Historically, the study of diabetes mellitus (DM) ocular manifestations has mostly been focused on the management and prevention of diabetic retinopathy and maculopathy as their ability to lead to irreversible blindness is well known [1]. However, ocular infections are also an important and sometimes underrated difficulty in diabetic people, because they greatly contribute to the patients' discomfort, decrease their quality of life, and are a reason for a big part of eye consultations in both primary and tertiary healthcare settings [1]. One of the main reasons why people with diabetes get infected is that the disease weakens the immune system and some of the ways it does this are through the improper functioning of neutrophils, slow movement to the infection site, and reduced uptake of the pathogens by the immune cells, and all these three factors together make the patients prone to a wide range of infections [2-4]. The affected ocular parts are eyelids, conjunctiva, cornea, and nasolacrimal duct, and there are many studies documenting that diabetic people are more prone to suffer from blepharitis, dacryocystitis, conjunctivitis, and infectious keratitis than non-diabetic individuals [5-7]. The presence of infections is more common in poorly controlled diabetes because continuous high blood sugar level breaks down the skin barrier, lowers the activity of the tear film and disrupts the normal ocular flora [8]. In spite of these findings, there is still a serious lack of systematically collected clinical data that would reveal the actual effect of diabetes on ocular infections. Very few small-scale studies have been done in this area with mixed results. In one study that included 328 subjects, no relationship between blood sugar level and superficial eye infections was found, while other studies claimed that diabetes that was not well controlled led to higher chances of getting conjunctivitis and to longer recovery times, but these findings were limited by small sample size and poor monitoring of blood sugar levels [9-10]. Bacterial conjunctivitis is still one of the most common external ocular infections globally, which is responsible for almost 1% of all consultations in primary care and tens of millions of visits to clinics and hospitals every year [11-12]. The conjunctiva, a semitransparent membrane that covers the sclera and the inside of the eyelids, acts as a physical and immunological barrier to microbes. The ocular surface normally harbors a commensal flora composed mainly of Gram-positive bacteria like Staphylococcus epidermidis and Corynebacterium spp. This flora plays a role in preventing the growth of pathogens by maintaining microbial balance [13-14]. Nevertheless, in the case of diabetic patients, this equilibrium is often upset which leads to the settlement of pathogenic bacteria and the increased danger of infection being caused by the bacteria already present [15-16]. Moreover, it has been demonstrated that patients with diabetes are more susceptible to postoperative endophthalmitis and other ocular surface infections, mainly due to the pathogenic abilities of the indigenous flora found in the conjunctiva, eyelid margin, and nasal mucosa [17-18]. Long-standing high blood sugar levels not only change the microbial community but also slow the healing of the epithelium which makes the patients more susceptible to an outbreak of the infection that would require a stronger antibiotic treatment due to the fact that it would be mixed and recurrent with a higher rate of complications [19]. Thus, it is of utmost importance to have a better grasp of the microbiological patterns, risk factors, antibiotic sensitivity profiles, and clinical outcomes of bacterial conjunctivitis in diabetic versus non-diabetic patients. The modification of glycemic control as an example of a modifiable factor might provide important clues for targeted prevention and better therapeutic approaches. This study was designed to assess the risk factors, microbial profiles, antibiotic susceptibility patterns, and clinical outcomes of bacterial conjunctivitis in diabetic and non-diabetic patients, and to evaluate the effect of glycemic control on infection severity, healing duration, and complication rates.

A prospective comparative study was performed at Al-Kindy Teaching Hospital and Imam Ali General Hospital, Baghdad, Iraq, lasting for 18 months from January 2024 to June 2025. The study aimed to assess and compare the factors, microbial spectrum, antibiotic susceptibility, and clinical outcomes of bacterial conjunctivitis in diabetic and non-diabetic individuals. A total of 200 patients who were clinically diagnosed with bacterial conjunctivitis were included in the study and were allocated into two equal groups, Group A (diabetic, n = 100) and Group B (non-diabetic, n = 100).

Inclusion Criteria

Participants between the ages of 20 and 70 years with classic signs of bacterial conjunctivitis—such as mucopurulent discharge, redness of conjunctiva, swelling of eyelids, and positive bacterial culture—were considered for inclusion.

Exclusion Criteria

Patients suffering from viral, allergic, or chlamydial conjunctivitis, those who had undergone recent ocular surgery or experienced trauma, and contact lens-related keratitis or treatment with topical or systemic antibiotics seven days prior were excluded from the study. Patients with immunocompromised conditions other than diabetes mellitus were also excluded in order to reduce the confounding effects.

Clinical Evaluation

At the time of enrollment, all patients received a thorough ophthalmologic evaluation that included best-corrected visual acuity, slit-lamp biomicroscopy to assess the degree of conjunctival hyperemia, corneal clarity, and the presence or absence of discharge or membrane formation. Each participant's detailed medical history was taken with a focus on duration of ocular symptoms, previous episodes of infections, associated systemic diseases, and medication history. In diabetic patients, HbA1c levels were measured to evaluate glycemic control and were categorized into well-controlled (≤8%) and poorly controlled (>8%) subgroups for correlation with clinical outcomes.

Microbiological Investigation 

Before the instillation of any topical medication, aseptic technique was used to obtain sterile conjunctival swabs from the inferior fornix. The specimens were subjected to Gram staining and the bacterial type was determined, and a culture was made on blood agar, chocolate agar, and MacConkey agar plates for isolating the organism. Organisms were identified by standard biochemical tests, including catalase, coagulase, oxidase, and indole reactions where appropriate. Susceptibility of the organism to antibiotics was measured using the Kirby–Bauer disk diffusion method in accordance with Clinical and Laboratory Standards Institute (CLSI) guidelines. The antibiotic panel included moxifloxacin, tobramycin, ciprofloxacin, erythromycin, and chloramphenicol, which were selected based on local prescribing practices and efficacy in ocular infections.

Treatment Protocol

Soon after sample collection, all patients were placed on empirical topical antibiotic therapy, and this was subsequently adjusted based on the culture and sensitivity results. Supportive treatment, including artificial tears and eyelid hygiene, was administered as necessary. During the treatment, the diabetic patients were instructed to keep their glycemic levels strictly under doctor's supervision.

Follow-Up and Outcome Measures

The patients were seen again on day 1, week 1, and week 3 following the commencement of treatment. At each consultation, improvement of symptoms, discharge clearance, color of the conjunctiva, and healing of the epithelium were recorded. The main outcome measures consisted of: Healing time (length of time till the discharge and redness completely disappeared). Recurrence rate (return of symptoms within 3 weeks of apparent cure) and Occurrence of complications like keratitis, corneal ulceration, or hospitalization due to severe infection.

Statistical Analysis

Data analysis was performed using SPSS version 26.0 (IBM Corp., Armonk, NY, USA). Means±standard deviation (SD) were used for quantitative variables and compared by Student’s t-test, while categorical variables were displayed as numbers and percentages and analyzed using the Chi-square test. A p-value of less than 0.05 was regarded as statistically significant.

Table 1 demonstrates notable demographic differences between diabetic and non-diabetic patients enrolled in the study. The mean age of the diabetic group (56.4±8.1 years) was significantly higher than that of the non-diabetic group (45.2±10.4 years; p<0.001), reflecting the greater prevalence of diabetes among older individuals and suggesting age-related vulnerability to ocular infections. Gender distribution showed a near-equal representation in both groups, with a slight male predominance—58 (58.0%) males among diabetics and 54 (54.0%) among non-diabetics—indicating no statistically significant difference (p = 0.52). Body mass index (BMI) was markedly higher in diabetic patients (28.6±3.9 kg/m²) compared to non-diabetics (25.1±3.3 kg/m²; p = 0.001), signifying an association between obesity and diabetes, both known risk factors for impaired ocular defense mechanisms.

Table 1: Demographic characteristics of patients

Table 2 outlines the comparative distribution of predisposing risk factors among diabetic and non-diabetic patients with bacterial conjunctivitis. Poor glycemic control, reflected by HbA1c levels >8%, was observed in 64 (64.0%) of diabetic patients, confirming the high prevalence of uncontrolled diabetes as a major systemic contributor to infection susceptibility. Contact lens use was more frequent among non-diabetic individuals—20 (20.0%) versus 12 (12.0%)—though the difference did not reach statistical significance (p = 0.11), likely reflecting lifestyle-related exposure rather than systemic factors. Conversely, chronic blepharitis was significantly more prevalent among diabetic patients (26.0% vs. 10.0%; p = 0.004), suggesting that chronic eyelid margin inflammation may serve as a reservoir for recurrent bacterial infection in this group. Similarly, dry eye syndrome occurred in 34 (34.0%) of diabetics compared to 18 (18.0%) of non-diabetics (p = 0.01), consistent with the known association between diabetes, tear film instability, and ocular surface dysfunction. Previous ocular surgery showed a higher but statistically insignificant frequency in diabetics (15.0%) compared with non-diabetics (7.0%; p = 0.09), implying that postsurgical ocular surface alterations may further predispose diabetic eyes to bacterial infection.

Table 2: Distribution of risk factors

Table 3 highlights the comparative clinical manifestations and severity of bacterial conjunctivitis in diabetic and non-diabetic patients. Purulent discharge was the most common symptom in both groups—92 (92.0%) among diabetics and 88 (88.0%) among non-diabetics—with no significant difference (p = 0.34), indicating that purulent secretion remains a universal hallmark of bacterial conjunctivitis regardless of systemic status. However, diabetics exhibited significantly higher rates of severe conjunctival redness (80.0% vs. 60.0%; p = 0.002) and eyelid edema (65.0% vs. 48.0%; p = 0.03), reflecting a more intense inflammatory response and possibly delayed resolution due to compromised vascular and immune functions. Importantly, corneal involvement was observed in 22 (22.0%) of diabetic patients compared with only 8 (8.0%) of non-diabetics (p = 0.01), suggesting that diabetes predisposes to deeper tissue extension and potential keratitis.

Table 3: Clinical presentation and severity

Table 4 presents the microbiological spectrum of bacterial conjunctivitis among diabetic and non-diabetic patients, revealing similarities in the predominant pathogens but notable differences in infection patterns. Staphylococcus aureus emerged as the leading causative organism in both groups—38 (38.0%) in diabetics and 40 (40.0%) in non-diabetics (p = 0.74)—confirming its universal role as a primary conjunctival pathogen. Streptococcus pneumoniae and Pseudomonas aeruginosa were the next most frequent isolates, with slightly higher detection in diabetics (16.0% and 14.0%, respectively) compared to non-diabetics (18.0% and 9.0%), although these differences were not statistically significant. The isolation of Klebsiella spp. and coagulase-negative Staphylococci was comparable between the two groups, indicating a shared background flora. However, a significant distinction was observed in the prevalence of mixed bacterial infections, which occurred in 28 (28.0%) of diabetic cases versus 12 (12.0%) of non-diabetic cases (p = 0.01). This finding suggests that diabetics are more prone to polymicrobial infections, likely due to impaired local immunity, reduced tear antimicrobial activity, and delayed epithelial healing.

Table 4: Distribution of bacterial isolates

Table 5 compares the antibiotic sensitivity profiles of bacterial isolates obtained from diabetic and non-diabetic patients with conjunctivitis, revealing subtle yet clinically relevant differences in antimicrobial response. Both groups demonstrated the highest sensitivity to broad-spectrum fluoroquinolones—moxifloxacin (90.0% vs. 94.0%) and tobramycin (88.0% vs. 90.0%) — with no tatistically significant difference (p = 0.40 and p = 0.68, respectively). These findings confirm the continued efficacy of these agents as first-line treatments for bacterial conjunctivitis in both populations. Ciprofloxacin sensitivity was slightly lower among diabetic isolates (74.0%) compared with non-diabetics (82.0%), although not statistically significant (p = 0.19), suggesting emerging resistance trends, possibly due to prior empirical antibiotic exposure. Notably, erythromycin resistance was more pronounced in diabetic patients, with only 46.0% sensitivity compared to 64.0% in non-diabetics (p = 0.03), indicating reduced effectiveness of macrolides in managing infections among diabetics. Chloramphenicol showed moderate sensitivity in both groups (58.0% vs. 72.0%), nearing statistical significance (p = 0.06).

Table 5: Antibiotic sensitivity pattern

Table 6 presents a comparative analysis of healing duration and post-infectious complications between diabetic and non-diabetic patients with bacterial conjunctivitis. The findings clearly indicate that diabetes exerts a significant negative impact on recovery and complication rates. The mean healing time among diabetic patients was 8.2±1.4 days, markedly longer than 5.6±1.2 days observed in non-diabetic individuals (p<0.001), reflecting delayed epithelial regeneration due to impaired microvascular circulation, reduced tear film stability, and weakened immune response associated with chronic hyperglycemia. Recurrent infection occurred in 14 (14.0%) of diabetics compared to 5 (5.0%) of non- diabetics (p = 0.03), suggesting that persistent bacterial colonization and poor glycemic control contribute to relapse. Furthermore, vision-threatening complications such as keratitis and corneal ulcer were significantly higher in diabetics—9 (9.0%) and 5 (5.0%), respectively—whereas these were rare or absent in non-diabetic patients (p = 0.04 and p = 0.02). Hospitalization was also more frequently required among diabetics (12.0% vs. 3.0%; p = 0.01), indicating more severe or refractory disease courses.

Table 6: Healing and complications

Table 7 examines the relationship between glycemic control and clinical outcomes among diabetic patients with bacterial conjunctivitis, underscoring the strong influence of metabolic regulation on infection behavior and therapeutic response. Patients with well-controlled diabetes (HbA1c ≤ 8%) demonstrated a significantly shorter healing period (6.5±1.2 days) compared with those with poor control (HbA1c > 8%, 9.3±1.5 days; p<0.001), indicating that hyperglycemia directly impairs ocular surface repair and immune defense mechanisms. Recurrent infections were notably higher among poorly controlled diabetics (20.0%) than in those with good control (6.0%; p = 0.01), reflecting persistent bacterial colonization and delayed epithelial recovery in a hyperglycemic environment. Similarly, keratitis occurred more frequently in patients with poor glycemic control (12.0% vs. 3.0%; p = 0.03), suggesting that elevated glucose levels promote corneal susceptibility to microbial invasion. Additionally, the antibiotic resistance was doubled in the poorly controlled subgroup (36.0% vs. 18.0%; p = 0.04), possibly due to recurrent infections necessitating repeated antibiotic exposure and impaired drug penetration in compromised ocular tissues.

Table 7: Effect of glycemic control (Diabetic group only)

The outcome of this strategic comparative inquiry is that the condition, diabetes mellitus, significantly hinders the risk, severity, and results of bacterial conjunctivitis. No matter which parameter was looked at—demographic, clinical, microbiological, or therapeutic—the status of being diabetic was associated with the occurrence of a more aggressive disease and slower recovery. Diabetes patients had a significantly higher age and body weight compared to non-diabetic patients in the study. These two factors can together impair the ocular surface defense and epithelial regeneration. Old age and other related metabolic problems have been associated with less tear film stability and changes in the conjunctival flora, which could lead to infection. Similar demographic trends were found by Adam et al. [1] and Baykal et al. [13], who pointed out that diabetes makes one more prone to ocular surface's microbial colonization, due to the presence of microangiopathy and immune dysfunction. Ocular surface disorders were dramatically more common among diabetic participants. Chronic blepharitis (26.0%) and dry eye syndrome (34.0%) were significantly more common than in the control group, which is corroborated by Karimsab and Razak [14] and Alkot et al. [9]. Such conditions change the composition of the tear film and make it easier for bacteria to adhere and become re-infected [18-22]. Although the non-diabetic group reported higher contact lens use, it is the systemic factors present in diabetics that—rather than the behavioral exposure—drive the increased risk of conjunctival infection [23,24,25]. From a clinical perspective, patients with diabetes had more significant symptoms and experienced similar disease marked by extreme redness (80.0%), swollen eyelids (65.0%), and involvement of the cornea (22.0%). These discoveries were in line with previous research demonstrating a more substantial inflammatory reaction and corneal involvement in diabetics [11,12,15]. Martins et al. [16] said that poorer corneal healing and lower tear antimicrobial activity due to continuous hyperglycemia promote bacterial infiltration and extend recovery. The larger corneal involvement is parallel to the mechanisms which Phillips and Tasman [17] described where delayed epithelial repair and neuropathy allow deeper infection spread. The microbiological profile in our study is comparable to that of other studies showing Staphylococcus aureus as the predominant isolate in both groups (26-29). However, the diabetic group showed a significantly greater rate of mixed bacterial infections (28.0% vs 12.0%) which is in agreement with the trend noted by Fernández-Rubio et al. [12] and Nahar et al. [13] who claimed biofilm formation, lowered immunity, and colonization at lid margins as the reasons for this. In concordance with past studies, Pseudomonas aeruginosa and Klebsiella spp. were more in number among diabetics indicating a greater likelihood of gram-negative colonization [17,19]. These are opportunistic and thus, more likely to cause corneal complications, especially when glycemic control is poor [29,30]. The analysis of antibiotic susceptibility revealed that fluoroquinolones and aminoglycosides were still very effective in both groups. Moxifloxacin (90–94%) and tobramycin (88–90%) exhibited strong activity, consistent with the ARMOR surveillance findings by Asbell et al. [2] and Thomas et al. [3]. The sensitivity of erythromycin, however, was notably lower in diabetic isolates (46.0% vs 64.0%), thereby reinforcing earlier reports of increasing macrolide resistance in diabetic ocular infections [4,5,6]. The decreased sensitivity of ciprofloxacin among diabetic isolates (74.0% vs 82.0%) may be connected to previous antimicrobial exposure and chronic colonization [22,28]. Hence, empirical macrolide monotherapy should not be employed in diabetic conjunctivitis, and culture-based selection is suggested instead [7,8]. Not only healing outcomes but also complication rates were indicative of the adverse effects of diabetes. Mean healing time was longer in diabetics (8.2±1.4 days) compared with non-diabetics (5.6±1.2 days), and recurrence was three times more frequent (14.0% vs 5.0%). Suto et al. [11] and Gupta et al. [20] have reported similar delays, explaining that diabetics suffer from prolonged epithelial defects and greater microbial persistence. The higher occurrence of keratitis (9.0%) and corneal ulcer (5.0%) is consistent with the pathophysiological role of hyperglycemia in impeding.

DM significantly worsens the clinical course of bacterial conjunctivitis, increasing polymicrobial infection, resistance, complications, and recovery time. Early culture-guided therapy plus strict glycaemic control are essential to optimise outcomes.

1. Adam, M., et al. “Conjunctival Flora in Diabetic and Nondiabetic Individuals.” Turk Journal of Ophthalmology, vol. 45, no. 5, 2015, pp. 193–196.

2. Asbell, P.A., C.M. Sanfilippo, and C.M. Pillar. “Antibiotic Resistance among Ocular Pathogens in the United States: Five-Year Results from the Antibiotic Resistance Monitoring in Ocular Microorganisms (ARMOR) Surveillance Study.” JAMA Ophthalmology, vol. 138, no. 6, 2020, pp. 557–564.

3. Thomas, R.K., et al. “Antibiotic Resistance among Ocular Pathogens: Current Trends from the ARMOR Study.” Clinical Ophthalmology, vol. 13, 2019, pp. 2223–2232.

4. Bispo, P.J.M., et al. “A Systematic Review of Multi-Decade Antibiotic Resistance Data for Ocular Pathogens.” PLoS ONE, vol. 17, no. 10, 2022, e0271234.

5. Haas, W., et al. “Monitoring Antibiotic Resistance in Ocular Microorganisms.” Archives of Ophthalmology, vol. 129, no. 6, 2011, pp. 795–801.

6. Manente, R., et al. “Prevalence and Antimicrobial Resistance of Causative Agents in Ocular Infections: A Five-Year Study.” Antibiotics, vol. 11, no. 4, 2022, 463.

7. Cabrera-Aguas, M., et al. “Antimicrobial Resistance in Ocular Infection – A Review.” Clinical & Experimental Ophthalmology, vol. 52, no. 3, 2024, pp. 389–398.

8. Chatterjee, T., et al. “Analytical Study of Conjunctival Bacterial Flora in Diabetic and Non-Diabetic Subjects and Their Antibiotic Sensitivity Pattern.” Journal of Clinical and Diagnostic Research, vol. 15, no. 7, 2021, pp. NC06–NC09.

9. Alkot, M., et al. “Conjunctival Flora in People with Type 1 or Type 2 Diabetes Mellitus.” International Journal of Medical Arts, vol. 6, no. 6, 2024.

10. Ashtamkar, S., R. Maheshgauri, B. Vadodaria, et al. “A Study to Compare the Conjunctival Flora of Non-Diabetic Individuals with That of Diabetic Patients.” Journal of Ophthalmology Research, 2024, e-publication.

11. Suto, C., et al. “Conjunctival Sac Bacterial Flora Isolated Prior to Cataract Surgery.” Infection and Drug Resistance, vol. 5, 2012, pp. 37–41.

12. Rubio, M.E., et al. “The Conjunctival Bacterial Pattern of Diabetics Undergoing Cataract Surgery.” Eye (London), vol. 24, 2010, pp. 825–834.

13. Baykal, O. “Conjunctival Flora in Patients with Type 1 or Type 2 Diabetes Mellitus.” Advances in Therapy, vol. 24, no. 5, 2007, pp. 1028–1035.

14. Karimsab, D., and S.K. Razak. “Study of Aerobic Bacterial Conjunctival Flora in Patients with Diabetes Mellitus.” Nepal Journal of Ophthalmology, vol. 5, no. 1, 2013, pp. 28–32.

15. Arbab, T.M., S. Qadeer, S. Iqbal, and M.A. Mirza. “Aerobic Bacterial Conjunctival Flora in Diabetic Patients.” Pakistan Journal of Ophthalmology, vol. 26, no. 4, 2010, pp. 177–181.

16. Martins, E.N., et al. “Aerobic Bacterial Conjunctival Flora in Diabetic Patients.” Cornea, vol. 23, no. 2, 2004, pp. 136–142.

17. Phillips, W.B., and W.S. Tasman. “Postoperative Endophthalmitis in Association with Diabetes Mellitus.” Ophthalmology, vol. 101, no. 3, 1994, pp. 508–518.

18. Zhang, Z., et al. “The Effect of Diabetes Mellitus on Chronic Rhinosinusitis and Sinus Surgery Outcome.” International Forum of Allergy & Rhinology, vol. 4, no. 4, 2014, pp. 315–320.

19. Shanmugam, P., et al. “The Bacteriology of Diabetic Foot Ulcers, with a Special Reference to Multidrug-Resistant Strains.” Journal of Clinical and Diagnostic Research, vol. 7, no. 3, 2013, pp. 441–445.

20. Gupta, A., et al. “Microbiology and Visual Outcomes of Culture-Positive Bacterial Endophthalmitis in Oxford, UK.” Graefe’s Archive for Clinical and Experimental Ophthalmology, vol. 252, no. 11, 2014, pp. 1825–1830.

21. Lim, H.W., et al. “Endogenous Endophthalmitis in the Korean Population: A Six-Year Retrospective Study.” Retina, vol. 34, no. 3, 2014, pp. 592–602.

22. Kurokawa, N., et al. “Increasing Ofloxacin Resistance of Bacterial Flora from Conjunctival Sac of Preoperative Ophthalmic Patients in Japan.” Japanese Journal of Ophthalmology, vol. 46, no. 6, 2002, pp. 586–589.

23. Sones, Y., J.P. Craig, and A. Tomlinson. “Importance of Lipid Layer in Tear Film Stability and Evaporation.” Optometry and Vision Science, vol. 74, no. 8, 1997, pp. 8–13.

24. McNatt, J., et al. “Anaerobic Flora of the Normal Human Conjunctival Sac.” Archives of Ophthalmology, vol. 96, no. 10, 1978, pp. 1448–1450.

25. Coşkun, M., et al. “Analyzing Normal Conjunctival Flora and Detecting Antibiogram Sensitivity to Fluoroquinolones and Penicillin Derivates.” Glo-Kat, vol. 2, 2007, pp. 167–170.

26. Singer, T.R., S.J. Isenberg, and L. Apt. “Conjunctival Anaerobic and Aerobic Bacterial Flora in Paediatric versus Adult Subjects.” British Journal of Ophthalmology, vol. 72, no. 6, 1988, pp. 448–451.

27. Comerie-Smith, S.E., J. Nunez, M. Hosmer, and R.L. Farris. “Tear Lactoferrin Levels and Ocular Bacterial Flora in HIV Positive Patients.” Advances in Experimental Medicine and Biology, vol. 350, 1994, pp. 339–344.

28. Hayashi, Y., T. Miyamoto, and S. Fujita. “Bacteriology of the Conjunctiva in Pre-Cataract Surgery Patients with Occluded Nasolacrimal Ducts and Their Operation Outcomes in Japanese Patients.” BMC Ophthalmology, vol. 17, 2017, p. 15.

29. Liao, H.R., H.W. Lee, and H.S. Leu. “Endogenous Klebsiella pneumoniae Endophthalmitis in Diabetic Patients.” Canadian Journal of Ophthalmology, vol. 27, no. 3, 1992, pp. 143–147.

30. Lloyd, J.C., and R. Braga-Mele. “Incidence of Postoperative Endophthalmitis in a High-Volume Cataract Surgical Centre in Canada.” Canadian Journal of Ophthalmology, vol. 44, no. 3, 2009, pp. 288–292.

31.