1 of 1

Long Term Visual Outcomes of Pars Plana Vitrectomy for Infectious Endophthalmitis

​

Revati Rashingkar, BS1; Viren K Govindaraju, MD2; Chi Phan, BS1, Adam J. Weiner, MD1,2, Sandeep Randhawa, MD2,

Lisa J. Faia, MD2, George A. Williams, MD2,3, Tarek S. Hassan, MD, FASRS1,2

1Oakland University William Beaumont School of Medicine, Rochester Hills, MI

2Associated Retinal Consultants, Royal Oak, MI

3Beaumont Eye Institute, Corewell Health, Royal Oak, MI

Introduction

  • Infectious endophthalmitis is a severe intraocular infection that can be a vision-threatening emergency, and rapid intervention is critical for favorable visual outcomes1.
  • Source control with either pars plana vitrectomy (PPV) and/or intravitreal antibiotic injection represents the current mainstay of endophthalmitis treatment1,2,3.
  • It is commonly thought that if patients present with NLP vision, there is no longer appropriate vascular perfusion or there is optic nerve atrophy resulting in permanent vision loss5,6.
  • There are currently no established and published guidelines specifically for treating eyes with NLP vision and endophthalmitis. Most studies have been performed in eyes with Light Perception (LP) or better vision, including the well-known Endophthalmitis Vitrectomy Study (EVS), which specifically excluded NLP eyes4.

​

Aims and Objectives

  • This study aims to assess whether performing standard interventions (intravitreal antibiotics and PPV) in NLP eyes should be advocated, which may challenge paradigms in current treatment algorithms.
  • Investigate visual acuity improvement after surgical intervention in NLP eyes and determine whether the magnitude of improvement is comparable to that in LPB eyes.

​

Methods

A retrospective study of eyes with endophthalmitis seen at Associated Retinal Consultants between January 2015 and January 2024.

​

Inclusion criteria: diagnosis of endophthalmitis, follow-up of at least 6 months, receiving intervention which included PPV, and visual acuity (VA) data

​

Exclusion criteria: cases with missing data, cases with follow up records of less than 6 months.

​

​

​

​

​

​

​

​

​

Continuous variables were analyzed using T-tests and Mann-Whitney U-Tests when appropriate.

Categorical variables were analyzed using Fisher’s Exact Tests. All tests used p value of 0.05.

​

​

Primary Outcome

​

12-month VA compared to presenting VA.

Secondary Outcomes

​

VA at 1, 3, and 6 months, and the proportion of VA change in each group

Results

  • 124 patients were included in the study. The eyes were stratified by having LP or better (LPB) vision or NLP vision prior to intervention. There was no significant difference between the two groups in terms of etiology (Table 1).

​

  • 107 LPB eyes received an anterior chamber paracentesis and injection of intravitreal vancomycin and ceftazidime prior to PPV intervention and 4 eyes received PPV combined with intravitreal antibiotics.

​

  • In the NLP group, all 13 eyes received intravitreal antibiotics and PPV concurrently (Table 1).

​

  • Among the 124 total eyes with endophthalmitis, there was no significant difference in organism types between the two groups (Table 2).

​

​

​

​

​

Table 1: Demographics

Table 2: Causative Organisms

Table 3: Visual Acuity

Visual Acuity Prior to Endophthalmitis

p=0.6

  • The NLP group had a mean logMAR VA of 0.4 (20/50) (+/- 0.4)
  • The LPB group had a mean logMAR of 0.5 (20/60) (+/- 0.48)

​

After Diagnosis Confirmed and Prior to Intervention

p<0.0001

  • The NLP group had a worse mean logMAR VA of 4.0
  • The LPB group had a mean logMAR VA of 2.0

​

12 months Post-Intervention

p=0.04

  • The NLP group still had a significantly worse VA with mean logMAR VA of 1.6
  • The LPB group had a mean logMAR VA of 1.0

​

Improvement Between Presentation and 12 Month Visit

​

  • The NLP group had a mean improvement in logMAR VA of 2.1 (p=0.008)
  • The LPB group had a mean improvement of logMAR 1.0 (p<0.001)

​

​

There were no significant differences in postoperative sequelae (p=0.77).

​

No eyes in either group progressed to phthisis.

​

Limitations

  • Being a single center study and the relatively small sample size of NLP cases limits generalizability.

​

  • As a retrospective study, it is limited by the possibility of missing data and non-standardized reporting.

​

  • Further studies with larger sample sizes are needed to better define indications for surgical intervention in NLP eyes and to provide stronger evidence supporting the treatment approach of offering surgery in this setting.

Conclusions

  • Patients presenting with infectious endophthalmitis and NLP vision can achieve significant improvement in VA following intravitreal antibiotics and PPV.

​

  • No patients progressed to requiring an enucleation or evisceration at 12 months after intervention.

​

  • These findings support an interventional approach in NLP endophthalmitis and underscore the need for evidence-based guidelines for this underrepresented patient population.

​

References

1. Ramachandran A, Das T, Pathengay A, Pappuru RR, Dave VP. Surgical approach to endophthalmitis: an overview. Eye (Lond). Sep 2024;38(13):2516–2521. doi:10.1038/s41433-024-03089-y

2. Lee JJ, Jo YJ, Lee JS. Clinical characteristics and risk factors for visual prognosis according to the types of infectious endophthalmitis. PLoS One. 2022;17(12):e0278625. doi:10.1371/journal.pone.0278625

3. Kritfuangfoo T, Chokchaitanasin R, Sujirakul T, Thongborisuth T, Tipsuriyaporn B. Two-Decade Retrospective Analysis of Postoperative Endophthalmitis at a Tertiary Care Academic Center: Microbial Spectrum, Treatment Modalities, and Visual Outcomes. Clin Ophthalmol. 2025;19:2729–2742. doi:10.2147/opth.S535966

4. VanderBeek BL, Chen Y, Tomaiuolo M, et al. Endophthalmitis Rates and Types of Treatments After Intraocular Procedures. JAMA Ophthalmology. 2024;142(9):827–834. doi:10.1001/jamaophthalmol.2024.2749

5. Garg A, Margolin E, Micieli JA. No Light Perception Vision in Neuro-Ophthalmology Practice. J Neuroophthalmol. Mar 1 2022;42(1):e225–e229. doi:10.1097/wno.0000000000001340

6. Lu X, Chen W, Xia H, et al. Atrophy of retinal inner layers is associated with poor vision after endophthalmitis: a spectral domain optical coherence tomography study. Eye (Lond). Oct 2017;31(10):1488–1495. doi:10.1038/eye.2017.100

​

Acknowledgements

I would like to thank all of the ARC faculty who supported this project and provided guidance throughout. I would also like to thank OUWB Biostatistician Jacob Keeley.