Prevalence of herpes simplex virus 1 and 2 in culture-negative suppurative keratitis using polymerase chain reaction – A pilot study

Résumé fourni par la source

Dear Editor, Around 8 million people are blind due to corneal diseases globally.[1] Corneal blindness, the third leading cause of blindness in India, is most commonly due to microbial keratitis.[2,3] Corneal scraping for smear and culture is routinely done to establish etiology. However, culture positivity is seen only in 35–55% of patients with suppurative keratitis.[4,5] Since there is a significant proportion of smear/culture negativity, it is necessary to explore other potential infectious agents other than bacteria/fungi, especially viruses like the Herpes Simplex Virus (HSV) that can mimic bacterial or fungal suppurative corneal ulcers. Using newer molecular diagnostic tools like polymerase chain reaction (PCR) can enhance the identification of viral etiology, such as HSV 1 and HSV 2, in patients with culture-negative suppurative keratitis. We report the prevalence of HSV 1 and 2 in culture-negative suppurative keratitis using qualitative real-time duplex PCR. To the best of our knowledge, there is no prospective study data on the presence of HSV in culture-negative suppurative keratitis. This prospective observational case series was conducted as a pilot study at the Departments of Ophthalmology, Clinical Microbiology, and Clinical Virology in a tertiary care hospital in South India and was approved by the institutional review board (IRB Min no 9265). We included patients who presented with scrapable suppurative corneal ulcers, from March 14, 2015 to September 30, 2015, that were smear and culture-negative. Informed consent was obtained by the Principal Investigator (PI) from all the study participants. Suppurative corneal ulcer was defined as the presence of a corneal epithelial defect with an underlying stromal infiltrate measuring at least 1 mm2 in size, with or without hypopyon. Patients were excluded if corneal scraping was not medically indicated, in the event of keratitis cases with impending perforation, or if the ulcer was smaller than 1 mm, as these conditions typically yield insufficient material for reliable testing. Children (under 15 years of age) and pregnant individuals were also excluded from the study. The time taken to heal was the time taken for complete closure of the epithelial defect and the absence of stromal inflammatory cells in the cornea. A 2-year retrospective audit (July 2012 to May 2014) examined 399 corneal scraping specimens from clinically nonviral, suppurative keratitis cases. The culture positivity rate was 44%, consistent with other studies (35–70%). Anticipating 30 culture-negative ulcers within 6 months among eligible adult patients, our pilot study with a sample size of 30 was conducted. This study aimed to determine the prevalence rate of HSV positivity with a 95% confidence interval (CI) among culture-negative suppurative keratitis in a hospital setting using PCR. Corneal ulcer scraping was then performed, and samples were collected for smear and culture. Simultaneously, corneal scrapings were obtained in all cases for PCR analysis. Real-time PCR assay was used to detect HSV as its sensitivity is in the range of 85–100%, with a specificity of 70–75%. The samples were stored at -80°C until required for virus detection by real-time PCR assay. When no growth was noticed, bacterial cultures were usually declared negative in 5 days, while fungal cultures were usually declared negative in 2 weeks. However, in this study, we had set a waiting period of 5 days for the culture report for both bacterial and fungal growth. Empirical therapy based on clinical suspicion was started during this period. After 5 days, preserved samples of negative smear and culture reports were processed for virus isolation by real-time PCR assay in these patients to avoid delay in treatment with antivirals if found positive. The time interval between scraping and availability of PCR results was the sum of 5 days of no response and additional days from the time PCR testing was initiated to the results being available. The culture-negative samples to be tested for PCR were arranged on a PCR rack with a negative control (water) included after every third sample to detect potential cross-contamination. An external quality control (EQC) sample consisting of a low positive previously tested sample was included in every test run to ensure that the PCR conditions did not miss low positive reactive samples. DNA extraction was done as per the manufacturer’s protocol of the Qiagen DNA Blood Mini kit (Qiagen, GmBH, Germany). The treatment strategy for culture and smear-negative cases in this study included an empirical approach with antibacterial (Cefazoline 5% and Fortified Gentamycin 1.4% eye drops) and/or antifungal (Natamycin 5% eye drops) agents for culture-negative corneal ulcers based on clinical presentation. If a patient tested positive for HSV 1 and 2 by PCR and showed no improvement or worsened with the initial treatment, they were additionally administered oral Acyclovir 400 mg (GlaxoSmithKline India) five times a day for 14 days and then 400 mg BD maintenance (tapering), Acyclovir ointment 3% (FDC Limited India) five times a day, tapered based on clinical response, and topical prednisolone acetate 1% (Allergan India) 4 hourly for a week and tapered depending on the response. Antifungals were stopped after starting antivirals and topical steroids. While we acknowledge the importance of other pathogens, this study was specifically designed to examine the role of HSV in culture-negative corneal ulcers. During the study period, 23,282 patients reported to the ophthalmology department for an eye check-up. Of them, 96 patients were diagnosed clinically with suppurative (clinically nonviral) keratitis. Thus, 0.412% of all patients had suppurative keratitis. 71 patients were included in the study as per inclusion/exclusion criteria. A detailed flow of the study is shown in Fig. 1A. Forty-three (61%) had confirmed infectious keratitis based on positive smear and/or culture results; fungi were identified in 31 (72%) and bacteria in 11 (26%). Acanthamoeba, a less common organism, was found in one patient (2% of positive cases). There was no discrepancy in the organism between the smear and culture reports. Out of 31 cases of fungal corneal ulcers, 21 were positive on both smear and culture, and 5 showed no hyphae on smear but grew on culture. Antivirals were also started empirically in 4 cases (5.6%). Some patients were on more than one medication. Among the four patients on antivirals in the current episode, 3 (75%) turned out to be positive on smear and/or culture for bacteria or fungus.Figure 1: (a-c) (A) Figure showing the detailed flow of the study patients. (B) Clinical photographs of patients who were both smear culture-negative and who tested positive on HSV PCRTwenty-eight smear and culture-negative samples were processed for duplex PCR (both HSV 1 and 2). Of these, 25 cases were negative for both HSV 1 and 2, while 3 cases were positive for HSV 1. None of the cases were positive for HSV 2. In this pilot study, we identified three cases that were HSV-positive, giving a prevalence of 10.7% (95% CI 2.3%–28.2%) using the Binomial (Clopper–Pearson) ‘exact’ method. The culture positivity rate in our study was 60%. All three HSV positive keratitis cases were clinically suppurative keratitis. None of them had any clinical features of viral keratitis, like old scarring with vascularization, significantly reduced corneal sensation, or rolled out margins. 48 (67.6%) patients received empirical antibacterial and 27 (38%) received empirical antifungal agents. None of these patients was on empirical antiviral therapy. Characteristics of the smear/culture-negative patients with suppurative keratitis are shown in Table 1. Past history of red eyes (P = 0.019), corneal ulcers (P = 0.006), and history of antiviral use were associated with HSV positivity (P = 0.001). The presence of a corneal scar was also significantly associated with HSV positivity

Ce résumé expose les affirmations des auteurs. BNTIC ne l’interprète pas comme une validation indépendante des résultats.