Abstract

     Aim: The objectives of this retrospective audit were to describe the clinical and microbiological characteristics of Bacterial Keratitis in existing practice at the Department of Ophthalmology of Hamad Medical Corporation (HMC).

     Methods: A simple descriptive retrospective study of the hospital reports of 70 patients were diagnosed as Bacterial Keratitis and treated at the Ophthalmology in-patient department of HMC. Patients with no corneal scraping, or culture and sensitivity findings, were excluded. Demographic and risk factors, clinical and microbiological data are reviewed.

     Results: Of the 70 patients with Bacterial Keratitis from January 2001 to October 2003, 42 (60%) were males, 28 (40%)females; half of them were native citizens, 39 (55.7%) had right eye involvement, 31(44.3%) had left eye involvement.

     The common predisposing factors were ocular trauma, 33 patients (47.1%), Contact lens was the second most common cause, 23 patients (32.9%), ocular surface diseases,11 patients (15.7%) and 1 patient (1.4%) for each of Lagophthalmos, steroid eye drops and infected corneal suture. Systemic risk factors were diabetes mellitus, 13 (18.6%), and immunosuppression 2 cases (2.8%). Eleven (15.7%) patients had poor visual outcome. Offending organisms could be isolated in 35(50%) cases only, pseudomonas was in 20/35 cases (57%), 6/35 cases (17%) were due to streptococcus pneumonia and the third common bacteria was staphylococcus aureus 4/35 (11%). Empirical topical antibiotic treatment was unchanged in 56 (80%) and modified by culture results in 14 (20%) patients. The mean hospital stay was 3 weeks. Topical steroid employed in 4 (5.7%) cases. 27/70 (38.5%) patients were also treated by systemic antibiotics. Eight patients (11%) required surgical intervention.

     Conclusion: In the patients admitted in Hamad Medical Hospital, Keratitis often occurs following ocular trauma and in contact lens wearers. Pseudomonas, streptococcus pneumoniae and staphylococcus aureus were the major causative organisms. Proper sampling and microbiological work are recommended to minimize the negative bacterial isolates.

Introduction

     Corneal infection is the leading cause of ocular morbidity and blindness worldwide, bacterial Keratitis is a potentially devastating infection that can rob a patient of sight. The spectrum of corneal pathogens is largely dictated by the local microbial flora, which account for disparate rates of various pathogens reported in series from different localities (1). The flora isolated in healthy individuals consist primarily of Staphylococcus epidermidis and diphthroids. Species of greater virulence, such as Staphylococcus aureus, Streptococcus pneumoniae, Pseudomonas aeruginosa, and even Neisseria meningitides, have been reported (2). The spectrum of bacterial Keratitis can also be influenced by geographic and climatic factors; many differences in Keratitis profile have been noted between populations living in rural or city areas, in Western, or in developing countries(3). The aim of the study was to identify predisposing factors, and to define clinical and microbiological characteristics of bacterial Keratitis in our current practice in Qatar.

Patients and Methods

     This is a retrospective audit. We have analyzed the records of 70 inpatients of all ages with bacterial Keratitis who were diagnosed, depending upon their clinical and microbiological results, and were treated in the inpatient department of Hamad Medical Corporation during the period January 2001 to October 2003. Patients with non-bacterial causes of Keratitis, and also those with no corneal scraping findings, were excluded.

     The following data was collected from each chart: patient's age, sex, and nationality. History and Examination were focused upon the following risk factors: corneal trauma, contact lens wear, ocular surface diseases, lagophthlamos, steroid eye drops and corneal foreign bodies or sutures as well as history of systemic diseases particularly diabetes mellitus. Corneal infiltrates classified to small (1mm or less) and large (>1mm) in size, and to central or peripheral. Data was reviewed for the visual acuity (VA) at presentation and final VA (best corrected at least 2-3 months after recovery). It was then Classified according to the WHO (World Health Organization)(4) ( see table 1), antibiotic treatment (empirical or culture guided), culture and sensitivity results, steroid drop therapy, and surgical interventions.
 

     The data was entered into the computer and analyzed with SPSS Windows version 9. Library Medline and Internet were the main databases used for the literature search conducted in preparation of this article.

Results

Clinical Consideration

     Patients have been classified according to their age into three groups: children (< 10 years) 7.1%, adolescents (10-20 years) 10% and adults (> 20 years) 82.9%. Sex distribution (F: M) was 1.5:1, 42 (60%) male and 28 (40%) females; half of them were native citizens.

     Predisposing factors along with the causative organisms are summarized in Table 2.
 

Table 2:  Culture results risk factors cross tabulation, polybacterial infection noted in 3 cases

     Trauma was the most common risk factor and this was encountered in 33 patients (47.1%), 11 (15.7%) of them were caused by pseudomonas. Contact lenses were the second most common cause 23 patients (32.9%), 6 (8.6%) of them were due to pseudomonas. Ocular surface diseases were present in 11 patients (15.7%) and 1 patient (1.4%) for each of Lagophthlamos, steroid eye drops, and infected corneal suture. Systemic risk factors were diabetes mellitus, 13 (18.6%), immunosuppression, 2 cases (2.8%).

     The visual outcome was relatively good as shown in Figure 1. The rise in the number of patients gaining good vision over a period of 2-3 months of follow up, was also supported by a significant decline in the number of cases with moderate and poor vision. Those with normal or near normal visual acuity 6/6 - 6/12 had increased by 16 cases (from 11 to 27 cases), those with moderate vision loss 6/18 - 6/36 diminished by 9 cases (from 24 to 15) and those with visual failure 6/60 or less had showed different outline. Those with 6/60 - 3/60 vision experienced an increase by 4 cases (from 6 to 10) and this could be explained on the basis that many cases had been checked for their vision unaided, since they were still intolerant to wear contact lenses or the new refraction was not done yet.
 

     While those with 2/60 - HM vision demonstrated reduction by 11 cases (from 20 to 9 cases).

     The final visual outcome in correlation to the predisposing factors showed diverse profile (Table 3). Visual impairment (6/60 or less) was found mostly in patients having surface ocular disease (9 out of 11 cases); this can be justified by the contribution of primary corneal disease to more visual impairment. Whereas normal or near normal vision (6/6 - 6/12) was found in the majority of the cases of contact lens wearers (10 out of 23) and in (15 out of 33) in traumatic cases.
 

     Keratitis involved the right eye in 39 patients 55.7% and the left eye in 31(44.3%). The location of the infiltrates was central in 41(58.5%) patients and peripheral in 29 (41.5%) cases. As the dimensions of the lesion were not measured for all patients, we didn't include this parameter in this study. Hypopyon was present in 15 cases (21.4%)

Microbiological consideration

     Table 2. In 70 cases, only 35 (50%) had bacteria identified from the corneal cultures. There were no obvious records about, whether antibiotic therapy had already been initiated or not, before hospital admission, so we didn't include this factor in the study.

     The bacterial spectrum is shown in Table 2, the main causative organism is pseudomonas, which was recovered in 20/35cases (57%), 6/35 cases (17%) were due to streptococcus pneumonia and the third commonest bacteria was staphylococcus aureus 4/35 (11%). One case was diagnosed for each of staphylococcus epidermidis, other streptococci and bacillus species and two cases for Moraxella Lacunata.

Treatment and Clinical Outcome

     All 70 patients in this study were initially managed empirically with first line fortified broad-spectrum antimicrobial treatment such as Keflex + Gentamycin (K+G) 47(67.1%) or Fortum+ Gentamycin (F+G) 23(32.9%); it was modified in 14 cases (20%) depending on the results of culture and sensitivity tests and it was continued unchanged on the initial treatment in 56 (80%) patients. (Figure 2A & 2B)

     The length of hospitalization ranged from 1 to 6 weeks with mean of 2.5 weeks and the prolonged hospital stay (3 weeks or more) mostly was associated with cases of pseudomonas infection 12/17 cases. The mean follow up was 3 months after discharge from the hospital. 4 (5.7%) received topical steroid during the course of their treatment. 27/70 (38.5%) patients received systemic antibiotic treatment.

     Two cases (2.9%) had penetrating keraoplasty for visually disabling residual corneal opacification; 3 cases (4.3%) ended with evisceration because of progression of the infection into unresolving endophthalmitis, one case needed conjunctival flap to assist the imminent perforating ulcer and two cases required removal of corneal suture. Pseudomonas is the commonest offending organism that caused Keratitis requiring surgical intervention. (Table 4).

Discussion

     Bacterial Keratitis is rare in the absence of predisposing factors. Until recently, most cases of microbial Keratitis were associated with ocular surface disease, but the widespread use of soft contact lenses has greatly increased the risk of bacterial Keratitis (5). In Jules Gonin Eye Hospital in Switzerland, the risk factors were identified as wearing contact lenses 36%, blepharitis 21%, trauma 20%, xerophthalmia 15%, keratopathies 8% and eye lid abnormalities 6%(2).

     Wearing contact lenses was the second most common cause of Bacterial Keratitis, accounting for 23(32.9%) of our cases, it was more prevalent among females and mainly caused by pseudomonas. (Table 2)

     Recent studies show a decrease of corneal ulcers following traumas, which is a far more common predisposing factor in rural areas, or low-income countries where it accounts for up to 77.5% of cases.

     In our collection of 70 cases of bacterial Keratitis, 33(47.1%) had bacterial Keratitis after ocular trauma (corneal abrasions, foreign bodies, wounds), such a high figure of traumatic etiology could be explained by the high incidence of occupational eye injuries that is associated with increasing constructional work in the last few years in Qatar.

     Extended wear is one of the main risk factors for complications in disposable soft contact lens use (6). Overnight wearing of contact lenses is the overwhelming risk factor for ulcerative Keratitis among contact lens users (7).

     The visual prognosis after bacterial Keratitis depends on the size, locality, and depth of the ulcer, as well as on the risk factors and the bacteria isolated(8). In this study 11(15.7%) patients had poor vision (hand motion perception or worse) and 6/11 were predisposed to corneal infection because of chronic ocular surface disorders (chronic trachoma, chronic herpetic Keratitis or ocular pemphigoid). The association of bacterial Keratitis is common in compromised cornea(9).

     Central locations of the ulcer are all major risk factors that can necessitate penetrating keraoplasty(8). Although 41(58.5%) presented with central corneal infiltration, only two cases required keraoplasty.

     The goal of initial antibiotic therapy for bacterial Keratitis is the proper selection of a drug, which has coverage for the aetiopathogen. Microscopic evaluation of corneal smear can provide insight into the identity of the pathogens(1). Although empirically guided therapy may suffice in cases of Keratitis caused by antibiotic-susceptible bacteria, there is a risk that resistant bacteria may result in unnecessarily poor visual outcome if the microbiological diagnosis is not made (10).

     Generally the isolation rate was estimated to be 70%(11), and T.Bourcier, et.al, have identified a high isolation rate of micro-organisms (68% in 201 cases) in a similar study conducted in Quinze-Vingte National Center of Ophthalmology, Paris, France (3). Moreover, it was 86% in Jules Gonin Eye Hospital, University of Lausanne in Switzerland.

     Our microbiological results of the corneal scraping samples were positive only in 50% of the cases, Pseudomonas 20 (28.6%), Streptococcus Pneumonia 6 (8.6%), staphylococcus Aureus 4(5.7%) while other microbes were found infrequently (Table 4), this microbial profile is completely different from other areas in the world. The causative organism identified in bacterial keratitis varies significantly based on the geographic location of patients. For example, it was Staphylococcus Aureus 49%, Streptococcus 9% and pseudomonas 8% in New York but it was 16%, 8% and 19% respectively in Florida(12).This low figure of bacterial isolation in our practice could be caused by one or more of the following: inaccurate sampling and inoculating technique, culturing complexity or due to the earlier antibiotic treatment.

     It is often not possible to culture fastidious or partially treated organisms successfully, even in the best series, up to 20% of cases of presumed infective keratitis remain culture negative (13). In the study by McDonnell and associates, 50% of all patients were diagnosed and treated without microbial culture, and after a survey, more than 80% of the respondent ophthalmologists prescribed non-fortified, commercial-strength antibiotics for the initial management of corneal ulcer. Furthermore, 6% of the survey respondents used topical Corticosteroids as part of their routine initial therapy(14).

     In this study, we found that the empirical treatment was successful in 80% of cases Figure (2A) and culture guided treatment was applied in 20%. This result is comparable to the previous studies, which have demonstrated that most community-acquired bacterial ulcers could be resolved with broad-spectrum empiric therapy (15).
 

     As corneal scraping is essential beyond its diagnostic value, it may accelerate disease resolution by enhancing antibiotic penetration and the therapeutic debridment of necrotic tissue (5), we would continue to do it for all patients with suspected Bacterial Keratitis, except for those with small or peripheral infiltrates.

     Although treatment regimens may vary between centers and individual practitioners, the gold standard has been the use of topical cefazoline and aminoglycosides. Vancomycin has been poorly adapted for topical use due to its pharmacokinetics and problems with pH that necessitate phosphate-buffered saline for reformulation. Fluoroquionlone antibiotics that are available for topical ophthalmic use demonstrate excellent broad-spectrum activity and low toxicity. Our Fortified antibiotics used for empirical treatment were (Keflex Gentamycin) used in 47(67%) cases and (Fortum Gentamycin) used in 23 (33%) of cases (Figure 2B). Clinical use of these agents is equivalent in effectiveness to combination of cefazoline and aminoglycosides (5).
 

     Stern and Buttross, in an extensive recent review of the debate surrounding the use of steroid treatment in microbial Keratitis, concluded that topical steroids should be reserved for cases of culture positive bacterial Keratitis in which resolution is incomplete after an initial period of intensive treatment with appropriate microbial antibiotic. We recommend that steroid medication is introduced, if appropriate, during phase of healing on initial therapy(13). Topical steroid therapy was applied only in 4 (5.7%) of our patients.

     Subconjuctival antibiotics were not used in our treatment strategy for bacterial Keratitis. These injections are unpleasant for the patient, introduce a new array of possible complications and do not enhance therapeutic efficacy where intensive topical treatment is used.

     Systemic antibiotics were prescribed in 27/70 (38.5%) patients 15/27 of them were with hypopyon and 12/27 had large or peripheral infiltration.  Ciprofloxacin given systemically (750 mg twice daily) is copiously secreted in the tears and has excellent intraocular penetration.  Other systemic antibiotics are not routinely used because they produce low corneal concentration ¹⁶.

     Surgical treatment were required in 8/70(11%) cases; 2 keratoplasties for residual scar, and 3 eviscerations due to corneal perforation and progression to endophthalmitis, one conjunctival flap for imminent corneal perforation and 2 corneal sutures removal for infected sutures (table 4).
 

    
     The unresponsiveness or poor responsiveness to antibiotic therapy in our practice may possibly be a result of polymicrobial infection, atypical organisms, inadequate antibiotic selection or application. Additionally the sensitivity tests were usually made only for particular antibiotics by the microbiology lab and unfortunately most of these antibiotics are not frequently recommended for the topical use. Therefore we extended the empirical treatment in most cases (80%).

Conclusion

1.  Despite not completely accurate representation of the population, however, this review study does give an outlook about the spectrum of bacterial Keratitis in Qatar.

2. Pseudomonas and, to a lessor extent, streptococcus pneumoniae bacteria were the major causes of Keratitis, principally following ocular trauma or contact lens wear.

3. Adequate ocular prevention, knowledge of the microbial pattern in given clinical practice, and prompt choice of appropriate fortified antibiotics constitute the management of bacterial Keratitis (2). This necessitates a further study to identify the pattern of microbial ocular flora in Qatar.

Recommendations

1. It is very important for clinical documentation to be full and accurate, as many clinical criteria were overlooked and this would make this work not comprehensive.

2. For better bacterial isolation, an appropriate corneal sampling and microbiological work is necessary, and this requires:

     a) Routine microscopy of smears is useful to help screening out non-bacterial infection (fungal or acanthameoba), which could be the cause for unresponsive cases with negative growth.

     b) It is practically more beneficial to work mutually with microbiologist to have sensitivity tests for those antibiotics used and prepared topically for the eye.

     c) Contact lens cultures may identify the causative organisms in most contact lens-related Keratitis (18).

3. A need to expand public education about the benefits of work protective goggles and also the proper use of contact lenses.

References

1. Bharathi M J, et al. In-vitro Efficacy of Antibacterial against Bacterial Isolates from Corneal Ulcers. Indian J Ophthalmol 2002; 50:109-114.

2. Schaefer F, et al. Bacterial Keratitis: a prospective clinical and microbiological study. B.J.O. 2001; 85:842-847.

3. Bourcier T, et. al. Bacterial Keratitis: predisposing factors, clinical and microbiological review of 300 cases. B.J.O. 2003; 87:834-838.

4. Crouch ER Jr, William PB. Trauma to the uveal tract: Traumatic hyphema. Duane's clinical Ophthalmology. Philadelphia: JB Lippincott 2001; 4.61:1-1.

5. Benson W H, and Lanier J D. Current diagnosis and treatment of corneal ulcers. Current opinion in ophthalmology 1998; 9,IV: 45-9.

6. Musch DC, et al. Demographic and Predisposing factors in Corneal Ulceration. Arch Ophthalmol 1983; 101:1545-8.

7. Lipener, Ribeiro AL. Bilateral Pseudomonas corneal ulcer in a disposable contact lens wearer. CLAO J 1999 Apr; 25(2): 123-4

8. Schein, et al. The impact of overnight wear on the risk of contact lens - associated ulcerative Keratitis -Arch Ophthalmol.994; 112(11): 1404-6.

9. Meidziak AI, et al. Risk factors in microbial Keratitis leading to penetrating keraoplasty. Ophthalmology 1999; 106:1166-71.

10. Goldstein MH, Gordon YJ. Emerging Fluoroquionlone Resistance in bacterial Keratitis; a 5 years review. Ophthalmology 1999; 106:1313-18.

11.Diamond J, et al. Corneal biopsy with tissue micro-homogenisation for isolation of organisms in bacterial Keratitis. Eye 1999; 13: 545-549

12. Waxman E, et al. Single culture media in infectious Keratitis. Cornea 1999; 18:257-61.

13. Morlet N, Daniell M. Empirical fluoroquionlone therapy is sufficient initial treatment. B.J.O. 2003; 87:1169-1172.

14. Allen B, John KG, Dart. Strategies for management of microbial Keratitis. B.J.O. 1995; 79:777-786.

15. Pepose J S and Wilhelmus K R. Divergent Approaches to the management of corneal ulcers. American Journal of ophthalmology 1992; 114(5): 630-1

16. McLeod S.D, et al. The role of smears, cultures and antibiotics sensitivity testing in the management of suspected infectious Keratitis. Ophthalmology 1996; (18): 23-28

17. Microbial Keratitis. Jack J.Kanaski. Clinical Ophthalmology. Third Edition 1995;(5) 105.

18. Gritz, D. C, J P Whitcher. Topical issues in the treatment of Bacterial Keratitis. International ophthalmology clinics 1998; 38(4): 107-114

19. Martin, EN. et al. Infectious Keratitis: correlation between corneal and contact lens cultures. CLAO J 2002; 28(3): 146-8

Other Topics:

Original Study # 1 -  Childhood Poisoning.  Is There a Greater Role for emergency Medicine?
Original Study # 2 -  Obstetric conditions requiring intensive care admission:  A five year survey
Original Study # 3 -  Eclampsia in Qatar:  Maternal and fetal outcomes, possible                                  preventive measures.