Abstract
Introduction
Rapid pathogen identification in severe infectious keratitis is critical for targeted therapy to prevent vision loss, but conventional methods are slow and only moderately sensitive. Our objective was to evaluate the diagnostic accuracy and turnaround time of combined FilmArray® Meningitis-Encephalitis (ME)/Blood Culture Identification (BCID) multiplex polymerase chain reaction (PCR) panels versus standard microbiology (culture + pathogen-specific PCR) in severe infectious keratitis.
Methods
This was a prospective pilot diagnostic study at a French tertiary center (July 2023–April 2025) including 23 adults with severe microbial keratitis. Corneal samples were analyzed using both conventional microbiological testing (culture and pathogen-specific PCR) and rapid multiplex PCR with the FilmArray® ME and BCID panels. Primary outcomes were diagnostic agreement between the FilmArray® multiplex PCR system and conventional microbiological methods, measured by Cohen’s κ coefficient. Turnaround times were compared using the Wilcoxon signed-rank test.
Results
Among 23 adult patients (mean age 61.0 [SD 21.2] years, range 24–99; 13 female patients [56.5%]), conventional microbiology identified pathogens in 16 cases (69.5%) versus 15 (65.2%) with multiplex PCR. Overall diagnostic agreement was moderate (κ = 0.50; 95% CI, 0.25–0.76), with perfect concordance (12/12, 100%) for monomicrobial infections detected by both methods. Multiplex PCR significantly reduced mean time to identification (2 h [no variation]) versus conventional methods (102 [SD 42.5] h; P < .001). Both methods were negative in six patients (26.1%) with comparable clinical severity.
Conclusion
Combined FilmArray® ME/BCID panels demonstrated complete concordance with standard microbiology for monomicrobial keratitis and reduced turnaround times by > 100 h. This strategy may accelerate targeted therapy, potentially improving visual outcomes.
Supplementary Information
The online version contains supplementary material available at 10.1007/s40123-025-01303-y.
Keywords: Fortified antibiotic, Cornea, Infectious keratitis, Microbial keratitis, Rapid multiplex PCR system, FilmArray
Plain Language Summary
Infectious keratitis is a painful and potentially serious eye infection that can eventually lead to significant vision loss. The agents responsible for these infections are most often bacteria, but fungi or parasites can cause similar symptoms, particularly in contact lens wearers. In order to identify the causative organism, samples are taken from the infected area for microbiological analysis. The treatment of severe infectious keratitis involves the intensive instillation of high doses of local antibiotics, sometimes during hospitalization, in order to limit the progression of the infection as much as possible. This local treatment will then be adapted to the germ identified by the microbiology laboratory. Severe infectious keratitis is frequently challenging to manage due to the inability to identify the causative pathogen in a substantial proportion of cases. To address this issue, we evaluated a novel microbiological detection approach utilizing multiplex polymerase chain reaction (PCR; FilmArray®, BioMérieux, France) in patients with severe infectious keratitis, aiming to expedite targeted therapy initiation. In this pilot study involving 23 patients, conventional microbiological methods identified the causative pathogens in 16 cases (69.5%), compared with 15 (65.2%) detected using the FilmArray® system, showing acceptable concordance between the two methods. The main advantage of FilmArray® is its shorter turnaround time. It provides results in about 2 hours, compared with 102 hours for traditional techniques. These promising results suggest that rapid molecular diagnostics may facilitate earlier adjustment of topical antibiotic therapy, potentially enhancing visual outcomes after such an infection. Nevertheless, validation in larger cohorts is required, and further research should assess the potential impact on long-term visual function.
Supplementary Information
The online version contains supplementary material available at 10.1007/s40123-025-01303-y.
Key Summary Points
| Severe infectious keratitis is frequently challenging to manage because the causative pathogen cannot be identified in a significant proportion of patients. |
| Multiplex polymerase chain reaction (PCR) showed moderate overall agreement with standard microbiology (κ = 0.50; 95% CI, 0.25–0.76) but 100% concordance for monomicrobial infections. |
| Time to identification was reduced from 102 h (using conventional microbiology, SD 42.5) to 2 h (no variation) (P<0.001). |
| For Pseudomonas aeruginosa, specificity and positive predictive value (PPV) were high (94.1% and 85.7%), while Streptococcus pneumoniae showed slightly lower PPV (50%) despite excellent sensitivity and negative predictive value (NPV). Herpes simplex virus 1 (HSV-1)/varicella-zoster virus (VZV) and Streptococcus spp. were consistently detected with 100% agreement across all parameters. |
| Combined multiplex PCR enables same-day pathogen detection in monomicrobial keratitis, potentially accelerating sight-saving therapy. |
| Larger clinical trials are required for the widespread adoption of this procedure in ophthalmology. |
Introduction
Microbial keratitis is a leading cause of blindness-inducing disease worldwide. Severe cases currently represent the most common reason for a hospitalization exceeding 5 days in the ophthalmology ward [1]. The incidence of infectious keratitis has been estimated at 2.5–799 cases per 100,000 population/year, with approximately 1.5–2 million cases annually [2, 3]. The spectrum of causative pathogens varies by continent, country, and climate [4, 5]. Although slit-lamp examination combined with fluorescein staining quickly confirms the diagnosis of keratitis based on symptoms of redness, photophobia, decreased vision, and severe pain in the eye [6], the determination of the microbial pathogen involved remains far more challenging and time-consuming. Routinely performed cultures and specific polymerase chain reactions (PCRs) on a corneal scraping usually require a few days to confirm the pathogen involved—mainly bacteria (Staphylococcus, Streptococcus, and Pseudomonas), but also fungi, viruses, and parasites [7, 8]. Corneal sampling may yield false-negative results due to reduced microbial inoculum, which varies with etiology and abscess size. Pain is also a limiting factor because it can impede adequate scraping, and pre-instillation of topical anesthetics can reduce PCR sensitivity [9]. Therefore, the time delay involved in these microbiological investigations could be deleterious, since the prognosis of severe microbial keratitis is correlated with prompt commencement of an efficient treatment [10, 11]. In the case of severe infectious keratitis, fortified topical antibiotics are commonly initiated hourly after the corneal scraping, with optional loading dose every 5–15 min [6, 12]. Therapy is then tailored to microbiology. The development of a specific rapid multiplex PCR system, as it is done for bacteremia or central nervous system infections, could be cost-saving and beneficial to the prognosis of microbial keratitis. Similar techniques have already been used in ophthalmology for endophthalmitis [13]. To the best of our knowledge, this rapid multiplex PCR methodology was first implemented for bacterial keratitis in our recently published case report [14]. In this pilot study, we prospectively assessed the performance—specifically concordance and turnaround time—of a rapid multiplex PCR strategy that integrates two widely available diagnostic panels for the microbiological diagnosis of severe infectious keratitis.
Patients and Methods
Participants
This study (ABCORFILM, NCT05888987) was prospectively conducted at a single institution (University Hospital of Reims, France) from July 1, 2023, to April 1, 2025. All patients aged ≥ 18 years and hospitalized in the ophthalmology ward for severe keratitis were consecutively included. Infectious keratitis was considered severe when at least one of the following criteria for the “1, 2, 3” rule was satisfied: anterior chamber cells ≥ 1 +; infiltrate ≥ 2 mm; ≤ 3 mm from center [15]. Suspected cases of severe herpetic viral keratitis, such as linear endotheliitis, extensive marginal ulcer, or necrotizing stromal keratitis, were also included. Patients who had non-severe keratitis or who had received antibiotics for more than 48 h before hospitalization were not included in this study according to the protocol. Clinical data were extracted from medical records.
Microbiological Data
First, two corneal specimens—one via spatula scraping and one via eSwab® swab—were collected per patient according to our standard protocol. The samples underwent standard-of-care testing: Gram staining for direct microscopy, bacterial/fungal cultures, and commercial PCR assays for viruses (herpes simplex virus 1 [HSV-1], herpes simplex virus 2 [HSV-2], varicella-zoster virus [VZV], BioFire® R-GENE™, bioMérieux) and Acanthamoeba (Acanthamoeba real-time PCR Kit, Vircell, Madrid, Spain) [7]. All standard testing was performed during routine laboratory hours (3 times/week).
Then, a second corneal swab (eSwab®) was taken on the same eye to perform rapid multiplex PCR assays. Due to the lack of a specific panel designed for all pathogens involved in microbial keratitis, two different FilmArray® PCR systems (Meningitis-Encephalitis [ME] and Blood Culture Identification [BCID] panels, BioMérieux, Lyon, France), based on a nested multiplex PCR and used internally for other indications, were performed in parallel to gather the main pathogens involved in microbial keratitis (Fig. 1).
Fig. 1.
Summary of our FilmArray® strategy. A List of pathogens and antimicrobial resistance markers tested combining the two commercially available meningitis-encephalitis (ME) and blood culture identification (BCID) panels. B From sampling to FilmArray testing, the ME and BCID panels were tested in parallel using two different machines
Clinical Management
Given the absence of international guidelines concerning the management of severe bacterial corneal ulcer, and following French recommendations [1], patients received a combination of local fortified antibiotic therapy with ticarcillin (5 mg/mL)/amikacin (40 mg/mL)/vancomycin (50 mg/mL), started at a rate of 1 drop/h day and night upon admission of the patient to our department and after microbiological sampling. In cases of suspected severe infectious keratitis linked to the herpes virus family, intravenous administration of acyclovir (10 mg/kg/8 h) or oral administration of valacyclovir (3 g/day) was initiated.
Ethics Statement
This study involving human participants was approved by the “Comité de protection des personnes Sud-Mediterranée II” (Ethics approval ID: 2023-A00521-44). In accordance with the principles of the Declaration of Helsinki, written consent was obtained from all participants before study enrollment.
Statistical Analysis
Inter-method agreement was assessed using Cohen’s κ (unweighted). The positive percent agreement (PPA), negative percent agreement (NPA), positive predictive value (PPV), and negative predictive value (NPV) of FilmArray® for each pathogen were reported, using culture as the reference for bacteria and conventional PCR for viruses.
Turnaround times were compared with the Wilcoxon signed-rank test (reported as mean ± SD). Statistical significance was defined as P < 0.05. Analyses were performed using GraphPad Prism 8.0 (GraphPad Software, San Diego, CA, USA) and MedCalc® 22.007 software (MedCalc Software Ltd, Ostend, Belgium).
Results
Twenty-three patients were enrolled in this study. The mean age at diagnosis was 61 years (± 21.2; range: 24–99), with a slight female predominance (n = 13; 56.5%). A substantial proportion of cases were associated with contact lens wear (n = 9; 39.1%). A history of recent ocular surgery was noted in three patients (13.0%). The mean diameter of the corneal infiltrates was 2.58 mm (± 2.13, standard deviation [SD]). The presence of endothelial decompensation and hypopyon were observed in 14 (60.9%) and four cases (17.4%), respectively (online supplementary material).
Conventional methods detected bacterial/viral pathogens in 16/23 cases (69.5%), with no amoebic DNA identified. The most frequently detected pathogen was Pseudomonas aeruginosa (n = 6), followed by HSV-1 (n = 3). Bacterial co-infection was evidenced in one case (Table 1).
Table 1.
Comparison of the results of our FilmArray® strategy with those obtained by culture and conventional PCR in patients with severe infectious keratitis
| Patient no. | Standard of care (SOC) | FilmArray results | ||||||
|---|---|---|---|---|---|---|---|---|
| Culture results | Conventional PCR | |||||||
| Pathogens | Time to growth (h) | Pathogens | Time of detection (h) | Pathogens | Panel | Time of detection (h) | ||
| 1 | Pseudomonas aeruginosa | 96 | Neg | Pseudomonas aeruginosa | BCID | 2 | FilmArray® +/SOC + | |
| 2 | Sterile | 240 | HSV-1 | 96 | HSV-1 | ME | 2 | FilmArray® +/SOC + |
| 3 | Pseudomonas aeruginosa | 72 | Neg | Pseudomonas aeruginosa | BCID | 2 | FilmArray® +/SOC + | |
| 4 | Pseudomonas aeruginosa | 96 | Neg | Pseudomonas aeruginosa | BCID | 2 | FilmArray® +/SOC + | |
| 5 | Sterile | 240 | VZV | 96 | VZV | ME | 2 | FilmAray® +/SOC + |
| 6 | Streptococcus pneumoniae | 96 | Neg | Streptococcus pneumoniae | ME | 2 | FilmArray® +/SOC + | |
| 7 | Sterile | 240 | Neg | Pseudomonas aeruginosa | BCID | 2 | FilmArray® +/SOC − | |
| 8 | Sterile | 240 | HSV-1 | 192 | HSV-1 | ME | 2 | FilmArray® +/SOC + |
| 9 | Pseudomonas aeruginosa | 72 | Neg | Pseudomonas aeruginosa | BCID | 2 | FilmArray® +/SOC + | |
| 10 | Sterile | 240 | Neg | Neg | 2 | FilmArray® −/SOC − | ||
| 11 | Staphylococcus epidermidis | 96 | Neg | Neg | 2 | FilmArray® −/SOC + | ||
| 12 | Sterile | 240 | Neg | Neg | 2 | FilmArray® −/SOC − | ||
| 13 | Streptococcus dysgalactiae | 48 | Neg | Streptococcus spp. | BCID | 2 | FilmArray® +/SOC + | |
| 14 | Sterile | 240 | Neg | Neg | 2 | FilmArray® −/SOC − | ||
| 15 | Pseudomonas aeruginosa | 120 | Neg | Pseudomonas aeruginosa | BCID | 2 | FilmArray® +/SOC + | |
| 16 | Klebsiella oxytoca + Pseudomonas aeruginosa | 96 | Neg | Pseudomonas aeruginosa | BCID | 2 | FilmArray® +/SOC + | |
| 17 | Staphylococcus epidermidis | 144 | Neg | Neg | 2 | FilmArray® −/SOC + | ||
| 18 | Moraxella catarrhalis | 72 | Neg | Streptococcus pneumoniae | ME | 2 | Mixed | |
| 19 | Sterile | 240 | Neg | Neg | 2 | FilmArray® −/SOC − | ||
| 20 | Sterile | 240 | Neg | Neg | 2 | FilmArray® −/SOC − | ||
| 21 | Sterile | 240 | Neg | Neg | 2 | FilmArray® −/SOC − | ||
| 22 | Moraxella catarrhalis | 192 | Neg | Haemophilus influenzae | ME | 2 | Mixed | |
| 23 | Staphylococcus aureus | 48 | HSV-1 | 48 | HSV-1 | BCID | 2 | FilmArray® +/SOC + |
SOC standard of care, PCR polymerase chain reaction, Neg negative, HSV-1 herpes simplex virus 1, VZV varicella-zoster virus, ME meningitis-encephalitis, BCID blood culture identification
The FilmArray® multiplex PCR system combining ME and BCID panels detected a bacterial and/or viral pathogen in 15 out of 23 patients (65.2%). The most frequently detected pathogen was Pseudomonas aeruginosa (n = 7), followed by HSV-1 (n = 3). No bacterial co-infection or viral plus bacterial co-infection was observed.
Cohen’s κ demonstrated moderate agreement between conventional methods and FilmArray® (κ = 0.50; 95% CI, 0.25–0.76). Overall concordance between FilmArray® and standard-of-care (SOC) diagnostics in identifying the predominant pathogen was 52.2% (12/23). This included 100% concordance among patients with a single pathogen detected by both methods (12/12). In one additional case, Pseudomonas aeruginosa was detected by FilmArray® but not identified by SOC (BioFire +/SOC −; 4.3%). Conversely, two cases of Staphylococcus epidermidis were isolated by culture but not detected by FilmArray® (BioFire −/SOC +; 8.7%) (Table 1).
Both diagnostic approaches returned negative results in six patients (26.1%). These patients appeared to have a smaller corneal abscess, but this was not significant (mean diameter: 2.30 mm ± 1.14 versus 3.36 mm ± 2.05, P = 0.20) (data not shown). Among them, one case of fungal filamentous keratitis was strongly suspected by in vivo confocal microscopy examination of the cornea.
The time to microbial identification was reduced with FilmArray®, delivering results in a mean of 2 h (± 0) across all samples (including preparation of the samples and analysis), whereas culture-based methods and conventional PCR required a mean of 102 h (± 42.5) (P < 0.0001) (Fig. 2).
Fig. 2.
Analysis of the concordance between results obtained by BioFire FilmArray® and SOC. SOC standard of care, BioFire BioFire FilmArray®
The PPA, NPA, PPV, and NPV were calculated for each bacterial or viral pathogen, using culture and conventional PCR as reference standard. FilmArray® demonstrated excellent diagnostic performance, with 100% sensitivity and NPV across all major pathogens tested present in the panel. For Pseudomonas aeruginosa, specificity and PPV were high (94.1% and 85.7%), while Streptococcus pneumoniae showed lower PPV (50%) despite excellent sensitivity and NPV. HSV-1/VZV and Streptococcus spp. were consistently detected with 100% agreement across all parameters. In contrast, the detection was limited for Staphylococcus epidermidis, Klebsiella oxytoca, Moraxella catarrhalis (not included in the panels), and Staphylococcus aureus (PPA 0%, with NPV 95.7%). Finally, Haemophilus influenzae was not detected despite a high NPA of 95.7% (Table 2).
Table 2.
Diagnostic performance of our FilmArray panels compared to SOC based on detected pathogens
| Pathogens | PPA (%) | NPA (%) | PPV (%) | NPV (%) |
|---|---|---|---|---|
| Bacteria | ||||
| Pseudomonas aeruginosa | 100 | 94.12 | 85.71 | 100 |
| Streptococcus pneumoniae | 100 | 95.45 | 50 | 100 |
| Haemophilus influenzae | § | 95.65 | 0 | 100 |
| Streptococcus spp. | 100 | 100 | 100 | 100 |
| Staphylococcus epidermidis | 0 | 100 | – | 95.65 |
| Klebsiella oxytoca | 0 | 100 | – | 95.65 |
| Moraxella catarrhalis* | 0 | 100 | – | 95.65 |
| Staphylococcus aureus | 0 | 100 | – | 95.65 |
| Viruses | ||||
| HSV-1 | 100 | 100 | 100 | 100 |
| VZV | 100 | 100 | 100 | 100 |
SOC standard of care, HSV-1 herpes simplex virus 1, VZV Varicella-zoster virus
§Cannot be determined as the organism never grew on culture
−Cannot be determined as the organism was not detected by FilmArray
*Bacteria not included in the panels
Discussion
In this prospective pilot study, we compared the diagnostic performance of a combined FilmArray® multiplex PCR panel with that of standard microbiological methods in severe microbial keratitis. The panel integrates two commercially available assays targeting principal pathogens implicated in this sight-threatening condition [16, 17]. Among monomicrobial infections, we observed 100% concordance between the FilmArray® and conventional methods, confirming its accuracy in detecting dominant pathogens. These findings align with prior studies reporting comparable positivity rates for the FilmArray® system in blood culture [18, 19]. Notably, the panel demonstrated good clinical specificity with no significant false-positive signals, minimizing overdiagnosis risk. However, off-panel bacterial pathogens were identified in four of 23 cases (17.4%) by conventional culture. In two cases, Moraxella catarrhalis was cultured while the multiplex PCR detected respiratory pathogens (Streptococcus pneumoniae or Haemophilus influenzae), suggesting either polymicrobial infection or potential misattribution of the causative agent. These cases of suspected polymicrobial infections account for 2–15% of infectious keratitis cases in the literature and consistently affected patients with chronic ocular surface disease in our study [2]. Furthermore, this discrepancy highlights the need for critical evaluation of pathogenicity when comparing PCR-based and culture-based methods. Conversely, in one case, Pseudomonas aeruginosa was detected exclusively by FilmArray®. Furthermore, two cases of Staphylococcus epidermidis, identified by culture, were not detected by FilmArray® (FilmArray −/SOC +). In these two specific clinical cases, these isolated positive culture results could be interpreted as false positives related to contamination during sampling, given the low clinical and biological concordance observed. In terms of diagnostic accuracy, FilmArray® demonstrated excellent performance, with 100% sensitivity and 100% negative predictive value (NPV) for the major pathogens tested. For Pseudomonas aeruginosa, the system showed high specificity (94.1%) and positive predictive value (PPV) (85.7%). However, the PPV for Streptococcus pneumoniae was lower (50%), despite excellent sensitivity and excellent NPV, probably due to the limited number of cases (including one suspected co-infection). For HSV-1, VZV, and Streptococcus spp., FilmArray® exhibited 100% agreement across all diagnostic parameters.
Nevertheless, the detection of Staphylococcus epidermidis, Klebsiella oxytoca, Moraxella catarrhalis, and Staphylococcus aureus by FilmArray® was more limited. Staphylococcus epidermidis was not detected by FilmArray® despite its isolation by culture (PPA 0%, NPV 95.7%). These findings indicate that certain pathogens, especially those rarely implicated in keratitis or absent from the FilmArray® panels, may not be detected by this approach. Due to the lack of international consensus on the treatment of severe bacterial keratitis [12], patients treated in our department received empirical treatment combining enhanced antibiotics based on ticarcillin (5 mg/mL), amikacin (40 mg/mL), and vancomycin (50 mg/mL) instilled hourly during the first 48 h of treatment. Treatment is then adjusted based on culture and antibiogram results and consists primarily of a topical combination of fluoroquinolone, rifamycin, or aminoglycoside. The American Academy of Ophthalmology recently reported that treatment with a single fluoroquinolone drug has been shown to be as effective as combination therapy, but that enhanced combination therapy is an alternative to consider, particularly for severe infections caused by highly virulent organisms such as Pseudomonas aeruginosa, Streptococcus pneumoniae, or Neisseria gonorrhoeae [6, 20]. However, the results of conventional microbiological diagnostic techniques are known on average between 2 and 3 days later, which could constitute a significant delay in treatment. Compliance with these new recommendations would therefore be facilitated by early identification of the causative organism, which would allow for better targeting of patients requiring aggressive treatment. With this in mind, several teams are currently developing molecular biology tools that comply with this paradigm [21, 22]. The development of a panel targeting the main pathogens responsible for severe keratitis would enable the diagnostic performance of these methods to be evaluated, including for the detection of rare pathogens such as Acanthamoeba or certain fungi (yeasts or filamentous fungi).
In our study the turnaround time was dramatically shorter with the FilmArray® multiplex PCR panel than with standard microbiology (2 h versus 102 h, respectively; P < 0.0001), taking care to only consider identification times for positive results for the SOC group so as not to disadvantage it. Therefore, the rapid identification of the causative pathogen using a multiplex PCR platform may enable earlier and more targeted antimicrobial treatment [22, 23], thereby reducing the time delay before appropriate treatment, shortening the use of broad-spectrum antibiotics, and minimizing ocular toxicity associated with inappropriate topical treatments [24]. Also, it may decrease the length of stay in the ophthalmology ward, further reducing healthcare cost. Such clinical, epidemiological, and economic impacts should be investigated in a randomized control trial. This pilot study presents several limitations. The prevalence of infections due to Staphylococcus aureus was low, and no fungal or parasitic pathogens were identified, either through conventional diagnostic methods or via our multiplex PCR approach. These findings can be partly attributed to the limited sample size, but more importantly to the inclusion criteria applied. Specifically, only patients who had not received topical antibiotic therapy, or who had received it for less than 48 h, were enrolled, with the aim of accurately assessing the diagnostic performance of multiplex PCR for severe infectious keratitis. Since fungal infections tend to be persistent, are frequently managed initially in primary care settings with prolonged topical antibiotic treatments, and are subsequently referred to tertiary centers due to unfavorable clinical progression, a selection bias is likely to have been introduced. Variability in sample collection techniques across multiple ophthalmologists may introduce heterogeneity in specimen quality, potentially reducing the sensitivity of the diagnostic panel—particularly in early infections or cases with low microbial biomass. In addition, restriction of sample processing to routine office hours may have limited enrollment and introduced selection bias, as specimens requiring urgent analysis outside this time window were excluded. Furthermore, we used the BCID and ME FilmArray® panels, two kits used in hospitals for other indications: although comprehensive, these panels have limited coverage, and the use of the Pneumonia Plus panel would have allowed us to identify Moraxella catarrhalis.
Conclusions
Our results indicate that the FilmArray® panel shows promise as a reliable and efficient diagnostic modality for the early identification of pathogens in cases of severe infectious keratitis. The system exhibits high concordance with conventional diagnostic techniques, particularly for monomicrobial infections, and offers a marked reduction in diagnostic turnaround time. However, additional investigation in the context of a randomized controlled trial is necessary to determine its effect on therapeutic choices and patient outcomes.
Supplementary Information
Below is the link to the electronic supplementary material.
Acknowledgments
Medical Writing/Editorial Assistance
Not applicable.
Author Contributions
Research conception and study design: Thomas Ferreira de Moura, Laurent Andreoletti, Alexandre Denoyer. Data acquisition and research execution: Thomas Ferreira de Moura, Sarah Gabison, Laurent Andreoletti, Alexandre Denoyer. Data analysis and interpretation: Thomas Ferreira de Moura, Sarah Gabison, Yohan N’Guyen, Laurent Andreoletti, Alexandre Denoyer. Manuscript preparation: Thomas Ferreira de Moura, Sarah Gabison, Yohan N’Guyen, Laurent Andreoletti, Alexandre Denoyer. Guarantor: Alexandre Denoyer.
Funding
The University Hospital of Reims (CHU de Reims) contributed to the funding of this study (Grant number AOL-2020). Prof. Andreoletti received funding (Grant number 0001–2024) from the BioMérieux Research Fund, Lyon, France. The journal’s Rapid Service Fee was funded by the University Hospital of Reims (“CHU de Reims”).
Data Availability
The datasets generated during and/or analyzed during the current study are available from the corresponding author on reasonable request.
Declarations
Conflict of interest
Prof. Laurent Andreoletti is a scientific consultant for BioMérieux Molecular systems, France. Thomas Ferreira de Moura, Sarah Gabison, Yohan N’Guyen, and Alexandre Denoyer have nothing to disclose.
Ethical Approval
This study involves human participants and was approved by the Comité de protection des personnes Sud-Mediterranée II (ethics approval ID 2023-A00521-44) and maintained adherence to the Declaration of Helsinki for research involving human subjects. Written consent was obtained from all participants before study enrollment.
Footnotes
Thomas Ferreira de Moura and Sarah Gabison have contributed equally to this work as first authors.
Laurent Andreoletti and Alexandre Denoyer have contributed equally to this work as last authors.
References
- 1.Rapport SFO 2015 - Surface oculaire. Accessed June 3, 2025. https://www.em-consulte.com/em/SFO/2015/html/file_100018.html
- 2.Ting DSJ, Ho CS, Deshmukh R, Said DG, Dua HS. Infectious keratitis: an update on epidemiology, causative microorganisms, risk factors, and antimicrobial resistance. Eye. 2021;35(4):1084–101. 10.1038/s41433-020-01339-3. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Stapleton F. The epidemiology of infectious keratitis. Ocul Surf. 2023;28:351–63. 10.1016/j.jtos.2021.08.007. [DOI] [PubMed] [Google Scholar]
- 4.Sharma P, Chipeta C, O’Kane K, et al. Five-year analysis of microbial keratitis incidence, isolates, and in vitro antimicrobial sensitivity in the South West of England: an epidemiological study. Microorganisms. 2025;13(7):1578. 10.3390/microorganisms13071578. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Prajna VN, Prajna L, Muthiah S. Fungal keratitis: the Aravind experience. Indian J Ophthalmol. 2017;65(10):912. 10.4103/ijo.IJO_821_17. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Rhee MK, Ahmad S, Amescua G, et al. Bacterial keratitis preferred practice pattern®. Ophthalmology. 2024;131(4):P87–133. 10.1016/j.ophtha.2023.12.035.38349295 [Google Scholar]
- 7.Ting DSJ, Ho CS, Cairns J, et al. 12-year analysis of incidence, microbiological profiles and in vitro antimicrobial susceptibility of infectious keratitis: the Nottingham Infectious Keratitis Study. Br J Ophthalmol. 2021;105(3):328–33. 10.1136/bjophthalmol-2020-316128. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Hoffman JJ, Dart JKG, De SK, Carnt N, Cleary G, Hau S. Comparison of culture, confocal microscopy and PCR in routine hospital use for microbial keratitis diagnosis. Eye. 2022;36(11):2172–8. 10.1038/s41433-021-01812-7. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Goldschmidt P, Rostane H, Saint-Jean C, et al. Effects of topical anaesthetics and fluorescein on the real-time PCR used for the diagnosis of Herpesviruses and Acanthamoeba keratitis. Br J Ophthalmol. 2006;90(11):1354–6. 10.1136/bjo.2006.098830. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Kerautret J, Raobela L, Colin J. Kératites bactériennes sévères : étude rétrospective clinique et microbiologique. J Fr Ophtalmol. 2006;29(8):883–8. 10.1016/S0181-5512(06)70108-5. [DOI] [PubMed] [Google Scholar]
- 11.Oldenburg CE, Lietman TM. The challenge of controlling infectious keratitis. Am J Ophthalmol. 2017;176:xv–xvi. 10.1016/j.ajo.2017.01.011. [DOI] [PubMed] [Google Scholar]
- 12.Bourcier T, Koestel E, Bertret C, Yaïci R, Borderie V, Bouheraoua N. Kératites bactériennes : rétrospective et prospective 2024. J Fr Ophtalmol. 2024;47(10):104335. 10.1016/j.jfo.2024.104335. [DOI] [PubMed] [Google Scholar]
- 13.Wang S, Liu Y, Li Y, et al. Rapid, multiplex and automated detection of bacteria and fungi in endophthalmitis via a microfluidic real-time pcr system. J Ophthalmic Inflamm Infect. 2024;14(1):64. 10.1186/s12348-024-00446-6. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.de Ferreira Moura T, Limelette A, Arndt C, Guillard T, Andreoletti L, Denoyer A. First use of a new rapid multiplex PCR system for the microbiological diagnosis and the clinical management of severe infectious keratitis: a case report. Am J Ophthalmol Case Rep. 2022;27:101601. 10.1016/j.ajoc.2022.101601. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Vital MC, Belloso M, Prager TC, Lanier JD. Classifying the severity of corneal ulcers by using the “1, 2, 3” rule. Cornea. 2007;26(1):16–20. 10.1097/ICO.0b013e31802b2e47. [DOI] [PubMed] [Google Scholar]
- 16.Caunedo-Jiménez M, Fernández-Colomer B, Fernández-Suárez J, et al. Clinical utility of the FilmArray® blood culture identification (BCID) panel for the diagnosis of neonatal sepsis. Microorganisms. 2023;11(3):732. 10.3390/microorganisms11030732. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Trujillo-Gómez J, Tsokani S, Arango-Ferreira C, et al. Biofire FilmArray Meningitis/Encephalitis panel for the aetiological diagnosis of central nervous system infections: a systematic review and diagnostic test accuracy meta-analysis. EClin Med. 2022;44:101275. 10.1016/j.eclinm.2022.101275. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Caméléna F, de Péan Ponfilly G, Pailhoriès H, et al. Multicenter evaluation of the FilmArray Blood Culture Identification 2 Panel for pathogen detection in bloodstream infections. Microbiol Spectr. 2023;11(1):e02547-22. 10.1128/spectrum.02547-22. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.García-Rivera C, Parra-Grande M, Merino E, Boix V, Rodríguez JC. Concordance of the filmarray blood culture identification panel 2 and classical microbiological methods in a bacteriemia diagnostic unit. Diagn Microbiol Infect Dis. 2022;104(4):115787. 10.1016/j.diagmicrobio.2022.115787. [DOI] [PubMed] [Google Scholar]
- 20.Garcin T, Hoarau G, Leger V, et al. Streptococcus pneumoniae keratitis: epidemiology and outcomes over 6 years in a French eye care center. Eur J Clin Microbiol Infect Dis. 2025;44(4):963–72. 10.1007/s10096-025-05057-0. [DOI] [PubMed] [Google Scholar]
- 21.Park S, Kim K, Lee Y, Ryoo N. Enhanced detection of bacterial ocular pathogens: a comparative study of broad-range real-time PCR and conventional culture methods. Diagnostics. 2025;15(8):966. 10.3390/diagnostics15080966. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Bispo PJM, Belanger N, Li A, et al. An all-in-one highly multiplexed diagnostic assay for rapid, sensitive, and comprehensive detection of intraocular pathogens. Am J Ophthalmol. 2023;250:82–94. 10.1016/j.ajo.2023.01.021. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Ung L, Belanger NL, Chodosh J, Gilmore MS, Bispo PJM. Novel molecular barcoding for rapid pathogen detection in infectious keratitis. Ophthalmol Sci. 2021;1(4):100066. 10.1016/j.xops.2021.100066. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Lin CP, Boehnke M. Effect of fortified antibiotic solutions on corneal epithelial wound healing. Cornea. 2000;19(2):204–6. 10.1097/00003226-200003000-00014. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Availability Statement
The datasets generated during and/or analyzed during the current study are available from the corresponding author on reasonable request.