Intravitreal dexamethasone as adjuvant in the treatment of postoperative endophthalmitis: a prospective randomized trial

  • Ivan M. Gan
  • Luana C. Ugahary
  • Jaap T. van Dissel
  • Eric Feron
  • Ed Peperkamp
  • Marc Veckeneer
  • Paul G. H. Mulder
  • Gert Jan Platenkamp
  • Jan C. van Meurs
Clinical Investigation

DOI: 10.1007/s00417-005-0133-1

Cite this article as:
Gan, I.M., Ugahary, L.C., van Dissel, J.T. et al. Graefe's Arch Clin Exp Ophthalmo (2005) 243: 1200. doi:10.1007/s00417-005-0133-1

Abstract

Purpose

To study whether intravitreal dexamethasone as adjuvant to intravitreal antibiotics improves the outcome in patients with suspected postoperative bacterial endophthalmitis.

Design

Prospective randomized clinical trial.

Setting

Tertiary referral center.

Patient population

Twenty-nine consecutive patients with suspected postoperative bacterial endophthalmitis within 6 weeks of cataract surgery.

Intervention

Patients underwent a vitreous biopsy followed by intravitreal injection of antibiotics (0.2 mg vancomycin and 0.05 mg gentamicin) and 400 μg dexamethasone or placebo. After 3–4 days the intravitreal injection of antibiotics and dexamethasone or placebo was repeated

Primary outcome measure

Snellen visual acuity at 3 and 12 months after treatment.

Results

In 20/29 patients (69%) the vitreous cultures were positive. 13/29 patients received dexamethasone. Seven patients had a functionally lost eye (final vision of hand movements or less), in four due to retinal detachment. Visual acuity tended to be better in the dexamethasone treated patients than in those given placebo, at both 3 months (P=0.055) and 12 months (P=0.080).

Conclusion

This small prospective, placebo-controlled series showed a trend towards a better visual outcome in patients with suspected bacterial endophthalmitis when treatment with intravitreal antibiotics was combined with intravitreal dexamethasone. Our findings justify a larger multicenter randomized study.

Keywords

Endophthalmitis Dexamethasone Adjuvant 

Introduction

Postoperative bacterial endophthalmitis is a rare, but severe complication of intraocular surgery. The infecting bacteria, by replication and release of toxins, and the inflammatory response of the host both contribute to the damage inflicted on the ocular structures [1]. In the treatment of bacterial meningitis, corticosteroids are of proven use, in addition to antibiotics, in reducing morbidity and mortality [4, 14, 19], presumably by limiting the recruitment of leucocytes, by stabilizing the blood-brain barrier (which is very similar in nature to the blood-retina barrier) and by producing cytoprotectants.

Indeed, corticosteroids have a beneficial effect on the outcome of endophthalmitis in most animal studies [11, 13, 15, 16]. The endophthalmitis vitrectomy study (EVS) [7], by far the largest prospective controlled clinical trial concerning treatment of postoperative endophthalmitis, demonstrated the effectiveness of intravitreal administration of antibiotics. As oral glucocorticosteroids were routinely administered to all patients in this study, their role could not be studied. Only two controlled trials have studied the effect of steroids as adjuvant to antibiotic treatment and their results were contradictory [3, 17].

In the present study, we investigated whether intravitreal dexamethasone as adjuvant to intravitreal administration of antibiotics improves the outcome in the treatment of patients with suspected postoperative bacterial endophthalmitis.

Materials and methods

Starting from April 1999, consecutive patients with a diagnosis of suspected bacterial post-cataract endophthalmitis presenting at the Rotterdam Eye Hospital were asked to participate in this study. The study protocol was approved by the Institutional Review Board. The patients were randomized to either intravitreal dexamethasone or placebo as adjunct to standard antibiotic treatment. The original sample size was 128 patients (64 per group), which allowed an effect on logMAR of −0.3 to be detectable with 80% power using a significance level of 0.05 (2-sided). The SD of logMAR was set at 0.6.

Suspected postoperative bacterial endophthalmitis was defined as follows: (1) severe, sudden visual deterioration and (2) an inflammatory response deemed excessive (relative to the anticipated course after surgery) with cells and hypopyon in the anterior chamber or posterior segment with loss of fundus detail. Informed consent was obtained from all subjects. Patients were excluded when (1) the cataract surgery had been performed more than 6 weeks previously, (2) cataract surgery had been performed without the expectation of a postoperative vision of 20/100 or better, (3) systemic or subconjunctival antibiotics had been administered, (4) there was a suspicion of fungal infection.

Our treatment approach has been described previously [9]. In short, a vitreous biopsy with a vitrectome was taken in patients with vision of hand movement or more, whereas a limited core vitrectomy with an anterior chamber infusion was performed in patients with light perception only. Undiluted vitreous was collected for Gram-staining and culture before the start of empiric antibiotic therapy; the vitreous was plated within 2 h and cultured aerobically and anaerobically. Immediately after the vitreous biopsy was taken, patients received an intravitreal injection of 0.2 mg vancomycin in 0.1 ml phosphate-buffered saline and 0.05 mg gentamicin in 0.1 ml phosphate-buffered saline, as well as 400 μg dexamethasone sodium diphosphate (Decadron 20 mg/ml with 1 mg/ml paraben as preservative, resulting in 0.025 mg paraben in 0.1 ml) or placebo in 0.1 ml phosphate buffered saline.

The intravitreal injection of 0.2 mg vancomycin and dexamethasone or placebo was repeated once after 3 or 4 days.

In the first 18 patients, the second injection was preceded by a vitreous biopsy with the vitrectome, to obtain a repeat bacterial culture and, if there was sufficient material left, for measurements of the intravitreal vancomycin, gentamicin and/or dexamethasone concentration. If the Gram-staining or culture of the first biopsy material yielded Gram-negative bacteria, 1 mg ceftazidime was immediately injected intravitreally, followed by continuous intravenous infusion of ceftazidime (6 g per day) [18].

After the first biopsy, patients used Predforte eye drops 6 times a day in a tapering schedule over the next 6 weeks and atropine 1% for 4 weeks. Additional procedures were allowed for the following indications, if the treating surgeon thought it to be in the patients' best interest: increasing inflammation, increasing pain, non-clearing opacities of the media, retinal detachment or macular pucker.

The primary outcome measure was Snellen visual acuity at 3 and 12 months. The difference between the dexamethasone and placebo group in vision at 3 and 12 months follow-up was tested for statistical significance with the Mann-Whitney test.

Results

Enrolment of patients

Twenty-nine patients were enrolled in the study between April 1999 and June 2000; in mid-June 2000 the study had to be terminated prematurely because the study drug, i.e. dexamethasone sodium diphosphate (Decadron, 20 mg/ml), was no longer available.

Thirteen patients received dexamethasone, 16 patients received placebo.

Patient characteristics

The intervals between cataract surgery, the first signs of endophthalmitis and the first injection of antibiotics and dexamethasone or placebo were similar in the two groups (Table 1). Two patients in the dexamethasone group presented with counting fingers vision, all other patients had hand movement vision or light perception (Table 2).
Table 1

Patient characteristics

Patient

Dexa or placebo

Age

m/f

Culture

Vision on admission

Vision at 3 mnth

Vision at 12 months

Remarks

1

P

47

m

Sterile

HM

0.25

0.25

Retinitis pigmentosa

2

D

52

m

Staph epidermidis

HM

1.0

1.0

 

3

P

79

v

Staph epidermidis

LP

0.1

0.15

 

4

D

73

m

Sterile

LP

CF

0.25

RD, vitrectomy and SO

5

D

81

m

Sterile

LP

0.15

0.5

 

6

P

75

v

Staph epidermidis

LP

LP

LP

RD, no further surgery

7

P

79

v

Sterile

HM

0.5

0.8

 

8

D

82

m

Staph epidermidis

HM

0.8

0.8

 

9

D

86

v

Staph epidermidis

CF

0.25

0.7

 

10

D

82

v

Sterile

LP

0.4

0.7

 

11

P

80

v

Staph epidermidis& moraxella spp.

LP

0.15

0.8

 

12

P

80

v

Staph aureus

HM

CF

0.1

Dropped IOL, vitrectomy+oil, RD, phthisis

13

D

83

m

Staph aureus

LP

0.15

?

 

14

P

72

v

Staph epidermidis

HM

HM

HM

Seclusio pupillae, surgical iridectomy

15

P

74

v

Staph epidermidis

LP

CF

HM

RD, vitrectomy and SO

16

P

81

v

Staph epidermidis

LP

CF

LP

Pucker, RD, vitrectomy and SO

17

P

63

m

Sterile

HM

0.7

0.8

 

18

D

82

v

Strep sanguis

LP

LP

NLP

Hypotony, vitrectomy and SO, RD and PVR

19

D

82

v

Staph aureus

CF

0.5

0.4

 

20

P

76

v

Sterile

HM

HM

HM

Vitrectomy, IOL removal, hypotony syndrome

21

D

75

m

Sterile

HM

0.4

0.4

 

22

P

72

m

Staph epidermidis

HM

0.7

0.5

 

23

D

78

m

Staph epidermidis

HM

1.0

0.8

 

24

P

79

m

Sterile

HM

LP

LP

 

25

D

59

m

Staph epidermidis

HM

1.0

1.0

RD, vitrectomy with gas

26

P

71

v

Staph epidermidis

HM

0.15

0.4

 

27

P

68

v

enterococc+Strep viridans+Staph.epidermidis

LP

0.15

0.15

 

28

D

67

v

Staph warneri

HM

0.15

0.25

Amblyopia

29

P

76

v

Staph warneri

HM

0.5

1.0

 

StaphStaphylococcus, Strep Streptococcus, , bd below detection, LP vision of light perception only, HM vision of hand movement, CF vision of counting fingers, RD retinal detachment, PVR proliferative vitreoretinopathy, IOL intraocular lens, SO silicone oil

Table 2

Baseline, preoperative and postoperative visual characteristics

 

Placebo

Dexamethasone

Baseline characteristics

Number of patients

16

13

Age (median age) in years

47–81 (76)

52–85 (82)

Male/ female

4/12

8/5

Interval cataract surgery and first treatment (median)

2–40 (7) days

1–29 (7)

Preoperative vision

  

 LP

6

5

 HM

10

6

 CF

2

Culture results

  

 Sterile

5

4

 Staph epidermidis

8

5

 Staph Warneri

1

1

 Staph aureus

1

2

 Strept sanguis

 

1

 Strept viridans

1

 

Further possible risk factors related to outcome

RD

4

3

LP

6

5

Second biopsy

9

9

Second biopsy and RD post-op

4

2

LP and second biopsy

4

5

LP, second biopsy and RD

3

2

Secondary procedure

4

3

Outcome variables

Postoperative vision at 3 months

  

 LP-HM

4

1

 CF

3

1

 0.1–0.25

5

4

 0.4–1.0

4

7

Postoperative vision at 12 months

  

 LP-HM

6

1

 CF

 0.1–0.25

5

2

 0.4–1.0

5

9

Microbiological findings

Before the start of treatment, 29 vitreous biopsies were obtained in 29 patients, of which 20 (69%) showed growth of micro-organisms. The relative number of positive cultures (i.e. 70%) did not differ between the placebo and dexamethasone group. Coagulase negative staphylococci (mostly Staphylococcus epidermidis) were most commonly isolated (n=15, Table 1). In this series no Gram-negative micro-organisms were cultured. Of note, all Gram-positive micro-organisms were found to be susceptible to Vancomycin.

Second biopsy

In 18 patients, a second vitreous biopsy was performed. In none of these biopsies was growth of micro-organisms observed, despite positive microbiological findings in 12 of them at first culture. Results on the concentrations of antibiotics and dexamethasone have been reported separately [8, 9].

Complications of endophthalmitis

In seven patients retinal detachment (RD) occurred (four in the placebo group). Seven patients (six in the placebo group) had a functionally lost eye (i.e. final vision of hand movements or less), in four due to RD, in one due to hypotony, in one due to seclusion of the pupil and in one because of persistent inflammation.

In all, seven patients underwent a secondary procedure: three in the dexamethasone group: vitrectomy and gas for RD (one), vitrectomy and silicone oil for persistent inflammation and PVR (one), RD (one) and four in the placebo treated group: vitrectomy and silicone oil for RD (three) and vitrectomy for hypotony to prevent phthisis (one).

Follow-up and recovery of vision

Patient 13 did not wish to return for her last follow-up at 12 months, but by telephone claimed an improvement compared with the 3-month visit. Her vision at 12 months is not included in the analysis.

On non-parametric analysis, there was a trend towards a better visual outcome at 3 months (P=0.055) and at 12 months (P=0.080) in patients treated with dexamethasone.

Discussion

The main finding of the present study was that in patients who presented with suspected postoperative bacterial endophthalmitis, the intravitreal administration of dexamethasone, together with antibiotics, may improve the visual outcome at long term follow-up, while being well tolerated and safe. This conclusion was based on the Snellen visual acuity determined at 3 and 12 months after the acute episode and the fact that dexamethasone did not inadvertently affect the pharmacokinetics of vancomycin administered intravitreally [8, 9], nor did it result in a delay in microbiological clearance.

A major limitation of our study, however, was its small sample size. Because the study dexamethasone (Decadron 20 mg/ml) was withdrawn from the market by the manufacturer (for reasons that had nothing to do with the present study) and the remaining commercially available formula (Decadron 5 mg/ml) contained 4 times as much of the preservative paraben (a concentration shown only just safe in rabbit eyes [12] and half the dose shown to be retinotoxic [10]), the study had to be stopped after enrolment of only 29 patients.

In bacterial endophthalmitis, the virulence of the infecting agent is an important factor in determining the amount of tissue damage. Virulent bacteria are more commonly encountered in bleb-related infections (Gram-negative bacteria) and in trauma (Bacillus). In post-cataract endophthalmitis patients (our study population), the organism was often commensal with a relatively low virulence. The results of our study (where all organisms were susceptible to the chosen antibiotics) therefore apply to post-cataract patients, but not necessarily to other types of endophthalmitis. When steroids are considered as an adjuvant, bacterial susceptibility to the antibiotics used should be a permanent concern and its monitoring a part of the treatment protocol.

Choice of antibiotics and steroids

We have tried to optimize the choice and dosage schedule of the antibiotics and dexamethasone used. Our empiric therapy consisting of a repeated lower but adequate dose of vancomycin (0.2 mg, instead of the usual dose of 1 mg) and gentamicin (0.05 mg, half the lowest dose ever reported to cause macular infarction: 0.1 mg [2]) resulted in antibiotic concentrations well above the minimal inhibitory concentration (MIC) for most micro-organisms for well over a week, while minimizing potential retinotoxic effects [9]. In addition, the combination of vancomycin and gentamicin acts synergistically against many micro-organisms. By stopping bacterial growth and helping to kill the bacteria, antibiotics contribute to the elimination of the infection, but cannot prevent tissue damage caused by bacterial components that act pro-inflammatory and are generated from growing and antibiotics-killed bacteria and the resulting recruitment of host defense factors. Glucocorticosteroids are the best-known agents to reduce tissue damage by limiting the recruitment of leucocytes, stabilizing the blood-retina barrier and the production of cytoprotectants. It is not clear which steroid concentration is needed for such drug effects. By repeating the injection, we achieved concentrations within the pharmacological range over approximately 1 week, despite the short estimated half-life of dexamethasone of 5.5 h we found in eyes with endophthalmitis [8]. With a physiological average cortisol level of 5.1 ng/ml [21] (equivalent to 0.17 ng/ml dexamethasone), the intravitreal dexamethasone concentrations we achieved ranged between 6×105 times the physiologic concentration (100 μg/ml peak values at injection) and 4×102 times the physiologic concentration (median trough values of 69 ng/ml after 3–4 and 6–8 days) [8]. Because preservative toxicity was shown to be the limiting factor in intravitreal dexamethasone administration [10, 12], we used a preparation with a preservative content 4 times lower than that reported as just safe. In fact, these data would support more frequent and lower dosing. This is not realistic, however, given the potential complications of repeated intravitreal injections. Obviously, longer acting, slow-release intravitreal antibiotics and steroids would increase treatment safety and possibly effectiveness.

Comparison with the EVS

When we compared the results of our study with those of the EVS, none of the patients in our study presented with a vision of 5/200 or more, compared with 13% of patients in the EVS. We performed a core vitrectomy in all patients with light perception only, without excluding those patients where media opacities would not allow visualization during a core vitrectomy, as was permitted in the EVS [6]. The percentages of EVS patients with a 9–12 month follow-up vision of greater than 20/40 was 53%, greater than 20/100, 74% and greater than 5/200, 85%, whereas they were only 35%, 47% and 75%, respectively, in our study. The percentage of positive cultures (69%) was similar in both studies, although we relied on vitreous cultures only, without the addition of anterior chamber tap culture or culture from the vitrectomy cassette [7]. Overall, our study population had more negative predictive base-line characteristics than the EVS had and, correspondingly, the results were less favourable. Another factor explaining our worse results was that the percentage of retinal detachments in our study (24%) was higher than in the EVS (9%) [5]. Our initial treatment protocol contained two actions that may have increased the risk for retinal detachment, such as a core vitrectomy in eyes with light perception only with opaque media where neither the vitrectomy tip nor a possible entry site retinal tear could be observed (this group of patients was allowed to be excluded in the EVS); the use of a repeat biopsy in the first 18 patients to assure the levels of antibiotics and steroids as well as the sterility of the culture. Therefore, in our current treatment regime we no longer perform a core vitrectomy in patients with light perception only, but limit our action to a vitreous biopsy. Nor did we continue to take a second vitreous biopsy after the first 18 patients.

All patients in the EVS received 30 mg prednisone orally twice a day for 5–10 days. Weijtens et al. [20] showed that an oral dose of 7.5 mg dexamethasone resulted in a vitreous concentration of 5.7 ng/ml. As the anti-inflammatory potential of dexamethasone is 7.5 times that of prednisone and if we assume that dexamethasone and prednisone reach the eye in comparable amounts, the EVS dosage would result in two daily peaks of 22 ng/ml prednisone over 5–10 days, compared with an at least 12 times greater minimal trough concentration during 8 days in our study patients given dexamethasone [8]. Steroid administration in the EVS in all patients may, however, be another explanation for its favorable results. However, by applying the steroids intravitreally, side effects of systemically administered glucocorticoids such as diabetes mellitus are avoided.

A prospective placebo-controlled trial in 63 patients with postoperative and post-traumatic endophthalmitis in Hyderabad showed a lower initial inflammation score in patients who had received 400 μg dexamethasone intravitreally, but did not report on final visual outcome [3]. A retrospective controlled study in Will's Eye Hospital on 57 patients with post-cataract endophthalmitis, however, revealed a negative effect on visual outcome in patients given 400 μg intravitreal dexamethasone [17].

We conclude that the role of steroids as an adjuvant to intravitreal antibiotics in the treatment of patients with suspected bacterial endophthalmitis is not yet clear. The results of our placebo controlled, prospective study, however, with a trend towards a beneficial effect of steroids, prompted us to initiate a second, multicenter study, now with the use of preservative-free dexamethasone.

Acknowledgements

We gratefully acknowledge the help of Diane A.E. Mertens, Peter J. Ringens, vitreoretinal surgeons and Jan Jansen, Westblaak Pharmacy, Rotterdam.

Copyright information

© Springer-Verlag 2005

Authors and Affiliations

  • Ivan M. Gan
    • 1
  • Luana C. Ugahary
    • 1
  • Jaap T. van Dissel
    • 2
  • Eric Feron
    • 1
  • Ed Peperkamp
    • 1
  • Marc Veckeneer
    • 1
  • Paul G. H. Mulder
    • 3
  • Gert Jan Platenkamp
    • 4
  • Jan C. van Meurs
    • 1
  1. 1.The Rotterdam Eye HospitalRotterdamThe Netherlands
  2. 2.Department of Infectious DiseasesUniversity of Leiden Medical CenterLeidenThe Netherlands
  3. 3.Department of Epidemiology & BiostatisticsErasmus University Medical Center RotterdamRotterdamThe Netherlands
  4. 4.Regional Medical Microbiological LaboratoryRotterdamThe Netherlands

Personalised recommendations