP). It is defined as an infection of the lung parenchyma in patients receiving invasive mechanical ventilation for at least 48 h. Conversely, community-acquired pneumonia (CAP) refers to episodes in patients with no recent healthcare exposure. In Europe, the estimated incidence of VAP is 18.3 episodes per 1000 ventilator-days [2,3]. In-ICU mortality attributable to VAP is limited but significant (from 1 to 6% according to case mix and methods), suggesting that mortality in these patients is mainly driven by their underlying conditions, as well as the severity of the disease [3,4,5]. However, VAP has been frequently associated with a longer duration of mechanical ventilation, ICU stay, prolonged hospitalization, and increased healthcare cost [4].
2. Multiplex PCR: Available Tests
Two US Food and Drugs Administration-approved CE-marked tests are currently available for pneumonia diagnosis. The Unyvero HPN (Curetis, Unyvero TM) detects 21 bacteria and one parasite, semi quantitatively (from + to +++), and identifies 15 resistance genes in approximately 5 h. The FilmArray Pneumonia+ Panel (BioFire, bioMérieux) detects 18 bacteria, among which are three atypical ones, quantitatively (from 104 to ≥107 genomic copies/mL), and identifies seven resistance genes, and eight viruses in approximately 90 min. The main targets of both tests are shown in the Table 1.
Table 1. Unyvero HPN and FilmArray Pneumonia+ Panel main targets.
| Unyvero HPN | FilmArray Pneumonia+ Panel | |
|---|---|---|
| Number of targets | 36 | 33 |
| Turnaround time | 5 h | 90 min |
| Type of detection | Semiquantitative (+ to +++) | Quantitative (104 to ≥107) |
| Included pathogens | ||
| Bacteria | ||
| Gram-positive cocci | ||
| Staphylococcus aureus | x | x |
| Streptococcus agalactiae | x | |
| Streptococcus pneumoniae | x | x |
| Streptococcus pyogenes | x | |
| Gram-negative cocci | ||
| Moraxella catarrhalis | x | x |
| Gram-negative bacilli | ||
| Haemophilus influenzae | x | x |
| Group 1 Enterobacterales | ||
| Escherichia coli | x | x |
| Proteus spp. | x | x |
| Group 2 Enterobacterales | ||
| Klebsiella oxytoca | x | x |
| Klebsiella pneumoniae | x | x |
| Klebsiella variicola | x | |
| Group 3 Enterobacterales | ||
| Enterobacter cloacae complex | x | x |
| Citrobacter freundii | x | |
| Enterobacter cloacae complex | x | x |
| Klebsiella aerogenes (Enterobacter aerogenes) | x | x |
| Morganella morganii | x | |
| Serratia marcescens | x | x |
| Non-fermenting bacteria | ||
| Acinetobacter baumannii complex * | x | x |
| Pseudomonas aeruginosa | x | x |
| Stenotrophomonas maltophilia | x | |
| Atypical bacteria | ||
| Chlamydia pneumoniae | x | x |
| Legionella pneumophila | x | x |
| Mycoplasma pneumoniae | x | x |
| Others | ||
| Pneumocystis jirovecii | x | |
| Resistance genes | x | x |
| Virus | x | x |
* Acinetobacter calcoaceticus-baumannii complex for FilmArray Pneumonia+ Panel.
3. Biological Accuracy
3.1. General Comments
Prior to any description of the available data on the diagnostic performance of these new tests, it should be highlighted that some pathogens potentially involved in HAP or VAP are not represented in the panels. In particular, their contribution to the diagnosis of HAP or VAP is hampered by their lack of detection of Hafnia alvei and Citrobacter koseri for both panels, and of Citrobacter freundii, Klebsiella variicola, Morganella morganii and Stenotrophomonas maltophilia for the FilmArray Pneumonia+ panel. Importantly, if the micro-organism is not detected or under the quantification threshold, the result of bacterial resistance genes is then unavailable.
3.2. Qualitative Comparison of mPCR versus Standard-of-Care
3.2.1. Viral and Bacterial Detection
Several studies have evaluated mPCR in pneumonia, revealing good diagnostic performances and high concordance with bacterial culture results. After reviewing bronchoalveolar lavage (BAL) specimens from 259 patients with suspected pneumonia, FilmArray Pneumonia+ Panel demonstrated a positive percentage agreement (PPA) of 96.2% and a negative percentage agreement (NPA) of 98.1% for the qualitative identification of 15 bacterial targets compared to routine bacterial culture. For viruses, the PPA has also done well (96.7%) as compared to monoplex PCR [11]. A recent meta-analysis including 30 observational studies and 8969 samples also revealed high diagnostic performance for the FilmArray Pneumonia+ Panel as compared to a standard culture [12]. The authors pointed out the lower diagnostic performance of mPCR for sputum, due to frequent contamination by oropharyngeal flora, leading to numerous false positive results.
4. Clinical Impact: Published Works
Recent works have tried to determine the optimal place of mPCR in the diagnosis process. A first United Kingdom (UK) study evaluated mPCR from 323 adults with radiologically-confirmed CAP explored mainly by sputum specimens (96%) [31]. Molecular testing achieved pathogen detection in 87% of CAP patients compared with 39% with culture-based methods. Among patients who had received antimicrobials within the 72 h prior to admission (85%), the pathogen detection rate was also clearly higher with mPCR (78%) compared to standard culture (32%; p < 0.001). The authors concluded that molecular testing could have an impact on antibiotic prescribing, with de-escalation in 77% of patients; however, the study was neither controlled nor randomized.
A retrospective multicenter study was conducted in four French university hospitals on 159 pneumonia episodes (HAP, CAP or VAP), with 81% of patients hospitalized in ICUs [32]. On the basis of the mPCR results, the multidisciplinary committee comprising an intensivist, an infectiologist, and a clinical microbiologist proposed a change in empirical treatment in 77% of cases. In microbiologically documented episodes, mPCR increased the appropriateness of empirical treatment to 87%, compared with 77% in routine care.
A single-center randomized controlled trial was conducted in UK on 200 critically ill adults with pneumonia (CAP 42%, HAP 35% and VAP 23%) [34]. Patients were allocated (1:1) to mPCR combined with an antibiotic stewardship strategy, or to routine clinical care. Eighty (80%) of patients in the interventional group received results-directed therapy, which was the primary outcome, compared with 29 (29%) of 99 in the control group (difference of 51%, 95% CI 39–63; p < 0.0001). In the mPCR group, 42% of patients had antibiotics de-escalated compared with 8% in the control group. Despite these major differences in therapeutic strategy, there were no major differences in clinical results or safety between the two groups.
In the particular context of the recent COVID-19 pandemic, another study attempted to evaluate an antibiotic de-escalation algorithm based on the combination of mPCR and procalcitonin (PCT) results in 194 patients who were critically ill with SARS-CoV-2 pneumonia [36]. As expected, respiratory bacterial co-infection rate was higher in mPCR group (45/93, 48.4%) than in standard-of-care group (21/98, 21.4%). The authors were unable to demonstrate a reduction in overall antibiotic exposure or a benefit in terms of clinical outcomes at day 28. These disappointing results from the intention-to-treat analysis could be partly explained by significant protocol deviations during this study, which was conducted in an exceptional context. Indeed, in the per-protocol analysis, the number of antibiotic-free days after randomization was two days higher at day seven in the intervention group than in the control group (4 days vs. 2 days; RR 1.38 (1.01 to 1.88)). Unfortunately, the antibiotics were reintroduced, and the results for antibiotic-free days on day 28 were no longer significant (14 days vs. 15 days; RR 0.98 (0.66 to 1.46)).
5. Conclusions
Multiplex PCR appears as a valuable tool in the management of severe bacterial pneumonia, as it is easy to perform, rapid and sensitive. However, there are some limitations for its use. First, it should not be used alone, and a prior knowledge of the product is required. Appropriate training, compliance with international recommendations and expert help are essential for the proper use of these tests. As always, it is essential to consider the pre-test probability of a particular etiology and resistance pattern in order to make an appropriate decision. Before performing these tests, the hypothesis and the question asked must be clearly stated in order to optimize the interpretation of the results.
In addition, as described above, the ongoing trials are still struggling to show a strong and clinically significant effect. At least partly because it’s a new tool, the technical aspects are not described exhaustively. The interpretation and evolution of copy number per milliliter under treatment, or the identification of a clinically relevant threshold, are areas that still need to be explored. In addition, it may be appropriate to limit the number of targets to a specific panel for each clinical situation, for example the CAP versus HAP panel or the immunocompetent versus immunocompromised patient panel.
Finally, clinical trial results are difficult to transpose to real-life situations, as they are produced by trained expert teams. Education and training of clinicians in microbiology are therefore fundamental for the adoption of these new techniques and their integration into routine practice.
This entry is adapted from the peer-reviewed paper 10.3390/antibiotics13010095
This entry is offline, you can click here to edit this entry!
