Clinical presentations, management outcomes, and diagnostic dilemma in Kocuria endophthalmitis

Kocuria species can mimic other coagulase-negative Staphylococci and cause diagnostic dilemma. Rarely, Kocuria can be a cause of endophthalmitis. In the current communication, we describe our experience of diagnosing and managing endophthalmitis due to Kocuria spp.

Kocuria are gram-positive coccoid bacteria of the family Micrococcaceae (order Actinomycetales, class Actinobacteria). They are found as tetrads or irregular clusters and are biochemically catalase-positive and coagulase-negative. Kocuria infections over the years have been reported largely as systemic infections in immunocompromised hosts [1, 2]. Though Kocuria has been known to be a part of the normal microbiota, it can cause infections like peritonitis, endocarditis, cholecystitis, pneumonitis, and urinary catheter-related infections [3, 4‐16]. Kocuria infection prevalence could be falsely low as phenotypic assays used for organism identification often misdiagnose Kocuria as Staphylococci [17]. In the current communication, we report the presentations, management outcomes, and diagnostic dilemma in a series of Kocuria endophthalmitis treated at our center.

This was a retrospective consecutive interventional case series. This was a retrospective chart review of cases with culture-positive Kocuria endophthalmitis. We included all cases with culture-positive Kocuria endophthalmitis from January 2013 to March 2018 at our tertiary eye care center. The study protocol was approved by the institutional review board. The medical records of the patients were retrieved from the medical records department and the microbiology laboratory database. The collected demographic information included the etiology of the infection, interval from the inciting event to the start of symptoms, interval between the start of symptoms and treatment intervention, comorbid systemic illnesses, visual acuity at presentation, detailed biomicroscopic examination including information on media clarity on indirect ophthalmoscopy, retinal examination, ultrasound findings, treatment provided, and final functional and anatomic outcomes.

The current case series included eight eyes of eight patients. There were five males and three females. The mean age at presentation was 31 ± 17.44 years (median 30 years). None of the patients had any systemic illness and all were immunocompetent. The inciting event was open globe injury in five cases and cataract surgery, microbial keratitis, and endogenous cause in one case each. The case with microbial keratitis had a corneal ulcer perforation and at presentation already had a primary procedure of a tissue adhesive application done elsewhere.

The ulcer finally resolved, and at the last visit, the patient showed a vascularized corneal scar. Of the three patients with post-trauma endophthalmitis who presented with a corneal tear, all show a favorable anatomic outcome while two also had a favorable visual outcome. One case presenting with endogenous endophthalmitis developed a quiet anterior chamber but had persistent vitritis at the last visit. The eye with an open globe injury with scleral involvement developed a recurrent retinal detachment with an overall poor anatomic and functional outcome. The clinical features, the type of intervention, the number of injections, and the medications used are summarized in Table 1. The presenting visual acuity was favorable in one case, while at the final visit, it was favorable in three cases. At the last visit, endophthalmitis was noted to be resolved in six cases. The tissue sample that tested positive for Kocuria sp. varied with the cases and is summarized in Table 2.

The direct microscopy of the vitreous fluid showed the presence of gram-positive cocci in pairs, groups, and chains (Figs. 1 and 2) under gram stain in four of eight cases in our series. Presence of Kocuria sp. was confirmed by growth of large, cream-colored colonies on chocolate and/or blood agar (Fig. 3) from the vitreous fluid in seven of eight cases and from AC Tap in one of eight cases. All of these isolates were catalase positive and coagulase negative. Vitek 2 identification revealed the presence of Kocuria rosea in four cases, K. kristinae in three cases, and K. varians in one case using the GP card. Overall antibiotic susceptibility to most commonly used antibiotics was found to be good (Table 3). Figures 4, 5, 6, and 7 depict clinical features at presentation and final visits of a few cases.

The current report describes the outcome of a series of cases of Kocuria endophthalmitis. In a retrospective study from China describing the causative organisms in infectious endophthalmitis, 1/330 cases reported was reported as Kocuria [17]. While cases of Kocuria keratitis have been reported previously with variable visual outcome [18, 19], this is the first detailed description of endophthalmitis caused by Kocuria sp. Kocuria is a member of the Micrococcaceae family and currently includes 18 species [1, 2, 20]. Though it is ubiquitous in nature and freely found as normal skin and oral flora in humans, only 5/18 known species are known to be opportunistic pathogens [1]. The diagnostic difficulties in Kocuria identification lie in the fact that the phenotypic assays used for microbe identification often misdiagnose Kocuria as coagulase-negative Staphylococci [21]. Further diagnostic difficulty occurs due to a differential behavior of Kocuria in vitro and in vivo especially in stress conditions [22, 23]. Pathological strains are known to be more phenotypically variable then saprophytic strains [24‐27]. Identification of Kocuria spp. remains elusive because most clinical microbiology laboratories have limited or no access to advanced molecular techniques. Laboratory identification of Kocuria spp. can be made conventionally only on a high index of suspicion. Properties such as morphological variability between these bacteria and other similar gram-positive cocci, as well as biochemical properties including the antimicrobial susceptibility patterns against selective antibiotics, could be used to presumptively identify Kocuria spp. Susceptibility towards bacitracin and lysozyme and resistance to nitrofurantoin, furazolidone, and lysostaphin can be used to separate this bacterium from Staphylococci. This however is not commonly done in routine work-ups. Increase in time-dependent pigmentation of the colonies can also lead to a diagnosis of Kocuria as the typical pigmentation of these colonies increase with time especially after a culture period of 48 h [1]. The introduction of the Vitek 2 GP-card database allows for a better identification of Kocuria in the recent times [28]. Though there are reports of coagulase-negative Staphylococci being wrongly identified as Kocuria by the Vitek 2 system [21], other more recent studies indicate a good reliability of the Vitek 2 system [29, 30]. Our case series includes cases that have been identified by Vitek 2 system alone. The clinical features of the cases in this series included findings like open globe trauma, hypopyon, vitritis, repeat intravitreal antibiotic injections, and an overall fair resolution of the vitritis. These clinical features largely mimic those of endophthalmitis following a Staphylococcus endophthalmitis.

For testing antibiotic susceptibilities, proper guidelines do not exist for Kocuria. Hence, susceptibility breakpoints for Staphylococci are used instead. This can cause a misdiagnosis of either the sensitivity or the resistance pattern in these organisms. Savini et al. [1], in their report, discussed the implications of this misdiagnosis which can lead to recurrent infections and a morbid clinical outcome. It has also been shown that most pathogenic species are susceptible to broad-spectrum antibiotics. In the current reported series, similarly, six of eight cases of Kocuria endophthalmitis reported resolution of infection and inflammation at the last visit. Among the three that did not, two had recurrent retinal detachment and one had corneal decompensation which possibly added to the persistence of the inflammation. This relatively favorable outcome could be attributed to a favorable antibiotic susceptibility pattern as seen in Table 3.

The current study has a few inherent limitations. The effect of various confounding factors could not be independently assessed due to the retrospective nature of the study. The limited sample size did not allow us to assess any parameters that could have influenced the outcome. A proportion of cases of endophthalmitis in this series were post trauma. Trauma itself is a confounding factor for a final unfavorable visual outcome. Thus, it would be difficult to clearly delineate in the post-traumatic subset, whether the unfavorable visual outcome is due to trauma or due to the subsequent endophthalmitis. Though 75% cases in the current study had resolution of endophthalmitis at the last follow-up, visual improvement is limited due to anterior segment opacities. Further corneal intervention for those patients is pending as this manuscript is being written. Following corneal intervention, further visual improvement can be expected. In conclusion, Kocuria endophthalmitis is a relatively rare but emerging cause of endophthalmitis and is often misdiagnosed as CoNS endophthalmitis. The occurrence of Kocuria spp. in patients with endophthalmitis should not be ignored as contaminants. Though clinical presentation can often be variably late, timely and appropriate management with multiple interventions as and when required can result in an acceptable visual and anatomic outcome.