Abstract
Purpose: To evaluate the risk factors for the development of concurrent or delayed-onset rhegmatogenous retinal detachment (RRD) related to endophthalmitis as well as the anatomic and visual outcomes with subsequent RRD repair. Methods: In this retrospective case study, a 2-tailed t test (continuous) and Fisher exact test were used to determine statistical significance of the observed findings. The relative risk (RR) and 95% CIs were calculated to assess statistical significance. Results: Of the 170 patients included, 22 were found to have a concurrent or subsequent RRD. Initial treatment with pars plana vitrectomy (PPV) (RR, 3.544; 95% CI, 1.650-7.614), aphakia (RR, 4.150; 95% CI, 1.434-12.011), endogenous endophthalmitis (RR, 2.684; 95% CI, 1.065-6.764), and posterior synechiae (RR, 3.026; 95% CI, 1.408-6.505) were statistically significant risk factors for RRD. Anatomically successful repair was achieved in 77.7% of patients. Conclusions: In addition to preexisting risk factors, the initial treatment of endophthalmitis may be a significant risk factor for RRD development, with a higher incidence of subsequent RRD in patients who have PPV as the initial treatment for endophthalmitis.
Keywords: endophthalmitis, retinal detachment, rhegmatogenous retinal detachment, vitreoretinal surgery, infection, microbiology
Introduction
Endophthalmitis is a severe, typically purulent intraocular inflammatory response to an endogenous or exogenous infectious source. The mainstays of treatment in cases of endophthalmitis are timely injection of intravitreal antibiotics and pars plana vitrectomy (PPV). Biopsy specimens can also be obtained via aqueous or vitreous tap. Gram stain, culture, and sensitivity testing are performed on biopsy specimens to ascertain the microbiologic etiology and to allow appropriate adjustments in therapy as indicated. Systemic antibiotics can also be beneficial, in particular in cases related to endogenous endophthalmitis or open-globe trauma. 1
A significant cause of morbidity in cases of endophthalmitis is the development of a rhegmatogenous retinal detachment (RRD), which can result from both the cause of and the treatment for the infection. The Endophthalmitis Vitrectomy Study (EVS) reported the incidence of RD to be 8.3% in the 420 patients enrolled in the study. An incidence of 2% to 37% has been reported in other studies, depending on the cause of endophthalmitis, and in general, the final visual outcomes in these patients are poor.2,3
At this time of this writing, there was limited information regarding the risk factors for the development of an RD in cases of endophthalmitis. The purpose of this study was to evaluate the potential risk factors, mechanisms, examination findings, and treatments related to the development of concurrent or delayed-onset RRD. In addition, the final visual outcomes and anatomic surgical success were analyzed and the risk factors for poor visual outcomes and anatomic failure examined.
Methods
A retrospective chart review was performed to identify patients who were treated for endophthalmitis at the University of Mississippi Medical Center (UMMC) from November 2006 to December 2020. Patients with endophthalmitis were reviewed for RRD development. The UMMC Institutional Review Board (IRB) approved this study under research protocol 2020-0348. Informed patient consent was deemed unnecessary by the IRB.
Inclusion criteria for patients in this study were those who were diagnosed clinically with endophthalmitis based on clinical suspicion and examination (surgical history and timing, hypopyon, vitritis, vision loss and/or pain, conjunctival injection, purulent discharge) rather than on microbiological confirmation. Eyes with endophthalmitis that were determined unsalvageable and thus enucleated or eviscerated as initial treatment were excluded.
The following descriptive variables were collected from the patient’s medical records: demographics (age, sex), lens status, presence of posterior synechia, previous vitrectomy, causes of endophthalmitis (acute postoperative status after cataract surgery, endogenous, postoperative corneal transplantation, postoperative glaucoma surgery, post-intravitreal injection (IVI), posttraumatic, or other miscellaneous causes), initial treatment (tap and IVI with vancomycin and ceftazidime for intravitreal antibiotics, fortified topical antibiotics, vitrectomy, PPV after intravitreal antibiotics), method of microbe collection before initial treatment, retinal attachment status at the last known follow-up, and the initial and last known best-corrected visual acuity (BCVA). The number of days from the endophthalmitis diagnosis to the last known follow-up was also evaluated. Culture-positive cases of endophthalmitis were categorized based on less virulent (coagulase-negative Staphylococcus and fungus) or more virulent (Staphylococcus aureus, Streptococcus species, Bacillus species, and gram-negative bacteria) microorganisms, with 7 patients excluded from this cohort because of no data.4,5
A 2-way frequency table was used to calculate the expected count for each variable. Specifically, a 2-tailed t test for continuous variables and a Fisher exact test for categorical variables were used to evaluate the statistical significance of differences between patients with RRD and those without RRD. These tests were specifically applied to compare the number of patients across the groups based on the examined risk factors. The relative risk (RR) and 95% CIs were calculated to assess the likelihood of developing an RRD in the presence of a specific risk factor. These risk factors included the endophthalmitis etiology, lens status, presence of posterior synechiae, type of initial treatment, attempted aqueous or vitreous tap, previous vitrectomy, initial and final BCVA, and microbiology findings. The threshold for statistical significance was set at P < .05 for all analyses. Data were analyzed using SPSS software (version 29, SPSS, Inc).
Results
Endophthalmitis As a Whole
The study included 170 cases of endophthalmitis. Table 1 shows the characteristics of the patients. At the time of diagnosis, 40 patients (23.5%) had a BCVA of light perception (LP) and 125 (73.5%) of hand motions (HM) or better; 5 patients did not have a recorded initial BCVA. In 51.4% of the cases, no growth was found on microbiologic testing. Of the cases with positive culture results, bacteria were the most common pathogen found followed by fungus (Table 1).
Table 1.
Characteristics of Patients With Endophthalmitis.
| Characteristic | Value |
|---|---|
| Age (y) | |
| Mean | 65.35 |
| Median | 68.0 |
| Sex, n (%) | |
| Male | 99 (58.2) |
| Female | 71 (41.8) |
| Race/ethnicity, n (%) | |
| White | 102 (60.0) |
| Black | 54 (31.8) |
| American Indian | 6 (3.5) |
| Hispanic | 1 (0.6) |
| Other/unknown | 7 (4.1) |
| Culture result, n (%) | |
| Bacterial | 68 (39.8) |
| Fungal | 7 (4.1) |
| No growth | 88 (51.4) |
| No culture data | 8 (4.7) |
| Initial treatment, n (%) | |
| Fortified topical antibiotics | 9 (5.3) |
| Tap and IVAB | 86 (50.6) |
| PPV after IVAB | 32 (18.8) |
| Vitrectomy only | 43 (25.3) |
Abbreviations: IVAB, intravitreal antibiotics; PPV, pars plana vitrectomy.
Acute postoperative endophthalmitis after cataract surgery was the most common mechanism (50 cases [29.4%]). Endophthalmitis occurred in 33 cases (19.4%) after IVI, 11 cases (6.5%) related to penetrating or perforating trauma, 15 cases (8.8%) after glaucoma surgery, 10 cases (5.9%) following corneal transplantation surgery; 13 (7.6%) were endogenous cases. At the time of diagnosis, 127 patients (74.7%) were pseudophakic, 39 (22.9%) were phakic, and 4 (2.4%) were aphakic.
The most common treatment was aqueous or vitreous tap with intravitreal antibiotics, typically vancomycin 1 mg/0.1 mL and ceftazidime 2.25 mg/0.1 mL (n = 86 [50.6%]). Vitreous tap was attempted in 35 patients who had a tap and inject procedure. Thirty-two patients (18.8%) had delayed PPV after failing to improve with tap and inject alone, and 43 patients (25.3%) had PPV with vitreous biopsy and subsequent intravitreal antibiotics as the initial treatment.
RRD After Endophthalmitis
Of the patients, 22 were found to have concurrent or subsequent RRD (Table 2). The mean number of days to development of an RRD was 109.3 ± 53.94 (standard error of mean). RD occurred within the first 6 weeks in 63.6% of cases (14/22). The overall incidence of RRD was 12.9% (22/170).
Table 2.
Characteristics of RRD After Endophthalmitis.
| Case | Mechanism of Endophthalmitis | Initial Endophthalmitis Treatment | Lens Status | Surgical Repair Attempted? | Anatomic Success After Surgery? | Initial BCVA | Final BCVA | Did Patient Develop PVR? | Days From Dx to LKFU |
|---|---|---|---|---|---|---|---|---|---|
| 1 | Postop CEIOL | Vitrectomy only | Phakic | Yes | No | HM | NLP | No | 1090 |
| 2 | Postop CEIOL | Vitrectomy only | Pseudophakic | No | NA | HM | LP | No | 74 |
| 3 | Post IVI | IVAB | Pseudophakic | Yes | Yes | HM | 20/40 | No | 73 |
| 4 | Postop CEIOL | Vitrectomy only | Pseudophakic | Yes | Yes | HM | CF | Yes | 189 |
| 5 | Endogenous | IVAB | Phakic | Yes | No | HM | HM | Yes | 62 |
| 6 | Post traumatic | IVAB | Phakic | No | NA | LP | NLP | No | 63 |
| 7 | Postop CEIOL | Vitrectomy only | Pseudophakic | Yes | Yes | LP | 20/400 | No | 432 |
| 8 | Endogenous | Vitrectomy only | Pseudophakic | Yes | Yes | 20/40 | 20/25 | No | 28 |
| 9 | Postop CEIOL | Vitrectomy only | Pseudophakic | Yes | Yes | CF | CF | No | 3370 |
| 10 | Postop CEIOL | Vitrectomy only | Pseudophakic | Yes | Yes | LP | HM | Yes | 299 |
| 11 | Postop CEIOL | Vitrectomy only | Pseudophakic | Yes | Yes | LP | 20/30 | No | 418 |
| 12 | Postop glaucoma sx | PPV after IVAB | Pseudophakic | Yes | Yes | LP | NLP | No | 2559 |
| 13 | Postop CEIOL | PPV after IVAB | Pseudophakic | Yes | Yes | HM | 20/200 | No | 385 |
| 14 | Other | Vitrectomy only | Pseudophakic | Yes | No | HM | Evisceration | Yes | 1365 |
| 15 | Post CEIOL | IVAB | Pseudophakic | Yes | Yes | HM | 20/200 | Yes | 302 |
| 16 | Post traumatic | IVAB | Aphakic | Yes | No | LP | NLP | Yes | 324 |
| 17 | Other | Vitrectomy | Aphakic | No | NA | LP | NLP | No | 1339 |
| 18 | Endogenous | Vitrectomy | Pseudophakic | Yes | Yes | HM | HM | Yes | 695 |
| 19 | Endogenous | IVAB | Pseudophakic | No | NA | NLP | NLP | No | 112 |
| 20 | Post CEIOL | Vitrectomy | Phakic | Yes | Yes | HM | 20/400 | Yes | 2088 |
| 21 | Postop glaucoma sx | PPV after IVAB | Pseudophakic | Yes | Yes | CF | 20/100 | No | 257 |
| 22 | Postop cornea transplant | IVAB | Pseudophakic | Yes | Yes | LP | LP | Yes | 126 |
Abbreviations: BCVA, best-corrected visual acuity; CEIOL, cataract extraction with intraocular lens implantation; CF, counting fingers; Dx, diagnosis; HM, hand motions; IVAB, intravitreal antibiotic; IVI, intravitreal injection; LKFU, last known follow-up; LP, light perception; NA, not applicable; NLP, no light perception; PPV, pars plana vitrectomy; PVR, proliferative vitreoretinopathy; sx, surgery.
Table 3 shows the potential risk factors for the development of an RD. The etiology resulting in the highest frequency of RRD was endogenous endophthalmitis (4/13 [30.8%]); this was statistically significant (RR, 2.684; 95% CI, 1.065-6.764). As for exogenous causes, the frequency of RRD was highest after cataract surgery (8/50 [16%]), which was not statistically significant (RR, 1.371; 95% CI, 0.614-3.063). Other etiologies for endophthalmitis were not found to be statistically significant. Patients who had a previous vitrectomy for any reason were not at a significantly increased risk for developing an RRD after endophthalmitis, regardless of the etiology (RR, 1.520; 95% CI, 0.504-4.581).
Table 3.
Relative Risk for Developing Rhegmatogenous Retinal Detachment in the Presence of Risk Factors.
| Potential Risk Factor | RR | 95% CI |
|---|---|---|
| Endophthalmitis etiology | ||
| Acute postop s/p cataract sx | 1.371 | 0.614-3.063 |
| Endogenous | 2.684 | 1.065-6.761 |
| Postop, s/p cornea transplant | 0.762 | 0.114-5.102 |
| Postop s/p glaucoma sx | 1.033 | 0.267-4.000 |
| Post IVI | 0.656 | 0.206-2.084 |
| Posttraumatic | 1.445 | 0.387-5.405 |
| Other | 0.347 | 0.085-1.420 |
| Lens status | ||
| Phakic | 0.746 | 0.268-2.076 |
| Pseudophakic | 0.903 | 0.377-2.160 |
| Aphakic | 4.150 | 1.434-12.011 |
| Posterior synechiae | 3.026 | 1.408-6.505 |
| Initial treatment | ||
| Fortified topical antibiotics | — | — |
| Tap and IVAB only | 0.456 | 0.196-1.061 |
| PPV after IVAB | 0.681 | 0.214-2.162 |
| PPV as initial treatment | 3.544 | 1.650-7.614 |
| Vitreous tap attempted vs aqueous tap | 0.857 | 0.310-2.371 |
| Vitreous tap attempted vs no tap | 0.857 | 0.310-2.371 |
| Patient had previous vitrectomy | 1.520 | 0.504-4.581 |
| Initial BCVA a | ||
| HM or better | 0.505 | 0.232-1.097 |
| LP or worse | 1.980 | 0.912-4.302 |
| Final BCVA | ||
| HM or better | 0.536 | 0.246-1.169 |
| LP or worse | 1.866 | 0.856-4.070 |
| Microbiology | ||
| Culture status | ||
| Bacteria | 1.397 | 0.643-3.034 |
| Fungal | — | — |
| No growth | 0.852 | 0.392-1.853 |
| Positive culture | ||
| More virulent | 2.889 | 0.830-10.055 |
| Less virulent | 0.346 | 0.099-1.205 |
Abbreviations: BCVA, best-corrected visual acuity; HM, hand motions; IVAB, intravitreal antibiotic; IVI, intravitreal injection; LP, light perception; PPV, pars plana vitrectomy; RR, relative risk; s/p, status post; sx, surgery.
Missing 4 eyes without initial BCVA.
More than half the patients in the RRD cohort had PPV as the initial treatment for endophthalmitis (n = 12 [54.5%]). Seven patients underwent tap and inject with intravitreal antibiotics alone, while 3 patients had PPV after failing to improve with intravitreal antibiotics. A vitreous tap was attempted in 4 patients in the RRD group and was not shown to increase the risk for RRD compared with an aqueous tap (RR, 0.857; 95% CI, 0.310-2.371) and with a vitreous tap (RR, 0.857; 95% CI, 0.310-2.371). The initial treatment was a statistically significant risk factor for the development of an RRD, with PPV showing a higher risk for RD than other treatments (RR, 6.544; 95% CI, 1.650-7.614).
Of the 22 patients who developed an RRD, 16 (72.7%) were pseudophakic at the time of diagnosis, 4 were phakic, and 2 were aphakic. Aphakia had a higher incidence of RRD and also was a statistically significant risk factor (RR, 4.150; 95% CI, 1.434-12.011). The presence of posterior synechiae was a statistically significant risk factor for RD development (RR, 3.026; 95% CI, 1.408-6.505), manifesting in 36% (8/22) in the RRD group.
There was no statistically significant difference in the rate of RRD between a BCVA of HM or better (RR, 0.505; 95% CI, 0.232-1.097) and a BCVA of LP or worse (RR, 1.980; 95% CI, 0.912-4.302). The same applied to the final BCVA comparing HM or better (RR, 0.536; 95% CI, 0.426-1.169) with LP or worse (RR, 1.866; 95% CI, 0.856-4.070). Further subdivisions of BCVA in relation to endophthalmitis etiology were not performed because the very small samples, making statistical analysis unreliable.
Of the 11 cases in the RRD cohort who had positive culture results, 8 (36.3%) were related to a more virulent organism (Table 4). Neither a positive culture result nor the presence of a more virulent organism was a statistically significant risk factor for RD (RR, 2.889; 95% CI, 0.830-10.06 and RR, 0.346; 95% CI, 0.099-1.205, respectively). Lack of culture growth was also not statistically significant (RR, 0.852; 95% CI, 0.392-1.853).
Table 4.
Potential Microbiology Associated With RD After Endophthalmitis.
| Number (%) | ||||
|---|---|---|---|---|
| Variable | RD Group (n = 22) |
No RD Group (n = 141 a ) |
RR | 95% CI |
| Culture status | ||||
| Bacteria | 11 (50) | 57 (40) | 1.397 | 0.643-3.034 |
| Fungal | 0 | 7 (5) | ||
| No growth | 11 (50) | 77 (55) | 0.852 | 0.392-1.853 |
| Positive culture | ||||
| More virulent | 8 (72) | 28 (44) | 2.889 | 0.830-10.055 |
| Less virulent | 3 (38) | 36 (56) | 0.346 | 0.099-1.205 |
Abbreviations: RD, retinal detachment; RR, relative risk.
Seven patients had no culture data and were not included.
The final mean logMAR BCVA in the RD group was 1.9 ± 1.0 (Snellen equivalent 20/1589). In the RD group 13.6% (n = 3) had a final BCVA of 20/40 or better, whereas 25.8% (n = 44) in the endophthalmitis group as a whole had a BCVA of 20/40 or better. In 18 of the 22 cases, surgical repair of the RD was attempted. Sixteen of these patients had PPV alone, while the other 2 had a combined PPV and scleral buckle procedure. Anatomically successful repair was achieved in 14 cases (77.7%), with an average final BCVA in the RD group with anatomically successful repair of 1.43 ± 0.94 logMAR (Snellen equivalent 20/538). Nine cases developed subsequent proliferative vitreoretinopathy (PVR); however, the development of PVR was not statistically significant for anatomic repair failure (RR, 3.000; 95% CI, 0.380-23.679). The final logMAR BCVA was 1.9 ± 1.0 (Snellen equivalent 20/1589) in the RD group and 1.37 ± 1.05 (Snellen equivalent 20/469) in the endophthalmitis group as a whole.
Conclusions
Infectious endophthalmitis, although rare, can lead to poor visual outcomes despite timely and proper management. 6 The development of RRD after endophthalmitis further increases the risk for a poor visual outcome. This phenomenon has been reported since the EVS and continues to be true even with surgical advances in minimally invasive vitrectomy. 7
The incidence of RRD in our study after endophthalmitis was 12.9%. This is consistent with the ranges of RRD after endophthalmitis reported in the literature (up to 25%).4,8,9 The statistically significant risk factors for RRD in our study were aphakia, endogenous etiology, posterior synechia, and initial treatment, with PPV showing the highest incidence of RRD postoperatively. Compared with other studies that analyzed the risk factors for RRD after treatment, this both confirms some previous findings and creates some interesting dichotomies in findings.10,11
Aphakia has been associated with RDs in various studies, especially in cases of endophthalmitis related to a ruptured or absent posterior capsule, leading to vitreous loss and traction.12,13 Despite the low number of aphakic patients, this was found to be a statistically significant factor and is in line with reports in the previous literature regarding this risk factor. 14 Thus, any event associated with RRD, such as endophthalmitis, may simply increase the already heightened risks for RRD in aphakic eyes.
Performing a vitreous or aqueous tap was not a statistically significant risk factor for detachment, nor was a vitreous tap vs no vitreous tap. The incidence of RRD was roughly the same between the 2 groups (18% vs 21%). Compared with an aqueous humor tap, aspirating a vitreous sample may theoretically cause undue vitreous traction and thus lead to a retinal break; however, our results did not reflect this theory. One potential explanation is that there may be preexisting or newly formed liquefied vitreous pockets that make aspiration safer. Another explanation is that the method and aggressiveness of vitreous tapping was not standardized. Hence, aggressive vitreous sampling could theoretically increase the risk for RRD.
Posterior synechia was a statistically significant risk factor for RRD as well. Interestingly, microorganism virulence was not a statistically significant risk factor for RRD. These findings are in line with those of Wang et al. 14 They also reported that the presence of a hypopyon was not a risk factor. One explanation for this seemingly dichotomous finding is that chronic inflammation, rather than the initial severity from the offending organism, may play a more significant role in RRD development. 15 Chronic inflammation may also play a role in endogenous endophthalmitis being a statistically significant factor associated with RRD. This subset of eyes is often diagnosed well after endophthalmitis begins, leading to prolonged infectious inflammation. In addition, with endogenous transmission of offending organisms through the choroid disrupting the structural retina–retinal pigment epithelium interface and causing overlying focal retinitis, the chronic inflammation may increase the risk for necrosis at these infiltrative areas and thus result in RRD. 11
The initial treatment choice was a statistically significant risk factor for RRD, with PPV having the highest risk. In addition, PPV after tap and inject had a much lower incidence than initial PPV. Interestingly, a presenting vision of HM or better vs LP or worse was not a statistically significant factor. Two thirds of patients who developed RRD after vitrectomy as an initial treatment had acute postoperative endophthalmitis; however, some of these eyes had HM, not LP, acuity at presentation. These findings may suggest that the initial surgery itself, not the virulence and aggressiveness of the microorganisms or the timeline in which endophthalmitis was diagnosed, may be related to RRD. Otherwise, both presenting vision and PPV as an initial treatment may have been statistically significant risks given that patients presenting with LP acuity received vitrectomy first, per the EVS guidelines.
These cases can be particularly challenging for the retina surgeon because visibility can be limited by corneal opacification, anterior chamber inflammation, or vitreous haze. Thorough vitrectomy can be difficult because of suboptimal visualization conditions. In addition, toxin-induced retinal necrosis can make the retinal tissue susceptible to breaks during vitrectomy. 16 Even in the absence of an iatrogenic break, chronic inflammation or residual hyaloid traction from incomplete vitrectomy can lead to an RRD in the postoperative period.
The finding of initial PPV being statistically significant for RRD is salient with respect to evolving trends in treating endophthalmitis. Recently, more literature has suggested that PPV should be performed sooner, if not as the initial treatment choice.17–19 Early vitrectomy may allow debulking of the infectious load and removing potentially vision-threatening endotoxin byproducts. The EVS guidelines came about during the era of 20-gauge vitrectomy; however, with advances in minimally invasive small-gauge PPV and improved viewing systems, initial PPV may now be less risky. In our study, initial PPV was performed with 23-gauge or 25-gauge instrumentation. With RRD being a prognostic indicator of a poor outcome in the long term, some precaution must to be taken before considering initial PPV on a regular basis. Other studies have not shown any superiority in outcomes between tap and inject vs initial PPV. Hence, any benefits that may come with initial PPV must be weighed against the potentially increased risks for RRD postoperatively.
Like other studies that examined this issue, patients who developed RRD after endophthalmitis had worse visual outcomes in our study. 20 PVR was not statistically significant for unsuccessful anatomic repair. This runs contrary to findings in other studies of the specific subject of outcomes after RRD repair, especially given that PVR itself is a bad prognostic indicator for RRDs alone. This difference is likely the result of the sample size and design of the study, which was not built to study this specific issue. These results and this explanation are included for the sake of academic transparency.
This study has limitations. It was a retrospective cohort study; thus, the clinical decisions and management may have varied per patient based on the attending’s judgment. Given the surgical nature of management for several patients, other nonquantifiable factors, such as surgeon factor or other outside issues, might also influence outcomes. The rarity of endophthalmitis also means sample sizes for all studied groups might not be uniform or might be small, which affects statistical analysis. Last, as stated previously, our study was not powered properly or designed to study risk factors and outcomes of patients who developed RRD after endophthalmitis. Thus, findings specific to that cohort of patients will not be as strong as those in other studies designed to analyze that specific issue.
In summary, endogenous endophthalmitis, aphakia, initial PPV as a treatment, and posterior synechia were associated with a higher rate of RRD in patients with endophthalmitis. Although anatomic success was possible in a significant majority of patients, the final visual outcomes were worse than those without RRD. With these findings, clinicians should be aware of these associations during clinical decision-making and more attention is warranted in exploring the link between the initial treatment and RRD in endophthalmitis.
Footnotes
Ethical Approval: This report was conducted in accordance with the Declaration of Helsinki. The collection and evaluation of all protected patient health information were performed in a US Health Insurance Portability and Accountability Act–compliant manner.
Statement of Informed Consent: This study was approved by the University of Mississippi Medical Center Institutional Review Board. Informed patient consent was not deemed necessary.
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding: The author(s) received no financial support for the research, authorship, and/or publication of this article.
ORCID iDs: Cary R. Baxter 
https://orcid.org/0009-0008-2705-1242
Albert L. Lin
https://orcid.org/0000-0001-6883-6937
References
- 1. Grzybowski A, Turczynowska M, Schwartz SG, Relhan N, Flynn HW., Jr. The role of systemic antimicrobials in the treatment of endophthalmitis: a review and an international perspective. Ophthalmol Ther. 2020;9(3):485-498. doi: 10.1007/s40123-020-00270-w [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2. Doft BM, Kelsey SF, Wisniewski SR. Retinal detachment in the endophthalmitis vitrectomy study. Arch Ophthalmol. 2000; 118(12):1661-1665. doi: 10.1001/archopht.118.12.1661 [DOI] [PubMed] [Google Scholar]
- 3. Lazzarini TA, Al-Khersan H, Echegaray JJ, Flynn HW., Jr. Management of concurrent and delayed-onset retinal detachment associated with endophthalmitis. Retin Physician. 2020;17:14-18. [Google Scholar]
- 4. Chiquet C, Aptel F, Combey-de Lambert A, et al. Occurrence and risk factors for retinal detachment after pars plana vitrectomy in acute postcataract bacterial endophthalmitis. Br J Ophthalmol. 2016;100(10):1388-1392. doi: 10.1136/bjophthalmol-2015-307359 [DOI] [PubMed] [Google Scholar]
- 5. Cornut PL, Thuret G, Creuzot-Garcher C, et al. Relationship between baseline clinical data and microbiologic spectrum in 100 patients with acute postcataract endophthalmitis. Retina. 2012; 32(3):549-557. doi: 10.1097/IAE.0b013e3182205996 [DOI] [PubMed] [Google Scholar]
- 6. Dong LK, Shields RA, Subramanian S, et al. Features and outcomes of eyes that underwent surgical repair of rhegmatogenous retinal detachments after being treated for acute endophthalmitis. Retina. 2021;41(8):1612-1617. doi: 10.1097/iae.0000000000003091 [DOI] [PubMed] [Google Scholar]
- 7. Soliman MK, Gini G, Kuhn F, et al. Visual outcome of early vitrectomy and intravitreal antibiotics in acute postsurgical and postintravitreal injection endophthalmitis: European Vitreo-Retinal Society Endophthalmitis Study report two. Retina. 2021; 41(2):423-430. doi: 10.1097/iae.0000000000002856 [DOI] [PubMed] [Google Scholar]
- 8. Feng HL, Robbins CB, Fekrat S. A nine-year analysis of practice patterns, microbiologic yield, and clinical outcomes in cases of presumed infectious endophthalmitis. Ophthalmol Retina. 2020; 4(6):555-559. doi: 10.1016/j.oret.2020.03.009 [DOI] [PubMed] [Google Scholar]
- 9. Lee CS, Khan M, Patrie J, Bajwa A, Shildkrot YE. Pars plana vitrectomy for endophthalmitis: microbiologic spectrum and clinical outcomes. Ocul Immunol Inflamm. 2021;29(5):871-876. doi: 10.1080/09273948.2019.1698750 [DOI] [PubMed] [Google Scholar]
- 10. Nanji K, Kherani I, Khan M, et al. A multicenter review evaluating the risk of rhegmatogenous retinal detachment post endophthalmitis: a Canadian retina research network investigation. Retina. 2022; 42(8):1503-1511. doi: 10.1097/iae.0000000000003486 [DOI] [PubMed] [Google Scholar]
- 11. Lee T, Robbins CB, Feng HL, Fekrat S. Rhegmatogenous retinal detachment in the setting of endophthalmitis: risk factors and visual outcomes. Ophthalmic Surg Lasers Imaging Retina. 2021; 52(5):257-262. doi: 10.3928/23258160-20210429-03 [DOI] [PubMed] [Google Scholar]
- 12. Petousis V, Sallam AA, Haynes RJ, et al. Risk factors for retinal detachment following cataract surgery: the impact of posterior capsular rupture. Br J Ophthalmol. 2016;100(11):1461-1465. doi: 10.1136/bjophthalmol-2015-307729 [DOI] [PubMed] [Google Scholar]
- 13. Bhardwaj G, Walker RJE, Ezra E, Mirza Z, Muqit MMK. A 21-year study of vitreoretinal surgery for aphakic retinal detachment: long-term surgical outcomes and complications. Ophthalmol Retina. 2019;3(9):784-790. doi: 10.1016/j.oret.2019.04.002 [DOI] [PubMed] [Google Scholar]
- 14. Wang T, Moinuddin O, Abuzaitoun R, et al. Retinal detachment after endophthalmitis: risk factors and outcomes. Clin Ophthalmol. 2021;15:1529-1537. doi: 10.2147/opth.S302757 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15. Kunavisarut P, Srisomboon T, Patikulsila D, et al. Risk factors for development of rhegmatogenous retinal detachment in patients with uveitis. Ocul Immunol Inflamm. 2019;27(4):681-685. doi: 10.1080/09273948.2018.1424343 [DOI] [PubMed] [Google Scholar]
- 16. Dave VP, Pathengay A, Relhan N, et al. Endophthalmitis and concurrent or delayed-onset rhegmatogenous retinal detachment managed with pars plana vitrectomy, intravitreal antibiotics, and silicone oil. Ophthalmic Surg Lasers Imaging Retina. 2017; 48(7):546-551. doi: 10.3928/23258160-20170630-05 [DOI] [PubMed] [Google Scholar]
- 17. Mautone L, Skevas C, Spitzer MS. Therapie der postoperativen Endophthalmitis: Operieren oder nur injizieren? [Treatment of postoperative endophthalmitis: Operate or only inject?]. Ophthalmologe. 2021;118(3):219-229. doi: 10.1007/s00347-021-01320-9. German. [DOI] [PubMed] [Google Scholar]
- 18. Dib B, Morris RE, Oltmanns MH, Sapp MR, Glover JP, Kuhn F. Complete and early vitrectomy for endophthalmitis after cataract surgery: an alternative treatment paradigm. Clin Ophthalmol. 2020;14:1945-1954. doi: 10.2147/opth.S253228 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19. Grzybowski A, Turczynowska M, Kuhn F. The treatment of postoperative endophthalmitis: should we still follow the endophthalmitis vitrectomy study more than two decades after its publication? Acta Ophthalmol. 2018;96(5):e651-e654. doi: 10.1111/aos.13623 [DOI] [PubMed] [Google Scholar]
- 20. Zheng Y, Casagrande M, Dimopoulos S, Bartz-Schmidt KU, Spitzer MS, Skevas C. Analysis of retinal detachment resulted from post-operative endophthalmitis treated with 23G pars Plana Vitrectomy. BMC Ophthalmol. 2021;21(1):414. doi: 10.1186/s12886-021-02175-z [DOI] [PMC free article] [PubMed] [Google Scholar]