Background

Nocardia are a heterogenous group of aerobic, branching, gram positive, weakly acid-fast bacteria commonly found in dust, decaying vegetable matter, and aquatic environments [1]. Ocular nocardiosis most often presents as keratitis [2, 3]. Nocardia keratitis is a rare, chronic, debilitating cause of keratitis historically associated with trauma [2,3,4,5,6]. Global prevalence is below 2% [2, 5]. It is difficult to diagnose and treat due to a combination of diverse species’ presentations and species-specific response to commonly used topical antibiotics.

Topical amikacin is the current standard of care for medical management of Nocardia keratitis [5, 7]. However, isolates are increasingly diverse and may differ by geography, patient population, and antimicrobial susceptibility [1, 2, 5]. Data on clinical presentation, risk factors, species diversity, and medical management have been reported predominantly for patient populations outside the United States. The purpose of the current study is to characterize and report Nocardia keratitis species diversity and in vitro antibiotic susceptibility, to identify contact lens wear as a risk factor among Nocardia keratitis patients, and to report visual acuity outcomes after treatment.

Methods

The current study is a retrospective, single center, clinical case series. Institutional Review Board (IRB) approval was obtained from the University of Miami Miller School of Medicine Sciences Subcommittee for the Protection of Human Subjects and the research adhered to the Tenets of the Declaration of Helsinki (IRB Protocol Study ID #20070960). Clinical data was collected and analyzed for 26 patients presenting with Nocardia keratitis between January 2014 and September 2021. Extracted data included patient demographics, risk factors, days from symptom onset to presentation, presenting best corrected visual acuity (BCVA), days to resolution, BCVA at last follow-up, and topical steroid use.

A combination of conventional (culture, biochemical assay), molecular (rDNA sequencing), and/or reference laboratories were used to confirm and speciate the Nocardia isolates. Antibiotic susceptibility was determined using a combination of Etests (BioMerieux, Raleigh, NC), commercial laboratories, and the Sensititre Rapmyco microdilution panel (Thermo Fisher Scientific, Waltham, MA). Minimal inhibitory concentrations (MIC) interpretive standards for susceptible and resistant strains were in accordance with manufacturers and Clinical Laboratory Standards Institute (CLSI, Wayne, PA) guidelines [8]. Nonsusceptibility included both intermediate and resistant isolates.

Results

The current study includes 26 eyes of 26 patients. Overall, the median age was 32.5 years (n = 26; range: 16–66 years) and included 17 male and 9 female patients. A diverse group (n = 13) of Nocardia species were recovered among this patient population. N. amikacinitolerans (Fig. 1) was the most frequent isolate (n = 11, 42.3%) followed by N. beijingensis (n = 3, 11.5%), N. arthritidis (n = 2, 7.7%), and one each (3.8%) of remaining 10 isolates detailed in Table 1. Patients with N. amikacinitolerans keratitis were younger with a median age of 24 years (n = 11; range: 16–56 years).

Fig. 1
figure 1

Patient with a Nocardia amikacinitolerans infiltrate. Classically described wreath-like, patchy lesions of Nocardia keratitis at six o’clock in a contact lens wearer

Full size image
Table 1 Clinical Characteristics of Nocardia Keratitis Cases (2014–2021)
Full size table

Complete susceptibility data is summarized in Table 2; in vitro susceptibility daya was not available for a total of one isolate. Amikacin nonsusceptibility was determined in 64% of isolates (n = 16/25). All 11 of the N. amikacinitolerans isolates were resistant to amikacin and constituted 73.3% (n = 11/15) of the amikacin nonsusceptible isolates documented during the study. Of note, 100% of isolates were susceptible to either trimethoprim sulfamethoxazole or linezolid.

Table 2 In vitro antibiotic susceptibility profiles of Nocardia keratitis isolates
Full size table

Mean presenting BCVA (n = 25) was 20/60 ± 2.3 lines. (Table 1). At presentation, 64% (n = 16/25) of the patients had a BCVA of 20/40 or better and a median time from symptom onset to presentation of 10 days (n = 25; range: 3–70 days). The mean post-treatment BCVA (n = 21) was 20/40 ± 2.7 lines with a median treatment duration of 44 days (n = 23; range: 3–190 days). A final post-treatment BCVA of 20/40 or better was achieved in 71.4% (n = 15/21) of patients. Overall, there was no significant difference in presenting versus last follow-up BCVA.

Contact lens wear was the leading identified risk factor for Nocardia keratitis among the study population (Table 1). A history of contact lens wear was present in88.5% (n = 23/26) of patients; the remaining non-contact-lens cases were either associated with trauma. Trauma-related Nocardia keratitis was documented in 15.4% (n = 4/26) of total cases. South Florida patients presenting with Nocardia keratitis were six times (23:4) more likely to be associated with contact lens wear than with trauma.

Discussion

The current study is the largest series to date on risk factors and amikacin-resistance among patient with Nocardia keratitis in the United States. The current series differs compared to reports from Asia by species diversity, risk factors, and amikacin susceptibility profiles [5, 7]. Compared to the largest reported series from India (n = 116) [5], patients in this current series were younger, presented earlier, had better presenting/final BCVA, and healed faster.

The true prevalence of Nocardia keratitis in the United States is unknown, but prior to this study, only one series of three or more patients with Nocardia keratitis had been reported in the United States [9]. Hirst reported on a series of eight patients in 1979. Since then, only sporadic cases (n = 17) have been reported from 10 states and Washington DC [9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25]. Overall, 72% (n = 18/25) have been reported from northern states with only six reported from southern states including Florida (n = 4), Georgia (n = 1), and Texas (n = 1). Nine of the 17 (52.9%) reports have been contact-lens associated supporting the evolving epidemiology in other parts of the United States.

Nocardia amikacinitolerans was the predominant identified Nocardia species among South Florida isolates in the current study resulting in keratitis; this is the second reported case series worldwide. Among more than 200 Nocardia keratitis cases reported from South India in the last three decades, none have been identified as N. amikacinitolerans [4, 5, 7].

Amikacin nonsusceptibility was found to be 64% in this case series. DeCroos and colleagues reported a resistance rate of 3% for their 116 Nocardia keratitis isolates over an 11-year period [5]. Sporadic, but increasing amikacin resistance have been reported for a diverse group of Nocardia keratitis isolates including N. tranvalensis [26], and members of the N. asteroides complex [24].

In vitro susceptibilities for Nocardia species are strain specific. It is important to run in vitro susceptibility testing to determine the most effective drugs for ocular Nocardia infections [1, 5, 27, 28]. Based on in vitro data in this current study, trimethoprim sulfamethoxazole and linezolid demonstrated 100% susceptibility rates. Given its wider availability, trimethoprim sulfamethoxazole may be the preferred antibiotic agent in treating Nocardia keratitis and specifically, amikacin-resistant cases of Nocardia keratitis. Data from the Steroids for Corneal Ulcer Trial (SCUT) study confirmed the correlation between increasing drug minimal inhibitory concentrations and patient’s outcomes.

Contact lens use was the leading risk factors identified among South Florida Nocardia keratitis patients. Contact lens wear was not a recognized risk factors among the 116 cases reported by DeCroos and colleagues nor among the 55 patients in the SCUT study [4]. However, contact lens associated Nocardia keratitis may be increasing worldwide and in the United States [28,29,30]. This infection should be considered with a higher index of suspicion in contact lens wearers with refractory corneal ulcers. Specific details regarding contact lens type or specific hygiene regimen surrounding contact lens use were unable to be determined in this study.

Conclusion

Nocardia keratitis is rare and its clinical presentation is diverse. Contact lens wear is the leading risk factor of Nocardia keratitis in South Florida and has been the most commonly associated risk factor in the United States for the last 10 years. First line therapy with amikacin alone may lead to clinical failure consider trimethoprim sulfamethoxazole. Early collaboration with a microbiology laboratory to speciate and perform susceptibility testing can lead to favorable visual outcomes.