January 2025
Volume 14, Issue 1
Open Access
Cornea & External Disease  |   January 2025
Clinical Characteristics and In Vivo Confocal Microscopic Study in Candida Keratitis
Author Affiliations & Notes
  • Yan Peng
    Beijing Institute of Ophthalmology, Beijing Tongren Eye Center, Beijing Tongren Hospital, Capital Medical University, Beijing, China
  • Qiankun Chen
    Beijing Institute of Ophthalmology, Beijing Tongren Eye Center, Beijing Tongren Hospital, Capital Medical University, Beijing, China
  • Yuan Wei
    Beijing Institute of Ophthalmology, Beijing Tongren Eye Center, Beijing Tongren Hospital, Capital Medical University, Beijing, China
  • Leying Wang
    Beijing Institute of Ophthalmology, Beijing Tongren Eye Center, Beijing Tongren Hospital, Capital Medical University, Beijing, China
  • Zijun Zhang
    Beijing Institute of Ophthalmology, Beijing Tongren Eye Center, Beijing Tongren Hospital, Capital Medical University, Beijing, China
  • Zhenyu Wei
    Beijing Institute of Ophthalmology, Beijing Tongren Eye Center, Beijing Tongren Hospital, Capital Medical University, Beijing, China
  • Jinding Pang
    Beijing Institute of Ophthalmology, Beijing Tongren Eye Center, Beijing Tongren Hospital, Capital Medical University, Beijing, China
  • Bo Peng
    Beijing Institute of Ophthalmology, Beijing Tongren Eye Center, Beijing Tongren Hospital, Capital Medical University, Beijing, China
  • Qingquan Shi
    Beijing Institute of Ophthalmology, Beijing Tongren Eye Center, Beijing Tongren Hospital, Capital Medical University, Beijing, China
  • Zhiqun Wang
    Beijing Institute of Ophthalmology, Beijing Tongren Eye Center, Beijing Tongren Hospital, Capital Medical University, Beijing, China
  • Yang Zhang
    Beijing Institute of Ophthalmology, Beijing Tongren Eye Center, Beijing Tongren Hospital, Capital Medical University, Beijing, China
  • Kexin Chen
    Beijing Institute of Ophthalmology, Beijing Tongren Eye Center, Beijing Tongren Hospital, Capital Medical University, Beijing, China
  • Xizhan Xu
    Beijing Institute of Ophthalmology, Beijing Tongren Eye Center, Beijing Tongren Hospital, Capital Medical University, Beijing, China
  • Qingfeng Liang
    Beijing Institute of Ophthalmology, Beijing Tongren Eye Center, Beijing Tongren Hospital, Capital Medical University, Beijing, China
  • Correspondence: Qingfeng Liang, Beijing Institute of Ophthalmology, Beijing Tongren Eye Center, Beijing Tongren Hospital, Capital Medical University, No. 17 Hougou Lane, Chongnei Street, Dongcheng District, Beijing 100005, China. e-mail: [email protected] 
Translational Vision Science & Technology January 2025, Vol.14, 23. doi:https://doi.org/10.1167/tvst.14.1.23
  • Views
  • PDF
  • Share
  • Tools
    • Alerts
      ×
      This feature is available to authenticated users only.
      Sign In or Create an Account ×
    • Get Citation

      Yan Peng, Qiankun Chen, Yuan Wei, Leying Wang, Zijun Zhang, Zhenyu Wei, Jinding Pang, Bo Peng, Qingquan Shi, Zhiqun Wang, Yang Zhang, Kexin Chen, Xizhan Xu, Qingfeng Liang; Clinical Characteristics and In Vivo Confocal Microscopic Study in Candida Keratitis. Trans. Vis. Sci. Tech. 2025;14(1):23. https://doi.org/10.1167/tvst.14.1.23.

      Download citation file:


      © ARVO (1962-2015); The Authors (2016-present)

      ×
  • Supplements
Abstract

Purpose: To clarify the clinical and imaging characteristics of Candida keratitis using in vivo confocal microscopy (IVCM) for improved early diagnosis and management.

Methods: A retrospective study of 40 patients with Candida keratitis at Beijing Tongren Hospital from January 2015 to December 2023 was conducted. Data included demographics, risk factors, clinical assessments, lab tests, and IVCM images. Ex vivo confocal microscopy and methylene blue staining of Candida colonies were also analyzed to complement the findings.

Results: Key risk factors identified were topical steroid use, intraocular surgery history, and systemic diseases. Common clinical signs included multifocal infiltration, cream-colored infiltration, and blurred boundary lesions. IVCM on 37 eyes (92.5%) consistently showed round, highly reflective Candida spores, and short rod-shaped spores in some cases. Spores exhibited two patterns: caviar-like clusters (51.4%) and sand-like dispersion (89.2%). Multifocal infiltration was significantly associated with a greater prevalence of clustered spores (75.0% vs. 33.3%; P < 0.05). Candida pseudo-hyphae appeared as beaded (91.7%) or lotus root-shaped (41.7%), highly reflective structures. These IVCM findings closely matched colony ex vivo confocal microscopy and light microscopy observations. After treatment, 45% of patients required surgery owing to minimal improvement in best-corrected visual acuity. Poor outcomes were linked to cream-colored infiltration, blurred lesions boundaries, hypopyon, high inflammatory cell density, and deep Candida infiltration (P < 0.05). Clustered spores suggested better outcomes, but lacked statistical significance (P > 0.05).

Conclusions: IVCM effectively identifies characteristic spores and pseudo-hyphae in Candida keratitis, facilitating early detection and timely management, particularly in cases with multifocal infiltration and blurred boundary lesions.

Translational Relevance: IVCM works well for the early diagnosis of Candida keratitis, especially in cases of deep corneal stromal infiltration or corneal interface infection after corneal transplantation.

Introduction
The global incidence of ocular infections is on the rise. Filamentous fungal keratitis (FK) is more prevalent in tropical climates, whereas yeast infections are increasingly common in milder regions, particularly among immunocompromised individuals.13 Although the clinical spectrum of filamentous FK is well-documented, data on Candida keratitis are sparse. This infection, caused by Candida species, has an annual incidence of approximately 1 case per 1 million individuals.4 Despite prompt treatment, Candida keratitis remains a clinical challenge, with high surgical intervention rates—76% in Vancouver and 48% in Houston.4,5 
Timely diagnosis and treatment may offer promising prospects for enhancing the prognosis of Candida keratitis.6 The gold standard for diagnosis involves microscopic examination and fungal culture. Nevertheless, fungal culture demonstrates a limited sensitivity rate, with fungal organisms frequently taking weeks to cultivate on culture media.7,8 In addition, the deep localization of infections in the corneal stroma makes diagnosis challenging, especially in postkeratoplasty cases.7 
In vivo confocal microscopy (IVCM), a noninvasive, high-resolution technique, allows real-time assessment of corneal structure and pathological changes. It is highly sensitive and specific for detecting fungal filaments, aiding early FK diagnosis. However, the absence of fungal filaments in Candida species complicates IVCM identification. Although some studies have noted spores or pseudo-hyphae in IVCM images of Candida keratitis,6,9 detailed descriptions are insufficient, leading to potential misdiagnosis risks. 
This study aimed to provide a comprehensive description of Candida keratitis' clinical characteristics, focusing on IVCM features, and supported by ex vivo confocal microscopy (EVCM) and light microscopy images of methylene blue-stained Candida colonies. 
Method
Participants
This study was conducted at the Beijing Institute of Ophthalmology and sanctioned by the Medical Ethics Committee of Beijing Tongren Hospital (TRECKY2015-KY09). A cohort of 40 patients with clinically and laboratory-confirmed Candida keratitis, evidenced by positive results in corneal scrapings, cultures, or histopathology, were included. The study spanned from January 2015 to December 2023, with patient data sourced from the electronic health records and laboratory records database of Beijing Tongren Hospital. Prognostic evaluation included follow-ups 6 six months after stabilization or in the postoperative period, with slit-lamp and IVCM images captured at 1-, 3-, and 6-month intervals. 
Demographics and Clinical Evaluation
Patient data were documented systematically using a standardized protocol that covered demographics, predisposing factors, symptom duration, initial evaluations, time to FK diagnosis, clinical features, treatment regimens, and best-corrected visual acuity (BCVA) before and after treatment, among other measures. 
After measurement of visual acuity, all patients were examined by slit-lamp biomicroscopy. The size of the corneal ulcer and stromal infiltration were measured on corneal photographs with ImageJ software (National Institutes of Health, http://rsb.info.nih.gov/ij) and recorded in millimeters. Large-size infiltration is defined as a corneal ulcer of greater than 6 mm in diameter, and deep infiltration is characterized as involving the posterior one-third of the corneal stroma. The presence or absence of hypopyon was recorded, and the height measured in millimeters. 
Laboratory investigations for Candida keratitis included corneal scrapings followed by optical microscopic observation using Gram and Giemsa staining, as well as cultures on potato dextrose agar plates. These procedures were carried out at the Department of Ocular Microbiology, Beijing Institute of Ophthalmology. Isolate identification was achieved through the matrix-assisted laser desorption/ionization time of flight mass spectrometry system (Bruker, Germany). Corneal tissue pathology and hematoxylin and eosin staining were performed at the Department of Ophthalmic Pathology, Beijing Tongren Hospital. 
IVCM Examination
IVCM images were acquired using the Rostock Cornea Module of the Heidelberg Retina Tomograph III (HRT III/RCM, Heidelberg Engineering, Germany). The device's 670-nm diode laser captured 384 × 384-pixel images over a 400 × 400-µm field. A drop of topical anesthetic (proparacaine hydrochloride, 0.5%) was applied before imaging. Eye positioning was adjusted and monitored via a color CCD camera. Two experienced ophthalmologists conducted IVCM in section mode, capturing images from the central cornea and the standard superior, temporal, inferior, and nasal quadrants, as well as the corneal ulcer and adjacent areas, ultimately acquiring 50 high-quality images per area, for approximately 10 images per layer.10 
To further elucidate the IVCM characteristics of Candida keratitis, EVCM evaluations were also conducted on Candida colonies using the same HRT III/RCM instrumentation. After a 3-day incubation period on potato dextrose agar plates, Candida colonies were stained with methylene blue and inspected under a light microscope, revealing the presence of spores and pseudo-hyphae. The objective was shielded by a sterile Tomo-Cap, and coupling between the cap and lens was facilitated by GelTears (0.2% carbomer 980). Depth-specific images of Candida colonies were captured and qualitatively analyzed for morphological characteristics. 
Image Analysis
Anonymized and representative IVCM images underwent selection by a masked observer, followed by independent analyses conducted by two experienced ophthalmologists. Interobserver discrepancies prompted the involvement of a third ophthalmologist if disparities exceeded 10%. Subsequently, the results obtained from the three independent evaluations were averaged. 
A total of 50 IVCM images representatively displaying Candida spores and 50 depicting Candida pseudo-hyphae were chosen. From each image, either the five largest spores or the five longest pseudo-hyphae were selected for measurement. Lossless image magnification and denoising were executed using Topaz (TopazLabs, Dallas, TX, USA), and ImageJ facilitated precise measurements of spore size, pseudo-hyphae length, tortuosity, and width. Tortuosity was determined by the arc length to chord length ratio (LC) using the formula LC = 1 – L/C, where L and C denote the linear distance and actual length between the two ends, respectively.11 The measurement process and detailed parameters for the size of Candida spores and pseudo-hyphae is illustrated in Figures 1A through 1H and Table 1. For comparison, an equal number of IVCM images from patients with filamentous FK showing hyphae (50 from Fusarium keratitis, 50 from Aspergillus keratitis, and 50 from Alternaria keratitis) were analyzed using the same method to compare with the length, tortuosity, and width of Candida pseudo-hyphae. Typical images are shown in Supplementary Figure 1
Figure 1.
 
Measurement of Candida structures and IC features in IVCM. (A) Original IVCM image displaying hyper-reflective Candida spores. (B) Enhanced, magnified, and denoised image of spores. (C, D) Area of hyper-reflective spores (yellow). (E) Original IVCM image of hyper-reflective Candida pseudo-hyphae. (F) Measurements of pseudo-hyphae lengths (red line) and tortuosity (red line/yellow line). (G, H) Measured widths of pseudo-hyphae. (I) Inactivated DCs. (J) Activated DCs. (K) ICs in a honeycomb distribution. (L) ICs in a nonspecific distribution.
Figure 1.
 
Measurement of Candida structures and IC features in IVCM. (A) Original IVCM image displaying hyper-reflective Candida spores. (B) Enhanced, magnified, and denoised image of spores. (C, D) Area of hyper-reflective spores (yellow). (E) Original IVCM image of hyper-reflective Candida pseudo-hyphae. (F) Measurements of pseudo-hyphae lengths (red line) and tortuosity (red line/yellow line). (G, H) Measured widths of pseudo-hyphae. (I) Inactivated DCs. (J) Activated DCs. (K) ICs in a honeycomb distribution. (L) ICs in a nonspecific distribution.
Table 1.
 
Definition of IVCM Parameters for Candida Keratitis
Table 1.
 
Definition of IVCM Parameters for Candida Keratitis
IVCM inflammatory cell (IC) features in Candida keratitis were defined by the presence or absence of specific characteristics,10,12 as detailed in Table 1 and Figures 1I through 1P. Furthermore, 15 IVCM images exhibiting maximum dendritic cell (DC) density from the epithelial layer to Bowman's layer, and another 15 images with the maximum IC density from the epithelial layer to the endothelial layer, were selected across five detected areas (three images in each area). Cell density was assessed utilizing the Cell Count Software on the HRT3/RCM. The mean value of 15 images was considered the mean corneal DC density or IC density (Table 1). 
Treatment and Prognosis
All patients diagnosed with Candida keratitis initially received topical voriconazole (1%) or natamycin (5%), administered hourly for the first week and subsequently tapered to four times daily over the following 4 weeks. Those with mid to deep stromal or interface infections were additionally prescribed oral VCZ (400 mg twice daily). Concomitant bacterial infections were treated with levofloxacin 0.5% eye drops (Santen Pharmaceutical Co., Ltd., Osaka, Japan), initially administered hourly and then reduced to three times daily once infection controlled. In cases of uncontrolled Candida keratitis with deep stromal infiltration or hypopyon, surgical interventions such as penetrating keratoplasty or deep anterior lamellar keratoplasty were performed, with regular follow-ups for all patients. 
Stable lesions were identified by the presence of corneal nebula, macula, or leukoma, along with the resolution of ciliary congestion and anterior segment inflammation. Active lesions were considered otherwise. The medically cured duration is the time from starting antifungal treatment to the formation of a stable lesion. Good outcomes were defined by stable lesions achieved with medical treatment, whereas poor outcomes required therapeutic keratoplasty owing to progression. 
Statistical Analysis
Statistical analysis was conducted using SPSS 26.0 (SPSS, Chicago, IL, USA). Descriptive statistics included means and standard deviations for continuous variables, and percentages for categorical variables. The Student t test was applied for normally distributed continuous data, and nonparametric tests for non-normally distributed data. Spearman correlation analysis examined the relationships between keratitis duration, Candida infiltration, and the density of ICs and DCs. The χ2 test compared clinical characteristics between patients with different keratitis durations and between those with good and poor outcomes. A P value of less than 0.05 was considered statistically significant. 
Results
Demographics
In this retrospective study, 40 patients with Candida keratitis from January 2015 to December 2023 in Beijing Tongren Hospital were reviewed. 23 males (57.5%) and 17 females (42.5%), aged 23 to 81 years (mean, 55.1 ± 14.8 years), were included. There were 22 cases (55.0%) northern China and 18 (45.0%) from other regions. Keratitis was unilateral in 39 cases and binocular in 1 case. For the bilateral case, the initially symptomatic eye was analyzed. 
Before disease onset, 21 patients (52.5%) had a history of corticosteroid use, and 19 cases (47.5%) had undergone intraocular surgery, primarily keratoplasty (n = 7 [36.8%]) and vitrectomy (n = 4 [21.1%]). Fourteen patients (35.0%) suffered from systemic diseases, including diabetes (n = 6 [42.9%]) and Sjogren's syndrome (n = 2 [14.3%]). Other systemic diseases (one case each) included leukemia, rheumatoid arthritis, systemic lupus erythematosus, hypokalemia, IgA nephropathy and syphilis. Additionally, 13 patients (31.7%) were diagnosed with ocular surface disease, most commonly blepharitis (n = 6 [46.2%]) and glaucoma (n = 3 [23.1%]). Other ocular surface diseases included Stevens–Johnson syndrome (n = 2 [15.4%]), uveitis (n = 1 [7.7%]), and mucous membrane pemphigoid (n = 1 [7.7%]). Ocular trauma was reported in 10 patients (25.0%), and 30 patients (75.0%) had two or more risk factors (Table 2). 
Table 2.
 
Population Characteristics of Candida Keratitis Cases
Table 2.
 
Population Characteristics of Candida Keratitis Cases
Clinical Profiles and Diagnosis
The diagnosis of Candida keratitis in this study ranged from 7 to 360 days (average, 94.2 ± 81.6 days). Early diagnosis, within 1 month, occurred in 12 eyes (30.0%); delayed diagnosis was noted in the remainder. Most cases (23 eyes [57.5%]) were diagnosed within 3 months, although eight eyes (20.0%) were diagnosed after 6 months (Table 2). Initially, 15 eyes (37.5%) were correctly diagnosed with FK, and others were misdiagnosed, including with bacterial keratitis (11 eyes [27.5%]) and herpes simplex keratitis (11 eyes [27.5%]). Before receiving the correct diagnosis, 11 eyes (27.5%) received antiviral therapy, 8 eyes were treated with antibiotics, and 11 eyes received steroids, either alone or combined with antibiotics. 
All patients reported decreased visual acuity or corneal irritation symptoms, such as pain, tearing, and photophobia. Central cornea infiltration was most common (17 eyes [42.5%]), followed by paracentral (13 eyes [32.5%]) and peripheral infiltration (5 eyes [12.5%]). Additionally, nearly complete corneal infiltration was observed in five eyes (12.5%). The average ulcer diameter was 5.5 ± 3.0 mm, with 40.0% having lesions larger than 6 mm and 60.0% within 6 mm. Clinical presentations (Fig. 2) included epithelial defects (32.5%), multifocal infiltration (42.5%), powdery deposits (20.0%), and plaque-like deposits (17.5%). Furthermore, a cream-colored infiltration was seen in 17 eyes (42.5%), cloud-like infiltration in 4 eyes (10.0%), and blurred boundary lesions in 20 eyes (50.0%). Neovascularization was common, occurring in 31 eyes (77.5%), and hypopyon was present in 13 eyes (32.5%). Corneal thinning was noted in 11 eyes, with 1 case progressing to perforation. 
Figure 2.
 
Clinical photograph of typical Candida keratitis signs. (A) Corneal epithelial defect. (B) Multifocal infiltration. (C) Powdery deposits. (D) Plaque-like deposits. (E) Cream-colored infiltration. (F) Cloud-like infiltration. (G) Burred corneal lesion boundary. (H) Neovascularization. (I) Hypopyon.
Figure 2.
 
Clinical photograph of typical Candida keratitis signs. (A) Corneal epithelial defect. (B) Multifocal infiltration. (C) Powdery deposits. (D) Plaque-like deposits. (E) Cream-colored infiltration. (F) Cloud-like infiltration. (G) Burred corneal lesion boundary. (H) Neovascularization. (I) Hypopyon.
The diagnosis of Candida keratitis was confirmed by detecting Candida spores or pseudo-hyphae through corneal scraping, cultures, or pathological examination. Of the 29 cases undergoing corneal scraping and subsequent optical microscopic evaluation, 25 (86.2%) exhibited positive findings for Candida spores or pseudo-hyphae. Fungal cultures from corneal samples of 36 patients yielded positive results in 30 cases (83.8%), and corneal tissue from 7 patients who had undergone corneal transplantation also tested positive for Candida, further supporting the diagnosis. In total, 37 cases were confirmed by positive cultures on potato dextrose agar plates. Candida parapsilosis was the most commonly isolated species (20 isolates), followed by Candida albicans (9 isolates), Candida tropicalis (5 isolates), and Candida glabrata (3 isolates). Bacterial cultures, conducted in 21 eyes (52.5%), revealed positive results for Pseudomonas aeruginosa or Staphylococcus aureus in 4 cases (Table 2). Additionally, hematoxylin and eosin staining of corneal tissue pathological sections revealed positive findings for Candida spores in one eye. 
Distinct clinical manifestations were noted for different Candida species (Figs. 4A–D). Neovascularization was significant, accounting for 90.0% in C. parapsilosis, 66.7% in C. albicans, 66.7% in C. glabrata, and 40.0% in C. tropicalis. Epithelial defects and multifocal infiltration were prevalent in C. parapsilosis keratitis (40.0% each). C. albicans keratitis showed high occurrences of multifocal infiltration (77.8%) and blurred boundary lesions (66.7%). C. tropicalis keratitis exhibited more blurred boundary lesions, cream-colored infiltration, and hypopyon (60.0% each). Additionally, C. glabrata keratitis demonstrated a high occurrence of cream-colored infiltration (66.7%). 
IVCM Morphologies of Candida Spores and Pseudo-hyphae
IVCM images were analyzed for 37 eyes (92.5%), revealing Candida spores in all cases and pseudo-hyphae in 12 eyes (32.4%). Candida spores or pseudo-hyphae were detected in 30 cases (81.1%) during the initial IVCM scan, whereas in 7 eyes (17.5%), Candida spores or pseudo-hyphae were only identified after multiple detailed scans and careful image analysis. 
Typical Candida spores, identified as round, highly reflective structures, were detected in all patients, with an average depth of 229.1 ± 174.7 µm (range, 60–630 µm) (Fig. 3B). These structures resembled those seen in EVCM of Candida colonies and methylene blue-stained spores (Fig. 3E). Additionally, short rod-shaped, highly reflective structures were found in nine eyes (24.3%, Fig. 3C), consistent with Candida blastoconidia, confirmed through EVCM and methylene blue staining (Fig. 3F). C. parapsilosis infections were more likely to display these rod-shaped structures. Candida spores appeared at different corneal lesion depths (Figs. 3G–K), either dispersed like sand or clustered in caviar-like aggregates. In superficial lesions, spores were distinct, but in deeper stroma or endothelium, they formed hazy, cloud-like clusters. Patients with multifocal infiltration were more likely to exhibit clustered spores compared with those without (75.0% vs. 33.3%; P < 0.05). 
Figure 3.
 
Appearance of Candida spores and pseudo-hyphae in IVCM and EVCM. (A) Clinical photograph showing multifocal infiltration. (B) Round, highly reflective spores. (C) Short rod-shaped, highly reflective spores. (D) Candida colony. (E and F) EVCM of colonies showing hyper-reflective spores, consistent with IVCM observations. (G) Scattered spores at 30 µm, sand-like appearance. (H) Clustered spores at 40 µm, caviar-like appearance. (I) Scattered spores at 90 µm. (J) Clustered spores at 110 µm. (K) Clustered spores at 560 µm, hazy cloud-like appearance. (L) Clinical photograph showing burred lesion boundary. (M) Beaded, highly reflective pseudo-hyphae. (N) Lotus root-shaped, highly reflective pseudo-hyphae. (O) Candida colony. (P and Q) EVCM of colonies showing hyper-reflective striped pseudo-hyphae, consistent with IVCM observations. (R) Scattered short pseudo-hyphae at 10 µm. (S) Clustered pseudo-hyphae at 50 µm. (T) Scattered pseudo-hyphae at 95 µm. (U) Clustered pseudo-hyphae at 112 µm. (V) Pseudo-hyphae at 360 µm, blurred appearance.
Figure 3.
 
Appearance of Candida spores and pseudo-hyphae in IVCM and EVCM. (A) Clinical photograph showing multifocal infiltration. (B) Round, highly reflective spores. (C) Short rod-shaped, highly reflective spores. (D) Candida colony. (E and F) EVCM of colonies showing hyper-reflective spores, consistent with IVCM observations. (G) Scattered spores at 30 µm, sand-like appearance. (H) Clustered spores at 40 µm, caviar-like appearance. (I) Scattered spores at 90 µm. (J) Clustered spores at 110 µm. (K) Clustered spores at 560 µm, hazy cloud-like appearance. (L) Clinical photograph showing burred lesion boundary. (M) Beaded, highly reflective pseudo-hyphae. (N) Lotus root-shaped, highly reflective pseudo-hyphae. (O) Candida colony. (P and Q) EVCM of colonies showing hyper-reflective striped pseudo-hyphae, consistent with IVCM observations. (R) Scattered short pseudo-hyphae at 10 µm. (S) Clustered pseudo-hyphae at 50 µm. (T) Scattered pseudo-hyphae at 95 µm. (U) Clustered pseudo-hyphae at 112 µm. (V) Pseudo-hyphae at 360 µm, blurred appearance.
Candida pseudo-hyphae, seen as striated, highly reflective structures, positioned at an average depth of 161.5 ± 109.1 µm (range, 72–400 µm) (Figs. 3M and 3N). Predominantly, they appeared beaded or lotus root-shaped. In rare cases, branching pseudo-hyphae or interconnected spores were seen. These forms matched EVCM and methylene blue staining findings (Figs. 3P and 3Q). Pseudo-hyphae were mainly present in C. albicans (5 eyes [41.7%]) or C. tropicalis (4 eyes [33.4%]). Pseudo-hyphae varied in presentation at different corneal lesion depths (Figs. 3R–V), appearing scattered in all examined eyes (Figs. 3R, 3T) or clustered in some (Figs. 3S and 3U). The presence of pseudo-hyphae in IVCM was higher in patients with blurred boundary lesions compared with those without (53.3% vs. 22.2%; P = 0.064). 
Under IVCM, Candida spores averaged 4.2 ± 1.7 µm² (range, 1.5–9.5 µm²). C. glabrata had the smallest spores (2.8 ± 0.9 µm²), followed by C. parapsilosis (3.5 ± 0.7 µm²), C. albicans (4.5 ± 1.6 µm²), and C. tropicalis (6.0 ± 1.5 µm²) (Fig. 4E). Pseudo-hyphae averaged 58.2 ± 33.0 µm in length (range, 5.5–243.8 µm), 2.1 ± 0.3 µm in width (range, 1.4–3.6 µm), and had a tortuosity of 1.16 ± 0.10 (range, 0.95–1.85). The width and length differences in pseudo-hyphae between C. albicans and C. tropicalis are minimal (2.1 ± 0.3 µm vs. 2.1 ± 0.2 µm, and 56.9 ± 32.8 µm vs. 62.7 ± 34.0 µm, respectively). However, C. albicans pseudo-hyphae exhibit slightly greater tortuosity than C. tropicalis (1.17 ± 0.11 vs. 1.13 ± 0.06; P < 0.05) (Figs. 4F–H). 
Figure 4.
 
Clinical signs and IVCM morphological characteristics of Candida species. (AD) Proportion of clinical signs caused by different Candida species. (EH) Distributions of size, widths, lengths, and tortuosity of Candida spores and pseudo-hyphae in IVCM.
Figure 4.
 
Clinical signs and IVCM morphological characteristics of Candida species. (AD) Proportion of clinical signs caused by different Candida species. (EH) Distributions of size, widths, lengths, and tortuosity of Candida spores and pseudo-hyphae in IVCM.
The average width of the hyphae was 2.5 ± 0.2 µm (range, 2.0–3.3 µm) for Fusarium spp., 2.6 ± 0.3 µm (range, 2.1–3.6 µm) for Aspergillus spp., and 2.7 ± 0.4 µm (range, 2.0–4.2 µm) for Alternaria spp. The average hyphal length measured 151.2 ± 61.7 µm (range, 48.5–351.5 µm) for Fusarium spp., 152.4 ± 65.1 µm (range, 44.5–402.4 µm) for Aspergillus spp., and 95.2 ± 46.5 µm (range, 31.4–297.7 µm) for Alternaria spp. Compared with the hyphae of filamentous fungi, including Fusarium spp., Aspergillus spp., and Alternaria spp., the pseudo-hyphae of Candida were significantly thinner and shorter (P < 0.05 for both parameters). In terms of tortuosity, Candida pseudo-hyphae exhibited slightly higher tortuosity than Aspergillus spp. (1.16 ± 0.10 vs. 1.14 ± 0.06; P < 0.05), comparable tortuosity with Fusarium spp. (1.16 ± 0.10 vs. 1.14 ± 0.07, P = 0.17), and lower tortuosity than Alternaria spp. (1.16 ± 0.10 vs. 1.19 ± 0.11; P < 0.05) (Supplementary Fig. 2). 
IC Features of Candida Keratitis With IVCM
IVCM images revealed immune cell characteristics in Candida keratitis. Inactivated DCs were present in all eyes, while activated DCs were observed in 10 eyes (27.0%). Four eyes (10.8%) exhibited ICs in a honeycomb pattern, and all eyes showed nonspecific ICs distribution. Honeycomb-distributed ICs were more common in patients with disease duration for longer than 3 months compared with shorter durations (23.1% vs. 4.2%). The detection rates of activated DCs did not significantly differ between shorter and longer disease durations. Notably, DCs or ICs were seldom seen near spores and pseudo-hyphae in both superficial and deep corneal lesions. 
Increased ulcer size was significantly associated with higher IC density (r = 0.51; P < 0.05) and DC density (r = 0.45; P < 0.05). DC density also positively correlated with Candida depth (r = 0.33; P < 0.05). No significant correlations were found between Candida depth, IC density, or DC density and disease duration (r = 0.04, P = 0.80; r = 0.09, P = 0.56; and r = 0.06, P = 0.68, respectively). IC density was not significantly correlated with Candida depth (r = 0.23; P = 0.17) (Fig. 5A). 
Figure 5.
 
IC response, Candida infiltration, and outcome predictors in Candida keratitis. (A) Scatterplot of correlation between keratitis duration, Candida infiltration, and density of ICs and DCs. (B) Typical cases: good vs. poor outcomes. (C) Features associated with poor outcomes.
Figure 5.
 
IC response, Candida infiltration, and outcome predictors in Candida keratitis. (A) Scatterplot of correlation between keratitis duration, Candida infiltration, and density of ICs and DCs. (B) Typical cases: good vs. poor outcomes. (C) Features associated with poor outcomes.
Treatment and Prognosis
Antifungal therapy for Candida keratitis averaged 55 ± 51 days, ending with inflammation resolution. Four cases with bacterial co-infections received intensive levofloxacin eye drops (0.5%, Santen). After treatment, 22 patients (55.0%) had resolution with residual scarring, and 18 patients (45.0%) required therapeutic keratoplasty owing to inadequate response, with graft sizes ranging from 7.75 mm to 11.75 mm. Postoperative care included topical voriconazole (1%) or natamycin (5%). Two recurrent cases required a second penetrating keratoplasty. Additionally, one case each of endophthalmitis and severe C. parapsilosis infection necessitated vitrectomy combined with keratoplasty and evisceration. 
Figure 5B illustrates Candida keratitis at different stages: before treatment, with stable lesions after medical therapy, and with clear grafts after penetrating keratoplasty (PK). Initially, 6 patients (15.0%) had a BCVA of 20/200 or better, and 34 patients (85.0%) had a BCVA of worse than 20/200. By final follow-up, BCVA improved to 20/200 or better in 8 patients (20.0%), with 21 patients (80.0%) still having a BCVA worse than 20/200. 
Further analysis indicates weak correlations between poor outcomes and factors such as ocular trauma, intraocular surgery, corticosteroid use, ocular surface diseases, and systemic diseases (Fig. 5C). Conversely, worse outcomes are significantly associated with features like cream-colored infiltration, blurred lesion boundaries, hypopyon, IC densities over 1000 cells/mm², and Candida penetration deeper than 200 µm (Fig. 5C) (P < 0.05). Although clustered spores in IVCM images may suggest better outcomes, they were not statistically significant (Fig. 5C) (P > 0.05). 
Discussion
This study summarized the clinical and IVCM features of Candida keratitis over a decade in a tertiary care hospital in northern China, a subtropical region, to facilitate early diagnosis and timely treatment. 
Candida, a commensal organism, is part of the normal human flora.13 Most Candida infections are opportunistic, primarily affecting immunocompromised individuals. Unlike filamentous FK, which often follows trauma, Candida keratitis typically occurs in compromised corneas,3 especially in patients with weakened ocular surface immunity owing to topical immunosuppression, preexisting ocular conditions, or a history of corneal surgery.14 In our study, the most frequent predisposing factors for Candida keratitis were corticosteroid use and intraocular surgeries, including keratoplasty and vitrectomy, followed by systemic diseases and ocular surface conditions, aligning with previous research on risk factors for Candida corneal ulcers.5,15,16 However, the overlap of steroid use with ocular surface conditions or after intraocular surgeries complicates its role as an independent risk factor. 
Filamentous FK typically presents with a dry texture and feathery margins, often accompanied by satellite lesions, corneal endothelial plaques, less stromal edema, and rare neovascularization.7,17 In contrast, Candida keratitis has distinct features. It often begins with epithelial defects, with delayed postkeratoplasty healing as a potential risk factor.6,18 Epithelial cell breakdown in ocular surface disease predisposes patients to Candida invasion.19 Corneal lesions can manifest as isolated punctate lesions, multifocal infiltration, powdery deposits, and plaque-like deposits.9,20 And blurred lesion boundaries, likely owing to Candida pseudo-hyphae, were noted in some cases. Advanced stages may present with cream-colored21 and cloud-like infiltration, often indicating a poor prognosis requiring surgical intervention. Neovascularization appeared in 77.5% of Candida keratitis cases, likely owing to its prolonged course, making it a significant indicator. Corneal melting were seen in 27.5% of cases, possibly related to Candida-released phospholipase and protease.22 
The predominant pathogens in Candida keratitis vary among studies, with C. albicans2325 and C. parapsilosis26,27 frequently reported. Our study indicates that C. parapsilosis is the most common species of Candida keratitis in northern China, followed by C. albicans. This finding aligns with trends in invasive candidiasis, where non-albicans Candida species are becoming more common.28,29 Furthermore, mixed or secondary infections with pathogens like P. aeruginosa,30 Corynebacterium macginleyi,31 and Acanthamoeba32 also occur, necessitating combined treatments. In our study, four cases involved P. aeruginosa or S. aureus
IVCM is widely used to assess fungal hyphae, as well as Acanthamoeba cysts or trophozoites, in infectious keratitis.33,34 Although Candida species lacks true hyphae, IVCM is capable of detecting their distinctive spores. Previous case reports have documented the spore morphology of C. parapsilosis and C. glabrata keratitis on IVCM, describing them as round or oval hyper-reflective structures.9,35 Our systematic analysis of IVCM images from 37 patients with Candida keratitis confirmed these observations, identifying round, highly reflective spores in all cases, with occasional rod-shaped structures. We validated these findings through external validation microscopy (by EVCM) and light microscopy. Our study provides a detailed characterization of spore arrangements and appearances at different corneal depths to aid clinical identification. Furthermore, pseudo-hyphae were observed in certain species such as C. albicans and C. tropicali, appeared as striped, highly reflective formations, often exhibiting a beaded or lotus root-like shape. Such morphological features align with Candida's documented growth characteristics3638 and correlate with findings from both EVCM and light microscopy. Taken together, certain IVCM patterns, such as clustered spores or pseudo-hyphae, seem to be specific indicators of Candida infection. These features are more readily recognized by clinicians and serve as valuable diagnostic markers, facilitating early antifungal treatment, particularly in patients with relevant risk factors and slit-lamp findings. 
This study optimized image processing by quantifying the size of Candida spores and pseudo-hyphae, addressing the resolution limitations of IVCM imaging and reducing errors from manual measurement of small pathogens. Candida spore size in our study is similar to, but slightly smaller than, in vitro spores, which typically measure 2 to 5 × 3 to 7 µm under optimal budding conditions.39 This difference may reflect variations in growth conditions and duration. Our comparative analysis showed Candida pseudo-hyphae are thinner and shorter than the hyphae of Fusarium, Aspergillus, and Alternaria species. Previous IVCM studies reported mean hyphal diameters of 3.09 ± 0.45 µm for Aspergillus spp. and 2.87 ± 0.39 µm for Fusarium spp.,40 whereas our measurements were slightly smaller. This discrepancy may stem from differences in measurement techniques or patient characteristics, such as duration and severity of the corneal infection and prior eye drop use in our tertiary referral hospital setting. However, unlike filamentous fungi such as Fusarium and Aspergillus,34 Candida pseudo-hyphae show minimal branching, precluding detailed assessment of branching angles. 
IVCM observations linked specific cellular features to micro-organisms in microbial keratitis.12 Honeycomb distributions of anterior stromal ICs were associated with FK, and stromal DCs with Aspergillus ulcers.12 Our study advances understanding of inflammatory responses in IVCM in Candida keratitis: inactivated DCs and nonspecific-distributed ICs appeared in all cases, and larger ulcers and deeper Candida penetration showed higher DC and IC density, indicating a stronger immune reaction but poorer pathogen clearance, complicating treatment outcomes. IVCM images can also provide insights into clinical outcomes of FK: honeycomb distributed ICs and DC presence at the final visit linked to worse outcomes.41 In this study, only four patients presented ICs in a honeycomb distribution, limiting further analysis and requiring a larger sample size. Nevertheless, our study clarifies that high IC density was significantly correlated with poor outcomes in Candida keratitis. 
The BCVA outcomes showed no significant difference before and after treatment, indicating the severity of keratitis and the limited efficacy of medical treatment. Our cohort had a 45.0% rate of surgical intervention, similar to the 48.3% in a U.S. study,4 but lower than the 76.2% in a Canadian study.5 Candida isolates are highly sensitive to Amphotericin B and Natamycin but resistant to fluconazole.42,43 Despite antifungal therapy, high surgical intervention rates may result from delayed disease recognition and treatment, and poor drug penetration. 
The limitations of this study include its retrospective nature and relatively small sample size. To accurately evaluate IVCM features in Candida keratitis, only microbiologically or pathologically confirmed cases were included, limiting direct comparison of IVCM and direct microscopy diagnostic efficiency in our cohort. However, to the best of our knowledge, this is the largest study to systematically detail IVCM features of Candida keratitis, correlating these findings with EVCM and light microscopy and comparing them to filamentous fungi hyphae to aid species identification. Additionally, the study explores the risk factors, clinical characteristics, and outcomes of Candida keratitis in northern China, contributing insights to the field. 
In conclusion, Candida keratitis is a severe disease with a high rate of surgical intervention and poor visual prognosis. Its clinical manifestations differ significantly from those of filamentous FK, necessitating accurate identification. IVCM serves as a valuable diagnostic modality, enabling early detection of Candida spores, pseudo-hyphae, and inflammatory cellular changes, thus facilitating prompt intervention. This is particularly critical in complex postoperative cases where sample collection is challenging or invasive procedures are impractical, limiting the ability to obtain reliable microbiological results. 
Acknowledgments
Supported by the Beijing Municipal Public Welfare Development and Reform Pilot Project for Medical Research Institutes (PWD&RPP-MRI, JYY2023-2026); National Key Research and Development Program, grant number 2021YFC2301000. 
Disclosure: Y. Peng, None; Q. Chen, None; Y. Wei, None; L. Wang, None; Z. Zhang, None; Z. Wei, None; J. Pang, None; B. Peng, None; Q. Shi, None; Z. Wang, None; Y. Zhang, None; K. Chen, None; X. Xu, None; Q. Liang, None 
References
Kredics L, Narendran V, Shobana CS, Vágvölgyi C, Manikandan P. Filamentous fungal infections of the cornea: a global overview of epidemiology and drug sensitivity. Mycoses. 2015; 58: 243–260. [CrossRef] [PubMed]
Leck AK, Thomas PA, Hagan M, et al. Aetiology of suppurative corneal ulcers in Ghana and south India, and epidemiology of fungal keratitis. Br J Ophthalmol. 2002; 86: 1211–1215. [CrossRef] [PubMed]
Tanure MA, Cohen EJ, Sudesh S, Rapuano CJ, Laibson PR. Spectrum of fungal keratitis at Wills Eye Hospital, Philadelphia, Pennsylvania. Cornea. 2000; 19: 307–312. [CrossRef] [PubMed]
Sun RL, Jones DB, Wilhelmus KR. Clinical characteristics and outcome of Candida keratitis. Am J Ophthalmol. 2007; 143: 1043–1045. [CrossRef] [PubMed]
Qiao GL, Ling J, Wong T, Yeung SN, Iovieno A. Candida keratitis: epidemiology, management, and clinical outcomes. Cornea. 2020; 39: 801–805. [CrossRef] [PubMed]
Zhang C, Li F, Liu H, Jia Z, Zhao S. Clinical and in vivo confocal microscopy characteristics of Candida keratitis following keratoplasty. BMC Ophthalmol. 2023; 23: 364. [CrossRef] [PubMed]
Mahmoudi S, Masoomi A, Ahmadikia K, et al. Fungal keratitis: An overview of clinical and laboratory aspects. Mycoses. 2018; 61: 916–930. [CrossRef] [PubMed]
Gopinathan U, Sharma S, Garg P, Rao GN. Review of epidemiological features, microbiological diagnosis and treatment outcome of microbial keratitis: experience of over a decade. Indian J Ophthalmol. 2009; 57: 273–279. [PubMed]
Scotto R, Forte P, Macrì A, Bonzano C, Traverso CE. In vivo confocal microscopy characterization of Candida parapsilosis keratitis. Eye Contact Lens. 2024; 50: 112–115. [CrossRef] [PubMed]
Wei Z, Cao K, Wang L, Baudouin C, Labbé A, Liang Q. Corneal changes in acanthamoeba keratitis at various levels of severity: an in vivo confocal microscopic study. Transl Vis Sci Technol. 2021; 10: 10. [CrossRef] [PubMed]
Wang L, Zhang Y, Wei Z, et al. Characteristics of toxic keratopathy, an in vivo confocal microscopy study. Transl Vis Sci Technol. 2021; 10: 11. [CrossRef]
Chidambaram JD, Prajna NV, Palepu S, et al. In vivo confocal microscopy cellular features of host and organism in bacterial, fungal, and Acanthamoeba keratitis. Am J Ophthalmol. 2018; 190: 24–33. [CrossRef] [PubMed]
Sudbery PE . Growth of Candida albicans hyphae. Nat Rev Microbiol. 2011; 9: 737–748. [CrossRef] [PubMed]
Donovan C, Arenas E, Ayyala RS, Margo CE, Espana EM. Fungal keratitis: mechanisms of infection and management strategies. Surv Ophthalmol. 2022; 67: 758–769. [CrossRef] [PubMed]
Masoumi A, Soleimani M, Azizkhani M, et al. Clinical features, risk factors, and management of Candida keratitis. Ocul Immunol Inflamm. 2023; 32: 1–6.
Villarrubia A, Cano-Ortiz A. Candida keratitis after Descemet stripping with automated endothelial keratoplasty. Eur J Ophthalmol. 2014; 24: 964–967. [CrossRef] [PubMed]
Wei Z, Wang S, Wang Z, et al. Development and multi-center validation of machine learning model for early detection of fungal keratitis. EBioMedicine. 2023; 88: 104438. [CrossRef] [PubMed]
Bahadir AE, Bozkurt TK, Kutan SA, Yanyali CA, Acar S. Candida interface keratitis following deep anterior lamellar keratoplasty. Int Ophthalmol. 2012; 32: 383–386. [CrossRef] [PubMed]
Naglik JR, Richardson JP, Moyes DL. Candida albicans pathogenicity and epithelial immunity. PLoS Pathog. 2014; 10: e1004257. [CrossRef] [PubMed]
Sati A, Wagh S, Mishra SK, Kumar SV, Kumar P. Post-corneal transplant Candida keratitis - incidence and outcome. Indian J Ophthalmol. 2022; 70: 536–541. [CrossRef] [PubMed]
Kanavi MR, Foroutan AR, Kamel MR, Afsar N, Javadi MA. Candida interface keratitis after deep anterior lamellar keratoplasty: clinical, microbiologic, histopathologic, and confocal microscopic reports. Cornea. 2007; 26: 913–916. [CrossRef] [PubMed]
Kantarcioglu AS, Yücel A. Phospholipase and protease activities in clinical Candida isolates with reference to the sources of strains. Mycoses. 2002; 45: 160–165. [CrossRef] [PubMed]
Ho JW, Fernandez MM, Rebong RA, Carlson AN, Kim T, Afshari NA. Microbiological profiles of fungal keratitis: a 10-year study at a tertiary referral center. J Ophthalmic Inflamm Infect. 2016; 6: 5. [CrossRef] [PubMed]
Belanger NL, Kim SJ, Bispo PJM. Molecular characterization of fungal endophthalmitis and keratitis caused by yeasts. Med Mycol. 2022; 61: myac099. [CrossRef] [PubMed]
Motukupally SR, Nanapur VR, Chathoth KN, et al. Ocular infections caused by Candida species: Type of species, in vitro susceptibility and treatment outcome. Indian J Med Microbiol. 2015; 33: 538–546. [CrossRef] [PubMed]
Bezerra FM, Rocchetti TT, Lima SL, et al. Candida species causing fungal keratitis: molecular identification, antifungal susceptibility, biofilm formation, and clinical aspects. Braz J Microbiol. 2023; 54: 629–636. [CrossRef] [PubMed]
Ranjith K, Sontam B, Sharma S, et al. Candida species from eye infections: drug susceptibility, virulence factors, and molecular characterization. Invest Ophthalmol Vis Sci. 2017; 58: 4201–4219. [CrossRef] [PubMed]
Franconi I, Rizzato C, Tavanti A, Falcone M, Lupetti A. Paradigm shift: Candida parapsilosis sensu stricto as the Most Prevalent Candida Species Isolated From Bloodstream Infections With Increasing Azole-Non-Susceptibility Rates: trends from 2015–2022 Survey. J Fungi (Basel). 2023; 9: 1012. [CrossRef] [PubMed]
Tóth R, Nosek J, Mora-Montes HM, et al. Candida parapsilosis: from genes to the bedside. Clin Microbiol Rev. 2019; 32: e00111–18. [CrossRef] [PubMed]
Paul D, Saha S, Singh N, Sengupta J, Mandal SM. Successful control of a co-infection caused by Candida albicans and Pseudomonas aeruginosa in keratitis. Infect Disord Drug Targets. 2021; 21: 284–288. [CrossRef] [PubMed]
Fejes I, Garcia-Hermosa P, Bodin A. Keratitis caused by Candida parapsilosis and subsequent Corynebacterium macginleyi infection: case report and short review. Eye Contact Lens. 2020; 46: e27–e29. [CrossRef] [PubMed]
Buchele MLC, Wopereis DB, Casara F, et al. Contact lens-related polymicrobial keratitis: Acanthamoeba spp. genotype T4 and Candida albicans. Parasitol Res. 2018; 117: 3431–3436. [CrossRef] [PubMed]
Chidambaram JD, Prajna NV, Larke NL, et al. Prospective study of the diagnostic accuracy of the in vivo laser scanning confocal microscope for severe microbial keratitis. Ophthalmology. 2016; 123: 2285–2293. [CrossRef] [PubMed]
Chidambaram JD, Prajna NV, Larke N, et al. In vivo confocal microscopy appearance of Fusarium and Aspergillus species in fungal keratitis. Br J Ophthalmol. 2017; 101: 1119–1123. [CrossRef] [PubMed]
Le Q, Wu D, Li Y, Ji J, Cai R, Xu J. Early-onset Candida glabrata interface keratitis after deep anterior lamellar keratoplasty. Optom Vis Sci. 2015; 92: e93–e96. [CrossRef] [PubMed]
Veses V, Gow NA. Pseudohypha budding patterns of Candida albicans. Med Mycol. 2009; 47: 268–275. [CrossRef] [PubMed]
Suzuki T, Imanishi Y, Iwaguchi SI, Kamihara T. Depolarized cell growth precedes filamentation during the process of ethanol-induced pseudohyphal formation in the yeast Candida tropicalis. Microbiology (Reading). 1998; 144(Pt 2): 403–410. [PubMed]
Kabir MA, Hussain MA, Ahmad Z. Candida albicans: a model organism for studying fungal pathogens. ISRN Microbiol. 2012; 2012: 538694. [CrossRef] [PubMed]
Silva S, Negri M, Henriques M, Oliveira R, Williams DW, Azeredo J. Candida glabrata, Candida parapsilosis and Candida tropicalis: biology, epidemiology, pathogenicity and antifungal resistance. FEMS Microbiol Rev. 2012; 36: 288–305. [CrossRef] [PubMed]
Tabatabaei SA, Soleimani M, Tabatabaei SM, Beheshtnejad AH, Valipour N, Mahmoudi S. The use of in vivo confocal microscopy to track treatment success in fungal keratitis and to differentiate between Fusarium and Aspergillus keratitis. Int Ophthalmol. 2020; 40: 483–491. [CrossRef] [PubMed]
Chidambaram JD, Prajna NV, Palepu S, et al. Cellular morphological changes detected by laser scanning in vivo confocal microscopy associated with clinical outcome in fungal keratitis. Sci Rep. 2019; 9: 8334. [CrossRef] [PubMed]
Galarreta DJ, Tuft SJ, Ramsay A, Dart JK. Fungal keratitis in London: microbiological and clinical evaluation. Cornea. 2007; 26: 1082–1086. [CrossRef] [PubMed]
Deorukhkar SC, Saini S, Mathew S. Non-albicans Candida infection: an emerging threat. Interdiscip Perspect Infect Dis. 2014; 2014: 615958. [CrossRef] [PubMed]
Figure 1.
 
Measurement of Candida structures and IC features in IVCM. (A) Original IVCM image displaying hyper-reflective Candida spores. (B) Enhanced, magnified, and denoised image of spores. (C, D) Area of hyper-reflective spores (yellow). (E) Original IVCM image of hyper-reflective Candida pseudo-hyphae. (F) Measurements of pseudo-hyphae lengths (red line) and tortuosity (red line/yellow line). (G, H) Measured widths of pseudo-hyphae. (I) Inactivated DCs. (J) Activated DCs. (K) ICs in a honeycomb distribution. (L) ICs in a nonspecific distribution.
Figure 1.
 
Measurement of Candida structures and IC features in IVCM. (A) Original IVCM image displaying hyper-reflective Candida spores. (B) Enhanced, magnified, and denoised image of spores. (C, D) Area of hyper-reflective spores (yellow). (E) Original IVCM image of hyper-reflective Candida pseudo-hyphae. (F) Measurements of pseudo-hyphae lengths (red line) and tortuosity (red line/yellow line). (G, H) Measured widths of pseudo-hyphae. (I) Inactivated DCs. (J) Activated DCs. (K) ICs in a honeycomb distribution. (L) ICs in a nonspecific distribution.
Figure 2.
 
Clinical photograph of typical Candida keratitis signs. (A) Corneal epithelial defect. (B) Multifocal infiltration. (C) Powdery deposits. (D) Plaque-like deposits. (E) Cream-colored infiltration. (F) Cloud-like infiltration. (G) Burred corneal lesion boundary. (H) Neovascularization. (I) Hypopyon.
Figure 2.
 
Clinical photograph of typical Candida keratitis signs. (A) Corneal epithelial defect. (B) Multifocal infiltration. (C) Powdery deposits. (D) Plaque-like deposits. (E) Cream-colored infiltration. (F) Cloud-like infiltration. (G) Burred corneal lesion boundary. (H) Neovascularization. (I) Hypopyon.
Figure 3.
 
Appearance of Candida spores and pseudo-hyphae in IVCM and EVCM. (A) Clinical photograph showing multifocal infiltration. (B) Round, highly reflective spores. (C) Short rod-shaped, highly reflective spores. (D) Candida colony. (E and F) EVCM of colonies showing hyper-reflective spores, consistent with IVCM observations. (G) Scattered spores at 30 µm, sand-like appearance. (H) Clustered spores at 40 µm, caviar-like appearance. (I) Scattered spores at 90 µm. (J) Clustered spores at 110 µm. (K) Clustered spores at 560 µm, hazy cloud-like appearance. (L) Clinical photograph showing burred lesion boundary. (M) Beaded, highly reflective pseudo-hyphae. (N) Lotus root-shaped, highly reflective pseudo-hyphae. (O) Candida colony. (P and Q) EVCM of colonies showing hyper-reflective striped pseudo-hyphae, consistent with IVCM observations. (R) Scattered short pseudo-hyphae at 10 µm. (S) Clustered pseudo-hyphae at 50 µm. (T) Scattered pseudo-hyphae at 95 µm. (U) Clustered pseudo-hyphae at 112 µm. (V) Pseudo-hyphae at 360 µm, blurred appearance.
Figure 3.
 
Appearance of Candida spores and pseudo-hyphae in IVCM and EVCM. (A) Clinical photograph showing multifocal infiltration. (B) Round, highly reflective spores. (C) Short rod-shaped, highly reflective spores. (D) Candida colony. (E and F) EVCM of colonies showing hyper-reflective spores, consistent with IVCM observations. (G) Scattered spores at 30 µm, sand-like appearance. (H) Clustered spores at 40 µm, caviar-like appearance. (I) Scattered spores at 90 µm. (J) Clustered spores at 110 µm. (K) Clustered spores at 560 µm, hazy cloud-like appearance. (L) Clinical photograph showing burred lesion boundary. (M) Beaded, highly reflective pseudo-hyphae. (N) Lotus root-shaped, highly reflective pseudo-hyphae. (O) Candida colony. (P and Q) EVCM of colonies showing hyper-reflective striped pseudo-hyphae, consistent with IVCM observations. (R) Scattered short pseudo-hyphae at 10 µm. (S) Clustered pseudo-hyphae at 50 µm. (T) Scattered pseudo-hyphae at 95 µm. (U) Clustered pseudo-hyphae at 112 µm. (V) Pseudo-hyphae at 360 µm, blurred appearance.
Figure 4.
 
Clinical signs and IVCM morphological characteristics of Candida species. (AD) Proportion of clinical signs caused by different Candida species. (EH) Distributions of size, widths, lengths, and tortuosity of Candida spores and pseudo-hyphae in IVCM.
Figure 4.
 
Clinical signs and IVCM morphological characteristics of Candida species. (AD) Proportion of clinical signs caused by different Candida species. (EH) Distributions of size, widths, lengths, and tortuosity of Candida spores and pseudo-hyphae in IVCM.
Figure 5.
 
IC response, Candida infiltration, and outcome predictors in Candida keratitis. (A) Scatterplot of correlation between keratitis duration, Candida infiltration, and density of ICs and DCs. (B) Typical cases: good vs. poor outcomes. (C) Features associated with poor outcomes.
Figure 5.
 
IC response, Candida infiltration, and outcome predictors in Candida keratitis. (A) Scatterplot of correlation between keratitis duration, Candida infiltration, and density of ICs and DCs. (B) Typical cases: good vs. poor outcomes. (C) Features associated with poor outcomes.
Table 1.
 
Definition of IVCM Parameters for Candida Keratitis
Table 1.
 
Definition of IVCM Parameters for Candida Keratitis
Table 2.
 
Population Characteristics of Candida Keratitis Cases
Table 2.
 
Population Characteristics of Candida Keratitis Cases
×
×

This PDF is available to Subscribers Only

Sign in or purchase a subscription to access this content. ×

You must be signed into an individual account to use this feature.

×