To report the clinical features, antibiotic susceptibilities and treatment outcomes in patients with endophthalmitis caused by nontuberculous mycobacterium.
Noncomparative, consecutive case series.
Retrospective chart review between December 1990 and June 2014.
In the 19 study patients, the clinical setting of endophthalmitis included post–cataract surgery (7/19, 36.8%), post–glaucoma implant (6/19, 31.6%), post–intravitreal injection (2/19, 10.5%), endogenous endophthalmitis (2/19, 10.5%), post–pars plana vitrectomy (1/19, 5.3%), and post–scleral buckle exposure (1/19, 5.3%). Chronic recurrent or persisting ocular inflammation was present in 15 of 19 patients (78.9%). The species isolated were Mycobacterium chelonae in 14 patients (73.7%), M fortuitum in 3 patients (15.8%), M triplex in 1 patient (5.3%), and M avium intracellulare in 1 patient (5.3%). Antibiotic susceptibilities to tested isolates were the following: amikacin (14/16; 87.5%) and clarithromycin (12/16, 75.0%). Intravitreal injections of amikacin (0.4 mg/0.1 mL) were given in 14 of 19 patients (73.7%) with an average of 7 injections per patient (range, 1–24 injections). Intraocular lens removal was performed for 6 of 7 patients (85.7%) with post–cataract surgery endophthalmitis. All the patients with glaucoma implant (6/6, 100%) underwent implant removal. At last follow-up, 6 of 19 patients (31.6%) had best-corrected visual acuity of 20/400 or better.
Endophthalmitis caused by nontuberculous mycobacterium often included chronic recurrent or persistent intraocular inflammation and frequently required removal of ocular device (intraocular lens, glaucoma implant, or scleral buckle). The majority of the isolates were susceptible to amikacin and clarithromycin. Visual outcomes in these patients even after treatment were generally poor.
The genus Mycobacterium encompasses more than 50 species, including typical (tubercular) and atypical mycobacteria (nontubercular). Atypical mycobacteria or nontuberculous mycobacterium (NTM) have been classified into 4 groups by Runyon in 1959 (Photochromogens, Scotochromogens, Non-chromogens, Rapidly growing mycobacteria). These are aerobic, nonmotile, non–spore forming, rapid-growing, acid fast-bacilli that are widespread in the environment. Rapidly growing mycobacteria from Runyon Group IV are associated with ocular infections including canaliculitis, dacryocystitis, scleritis, keratitis, uveitis, orbital cellulitis, panophthalmitis, scleral buckle infections, and endophthalmitis.
Endophthalmitis caused by nontuberculous mycobacterium was first reported in 1989 and was associated with poor visual outcomes. The most common nontuberculous mycobacterium species isolated from patients with endophthalmitis are Mycobacterium fortuitum and M chelonae . Rarely, endophthalmitis has been reported to be caused by other species of nontuberculous mycobacterium, such as M hemophilum , M goodii , M avium , and M gordonae . The purpose of this study is to report the clinical features, antibiotic susceptibilities, and treatment outcomes of nontuberculous mycobacterium endophthalmitis cases from our institution.
The current study is a noncomparative, consecutive case series. A search identifying all culture-positive mycobacteria specimens between December 1990 and June 2014 was performed within a computerized database at the microbiology laboratory of Bascom Palmer Eye Institute. A retrospective chart review of medical and microbiological records of these patients was performed. Nontuberculous mycobacterium was isolated in 19 of 2173 culture-positive endophthalmitis patients during this time period. The study was approved by the Institutional Review Board at the University of Miami, Miller School of Medicine, Miami, Florida. The study was performed in accordance with HIPAA regulations. All patients gave informed consent for treatment.
Patient data including age at presentation, sex, history of immunosuppression, etiology, Snellen best-corrected visual acuity at presentation and last visit, clinical features, microbiology findings, management, and outcomes were recorded. Clinical diagnosis was confirmed by positive culture of intraocular fluid sample (anterior chamber fluid or vitreous) or cornea by standard laboratory protocol, including smear preparation (Gram, Geimsa, 10% KOH wet mount, and Ziehl-Neelsen stain if nontuberculous mycobacterium was suspected) and inoculation of culture media (5% sheep blood agar, chocolate agar, brain heart infusion broth, Sabouraud’s dextrose agar, and thioglycollate broth and Lowenstein-Jenson medium if nontuberculous mycobacterium was suspected). Antibiotic susceptibility was determined by obtaining minimum inhibitory concentrations of various antibiotics using the broth microdilution method (as per National Committee for Clinical Laboratory Standards or Clinical and Laboratory Standards Institute guidelines).
The current study included 19 patients with a mean age at presentation of 62.2 years (range, 23–86 years) and no specific sex predilection (male-to-female ratio 10:9) ( Table 1 ). The mean follow-up duration was 50.4 months (range, 1–180 months) ( Table 2 ). The clinical setting of endophthalmitis was as follows: post–cataract surgery (7/19, 36.8%), post–glaucoma implant (6/19, 31.6%), post–intravitreal injection (2/19, 10.5%), endogenous endophthalmitis (2/19, 10.5%), post–pars plana vitrectomy (1/19, 5.3%), and post–scleral buckle exposure (1/19, 5.3%). Mean duration of symptoms before presentation was 9 weeks (range, 2 days to 1 year) before the diagnosis of endophthalmitis was made. Chronic recurrent or persisting ocular inflammation was present in 15 of 19 patients (78.9%). Associated clinical findings included chronic hypotony (6 patients), retinal detachment (1 patient), corneal infiltrates (2 patients), scleral involvement (2 patients), and hypopyon ( Figure 1 and Table 1 ).
|Patient No.||Age/Sex||Etiology||Presentation (Acute/Chronic)||Mycobacterium Species a (Isolated From)||Associated Systemic Illness|
|1||76/M||Post-cataract surgery||Chronic (H)||M chelonae (Vit)||–|
|2 b||68/F||Post-GDI and PK||Chronic||M chelonae (Vit)||–|
|3||76/F||Post-GDI||Chronic (H)||M chelonae (AC)||Asthma, arthritis, anemia, hysterectomy, cancer|
|4||80/F||Post-cataract surgery||Acute||M chelonae (Vit)||–|
|5||80/M||Post-scleral buckle removal||Chronic||M chelonae (AC)||–|
|6||23/M||Post-GDI||Chronic (H)||M chelonae (Vit)||–|
|7||65/M||Post-GDI||Chronic||M chelonae (Vit)||–|
|8||71/F||Post-cataract surgery||Chronic (H)||M chelonae/abscessus (Vit)||Arthritis|
|9 b||62/M||Post-IVTA||Chronic||M chelonae/abscessus (Vit)||DM, HTN|
|10||69/F||Post-cataract surgery||Chronic||M chelonae/abscessus (Vit)||RT, CT, low platelet, arthritis, asthma|
|11||69/F||Post-intravitreal bevacizumab||Chronic (H)||M chelonae/abscessus (AC)||Low-grade leukemia, CT, RT, lung lobectomy|
|12||53/F||Post-cataract/pupilloplasty surgery||Chronic||M chelonae/abscessus (Vit)||HTN|
|13||54/M||Post-cataract surgery (traumatic cataract)||Chronic||M chelonae/abscessus (AC)||HTN, arthritis, DM, insulin|
|14||74/M||Post-cataract surgery||Acute||M chelonae/abscessus (AC)||–|
|15||35/F||Post-GDI||Chronic||M fortuitum (Vit)||Kidney & pancreas transplant|
|16||55/F||Post-vitrectomy||Chronic||M fortuitum (Vit)||Breast CA, CT and RT, hysterectomy|
|17||86/M||Post-GDI||Chronic (H)||M fortuitum (AC)||–|
|18||33/M||Endogenous||NA||M triplex (Vit)||–|
|19||53/M||Endogenous||Acute||M avium intracellulare (Vit)||HIV and CMV retinitis|
a Mycobacterium species isolated from vitreous or anterior chamber fluid or intraocular lens.
b Patients 2 and 9 were previously were reported by Scott and associates.
|Patient No.||Systemic Therapy (Duration)||Primary Treatment||Implant Removal a||Number of Additional Surgeries b||Number of Intravitreal Amikacin|
|1||None||PPV+ capsulectomy+ IOAB (V+D)||IOL-R||3 c||1|
|2 d||None||GDI-R + IOAB (V+C)||GDI-R||1||0|
|3||Oral clarithromycin (4 weeks)||Revision of GDI + IOAB (V+A)||GDI-R||1 c||24|
|4||Oral clarithromycin (4 weeks)||PPV + IOAB (V+C)||IOL-R + PK||1 c||7|
|5||Oral clarithromycin (4 weeks)||PPV + lensectomy + gas exchange||SB-R||1 c||3|
|6||Oral clarithromycin (1 week)||Tube revision + AC tap||GDI-R||1||0|
|7||None||AC tap + IOAB (V+C+D)||GDI-R||2||2|
|8||None||PPV + IOAB (A)||IOL-R||1 c||7|
|9 d||Intravenous amikacin (1 week)||PPV||Enucleation||2 c||5|
|10||Oral clarithromycin (3 weeks)||PPV + capsulectomy + IOAB (V+C+D)||IOL-R|
|Oral rifampin (2 weeks)|
|Oral azithromycin (4 weeks)||2 c||9|
|11||Oral clarithromycin (4 weeks)||PPV + IOAB (V+G)||–||3 c||8|
|12||Oral azithromycin (1 week)||VIT tap + IOAB||IOL-R||4 c||13|
|13||Oral azithromycin (2 weeks)||AC tap + IOAB (V)||–||2||2|
|14||None||AC wash and biopsy||IOL-R||2 c||0|
|15||Oral ciprofloxacin (2 weeks)||Vitreous tap + IOAB (V+C+D)||GDI-R + IOL-R||1 c||4|
|16||None||Vitreous tap + IOAB (V+C)||–||1 c||9|
|17||Intravenous levofloxacin (1 week)||GDI-R + IOAB (V+C+D)||GDI-R||6 c||1|
|19||Oral clarithromycin (3 weeks)||PPV (diagnostic)||Retinal biopsy||1 c||0|
a Removal of implant such as intraocular lens, glaucoma drainage implant, or scleral buckle.
b Additional procedures excluding IOAB (14/19 patients underwent PPV).
c Patients who underwent PPV during the course of treatment.
d Patients 2 and 9 were previously reported by Scott and associates.
History of systemic illness was present in 6 of 19 patients (31.5%) ( Table 1 ): 5 (5/19; 26.3%) with a history of malignancy and treatment with radiotherapy and/or chemotherapy, and 1 (1/19; 5.3%) with associated cytomegalovirus retinitis. Two patients (10.5%) had received sub-Tenon triamcinolone and may have had a local immunosuppressed environment. Systemic history of diabetes (2/19 patients, 10.5%), arthritis (4 patients, 21.0%), hypertension (3 patients, 15.8%), and cone dystrophy (1 patient, 5.3%) was noted.
All 10 study patients had a positive culture from intraocular fluid (vitreous: 13; anterior chamber fluid: 6). The causative organisms isolated were Mycobacterium chelonae in 14 patients (73.7%), M fortuitum in 3 patients (15.8%), M triplex in 1 patient (5.3%), and M avium intracellulare in 1 patient (5.3%). Using DNA sequencing, 7 subspecies of M chelonae/abcessus were identified. Antibiotic susceptibility was determined for 17 of 19 patients (89.5%) (test not performed for other 2 patients). Table 3 shows the rate of antimicrobial susceptibility, intermediate susceptibility, and resistance in this cohort of patients. The mycobacterial isolates were susceptible to amikacin in 14 of 16 isolates (87.5%) and clarithromycin in 12 of 16 isolates (75.0%). However, resistance was seen to ciprofloxacin in 12 of 15 isolates (80%), to doxycycline in 11 of 12 isolates (91.7%), and to sulfamethoxazole in 9 of 11 isolates (81.8%).
|%||n/N a||%||n/N a||%||n/N a|