Abstract
Purpose
To report on the presence of 4 different structures visualized by confocal microscopy in patients whose clinical presentation suggested infection by Acanthamoeba.
Methods
Data and charts of 28 consecutive patients were analyzed in a retrospective study. Four types of structures were recognized by confocal microscopy performed with HRT II Rostock Cornea Module: trophozoites, double-walled cysts, signet rings, and bright spots. The 28 patients (mean age 30.8 years, range 17-61 years, 10 male, 18 female) were divided into 4 groups according to the diagnosis established later by microscopic examination of smear, culture, response to therapy, and the course of keratitis. The 4 groups were Acanthamoeba keratitis (AK), Acanthamoeba suspect (AK-suspect), bacterial keratitis (BK), and fungal keratitis (FK).
Results
The rate of patients in AK, AK-suspect, FK, and BK groups where bright spots were found were 100%, 100%, 40%, and 55%, respectively. The sensitivity of presence of bright spots in the in vivo confocal microscopy in Acanthamoeba keratitis was 100% (95% confidence interval [CI] 73.5% to 100.00%) and specificity was 50% (CI 24.7% to 75.4%). When cases where the only signs of Acanthamoeba were bright spots were excluded, and only those cases were counted where any of cysts, trophozoites, or signet rings were also found, the sensitivity was 67% (95% CI 34. 9% to 90.1%) and the specificity was 94% (95% CI 69.8% to 99.8%).
Conclusions
The relatively high rate of bright spots in non-Acanthamoeba keratitis challenges the assumption that bright spots seen by confocal microscopy are a specific indication of Acanthamoeba keratitis.
Introduction
Acanthamoeba is a ubiquitous pathogen, a dreaded cause of contact lens keratitis. Acanthamoeba has a 2-stage life cycle: an active and infective trophozoite stage and a dormant cystic stage with minimal metabolic activity. The ability to transform into cysts contributes to the resistance of Acanthamoeba to chemotherapeutic drugs (1). The diagnosis of Acanthamoeba keratitis (AK) is seldom straightforward and it is often misdiagnosed as bacterial, viral, or fungal keratitis. Delayed diagnosis can result in delay in treatment. Since early treatment of AK harbors a better prognosis, a rapid and accurate diagnostic method is crucial (2).
Contact lens wear and excruciating pain experienced by the patient are strongly indicative of Acanthamoeba infection. Typically the clinical features of AK vary according to the duration of symptoms before presentation. Early disease is characterized by an epitheliopathy including punctate keratopathy, pseudodendrites, epithelial or subepithelial infiltrates, and perineural infiltrates with ring stromal infiltrates. Later disease is characterized by ring infiltrates, ulceration, and a secondary sterile anterior uveitis, sometimes with hypopyon (2, 3).
Corneal scrapings stained with Giemsa, periodic acid-Schiff (PAS), Wright's, calcofluor white, or acridine orange demonstrate the cyst and trophozoite in early disease with good effectivity. Once deeper involvement of the stroma occurs, a corneal biopsy may be necessary. Culture of Acanthamoeba and polymerase chain reaction (PCR) would definitively confirm the diagnosis; however, it is not always available (2, 3). In vivo confocal microscopy (IVCM) is a recently developed tool for investigating different forms of keratitis. That potentially provides an instant and reliable diagnosis. Trophozoites are 25-40 µm in diameter and appear as hyper-reflective and ovoid structures on IVCM. Double-walled cysts appear as hyper-reflective, spherical, occasionally ovoid structures, with a range of 15-28 µm in diameter (4–5–6–7). Another presentation is bright signet ring images (8). However, the double wall may not always be apparent by IVCM, and so the cysts are present as bright spots, making it occasionally difficult to differentiate cysts from leukocytes or epithelial nuclei (6, 8). This fact undermines the specificity of this method in AK. Our aim was to determine the presence of these different structures in patients with keratitis with a suspicion of Acanthamoeba infection based on the clinical picture.
Methods
Patient selection, patient groups
This is a retrospective study. We selected all the patients with keratitis who had been examined with confocal microscope between 2008 and 2014 at our institution in search of signs of Acanthamoeba infection. We went through the electronic files of the patients to find out the final etiologic diagnosis. The keratitis was diagnosed as AK if cysts or trophozoites were detected in the corneal scraping material with PAS (AK group). Acanthamoeba keratitis suspect diagnosis was made if 4 conditions coexisted: the patient was a contact lens wearer with severe pain, the patient had branching epithelial keratitis or perineuritis or ring infiltrate, the anti-Acanthamoeba treatment was effective, and the culture did not show fungus or bacteria (AK-suspect group). In this group, scrapings yielded no results. The diagnosis was fungal keratitis if fungal filaments or yeasts were present in the corneal smear or fungal cultures were positive (FK group). The keratitis was regarded as bacterial if bacteria were found in the scraping material, or bacterial culture was positive, or final healing with topical antibiotics was achieved within 2 weeks (BK group). We excluded patients without a definitive diagnosis.
Evaluation of IVCM pictures
A HRT II Rostock Cornea Module (Heidelberg Engineering, Heidelberg, Germany) was used for the confocal microscopy examinations. Scans were taken from the surface of the epithelium to the deepest stroma and pictures were made at regular intervals. Special attention was focused on the epithelial and subepithelial layers. Four types of structures were registered: trophozoites, double-walled cysts, signet rings, and bright spots. Trophozoites were defined as hyperreflective and ovoid structures, 25-40 µm in diameter, sometimes with a less reflective round nucleus. Double-walled cysts were oval or round highly refractive structures with a polygonal inner shape, surrounded by a slightly less reflective ring, varying in size between 12 and 30 µm. The definition of signet rings was round or oval structures with thin, highly reflective wall at some point including a bright point; inside the ring, the reflectivity was less pronounced. Finally, bright spots were round or oval, evenly highly reflective particles. The size of signet rings and bright spots was about the same as the size of the double-walled cysts. These definitions were determined according to previous descriptions (5, 6, 8).
Statistical analysis
Sensitivity, specificity, positive and negative likelihood ratio, and positive and negative predictive value of the different confocal microscopic structures (bright spots, double-walled cysts, trophozoites, and signet rings) for Acanthamoeba keratitis was counted with an Internet calculator (www.medcalc.org/calc/diagnostic_test.php).
Results
A total of 32 patients had been examined with the HRT II Rostock Cornea Module confocal microscope. Four of them were excluded because the etiology of the corneal inflammation could not be elucidated. The mean age of the remaining 28 patients was 30.8 ± 9.38 years. There were 10 men and 18 women. Eight patients belonged to the AK group (mean age: 27.8 ± 7.48 years, 4 men, 4 women), 4 to AK-suspect group (mean age: 28.8 ± 2.36 years, 4 women), 5 to FK group (mean age: 39.6 ± 14.99 years, 2 men, 3 women), and 11 to BK group (mean age: 29.8 ± 7.61 years, 4 men, 7 women). Patients’ demographic data, contact lens wear, the results of bacterial culture and microscopic examination of corneal smear, and the structures found by confocal microscopy are shown in Table I. Some samples of confocal microscopic pictures are presented in Figure 1. The rate of patients in the AK, AK-suspect, FK, and BK group who exhibited bright spots was 100%, 100%, 40%, and 55%, respectively (Fig. 2). In AK and AK-suspect groups, the rate of those where bright spots were present together with double-walled cysts or trophozoites or signet rings were 63% and 75%, respectively. If we consider AK and AK-suspect group together, this rate is 67% (Fig. 2).
Demographic data and confocal microscopy results
AK = Acanthamoeba keratitis; antibiotics − = keratitis did not react to antibiotics treatment; antibiotics + = keratitis reacted to antibiotics treatment; b = bacterial culture negative; B = bacterial culture positive; BK = bacterial keratitis; FK = fungal keratitis; NA = no data; p = smear negative for Acanthamoeba and fungus; P = smear positive for Acanthamoeba; Pf = smear positive for fungus.
In vivo confocal microscopy (HRT II Rostock Cornea Module) samples from the different groups. (
The rate of eyes with bright spots in in vivo confocal microscopy in the different groups. Black = bright spots are present together with any of the other 3 (double-walled cyst, trophozoite, signet ring) structures. Gray = only bright spots are present. White = none of the 4 structures could be found.
We calculated the sensitivity, specificity, positive and negative likelihood ratio, and positive and negative predictive value of the different IVCM structures in Acanthamoeba keratitis (Tab. II). Taking into consideration the cases where bright spots (without or with any of the other 3 structures) were present in the IVCM, sensitivity was 100% (95% confidence interval [CI] 73.5% to 100.00%) and specificity was 50% (CI 24.7% to 75.4%). When excluding the cases where the only signs of Acanthamoeba were the bright spots, and only the cases where cysts, trophozoites, or signet rings were also present were considered, the sensitivity was 67% (95% CI 34.9% to 90.1%) and the specificity was 94% (95% CI 69.8% to 99.8%) (Tabs. II and III).
Diagnostic accuracy parameters of the different IVCM structures in Acanthamoeba keratitis
B = bright spots; CI = confidence interval; D = double-walled cysts; IVCM = in vivo confocal microscopy; NaN = not a number; S = signet rings; T = trophozoites.
B, D, T, S: any of B, D, T, or S is present.
D, T, S: any of D, T, or S is present.
Confocal microscopy results comparing Acanthamoeba (AK + AK suspect) vs non-Acanthamoeba (BK + FK) patients with and without bright spots
AK = Acanthamoeba keratitis; B = bright spots; BK = bacterial keratitis; D = double-walled cysts; FK = fungal keratitis; S = signet rings; T = trophozoites.
Sensitivity is 100% with bright spots and 67% without bright spots. Specificity is 50% with bright spots and 94% without bright spots.
Positive (B/D/T/S): any of B, D, T, or S is present.
None of the 4 structures is present.
Positive (D/T/S): any of D, T, or S is present.
None of D, S, or T is present.
Discussion
The purpose of our study was to determine whether the presence of bright spots, double-walled cysts, trophozoites, and signet rings is indicative of Acanthamoeba infection. We found that bright spots—round or oval evenly highly reflective particles of 12 to 30 µm diameter—occurred not only in Acanthamoeba keratitis, but were present in 40% of fungal and 55% of bacterial infections as well. These relatively high rates in non-AK cases seriously questions the specificity of bright spots seen on IVCM as indicative of AK.
The specificity and sensitivity of in vivo confocal microscopy in Acanthamoeba keratitis was examined in several studies. The sensitivity was between 100% and 55.8% and the specificity between 100% and 77.3% (4, 5, 7, 9). This wide range can be explained with the variably experienced examiners and the lack of consensus about the signs that confirm the presence of Acanthamoeba cysts and trophozoites (6, 9). In our small retrospective study, we found 100% sensitivity of bright spots in Acanthamoeba keratitis, but the specificity, the positive likelihood ratio, and the positive predictive value were relatively low (50%, 2.0%, and 60%, respectively). This implies the uncertainty of treating bright spots as a special sign of AK. On the contrary, although double-walled cysts, trophozoites, and signet rings have relatively high specificity, positive likelihood ratio, or positive predictive value, their sensitivity (33.3%, 58.3%, and 8.3%, respectively) is low. Similarly, if we count those cases where double-walled cysts, trophozoites, or signet rings were observed, the sensitivity was only 67%, although the specificity was as high as 94%. This means that although the double-walled cysts, trophozoites, or signet rings were a specific sign of Acanthamoeba keratitis, we had high rates of false-negative results.
We found a trophozoite-like structure in the corneal stroma of one BK patient (Fig. 1d). This patient was healed by antibiotic drops in 1 week, and no recurrence was experienced. The exact origin of this large, highly reflective body remained unknown.
Although recently the possibility has arisen in some publications that IVCM may be an equal or a more sensitive diagnostic method than corneal scrapings or culture in corneal infections, microscopic examination of the scrapings and culture are still the gold standard of diagnosis (4, 6). The benefits of light microscopic examination of scrapings over the HRT II confocal microscope are that the resolution of the light microscope is about 5 times higher (0.2 µm vs 1 µm), and contrast and specificity can be ameliorated by applying different stainings, which is not possible with the IVCM. The sample collection method could also have an implication on the sensitivity value of the confocal microscopy of the diagnosis of AK. With scraping—unlike corneal biopsy—we can take samples only superficially, and only from a small area. On the other hand, with a confocal microscope, wide areas of the cornea can be examined as deep as the posterior stroma. This incongruity can be a weakness of our study. However, we found almost all of the examined structures with IVCM in the epithelium or just under the epithelium, which is the place from where the scraping sample is taken.
Our laboratory diagnostic method for direct detection of the causative agent in the clinical samples was microscopic examination of the scraping material alone, because we had no possibility to perform culture or PCR for detection of Acanthamoeba. As the specificity of microscopic examination of stained smear is 100% (10), it is unlikely that we categorized a patient into the AK group mistakenly. On the other hand, although its sensitivity is not 100% (10), according to the fact that the keratitis classified in the BK group and FK group healed after antibiotic and antifungal therapy, respectively, it seems that the diagnosis was correct in all cases.
In our study, the first step was to evaluate the IVCM pictures without knowing the diagnosis from the patient's file. In this way, the diagnosis did not influence us in noticing and categorizing the 4 examined structures. This supports the objectivity of our results.
With our study, we aimed to draw attention to the possibility of misinterpretation of IVCM in keratitis. New technologies, for example multi-photon imaging or second harmonic imaging (11, 12), may help with better differentiation between the different components of inflammation in infectious keratitis.
Footnotes
Acknowledgment
The authors thank Dr. Dipika Patel for advice on evaluation of some IVCM pictures.
Financial support: No financial support was received for this submission.
Conflict of interest: None of the authors has conflict of interest with this submission.