Downloaded from http://bjo.bmj.com/ on September 1, 2016 - Published by group.bmj.com Clinical science Corynebacterium spp as causative agents of microbial keratitis Sujata Das,1 AV Subba Rao,1 Srikant K Sahu,1 Savitri Sharma2 1Cornea and Anterior Segment ABSTRACT During the past few years, there have been an Service, LV Prasad Eye Purpose To report the clinical and microbiological increased number of ocular infections due to cory- Institute, Bhubaneswar, fi 81014 Odisha, India pro le of keratitis caused by Corynebacterium spp. neform bacteria. At L V Prasad Eye Institute, 2Ocular Microbiology Service, Methods The medical and microbiology records of 22 Bhubaneswar, a large number of patients are seen LV Prasad Eye Institute, patients, who had presented at the L V Prasad Eye with microbial keratitis. In this report, we share the Bhubaneswar, Odisha, India Institute, Bhubaneswar, between June 2009 and clinical and microbiological profile of keratitis December 2012, and whose corneal scrapings had caused by Corynebacterium spp. Correspondence to fi Dr Sujata Das, Cornea and yielded signi cant growth of Corynebacterium spp, were Anterior Segment Service, LV retrospectively reviewed. A detailed ocular examination MATERIALS AND METHODS Prasad Eye Institute, was performed before the respective corneal scraping The medical and microbiology records of all Bhubaneswar 751024, Odisha, was sent for a microbiological work-up. The data India; [email protected] patients, who had presented at the LV Prasad Eye collected from each record included age, gender, Institute, Bhubaneswar, between June 2009 and Received 8 February 2015 predisposing factors (ocular and systemic), clinical December 2012, and whose corneal scrapings had Revised 24 August 2015 presentation, management and outcome of treatment. yielded significant growth of Corynebacterium spp, Accepted 3 October 2015 Results The mean age of the patients was 43.8 were retrospectively reviewed. The following data Published Online First ±24.4 years. Ocular predisposing factor was present in 13 November 2015 were collected from each record: age, gender, pre- 14 (63.6%) eyes. Surgical intervention was required in disposing factors (ocular and systemic), clinical – 12 (54.5%) patients. In vitro susceptibility (Kirby Bauer presentation, management and outcome of disc diffusion method) results of Corynebacterium spp to treatment. vancomycin (17/19, 89.5%), cefazolin (16/20, 80%), A detailed ocular examination was performed fl chloramphenicol (11/20, 55%), o oxacin (13/19, before collecting the sample for microbiological fl fl 68.4%), cipro oxacin (10/20, 50%) and gati oxacin investigation (figure 1). As per our institutional (10/19, 52.6%) were variable. Drug resistance (more protocol, all cases of clinically non-viral microbial than one drug) was seen in nine (40.9%) keratitis require microbiological investigation, Corynebacterium isolates, of which, two (22.2%) which consists of smear examination and culture of showed multidrug resistance to three or more classes of the corneal scrapings. Corneal scraping was done antibiotics. using #15 surgical blades on a Bard-Parker handle Conclusions Corynebacteria can cause severe corneal under topical anaesthesia (0.5% proparacaine infection requiring surgical intervention. hydrochloride). Microbiological processing included Gram stain, potassium hydroxide/calco- fluor white (KOH+CFW) mount and inoculation INTRODUCTION on appropriate media (5% sheep blood agar, 5% Bacterial keratitis can lead to significant loss of sheep blood chocolate agar, Sabouraud dextrose vision.1 The severity of corneal infection is depend- agar (SDA), non-nutrient agar with Escherichia coli, ent on the virulence of the organism and the under- thioglycollate broth and brain heart infusion lying condition of the cornea.23It occurs when broth). All media were incubated aerobically at 37° microorganisms overcome host defences. Under C except chocolate agar (incubated in 5% CO2 at conditions, such as trauma or systemic infection, the 37°C) and SDA (incubated at 27°C). The media normal flora too can contribute to ocular infection. were observed for 14 days for any growth. The predominant resident normal commensals of Colonies growing on the site of inoculum were the external ocular surface are Staphylococcus epi- considered for further processing and identifica- dermidis and Corynebacterium xerosis. Although tion. Characteristic colonies on solid media and Corynebacterium spp, which are also referred to as turbidity in liquid media were subjected to Gram Editor’s choice Scan to access more diphtheroids, are regarded as non-pathogenic stain and standard biochemical tests for the con- free content organisms, they have been recognised as the cause firmation of Corynebacterium spp. A culture was – of serious systemic and ocular infection.4 6 considered significant when there was growth of Corynebacterium spp are associated with conjunc- the same organism on two or more media and/or tivitis,7 keratitis89and endophthalmitis.10 11 confluent growth at the site of inoculation on one Corynebacterium spp occur commonly in nature solid medium and/or growth in one medium with in the soil, water, plants and food products.12 consistent (matches with) direct microscopy find- Some species are known for their pathogenic ings of the corneal scraping. In few cases, analytical effects in human beings and other animals. The profile index (bioMerieux, USA) was done to iden- To cite: Das S, Rao AVS, most notable infection is diphtheria that is caused tify Corynebacterium species. Antibiotic sensitivity Sahu SK, et al. Br J by C. diphtheriae. Other pathogenic species in of Corynebacterium spp was done by disc diffusion Ophthalmol 2016;100: human beings include C. amycolatum, C. striatum, Kirby–Bauer method on blood agar as per Clinical – 939 943. C. jeikeium, C. urealyticum and C. xerosis.13 and Laboratory Standards Institute guidelines. Das S, et al. Br J Ophthalmol 2016;100:939–943. doi:10.1136/bjophthalmol-2015-306749 939 Downloaded from http://bjo.bmj.com/ on September 1, 2016 - Published by group.bmj.com Clinical science Figure 1 Slit lamp photograph showing full-thickness corneal infiltrate and hypopyon. The initial treatment was based on the microscopy results of the corneal scraping (figure 2A). This was continued or modi- fied depending on antibiotic sensitivity report after growth in culture (figure 2B) and/or clinical response. Surgical intervention was planned in cases of non-response to medical management, large infiltrate, thinning and perforation. Figure 2 (A)Corneal scraping showing gram-positive bacilli arranged in Chinese letter pattern suggestive of Corynebacterium spp (Gram RESULTS stain, ×1000). (B) Tiny, confluent, white, opaque colonies on chocolate During the study period, of the 1074 clinical samples that were agar. culture positive, Corynebacterium spp was isolated from 101 patients. The majority of the isolates (n=77; 76.2%) were from corneal scrapings of patients with keratitis. Twenty-two cases fluoroquinolone. Surgical intervention was required in 12 (M:F :: 15:7) showing significant growth of Corynebacterium (54.5%) patients. There was no recurrence during follow-up spp as sole organism were included in this analysis. Majority of visits. The mean visual acuity (LogMAR) was 2.46 and 2.24 at the samples yielded mixed growth. In 15/55 cases, growth of presentation and at the last follow-up, respectively. Five patients corynebacteria as single organism was not significant. In 37/55 had visual acuity >20/200 at the last follow-up visit. cases, there was mixed growth with corynebacteria that was In vitro susceptibility results (table 2)ofCorynebacterium spp either significant or not significant. In 3/55 cases, both coryne- to vancomycin (17/19, 89.5%), cefazolin (16/20, 80%), chlor- bacteria and other organism were not significant. Analytical amphenicol (11/20, 55%), ofloxacin (13/19, 68.4%), ciprofloxa- Profile Index was done for 6/22 samples. The species identified cin (10/20, 50%) and gatifloxacin (10/19, 52.6%) were variable. were C. propinquum (n=3), C. bovis (n=1), C. striatum (n=1) Out of four patients who were found to have drug resistance to and C. pseudodiphtheriticum (n=1). The mean age of the fluoroquinolones, two (case #8 and #9) underwent therapeutic patients was 43.8±24.4 years (range: 2 months to 82 years). keratoplasty after few days of presentation as they worsened. The mean duration of symptoms before presentation was 15.9 The other two patients were continued with gatifloxacin as they ±12.2 days (range, 1–30 days). were responding to treatment clinically. While one (case #9) An ocular predisposing factor was present in 14 (63.6%) eyes had resistance to chloramphenicol, cefazolin and fluoroquino- (table 1). Two patients had diabetes mellitus. Hypopyon was lone (ciprofloxacin and gatifloxacin), the other (case #20) had present in six (27.3%) eyes at the initial presentation. One resistance to chloramphenicol, cefazolin and vancomycin. patient had associated endophthalmitis. Case #6 had undergone therapeutic penetrating keratoplasty (TPK) for Pseudomonas DISCUSSION keratitis 9 months before presentation that had failed subse- The genus Corynebacterium was created by Lehmann and quently. Case #7 had presented as graft infiltrate after 16 days Neumann in 1896 as a taxonomic group to contain the bacter- of TPK due to fungal aetiology. Case #8 had clear graft during ial rods responsible for causing diphtheria. The principal fea- infection. All patients were started with intensive topical tures of the Corynebacterium genus were described by Collins 940 Das S, et al. Br J Ophthalmol 2016;100:939–943. doi:10.1136/bjophthalmol-2015-306749 Downloaded from http://bjo.bmj.com/ on September 1, 2016 - Published by group.bmj.com Clinical science Table 1 Epidemiology, risk factor and clinical profile of patients with keratitis caused by Corynebacterium spp Risk factor Size of Visual acuity Age infiltrate Surgical Sl. (years) Sex Eye Ocular Trauma(mm×mm) Hypopyon intervention Presenting Final Final outcome 1. 10 F OS Lagophthalmos – 1.2×1.2 Yes – CF: 1 m 20/100 Corneal scar (S/p ptosis surgery) 2. 28 M OS Mooren’s ulcer – 1×2 No TA+BCL CF: 1.5 m HM Corneal scar 3.