|Year : 2004 | Volume
| Issue : 4 | Page : 231-237
Ocular infections due to Non-Tuberculous Mycobacteria
P Lalitha , SR Rathinam , M Srinivasan
Department of Ocular Microbiology, Aravind Eye Hospital and Post Graduate Institute of Ophthalmology, Anna Nagar, Madurai - 625 020, Tamil Nadu, India
|Date of Submission||16-Apr-2004|
|Date of Acceptance||02-Jul-2004|
Department of Ocular Microbiology, Aravind Eye Hospital and Post Graduate Institute of Ophthalmology, Anna Nagar, Madurai - 625 020, Tamil Nadu, India
PURPOSE: To investigate the types and causes of non-tuberculous ocular infections and study their response to topical antibiotic therapy. METHOD: A single center, retrospective review of 18 patients with non-tuberculous mycobacterial ocular infections, seen over a 3 year period was done. Laboratory diagnosis was established by growth on blood agar, LJ medium and Ziehl-Nielsen acid fast stain. RESULTS: Out of 18 patients, six had post corneal graft infection, six had corneal ulcers, three had endogenous endophthalmitis, one had post operative endophthalmitis and two cases were of post surgical wound infection. History of trauma was reported in two cases and surgery in nine cases. M.chelonae was grown in blood agar for all patients. For corneal infections fortified genatmicin and fortified amikacin topical eye drops were given while the cases of endophthalmitis received intravitreal amikacin. Response to treatment was poor in 16 cases (88.9%). Only two cases of corneal ulcer improved after prolonged treatment. There was a misdiagnosis of Corynebacterium spp. on Gram stain in the initial cases. Majority of the isolates were sensitive to gentamicin (72.2%) followed by amikacin (44.4%). CONCLUSIONS: Early clinical recognition and prompt laboratory diagnosis together with aggressive topical antibiotic therapy may shorten morbidity and improve the clinical outcome of non-tuberculous mycobacterial ocular infection.
|How to cite this article:|
Lalitha P, Rathinam S R, Srinivasan M. Ocular infections due to non-tuberculous mycobacteria. Indian J Med Microbiol 2004;22:231-7
|How to cite this URL:|
Lalitha P, Rathinam S R, Srinivasan M. Ocular infections due to non-tuberculous mycobacteria. Indian J Med Microbiol [serial online] 2004 [cited 2021 Jan 19];22:231-7. Available from: https://www.ijmm.org/text.asp?2004/22/4/231/12813
Non-tuberculous mycobacteria (NTM) or atypical mycobacteria are wide spread in the environment and were once thought to be non pathogenic but are now known to cause a variety of human diseases including infections of the eye. They have been implicated in ocular infections as early as 1965 with the first report by Turner et al of a case of chronic keratitis caused by Mycobacterium fortuitum, following the removal of a corneal foreign body. Infections caused by NTM range from external adnexal infections to keratitis,, scleritis, uveitis, endophthalmitis,,, panophthalmitis, and following refractive surgeries like lasik and radial keratotomy. The recent reports on ocular infections caused by NTM show that this group of bacteria is the commonest cause of infective keratitis associated with laser-in-situ keratomileusis., There is one case report of post cataract surgery endophthalmitis caused by NTM. The species of NTM usually associated with ocular infection are M.chelonac and M.fortuitum.,,, Corneal infections with M. flavescens, M avium-intracellulare M.gordonae  and M. marinum have also been reported but appear to be infrequent in comparison.
Most reported ocular infections with NTM are associated with trauma or surgery. The NTM show varying degree of susceptibility to the commonly used antibiotics including the flouroquinolones, aminoglycosides and erythromycin. Amikacin, ciprofloxacin and ofloxacin are considered the drugs of choice.
We report herein 18 culture proven cases of ocular infections caused by NTM. The purpose of this study was to investigate the risk factors, microbiological data and the type of ocular infections and clinical findings in non-tuberculous ocular infections and to study their response to topical therapy and the clinical outcome.
| ~ Materials and Methods|| |
Between January 2000 and June 2003 we investigated 18 consecutive cases of NTM ocular infections. The study included review of the patients' records for clinical details, mode of presentation and clinical outcome. Microbiological records of all ocular specimens processed at the ocular microbiology and immunology laboratory at a tertiary eye care center were also reviewed.
The type of specimens that were processed included corneal scrapings from patients with keratitis and vitreous humour and or aqueous humour from patients with endophthalmitis. The procedure for collection of ocular specimens was as follows: for keratitis, under aseptic techniques, after instilling local anesthesia, the corneal ulcer was scraped using a heat sterilized platinum tipped Kimura's spatula, under slit lamp magnification and the scrapings were inoculated on to culture media and smeared on glass slides directly. Transport medium was not employed. In case of endophthalmitis, either aqueous or vitreous humour was collected aseptically in the operation theater and transported within 15 minutes to the laboratory in a tuberculin syringe, where the culture media were inoculated and smears were made.
The endophthalmitis patients were subjected to routine investigation like random blood sugar levels, blood pressure recording, chest X-rays, HIV testing, collagen vascular disease or pulmonary infections. History was taken pertaining to any past or present systemic illness or medication.
In most cases treatment was started based on smear report. If response to medication was inadequate, the treatment was modified according to the culture and report pattern obtained from the laboratory.
Culture methods included the use of blood agar, chocolate agar, brain heart infusion broth, thioglycollate broth and Sabouraud dextrose agar. Whenever indicated non-nutrient agar with Escherichia More Details coli confluent overlay, and Lowenstein-Jensen medium were included. Smears were made for 10% KOH wet mount and Gram stain. Acid-fast stain was not done as a routine and was included only if the clinical picture was suggestive of NTM infection. The blood agar, chocolate agar and brain heart infusion broth were incubated at 37șC for a period of 7 days, and declared culture negative if there was no growth of bacteria or fungus. If there was positive culture it was identified by colony morphology and standard biochemical methods. A positive culture was defined as growth on two solid medium, or one solid and one liquid medium or positive microscopy and growth on one solid medium. NTM was identified by the colony morphology, rate of growth within 48 hours and ability to grow on blood agar. If the colony on blood agar showed a poorly staining gram positive bacilli, a Ziehl-Nielsen stain was done to confirm whether the organism was acid fast or not. All the isolates were further confirmed to species level by a reference laboratory (Department of Microbiology, Christian Medical College, Vellore). The antibiotic sensitivity testing was done in our laboratory by the disc diffusion Kirby-Bauer method and the results were read after 48 hours.
| ~ Results|| |
Demographics and clinical details
A total of 18 culture positive NTM cases were analyzed. All the isolates were from ocular infections [Table - 1] comprising of corneal graft infection following penetrating keratoplasty (eye nos. 1-6) corneal ulcers (eye nos. 7-12) endogenous endophthalmitis (eye nos. 13-15) and two cases of post cataract surgery wound infection (eye nos. 16,17) and one case of postoperative endophthalmitis (eye no. 18).
The patients' age ranged from 17 to 71 years with the average age being 44 years [Table - 1]. There were eight male subjects (44.4%) and 10 female subjects (55.5%). The time interval between the onset of infection and the initial presentation ranged from 1 to 3 months. A predisposing factor relating to trauma could be identified in only two cases (1%) out of 18. Of the remaining 16, there was a history of surgery in nine patients (50%), six penetrating keratoplasty and three cataract surgery. The main presenting features in all the cases were pain, redness and defective vision. The six cases of graft infections were operated at our hospital around the same time period and developed infection with in 2 to 4 months of surgery. The clinical findings of the graft infections were similar with a small infiltrate at the graft host junction in all six cases. Treatment was started with topical fortified amikacin eye drops. Resolution of the infection took four weeks in all the six cases. There was complete healing in one case and the remaining five cases had graft failure and needed a regrafting.
The corneal ulcers developed insidiously in all cases of which history of trauma was present only in two patients. The size of the ulcer was small to medium sized. Only one patient had a history of previous treatment with antibiotics for two weeks. A complete healing was seen in two patients who regained a visual acuity of 6/60 or better. In four patients, the ulcer healed with scarring and vascularisation. None of the ulcers perforated.
There were three cases, which presented with unilateral endogenous endophthalmitis [Figure - 1]. The common mode of presentation was pain and defective vision. The infection evolved rapidly and was characterized by a marked anterior chamber reaction, hypopyon and a moderate vitreous humor inflammatory reaction. One patient had a granuloma in the anterior chamber. Vitreous and aqueous tap was done for all three patients. Culture was positive for NTM only from the anterior chamber fluid. Sputum, blood and urine were also cultured from these patients to rule out the possibility of foci of infection elsewhere in the body and it was found negative. One eye had to be enucleated because of severe pain and two eyes become phthisical.
Three cases of postoperative endophthalmitis of clinically suspected Nocardia etiology showed a positive growth of NTM from vitreous culture in one case and from the site of incision in the other two cases. There was no globe preservation in the postoperative endophthalmitis case. The two cases of post cataract surgery wound infections showed healing but there was no regain of visual acuity.
A 10% KOH wet mount was done for 11 cases (6%) and was negative for any organism. Gram stain was done on the direct ocular specimen for 15 cases (83%) and 7 cases (46.6%) showed gram positive bacilli, three (20%) had gram positive cocci in pairs, one case (6.6%) had gram negative bacilli and four cases (26%) had no organisms. Ziehl-Nieelsen acid-fast stain was done for 10 cases, out of which eight cases (80%) showed acid-fast bacilli [Figure - 2] and two cases (20%) were negative [Table - 2].
NTM was grown on blood agar [Figure - 3] in all the 18 cases. Lowenstein Jensen medium was included only in three cases of endogenous endophthalmitis, all of which had positive growth [Figure - 4]. There were no associated bacterial or fungal infections. The average time duration for culture to grow was five days in 16 cases and only in two cases of corneal infection it grew with in 48 hours. The sensitivity pattern of the 18 isolates was as follows: 13(72.2%) isolates were sensitive to gentamicin, 8(44.4%) to amikacin, 5(27.8%) to ofloxacin and 6(33.3%) to tobramycin, 4(22.2%) to chloramphenicol, 3(16.7%) to cefazolin and 4(22.2%) to cephotaxime [Figure - 5]. The patients were started on 2% fortified amikacin eye drops for a period of one month. Fortified amikacin was prepared from parental solution in 2% concentration based on the literature. Of the 18 isolates, only one (postoperative endophthalmitis) was resistant to all the drugs.
| ~ Discussion|| |
Non-tuberculous mycobacteria are aerobic, non-sporing, non-motile bacilli. Mycobacterium is the only genus in the family Mycobacteriaceae and contains over 50 species including M.tuberculosis and M. leprae, the causal agents of tuberculosis and leprosy. Members of other species have been termed atypical, anonymous, non-tuberculous, and mycobacteria other than tuberculosis (MOTT). Owing to the many differences in characteristics from the "typical" M. tuberculosis the term NTM and MOTT have gained increasing favour. Runyon proposed a classification of atypical mycobacteria into four groups based on pigment production and rate of growth. Group IV comprises of the rapid growers that form colonies in culture at room temperature within three to five days instead of the 2 to 3 weeks required by the other mycobacteria.
Although isolated infrequently from clinical specimens in the past, the relative importance on these other mycobacterial species have progressively increased in the recent years. This change is associated with improved microbiological diagnostic methods and increased number of patients with abnormal local host defense or systemic deficits of immunity. NTM are ubiquitous in nature and are present as normal flora of human sputum, gastric contents and urinary tract. Environmental exposure to NTM is the most common route of inoculation and many patients have a history of antecedent trauma or prior surgery.The precise source of surgical contamination has sometimes been ambiguous.
In the present series, there were six cases of corneal ulcers of which only two had a history of trauma with stick. The other four cases did not have any predisposing factor for the ulcers. Only two cases of corneal ulcers had a good response to topical medication with fortified amikacin and fortified gentamicin. The treatment was prolonged for a period of one month. Most of the previous reports support this finding of high rate of resistance and the need for prolonged treatment.,,
The six cases of post corneal graft infection were operated at the same time period. The source of infection could be traced to a common eye collection center, where contaminated tap water may have been used. The source of infection could be from the donor eye. External factors like trauma, contact lens and usage of corticosteroid eye drops have also been reported as source of infection. Aylward et al, in their report of one case of corneal graft infection, speculated the possible source of infection to be contact lens, facilitated by the use of postoperative topical steroids. In the report by Sudhesh et al, the patient developed NTM infiltrate seven years later after penetrating keratoplasty following a mild trauma to the eye. Laflamme et al reported M.chelonae keratitis following penetrating keratoplasty with poor response to treatment. Our six patients also had a poor outcome with topical and systemic amikacin and had graft failure and required a regraft.
There were three cases of postoperative infections out of which one case had a delayed presentation with features suggestive of chronic uveitis. Vitreous and aqueous aspirate grew M.chelonae, which was resistant to all antibiotics. Out of the other two cases, one developed wound infection at the site of incision after a period of one month, and the other had a suture abscess. These also had a poor response to treatment. In the report by Ingrid Scott et al, three cases of postoperative endophthalmitis out of a series of five, showed M.chelonae to be an uncommon cause of endophthalmitis with the chronic course of marked intraocular inflammation with generally poor visual outcome.
The literature shows only one report of endogenous endophthalmitis due to M.chelonae in an immunocompromised 67 year old man with mycobacteraemia in association with osteomyelitis of both calcanei. The three cases of unilateral endogenous endophthalmitis in this series occurred in immunocompetent adult females with no foci of infection elsewhere. The response to treatment was uniformly poor in all the three patients. One eye had to be enucleated and two eyes went in for phthisis. The source of infection could not be determined. Antituberculous treatment was started in only one patient, as there was a danger of the other eye being involved.
We did not encounter any case of other ocular infections positive for NTM such as scleritis, adnexal infection, cellulitis, canaliculitis or post keratectomy infections. Post lasik infections with different species of NTM with possible contamination from the environment or water have been reported.
Owing to the slow rate of growth of most NTM, many of these infections may remain undiagnosed if culture plates are not maintained for an adequate time. This period may be as long as eight weeks for the slow growing species. Lowenstein-Jensen and Middle brook agar provide excellent culture media, although these bacteria can also grow on blood agar. Because these bacteria are gram positive and encountered infrequently, they have at times been mistakenly identified on smears as Corynebacterium or Nocardia spp. The initial cases of graft infection in this series were actually misdiagnosed as Corynebacterium spp. based on gram stain but culture grew NTM. Acid fast and/or fluorescein conjugated stains can be helpful in providing an accurate diagnosis. In our experience, after the initial few cases of misdiagnosis, the typical corneal appearance of delayed onset, paucity of suppuration, stromal abscess and multifocal infiltrates were recognized and acid fast stain was done. Out of 10 cases, eight showed acid fast organisms.
To summarize, although NTM ocular infections are rare they are being reported with increasing frequency. Clinicians should suspect NTM infections when treating a patient with a non-healing traumatic ulcerative keratitis especially if associated with the use of corticosteroids or if direct microscopy shows possible diphtheroids. Microbiologists must be aware of these organisms especially when dealing with ocular specimens. Recognition of the colonies on blood agar and confirming any poorly staining gram positive bacilli on gram stain with a Zhiel Neelsen stain would be helpful in establishing the etiology. Better awareness of the disease and early diagnostic and specific therapeutic methods may help in reducing the incidence of debilitating ocular damage from this infection.
| ~ Acknowledgement|| |
Acknowledgement is made to Dr. Savitri Sharma, L.V. Prasad Eye Institute, Hyderabad and Christian Medical College, Vellore for helping us in the speciation of the organisms.
| ~ References|| |
|1.||Turner L, Stinson I. Mycobacterium fortuitum as a cause of corneal ulcer. Am J Ophthalmol 1965;60:329-331. |
|2.||Katowitz JA, Kropp TM. Mycobacterium fortuitum as a cause for nasolacrimal obstruction and granulomatous eyelid disease. Ophthalmic Surg 1987;18:97-99. |
|3.||R ichardson P, Crawford GJ, Smith DW, Xanthis CP. Mycobacterium chelonae keratitis. Aust NZ J Ophthalmol 1989;17:195-196. |
|4.||Lazar M, Nemet P, Bracha R, Campus A. Mycobacterium fortuitum keratitis. Am J Ophthalmol 1974;78:530-532. |
|5.||Lavenson DS, Harrison CH. Mycobacterium fortuitum corneal ulcer. Arch Ophthalmol 1966;75:189-191. |
|6.||Pope Jr J, Sternberg Jr P, Mc Lane NJ, David W.Potts, Doyle R.Stutting. Mycobacterium chelonae scleral abscess after removal of a scleral buckle. Am J Ophthalmol 1989;107:557-558. |
|7.||Weber JC, Schlaegel TF. Atypical Mycobacteria and uveitis. Am J Ophthalmol 1971;72:167-170. |
|8.||Ambler JS , Meisler DM, Zakor ZN, Gerris S.Hall, Thomas J.Spech. Endogenous Mycobacterium Chelonae endophthalmitis, Am J Ophthalmol 1989;108:338-339. |
|9.||Roussel TJ, Stern WH, GoodmanDF, Whitcher JP. Postoperative mucobacterial endophthalmitis, Am J Ophthalmol 1989;107:403-406. |
|10.||Ingrid U.Scott , Douglas F.Liedb, Harry W.Flynn, Amar Dessoulsi, Timothy G.Murthy, Darlene Miller. Endophthalmitis caused by Mycobacterium Chelonae: Selection of antibiotics and outcome of treatment. Arch Ophthamol 2003;121:573-576. |
|11.||Fulcher SFA, Fader RC, Rosa Jr. RH, Holmes GP. Delayed - onset Mycobacterial keratitis after lasik. Cornea 2002;21(6):546-554. |
|12.||Runyon EH. Anonymous mycobacteria in pulmonary disease. Med Clin North Am 1959;43:273-289. |
|13.||Moore MB, Newton C, Kaufman HE. Chronic Keratitis caused by Mycobacterium gordonae. Am J Ophthalmol.1986;102:516-521. |
|14.||Knapp A. Stern GA, Hood CI. Mycobacterium avium-intracellulare corneal ulcer. Cornea 1987;6:175-180. |
|15.||Schonherr U, Naumann GOH, Lang GK, Bialasiewicz AA. Sclerokeratitis caused by Mycobacterium marinum. Am J Ophthalmol 1989;108:607-608. |
|16.||Samuel CM, Huang, H Kaz Soong, Jen-Shiang Chang, Yu-Sung Liang. Non Tuberculous Mycobacterial kertitis: a study of 22 cases. Br J Ophthalmol 1996;80: 962-968. |
|17.||Helm CJ, Holland GN, Lin R, Berlin OG, Bruckner DA. Comparison of topical antibiotics for treating Mycobacterium fortuitum keratitis in an animal model. Am J Ophthalmol 1993;116:700-707. |
|18.||Wilhelmus KR, Liesegang TJ, Osato MS, Jones DB. Cumitech 13A, Laboratory Diagnosis of Ocular infections, Washington DC: American Society for Microbiology:1994 |
|19.||Aylward GW, Stacey AR, Marsh RJ. Mycobacterium chelonae infection of a corneal graft. Br J Ophthalmol 1987;71:690-693. |
|20.||Laflamme MY, Poisson M, Chehade N. Mycobacterium chelonae Keratitis following penetrating keratoplasty. Can J Ophthalmol 1987;22:178-180. |
|21.||Bullington Jr. RH, Lanier JD, Font RL. Non Tuberculous Mycobacterial keratitis. Arch Ophthalmol 1992;110. |
|22.||Garg P, Athmanathan S, Rao GN. Mycobacterium chelonae masquerading as Corynebacterium in a case of infectious keratitis: a diagnostic dilemma. Cornea 1998:17:230-232. |
|23.||Solomon R, Donnenfeld ED, Azar DT, Holland EJ, Palmon FR, Pflugfelder SC, Rubenstein JB. Infectious keratitis after laser in situ keratomileusis: results of an ASCRS survey. J Cataract Refract Surg 2003;29(10):2001-2006. |
|24.||Daines BS, Vroman DT, Sandoval HP, Steed LL, Solomon KD. Rapid diagnosis and treatment of mycobacterial keratitis after laser in situ keratomileusis. J Cataract Refract Surg 2003;29(5):1014-1018. |
|25.||Ramaswamy A, Biswas J, Baskar V, Gopal L, Rajagopal, Madhavan HN. Postoperative Mycobacterium chelonae endophthalmitis after extracapsular cataract extraction and posterior chamber intraocular lens implantation. Ophthalmology 2000; 107(7):1283-1286. |
|26.||Abshire R, Cockrum P, Crider J, Schlech B. Topical antibacterial therapy for mycobacterial keratitis: potential for surgical prophylaxis and treatment. Clin Ther 2004;26(2):191-196. |
|27.||Sudhesh S, Cohen EJ, Schwartz LW, Myers JS. Mycobacterium chelonae infection in a corneal graft. Arch Ophthalmol 2000;118:294. |