Home About us Editorial board Ahead of print Current issue Search Archives Submit article Instructions Subscribe Contacts Login 
  • Users Online: 1594
  • Home
  • Print this page
  • Email this page

   Table of Contents      
Year : 2006  |  Volume : 54  |  Issue : 1  |  Page : 5-10

Clinico-microbiological profile and visual outcomes of post-traumatic endophthalmitis at a tertiary eye care center in South India

Aravind Eye Hospital and PG Institute of Ophthalmology, 1, Anna Nagar, Madurai, TamilNadu-625020, India

Correspondence Address:
Vasumathy Vedantham
Aravind Eye hospital and PG Institute of Ophthalmology, 1, Anna Nagar, Madurai, Tamil Nadu 625020
Login to access the Email id

Source of Support: None, Conflict of Interest: None

DOI: 10.4103/0301-4738.21607

Rights and Permissions

Purpose: To analyze the clinical and microbiological characteristics of eyes with post-traumatic endophthalmitis and factors influencing the visual outcomes in these cases.
Methods: We performed a retrospective chart analysis of the clinical and microbiological data of 97 consecutive patients with post-traumatic endophthalmitis presenting to a tertiary eye care hospital in South India.
Results: Thirty-nine (40.2%) cases were culture-positive, Gram-positive cocci (n=24/42, 57.1%) being the commonest isolates. Staphylococcus epidermidis and Pseudomonas aeuruginosa were the commonest single isolates (n=10; 23.8%). Majority of the organisms were sensitive to chloramphenicol (n=27) or ciprofloxacin (n=26). Susceptibility to vancomycin (n=3) and amikacin (n=4) was poor. Cases with negative cultures at presentation were more likely to have improvement in visual acuity compared with culture-positive cases (on multivariate analysis, OR: 3.2, 95% CI: 1.1, 9.0).
Conclusions: In this series of post-traumatic endophthalmitis, a high prevalence of resistance of the culture isolates to vancomycin and amikacin was observed.

Keywords: Post-traumatic endophthalmitis, Outcome, Microbiology, Susceptibility profile, Vitrectomy

How to cite this article:
Vedantham V, Nirmalan PK, Ramasamy K, Prakash K, Namperumalsamy P. Clinico-microbiological profile and visual outcomes of post-traumatic endophthalmitis at a tertiary eye care center in South India. Indian J Ophthalmol 2006;54:5-10

How to cite this URL:
Vedantham V, Nirmalan PK, Ramasamy K, Prakash K, Namperumalsamy P. Clinico-microbiological profile and visual outcomes of post-traumatic endophthalmitis at a tertiary eye care center in South India. Indian J Ophthalmol [serial online] 2006 [cited 2021 Mar 9];54:5-10. Available from: https://www.ijo.in/text.asp?2006/54/1/5/21607

Click here to view

Click here to view

Click here to view

Click here to view

Click here to view

Click here to view

Click here to view

Click here to view

Click here to view

Click here to view
Endophthalmitis after penetrating ocular trauma is a distinct subset of intraocular infections in which the precipitating injury, the attendant inflammation, and virulent organisms often result in a less than satisfactory visual outcome. [1],[2],[3],[4],[5],[6] Polymicrobial infections in such cases are very common,[5],[7] but species of Staphylococcus and Bacillus are the commonest reported pathogens.[1],[2],[3],[4],[5],[6],[7]Although many reports on microbial profiles in post-traumatic endophthalmitis are available, susceptibility profiles of the microbial isolates have been discussed only once previously.[7] In addition, there are very few reports from India on this condition.[7],[8] We report the clinical and microbiological data including the susceptibility profiles of a large, retrospective series of eyes with post-traumatic endophthalmitis.

  Materials and Methods Top

We reviewed the medical records of 97 consecutive cases of post-traumatic endophthalmitis managed at the retina-vitreous service of a tertiary eye care institute between January 2000 and December 2001. Information on demographic characteristics of the patients, the time to diagnosis of endophthalmitis following the precipitating event, the presence of associated injuries, and agents causing trauma were retrieved from the medical records. All patients had been subject to a comprehensive ocular examination at presentation, which included visual acuity estimation and refraction (whenever feasible), slit-lamp bio-microscopic examination of the anterior segment, and posterior segment examination with an indirect ophthalmoscope. B-scan ultrasonography was performed for all cases, irrespective of media clarity, to assess the associated injuries and choroidal thickness. The diagnosis of endophthalmitis was made based on subjective symptoms such as pain, and diminished vision in the presence of objective signs such as lid edema, anterior chamber reaction, hypopyon, and/or vitreous exudates.

Undiluted vitreous samples (obtained by aspiration by a syringe connected to the suction port of the vitreous cutter at the beginning of vitrectomy, or with a 23G needle as a tap) and anterior chamber aspirates (collected under aseptic conditions in a tuberculin syringe with a 30G needle) were subjected to microbiological analysis. Eyes with associated infectious keratitis underwent additional corneal scrapings that were sent for microbiological analysis. The samples were plated on blood and chocolate agar, and thioglycollate broth (at 37C) and Sabouraud's dextrose agar (25C) for detection of bacteria (aerobic and anaerobic) and fungi. Gram, Giemsa, and KOH stains were performed on all specimens. A positive culture was defined as growth of the same organism on two or more media or confluent growth on at least one solid medium. Subcultures and antibiotic susceptibility testing by disk diffusion method were performed on all culture-positive cases. Testing for susceptibility for ceftazidime was not carried out owing to nonavailability of the disks. An organism was described as being susceptible to an antimicrobial agent if it were either sensitive or demonstrated intermediate sensitivity, as mentioned previously.[7]

Prior to surgical interventions, subjects received intravenous antibiotics (gentamicin 4 mg/kg body weight and cefazolin 50 mg/kg body weight in divided doses per day), hourly application of topical antibiotics (14 mg /ml fortified gentamicin and 0.3% oflaxacin), hourly application of steroid eye drops (1% prednisolone acetate), and 8-hourly application of cycloplegic eye drops (1% cyclopentolate).

The type of surgical intervention was decided by the operating surgeon based on several factors, including the severity and duration of the infection, type of predisposing injury, and presence of associated injuries such as corneal or scleral tear, and ultrasonography results. Eyes with a small hypopyon, mild vitritis, and visual acuity 20/200 received intraocular antibiotics (IOAB) following a vitreous tap (with a 23G needle) or vitrectomy. Eyes with a moderate-severe presentation, with associated traumatic cataract, and exudates sticking to the back of the lens underwent a lensectomy in addition to the vitrectomy and injection of IOAB. Associated injuries such as corneal or scleral tears and intraocular foreign body (IOFB) were dealt with at the time of primary surgical intervention. Secondary surgical interventions included repeat intravitreal antibiotic injection, repeat vitrectomy, cataract surgery with or without intraocular lens, and retinal reattachment surgery.

We used stata version 7.0 (College Station, Texas, USA) for statistical analysis. Final treatment outcome (visual and anatomic) at the last follow-up was considered for analysis. We defined improvement as a betterment of two or more lines over the baseline visual acuity, whereas worsening was a loss of more than two lines from the baseline visual acuity. Cases that had neither improvement nor worsening based on this criterion were considered as showing no change (status quo). Phthisis bulbi was defined as an eye with an intraocular pressure less than 6 mmHg on applanation tonometry and without any vision. We performed X 2sub-test or Fisher's exact test, as appropriate, to test significance for categorical variables. We performed bivariate analysis to explore the associations between final treatment outcomes and other variables including the setting and agent of injury, presence of associated injuries such as tears, infectious keratitis, IOFB, duration of symptoms, and microbiology profile. Variables that were significant on bivariate analysis were included in a multivariate regression model. Odds ratios (ORs) and 95% confidence intervals (CIs) are presented. We considered P < 0.05 to be significant for this analysis.

  Results Top

The mean age of patients in this series was 25.417.3 years (median: 22.0 years, range: 1-75 years), and 69 (71.1%) were males [Table - 1]. The mean duration between injury and presentation to the center was 7.312.6 days (median: 4.0 days, range: same day to 90 days). Most of the injuries occurred at the workplace ( n= 42, 43.3%), and were mostly owing to vegetable matter ( n= 48, 49.5%). Fifty-eight (59.8%) patients gave a history of prior treatment from an ophthalmologist before presenting to our center. One subject had used traditional medications. Eighty-seven (91.6%) eyes had best corrected visual acuity worse than 20/200, with three eyes having no perception of light at presentation.

A wound of entry was visible in 81 (83.5%) eyes, 74.1% (60/81) of which were self-sealed wounds. Most of the injuries ( n= 69, 87.1%) involved the cornea. Anterior segment findings included infection of the cornea ( n= 31, 32.0%), hypopyon ( n= 74, 76.3%), hyphema ( n= 14 eyes, 14.4%), and cataract ( n= 42, 43.3%). Lens capsule disruption was seen in three (3.1%) eyes. The status of the lens could not be determined in 37 (38.1%) eyes. The status of the retina could not be determined by indirect ophthalmoscopy in 75 (77.3%) eyes. IOFB were present in 14 (14.4%) eyes, primarily in the posterior segment ( n= 13, 92.9%) [Table - 1].

An undiluted vitreous sample was obtained from 92 cases and anterior chamber aspirates from four eyes. Material from corneal scrapings was analyzed for one eye. (This patient refused to undergo any form of surgical intervention.) Thirty-nine (40.2%) cases were culture-positive and the total number of isolates including three polymicrobial cases was 42. These included Gram-positive cocci ( n= 24, 57.1%), Gram-positive bacilli ( n= 7, 16.7%), and Gram-negative bacilli ( n= 11, 26.2%). Staphylococcus epidermidis and Pseudomonas aeuruginosa were the most common amongst single isolates ( n= 10; 23.8%) [Table - 2]. Majority of the organisms were susceptible to chloramphenicol ( n= 27) or ciprofloxacin ( n= 26). Susceptibility to vancomycin ( n= 3) and amikacin ( n= 4) was poor. Intravitreal antibiotics were administered in 91 eyes and 26 eyes underwent pars plana vitrectomy with or without additional procedures. The commonest antibiotics to be injected intravitreally were vancomycin and amikacin (53 cases). [Table - 3] presents results of the antibiotic susceptibility profile for these 39 positive cultures.

The mean duration of follow-up was 7.1 8.6 months (median: 4.0 months, range: 7 days to 36 months).Twelve (12.4%) cases had retinal detachment that was considered inoperable and 24 (24.7%) cases were phthisical at the last follow-up. Thirty-one (32%) eyes had a best corrected visual acuity better than 20/200 at final follow-up [Figure - 1]. Improvement in visual acuity was observed for 31 (32.0%) eyes, 19 (19.6%) eyes worsened and 45 (46.4%) eyes showed no change in vision status. The use of amikacin and vancomycin combination showed no significant association with either worsening (reference category: vision status remained the same post-treatment; OR: 0.5, 95% CI: 0.2, 1.2) or improvement (reference category: vision status remained same post-treatment, OR:1.1, 95% CI: 0.4, 3.5) of visual activity after treatment.

On bivariate analysis, improvement in visual acuity was significantly associated only with negative cultures (OR: 2.7, 95% CI: 1.0, 7.1). The time interval to presentation ( P =0.5), agent of injury (reference category: vegetable matter, OR: 0.7, 95% CI: 0.3, 1.7), visual acuity at presentation 20/200 (reference category: 20/200, OR: 0.5, 95% CI: 0.1, 2.0), presence of IOFB (OR: 3.0, 95% CI: 0.6, 14.9), or presence of hypopyon (OR: 0.7, 95% CI: 0.2, 2.0) were not associated with improvement in vision on bivariate analysis. On multivariate analysis [Table - 4], negative cultures at presentation were associated with improvement in visual acuity (OR: 3.2, 95% CI: 1.1, 9.0).

  Discussion Top

Similar to the other reported studies,[1],[2],[3],[4],[5],[6],[7] we found Gram-positive organisms to be the most common organisms to be implicated in post-traumatic endophthalmitis. Consistent with a major review on post-traumatic endophthalmitis,[6] S. epidermidis was one of the two commonest isolates in our series. Organisms of the Streptococcal species accounted for 31% of isolates in our series. A high incidence of Streptococcal isolates has previously been reported from post-traumatic cases of pediatric age groups[9] and in a previous study from India.[7] However, the number of polymicrobial cases was lower in our series than other reports [Table - 5]. The occurrence of P. aeuruginosa as another commonest isolate is also interesting, and a relatively high predominance of infections owing to these organisms has been reported previously from India.[7]

The reported rate of positive cultures in our series is lower than reported from other studies published previously.[2],[5],[7] The possibility that the lower rate of positive cultures might be related to prior treatment with antibiotics was entertained, especially because 58 (59.8%) patients gave a history of prior treatment from an ophthalmologist before presenting to our center. However, we did not find any statistically significant association ( P =0.28) for prior treatment with antibiotics and positive cultures. The association of negative cultures at presentation with good outcomes suggests that these may potentially be associated with a lower initial microbial load.

Antibiotic susceptibility testing of positive cultures revealed several interesting findings. Of great interest is the poor susceptibility of organisms, especially the Gram-positive cocci and Bacillus species, to vancomycin, which is the commonest intravitreal antibiotic used for management of post-traumatic endophthalmitis. Resistance of this magnitude to vancomycin and amikacin (another commonly administered antibiotic) has not previously been reported in the literature. A previous study from South India[7] had reported higher susceptibility to vancomycin (84.0%) and amikacin (89.6%) compared with our series. It is possible that the widespread use of these two drugs might have lead to the emergence of microbial resistance against them; but, whether their widespread use under diverse clinical conditions provides the necessary selection pressures to promote resistance is debatable.

Although the increasing use of ceftazidime instead of amikacin might address the issue of resistance to amikacin, we could not evaluate the susceptibility to ceftazidime in our series owing to a lack of the disks, and hence are unable to comment. We also explored the possibility as to whether the resistance of the isolates to vancomycin and amikacin was causative for the worsening noted in 19 eyes. Of the 13 culture-positive worsened eyes, 8 had received the said combination. We did not find any statistically significant association between the use of vancomycin and amikacin combination and visual outcomes in these eight eyes. However, this could simply imply that in vitro resistance does not necessarily translate into in vivo ineffectiveness.

We found most organisms isolated to be susceptible to chloramphenicol ( n= 27) and ciprofloxacin ( n= 26), suggesting the potential to use these medications as an alternative to vancomycin and amikacin, respectively. Our study shows a significantly lower rate of resistance of Gram-positive organisms to ciprofloxacin than that found in the Early Vitrectomy Study. Another study from India too has found similar results with 88.4% of Gram-positive organisms being sensitive to ciprofloxacin.[8] Systemic ciprofloxacin has been shown to have good intraocular penetration and may be considered for the management of post-traumatic endophthalmitis.[10],[11] However, the use of either ciprofloxacin as an intravitreal medication needs to be studied further. The efficacy of intravitreal ciprofloxacin has been inconclusive in a previous animal study.[12] Recently, Hui et al .,[13] in a study have demonstrated the greater efficacy of intravitreal ciprofloxacin when compared with ceftazidime, by demonstrating that ciprofloxacin has a much lesser rate of precipitation, irrespective of the medium used to dilute the drug for intravitreal usage, thus accounting for its greater efficacy. There have, however, been recent reports of emerging resistance of Pseudomonas species to ciprofloxacin from India.[14]

There are few published reports on intravitreal use of chloramphenicol. This drug has fallen into disuse in India owing to widespread reports of resistance and this could have resulted in the re-emergence of susceptible isolates. Intravitreal chloramphenicol injected into a rabbit eye model had resulted in the formation of small, posterior subcapsular cataracts but a good efficacy in the eradication of infection when administered early,[15] but one study reported that one human eye with post-traumatic endophthalmitis that received intravitreal chloramphenicol became phthisical.[2] Further studies are required to evaluate efficacy and safety of chloramphenicol for intravitreal use.

Previous studies have reported achievement of a final visual acuity better than 20/400 in 9-50% of persons with post-traumatic endophthalmitis.[1],[2],[16] We found nearly similar results in our series with 32% eyes achieving a final visual acuity better than 20/200. Besides negative cultures at presentation, we could not find any association for improvement in visual acuity. The possibility that the most important determinant of final visual outcome might be related to initial microbial load has emerged in this series. Further studies in a prospective manner are needed to establish the prognostic factors and the most effective antibiotics in the management of post-traumatic endophthalmitis. We also suggest the reintroduction of chloramphenicol into the ophthalmologist's armamentarium and studies to find method to prolong the half-life of intravitreal ciprofloxacin.

  Acknowledgment Top

The work on ocular trauma was supported by the Aravind Medical Research Foundation, Madurai, India.

  References Top

Brinton GS, Topping TM, Hyndiuk RA, Aaberg TM, Reeser FH, Abrams GW. Posttraumatic endophthalmitis. Arch Ophthalmol 1984;102:547-50.  Back to cited text no. 1
Affeldt JC, Flynn HW, Jr, Forster RK, Mandelbaum S, Clarkson JG, Jarus GD. Microbial endophthalmitis resulting from ocular trauma. Ophthalmology 1987;94:407-13.  Back to cited text no. 2
Parrish CM, O'Day DM. Traumatic endophthalmitis. Int Ophthalmol Clin 1987;27:112-9.  Back to cited text no. 3
Boldt HC, Pulido JS, Blodi CF, Folk JC, Weingeist TA. Rural endophthalmitis. Ophthalmology 1989;96:1722-6.  Back to cited text no. 4
Alfaro DV, Roth D, Liggett PE. Posttraumatic endophthalmitis. Causative organisms, treatment and prevention. Retina 1994;14:206-11.  Back to cited text no. 5
Reynolds DS, Flynn HW, Jr. Endophthalmitis after penetrating ocular trauma. Curr Opin Ophthalmol 1997;8:32-8.  Back to cited text no. 6
Kunimoto DY, Das T, Sharma S, Jalali S, Majji AB, Gopinathan U, et al . Microbiologic spectrum and susceptibility of isolates: part II. Posttraumatic endophthalmitis. Endophthalmitis Research Group. Am J Ophthalmol 1999;128:242-4.  Back to cited text no. 7
Anand AR, Therese KL, Madhavan HN. Spectrum of aetiological agents of postoperative endophthalmitis and antibiotic susceptibility of bacterial isolates. Indian J Ophthalmol 2000;48:123-8.  Back to cited text no. 8
Alfaro DV, Roth DB, Laughlin RM, Goyal M, Liggett PE. Pediatric posttraumatic endophthalmitis. Br J Ophthalmol 1995;79:888-91.  Back to cited text no. 9
Keren G, Alhalel A, Bartov E, Kitzes-Cohen R, Rubinstein E, Segey S, et al . The intravitreal penetration of orally administered ciprofloxacin in humans. Invest Ophthalmol Vis Sci 1991;32:2388-92.  Back to cited text no. 10
Alfaro DV, Hudson SJ, Rafanan MM, Moss ST, Levy SD. The effect of trauma on the ocular penetration of intravenous ciprofloxacin. Am J Ophthalmol 1996;122:678-83.   Back to cited text no. 11
Alfaro DV, III, Hudson SJ, Offele JJ, Bevin AA, Mines M, Laughlin RM, et al . Experimental posttraumatic Bacillus cereus endophthalmitis in a swine model. Efficacy of intravitreal ciprofloxacin, vancomycin, and imipenem. Retina 1996;16:317-23.  Back to cited text no. 12
Hui M, Kwok AK, Pang CP, Cheung SW, Chan RC, Lam DS, et al . An in vitro study on the compatibility and precipitation of a combination of ciprofloxacin and vancomycin in human vitreous. Br J Ophthalmol 2004;88:218-22.  Back to cited text no. 13
Garg P, Sharma S, Rao GN. Ciprofloxacin-resistant Pseudomonas keratitis. Ophthalmology 1999;106:1319-23.  Back to cited text no. 14
Koziol J, Peyman GA. Intraocular chloramphenicol and bacterial endophthalmitis. Can J Ophthalmol 1974;9:316-21.  Back to cited text no. 15
Nobe JR, Gomez DS, Liggett P, Smith RE, Robin JB. Post-traumatic and postoperative endophthalmitis: a comparison of visual outcomes. Br J Ophthalmol 1987;71:614-7.  Back to cited text no. 16

This paper received the Rakesh Sharma memorial award and the certificate of merit for the best paper read at the free paper session on Trauma at AIOS Bhubaneshwar 2005.


  [Figure - 1]

  [Table - 1], [Table - 2], [Table - 3], [Table - 4], [Table - 5]

This article has been cited by
1 Could chloramphenicol be used against ESKAPE pathogens? A review ofin vitrodata in the literature from the 21st century
Rok Civljak,Maddalena Giannella,Stefano Di Bella,Nicola Petrosillo
Expert Review of Anti-infective Therapy. 2014; 12(2): 249
[Pubmed] | [DOI]
2 Causative organisms of post-traumatic endophthalmitis: a 20-year retrospective study
Chongde Long,Bingqian Liu,Chaochao Xu,Yuan Jing,Zhaohui Yuan,Xiaofeng Lin
BMC Ophthalmology. 2014; 14(1): 34
[Pubmed] | [DOI]
3 Polymicrobial endophthalmitis: prevalence, causative organisms, and visual outcomes
Animesh Jindal,Mayur R Moreker,Avinash Pathengay,Manav Khera,Subhadra Jalali,Ajit Majji,Annie Mathai,Savitri Sharma,Taraprasad Das,Harry W Flynn
Journal of Ophthalmic Inflammation and Infection. 2013; 3(1): 6
[Pubmed] | [DOI]
4 Endophthalmitis: A review of recent trends
Janice R. Safneck
Saudi Journal of Ophthalmology. 2012; 26(2): 181
[Pubmed] | [DOI]
5 A multicentre prospective study of post-traumatic endophthalmitis
Acta Ophthalmologica. 2012; : no
[VIEW] | [DOI]
Hooshang Faghihi, Fedra Hajizadeh, Mohammad Riazi Esfahani, Seyed Ahmed Rasoulinejad, Alireza Lashay, Ahmad Mirshahi, Reza Karkhaneh, Ali Tabatabaey, Mehdi Khabazkhoob, Shaahin Faghihi
Retina. 2012; 32(1): 146
[VIEW] | [DOI]
Retina. 2011; 31(9): 1806
[VIEW] | [DOI]
8 Guidelines for the Prevention of Infections Associated With Combat-Related Injuries: 2011 Update : Endorsed by the Infectious Diseases Society of America and the Surgical Infection Society
The Journal of Trauma Injury Infection and Critical Care. 2011; 71: S210
[VIEW] | [DOI]
9 Post-traumatic Infectious Endophthalmitis
Neelakshi Bhagat, Saya Nagori, Marco Zarbin
Survey of Ophthalmology. 2011; 56(3): 214
[VIEW] | [DOI]
10 Isolates and Antibiotic Sensitivity of Eighty Culture-Proven Endophthalmitis Cases from Istanbul
Eser, I. and Kapran, Z. and Altan, T. and Ozel Karatas, M. and Aydin, D. and Okaygun, E. and Yilmaz, O.F.
Ophthalmologica. 2008; 222(3): 157-160
11 Microbial cultures in open globe injuries in southern India
Gupta, A., Srinivasan, R., Kaliaperumal, S., Setia, S., Kaliaperumal, S.
Clinical and Experimental Ophthalmology. 2007; 35(5): 432-438
12 Ocular Trauma in Children and in Elderly Patients
Kuhn, F.
Ocular Traumatology. 2007; : 417
13 Microbial cultures in open globe injuries in southern India
Arvind Gupta,Renuka Srinivasan,Subashini Kaliaperumal,Sajita Setia
Clinical & Experimental Ophthalmology. 2007; 35(5): 432
[Pubmed] | [DOI]


    Similar in PUBMED
   Search Pubmed for
   Search in Google Scholar for
 Related articles
    Access Statistics
    Email Alert *
    Add to My List *
* Registration required (free)  

  In this article
Materials and Me...
Article Figures
Article Tables

 Article Access Statistics
    PDF Downloaded539    
    Comments [Add]    
    Cited by others 13    

Recommend this journal