Serum IL-34 and IL-17A as novel biomarkers for diabetic retinopathy in type 2 diabetes: A cross-sectional study.

Ginkgo biloba extract (EGb), a complementary and alternative medicinal option, has gained extensive application in addressing conditions like cerebrovascular and peripheral vascular disorders. We aim to assess the neuroprotective efficacy of EGb for retinal disorders and to clarify its potential mechanisms of action through a systematic review. We searched original literature about laboratory experiments from four databases which were released until April 2024. The methodological quality of included in vivo studies was assessed using the SYRCLE risk of bias tool. The results showed that out of the 398 studies initially collected, 26 articles met the requirements for full-text review. 20 of them presented in vivo data, 2 detailed both in vitro and in vivo evidence, and 4 were in vitro experiments. Results demonstrated the protective effects of EGb against several retinal disorders, including retinal ganglion cell injury, retinal degeneration, ischemia, vitreo- or pre-retinal proliferative disorder, uveitis, and diabetic retinopathy. Based on SYRCLE's evaluation of bias risk, the in vivo studies' quality scores varied from 4 to 7 points. The data indicated that EGb preserved visual function by maintaining retinal morphology and structure in preclinical models. Its action mechanism may be associated with suppressing apoptosis, attenuating oxidative stress, reducing inflammation, inhibiting angiogenesis, and suppressing proteolysis. These preclinical findings suggest that EGb may be a promising neuroprotective agent for retinal disorders. However, the methodological limitations of the included studies necessitate cautious interpretation; to demonstrate the effectiveness and safety of EGb, more extensive clinical randomized controlled trials are required.

The association between diabetes mellitus and the risk of anal cancer: Results from a large nationwide cohort study in Taiwan.

Proliferative diabetic retinopathy (PDR) is one of the leading causes of blindness in working-age adults. We have previously shown that the risk of PDR is significantly elevated in individuals with intrauterine exposure to famine. However, the genetic mechanisms mediating this association remain unknown. The aim of the current study was to investigate the molecular underpinnings of famine-related PDR by performing genome-wide association (GWAS) and interaction studies (GWIS). We analysed n = 2925 patients with type 2 diabetes from the DOLCE cohort of Northern Ukraine, of whom n = 1364 were born during historical famine periods (1929-1949, including the Holodomor and World War II). PDR cases were defined as individuals with either diagnosed proliferative retinopathy, laser-treated diabetic retinopathy (DR) or blindness in either eye. GWAS and GWIS were performed using linear mixed model (LMM) adjusted for established risk factors and genetic relationship matrix. GWAS identified rs3795299 in IL22RA1 as the top signal (pLMM = 1.05 × 10-6), which was also the strongest gene in the gene-based analysis (p = 3.19 × 10-5), with suggestive enrichment of response to ketones (GO) and base excision repair (KEGG) pathways. In the GWIS, the strongest signal was rs1506783 in PAPPA2 (pLMM = 1.29 × 10-7). A second biologically credible candidate was rs2230805 in ABCA1 (pLMM = 4.44 × 10-6), reaching borderline genome-wide significance in gene-based analysis (p = 4.31 × 10-6). Interaction analyses showed suggestive enrichment for nucleosomal DNA binding (GO), tryptophan metabolism and glycerolipid metabolism (KEGG) pathways. Furthermore, at nominal significance, we validated variants in previously reported diabetic retinopathy-associated genes, including TCF7L2, SLC2A1, SLC2A11 and VDR in the GWAS, as well as 13 variants in genes including VEGF, VEGFR1, ANGPT1, PLXDC2, SELP and PON2 in the GWIS. Our findings suggest that famine-related PDR susceptibility involves distinct developmental programming mechanisms, including altered insulin-growth signalling (PAPPA2) and lipid metabolism (ABCA1), whereas immune-related pathways (IL22RA1) may contribute to the conventional glycaemia-driven route to PDR through VEGF-mediated angiogenesis. These genes represent potential therapeutic targets and emphasize the importance of the perinatal environment in lifelong vascular health and disease.

This study aims to assess the concentrations of angiopoietin-1, angiopoietin-2, and Tie-2, which are implicated in the pathophysiology of proliferative diabetic retinopathy (PDR), in the vitreous fluid and to evaluate the expression profiles of microRNA (miR)-145-5p, miR-542-3p, miR-5195-3p, miR-126-3p, miR-211-3p, and miR-204-5p. The study included 25 patients with PDR and 25 non-diabetic individuals as controls. Vitreous angiopoietin-1, angiopoietin-2, and Tie-2 levels were measured using enzyme-linked immunosorbent assay (ELISA). miRNA expression levels were evaluated using real-time polymerase chain reaction. Vitreous angiopoietin-1 and 2 levels were significantly lower in the PDR group when compared to controls (p<0.05). The PDR group also had a lower angiopoietin-1/angiopoietin-2 ratio and higher Tie-2 levels, but these differences did not reach statistical significance (p>0.05). Significantly higher levels of miR-126-3p and miR-204-5p were detected in the PDR group (p<0.05), whereas miR-211-3p, miR-5195-3p, miR-542-3p, and miR-145-5p did not show statistically significant differences (p>0.05). Our data demonstrate that increased miR-204-5p and miR-126-3p expression may be associated with angiogenesis-related alterations in PDR. These findings provide insight into PDR-related angiogenesis and suggest that these microRNAs may represent potential biomarkers of disease-related vascular alterations.

Outpatient ophthalmology services face rising demand, and delays in follow-up can result in avoidable sight loss. The problem is often framed as a lack of capacity, but a more accurate description is demand-capacity mismatch. We aimed to develop a generalisable framework to quantify how follow-up interval choice influences service demand, using diabetic retinopathy as a case study. We framed follow-up scheduling as a three-way trade-off between appointment demand, disease detection delay, and risk of progression. The risk of proliferative diabetic retinopathy (PDR) progression at 12 months was taken from Early Treatment Diabetic Retinopathy Study (ETDRS) data. In the absence of intermediate time-point data, we interpolated cumulative PDR risk for months 1-12 using a constant hazard model, which assumes a uniform instantaneous risk of progression across time and provides a transparent way to estimate risk at shorter intervals. We applied this framework to the diabetic retinopathy service of a UK district general hospital and considered other service examples. Small changes in follow-up interval produced large shifts in demand. In our service, increasing the mean interval from 4.7 to 6.2 months reduced demand by an estimated 700 appointments per 1000 patients annually. Illustrative examples in injection and glaucoma services showed similar effects. Follow-up interval is a powerful, modifiable determinant of outpatient demand. Quantifying trade-offs between demand, detection delay, and disease progression risk supports decision-making regarding follow-up interval selection. This framework complements, rather than replaces, capacity expansion and is generalisable across other chronic disease services.

Pathological tissue remodeling, including choroidal neovascularization (CNV) and fibrosis, is central to the development of retinal diseases such as age-related macular degeneration; however, its molecular mechanisms remain incompletely defined. We examined the intraocular expression patterns of Meflin encoded by the ISLR gene, a mesenchymal stromal cell marker with anti-fibrotic properties, in human surgical specimens and mouse retinal tissues, and evaluated its involvement in a retinal injury mouse model. Meflin expression was detected in fibroblast-like stromal cells within human proliferative vitreoretinopathy / proliferative diabetic retinopathy fibrovascular membranes and the ciliary body, lens epithelium, retinal pigment epithelium, optic nerve meningeal cells, and choroidal perivascular fibroblast-like cells of the mouse eye. Lineage-tracing using a Meflin reporter mouse line revealed accumulation of Meflin-lineage cells within laser-induced CNV lesions. Notably, adeno-associated virus-mediated Meflin overexpression increased CNV volume in the acute phase of retinal injury but suppressed subretinal fibrosis after the CNV growth phase. These findings suggest that Meflin overexpression modulates angiogenic and fibrotic remodeling after retinal injury in a phase-dependent manner, consistent with reported roles in cancer and fibrotic diseases. These results improve understanding of retinal disease pathology and identify Meflin as a potential target for future therapeutic studies on proliferative and fibrotic retinal diseases.

To explore the relationship between the number of retinal lesions observed on ultra-wide field fundus photography (UWF-FP) and the non-perfusion index (NPI) quantified on UWF fluorescein angiography (UWF-FA) in patients with proliferative diabetic retinopathy (PDR). The NPI was measured on early-phase UWF-FA in 100 eyes from 100 patients with PDR, including both type 1 and type 2 diabetes. Eyes were divided into quartiles based on NPI values. Red dots (microaneurysms, retinal hemorrhages, and intraretinal microvascular abnormalities), venous beading, cotton-wool spots (CWS), and white thread-like arterioles were counted on UWF-FP. Mean NPI values for quartiles 1 to 4 were 2.8 ± 1.1%, 8.5 ± 2.1%, 17.3 ± 4.5%, and 42.8 ± 12.2%, respectively. Eyes in the highest NPI quartile (quartile 4) had significantly fewer retinal lesions on UWF-FP (76.1 ± 50.6) and fewer CWS compared with those in the lowest NPI quartile, which showed 253.0 ± 165.5 retinal lesions (p < 0.001). White thread-like arterioles were present in 80.0% of eyes in quartile 4 and none in quartile 1. Patients in quartile 4 were predominantly type 2 diabetics with higher rates of end-stage kidney disease, peripheral neuropathy, lower-extremity peripheral artery disease, and prior limb amputation or revascularization. In patients with poorly controlled, long-standing type 2 diabetes complicated by renal failure, peripheral neuropathy, peripheral arterial disease, or lower-limb amputation, fluorescein angiography should be considered when highly ischemic PDR is suspected, particularly in the presence of white thread-like arterioles, as it may otherwise appear milder on color fundus photography.

Retinal neovascularization (RNV) is a potential vision-threatening process characterized by the abnormal growth of retinal vessels. Despite its clinical prevalence, the precise mechanisms governing RNV initiation and progression remain incompletely defined. Here we identify myeloid-derived growth factor (MYDGF) as a critical regulator of retinal vascular dynamics. Using single-cell RNA sequencing and human patient validation, we show that MYDGF is upregulated in retinal endothelial cells during proliferative diabetic retinopathy and mouse models of pathological neovascularization. In vitro, MYDGF promotes retinal endothelial cell proliferation, migration, and sprouting. In vivo, endothelial-specific MYDGF depletion inhibits both normal vascular development and pathological neovascularization in neonatal mice, and disrupting adult vascular homeostasis in male mice. MYDGF drives angiogenesis by activating the Akt-mTOR cascade through the Gαi1/3-Gab1 signaling complex; genetic depleting or mutation these components suppress MYDGF-induced Akt-mTOR activation and angiogenic responses. Together, MYDGF promotes retinal angiogenesis and maintains vascular homeostasis via the Gαi1/3-Gab1-Akt-mTOR signaling axis.

To synthesise the bidirectional evidence on the association between obstructive sleep apnea (OSA) and diabetic retinopathy (DR), separating studies framing OSA as the exposure for DR (Arm 1) from studies framing DR as the exposure for OSA (Arm 2). Systematic review and meta-analysis of observational studies (PROSPERO: CRD420251135427). PubMed, Scopus, Web of Science, and Google Scholar were searched from inception to July 13, 2025, supplemented by structured re-screening of five prior systematic reviews and forward and backward citation searching. Eligible studies enrolled individuals with diabetes mellitus and reported binary OSA (or, in Arm 2, DR) exposure with an unexposed comparator. Within each arm, exposure and outcome were severity-stratified; unadjusted and adjusted odds ratios (ORs) were pooled separately using random-effects models in MetaAnalysisOnline.com. Risk of bias was assessed with the Newcastle-Ottawa Scale (NOS) and certainty of evidence with the Grading of Recommendations Assessment, Development, and Evaluation (GRADE) framework. Twenty-six studies met eligibility criteria. In Arm 1 (OSA → DR), the unadjusted pooled OR for any DR with OSA classified as a binary exposure was 2.41 (95% confidence interval [CI] 1.53-3.80; 10 studies; very low certainty), but attenuated to 1.03 (95% CI 0.64-1.66; 2 studies; low certainty) in confounder-adjusted analyses. OSA-severity-stratified Arm 1 estimates for mild, moderate, and severe OSA were directionally consistent but not statistically significant in either unadjusted or adjusted pooling. DR-stage-specific Arm 1 analyses (background DR, mild/moderate/severe non-proliferative DR [NPDR], proliferative DR [PDR]) were directionally aligned with the overall Arm 1 finding but imprecise. In Arm 2 (DR → OSA), pooled estimates showed inverse or null associations across DR severity strata; severe OSA as an outcome of DR yielded a pooled OR of 1.24 (95% CI 0.92-1.68; 4 studies; low certainty). The OSA-DR association is sensitive to the direction of the framed exposure-outcome relationship, to confounder adjustment, and to severity classification. Substantial attenuation of the unadjusted Arm 1 signal after adjustment, together with the inverse or null Arm 2 pattern, indicates that previously reported pooled estimates were influenced by confounding and by directional misclassification. Polysomnography-confirmed prospective cohorts with uniform DR grading and individual-participant-level adjustment for body mass index, diabetes duration, and metabolic comorbidities are needed to clarify the direction and magnitude of the relationship.

Predicting the progression of proliferative diabetic retinopathy: Pathophysiology, imaging phenotypes, and determinants of disease persistence despite therapy.

Suprachoroidal injection enables targeted posterior segment drug delivery with reduced anterior segment exposure. Off-label use of the dexamethasone implant (Ozurdex) via the suprachoroidal route may benefit patients with macular edema who have contraindications to intravitreal therapy. This article aims to introduce a two-step surgical technique for suprachoroidal Ozurdex injection and reports its feasibility and short-term outcomes in an illustrative case. A custom‑made injector assembled from a 26‑gauge hypodermic needle and an intravenous catheter. The technique involves: (1) injection of 0.3mL sodium hyaluronate viscoelastic into the suprachoroidal space at 4.0mm posterior to the limbus; (2) injection of the dexamethasone implant through the same site. The technique was performed in a 37‑year‑old pregnant woman at 24 weeks of gestation with vitreous hemorrhage and macular edema secondary to proliferative diabetic retinopathy in the left eye immediately following pars plana vitrectomy. Main outcomes included feasibility, safety, best‑corrected visual acuity (BCVA), intraocular pressure (IOP) and imaging findings on ultrasound biomicroscopy (UBM) and optical coherence tomography (OCT). The procedure was completed without intraoperative complications. UBM at postoperative week 1 confirmed Ozurdex within an anechoic suprachoroidal space, which resolved by week 12. BCVA improved from decimal 0.2 (Snellen 4/20) at baseline to 0.4 (20/50) at 12 weeks. IOP remained stable (21-23 mmHg), with no cataract progression or significant IOP elevation. OCT showed resolution of intraretinal cysts. This two-step technique for suprachoroidal Ozurdex delivery using a custom‑made injector with viscoelastic pre‑injection is feasible and showed no major safety concerns in this case. It may provide an alternative delivery method for selected patients with non‑infectious macular edema who have relative contraindications to intravitreal anti‑vascular endothelial growth factor (VEGF) or corticosteroid injections. Further studies are needed to establish safety and efficacy.

This study investigates the relationship between retinal changes in proliferative diabetic retinopathy (PDR) patients and the severity of internal carotid artery (ICA) stenosis, using optical coherence tomography angiography (OCTA). The goal is to explore potential correlations between ICA stenosis and specific retinal parameters, aiding in risk factor identification. A retrospective analysis was conducted on 73 PDR patients (146 eyes), assessing carotid artery stenosis (CAS) severity via neck CT angiography and carotid doppler ultrasonography. Patients were divided into two groups based on ICA stenosis: normal to mild (Group I) and moderate or above (Group II). Retinal parameters, including the foveal avascular zone (FAZ) area and vessel density in both superficial and deep capillary plexuses, were analyzed using OCTA and Image J software. Statistical adjustments were made for age differences between the groups, and appropriate statistical methods were employed to account for the non-independence of eyes within the same patient and small sample size in Group II. The average age was 64.84±9.4 years. Group II patients were on average about 5 years older than Group I, and this age difference was significant. After adjusting for age, significant differences remained between the groups in macular vessel density of the superficial plexus and FAZ area, with Group II showing decreased vessel density and increased FAZ area in severe ICA stenosis cases. However, deep capillary plexus vessel density showed no significant variation. OCTA demonstrates that increased ICA stenosis in PDR patients correlates with decreased macular vessel density and expanded FAZ area, underscoring the need for carotid assessments. Early microvascular changes detectable by OCTA, even at mild NPDR stages, may indicate disease progression, positioning OCTA as a crucial non-invasive monitoring tool.

A woman aged in her 50s presented with sudden onset of vertigo, headaches and epistaxis, initially attributed to her background of chronic systemic hypertension. Ophthalmological history was significant for bilateral proliferative diabetic retinopathy and retinal detachment managed with vitrectomy, endolaser and intraocular silicone oil 5 years prior to the current presentation. Review of the initial non-contrast CT by our neuroradiology department identified right subretinal silicone oil, tracking along the subarachnoid space of the right optic sheath, right suprasellar cistern and non-dependent lateral ventricles. Initially interpreted as acute haemorrhage at an external facility, the patient underwent repeat imaging over a period of 6 months that failed to demonstrate expected temporal resolution of blood products. A subsequent unifying diagnosis of disseminated subarachnoid and intraventricular silicone oil was established. The following case highlights the relevant clinical history when requesting imaging and pertinent radiological findings when reporting patients postretinal detachment surgery.

To evaluate photoreceptor changes and their correlation with microvascular alterations in different stages of diabetic retinopathy (DR). Thirty-two treatment-naïve DR patients (50 eyes) without clinically significant macular edema were graded into mild, moderate, and severe non-proliferative DR (NPDR) and proliferative DR (PDR). The macula was imaged using adaptive optics scanning laser ophthalmoscopy (AO-SLO) at five subsections. Mean/minimum photoreceptor density, cell regularity, dispersion, spacing, and dark areas were recorded. Central macular thickness (CMT) and macular leakage index (MLI) were measured. The percentage of dark areas increased gradually with DR severity: 1.86 ± 1.52 (mild NPDR), 3.32 ± 1.45 (moderate NPDR), 4.22 ± 2.11 (severe NPDR), and 5.92 ± 3.06 (PDR) (P = 0.007). MLI also progressively increased (P = 0.01): 0.31 ± 0.60%, 3.47 ± 4.70%, 6.92 ± 5.44%, and 8.98 ± 3.58%, respectively. MLI showed a positive correlation with the percentage of dark areas (r = 0.36, P = 0.01). No significant changes were observed in photoreceptor density, regularity, dispersion, spacing, CMT, or visual acuity. Dark areas observed by AO-SLO suggest photoreceptor abnormalities, indicating decreased waveguiding ability and localized dysfunction as DR progresses. The positive correlation between dark areas and macular vascular leakage implies early dysfunction of the retinal neurovascular unit (NVU) in DR, warranting further investigation.

Comparative evaluations of commercially available artificial intelligence (AI) systems for use in diabetic retinopathy (DR) screening, particularly studies that identify systems by name, are limited, constraining procurement and implementation. This study aimed to identify commercially available AI systems potentially suitable for DR screening in a low-resource Tanzanian setting and compare their accuracy in detecting referable DR. Through a scoping review and expert consultation, we identified AI systems potentially suitable for implementation. Systems confirmed as suitable, and whose developers agreed to participate, were evaluated. Performance was assessed on a data set of retinal images collected from a Tanzanian DR screening program. The primary outcomes were sensitivity and specificity in detecting referable DR. Additional implementation data, including regulatory approval, referral thresholds, and additional product features, were also collected. Four commercially available AI systems (Medios AI/Remidio, MONA, Ophtai, and SELENA+) were evaluated. Among 689 people included in the test data set, 379 (55.0%) had referable DR and 93 (13.5%) had proliferative DR. Sensitivity in detecting referable DR ranged from 83.9% to 93.7%, with lower specificity ranging from 70.3% to 79.0%. Sensitivity in detecting proliferative DR exceeded 98% in all four AI systems. All the evaluated AI systems were Conformité Européenne-marked medical devices; one system (Medios AI/Remidio) functions offline as standard. Several commercially available AI systems demonstrated high sensitivity in detecting referable and proliferative DR, supporting their potential implementation. Consensus on minimum performance thresholds and consideration of implementation factors such as regulatory approval and offline functionality are needed.

To evaluate diabetic retinopathy (DR) severity, vision-threatening complications (VTCs), treatment patterns, and longitudinal outcomes among unhoused individuals with diabetes compared with stably housed controls. Using a large multicenter electronic health records network, patients with type 1 or type 2 diabetes were stratified by unhoused status via ICD-10 coding. Baseline DR severity, VTCs, and initial treatments were assessed across 9,643 unhoused and 1,180,509 housed patients. A secondary analysis requiring ≥6 months of ophthalmic follow-up yielded propensity score-matched cohorts of 2,911 patients each, with incident complications and treatments evaluated at one year. Multivariate Cox regression assessed the independent association between unhoused status and proliferative diabetic retinopathy (PDR) risk relative to other social determinants of health. At presentation, unhoused patients had significantly higher rates of PDR (5.24% vs 1.46%), vitreous hemorrhage (VH, 4.04% vs 1.03%), and tractional retinal detachment (TRD, 1.34% vs 0.24%) (all P<.01), and were less likely to achieve at least 6 months of follow-up (30.3% vs 55.2%; P<.01). In matched analysis, unhoused status increased 1-year risks of VH (hazard ratio [HR], 1.35; 95% confidence interval [CI], 1.08-1.69), diabetic macular edema (DME, HR, 1.28; 95% CI, 1.17-1.40), and PDR (HR, 1.25; 95% CI, 1.08-1.44). In multivariate analysis incorporating social determinants, unhoused status independently increased PDR risk (HR, 1.55; 95% CI, 1.41-1.71; P<.0001). Unhoused individuals present with more advanced DR, higher VTC rates, altered treatment patterns, and markedly reduced follow-up. These findings highlight unhoused status as a critical determinant of DR outcomes and underscore the need for targeted screening and socially informed ophthalmic care.

PurposeTo evaluate and compare the clinical validity of synthetic ultra-widefield fluorescein angiography (UWFA) images generated from ultra-widefield fundus photography (UWFP) using generative adversarial networks (GANs) in patients with diabetic retinopathy (DR).MethodsTwo GAN-based models, RegGAN and UWAFA-GAN, were trained to generate synthetic UWFA images from corresponding UWFP acquired using Optos California P200DTx. The dataset included 2084 image pairs (no DR: 124; nonproliferative DR: 795; severe NPDR: 770; proliferative DR: 395). Technical image similarity was assessed using peak signal-to-noise ratio (PSNR) and structural similarity index (SSIM). Clinical similarity was assessed by expert graders using the DR severity scale, comparing the synthetic images with corresponding real UWFA images.ResultsBoth models demonstrated acceptable performance in generating synthetic UWFA images from UWFP (F1-score range: 0.281–0.709; SSIM: 0.43–0.585; PSNR: 18.15–19.90). UWAFA-GAN achieved higher quantitative similarity metrics (SSIM and PSNR), indicating superior overall image fidelity, whereas RegGAN showed stronger clinical correlation with ground-truth UWFA, achieving higher diagnostic accuracy (66.7% vs. 49.2%) and more balanced precision, recall, and F1-scores across DR severity levels. This suggests that while UWAFA-GAN excels in pixel-level resemblance, RegGAN better preserves diagnostically relevant features.ConclusionsGAN models successfully generated UWFA images from UWFP in patients with DR. With further improvements, such models may serve as a complementary noninvasive imaging tool in selected clinical settings and may extend accessibility in resource-limiting settings.Translational RelevanceGAN-based synthetic angiography may serve as a noninvasive alternative to UWFA, particularly when conventional angiography is contraindicated or unavailable.

This study aimed to identify aqueous humor (AQ) proteins that correlate with vitreous humor (VH) levels in proliferative diabetic retinopathy (PDR). Using paired VH and AQ samples from 41 patients undergoing vitrectomy for PDR, we measured 92 protein biomarkers via a high-throughput immunoassay. We assessed Spearman correlations between AQ and VH levels for each protein and associations with PDR severity parameters. Thirty-two proteins showed positive AQ-VH correlations. A stepwise selection process, based on detectability, consistency and correlation strength narrowed this down to 21 proteins. An exploratory analysis of associations with disease severity parameters suggested that of these proteins, CXCL-10, IL-8, and MCP-4 were most strongly associated with fibrosis, while GAL-9 and PGF were most strongly associated with the extent of neovascularization, supporting the potential of these proteins as useful biomarkers. In conclusion, 21 AQ proteins show significant correlation with patient-matched VH levels in PDR patients, indicating that relative patient-level differences in AQ reflect relative differences in VH for these proteins. This methodological framework provides a foundation for future studies aiming to identify diagnostic or treatment stratification biomarkers for PDR and related complications such as diabetic macular edema.

Purpose: To assess differences in postoperative outcomes between combined phacovitrectomy and vitrectomy alone in patients undergoing surgery for complications of proliferative diabetic retinopathy (PDR). Methods: This retrospective analysis examined patients with PDR undergoing pars plana vitrectomy (PPV) at Boston Medical Center between 2019 and 2025. Patients were stratified by lens status and surgical approach (group 1 = phakic lens and PPV alone [n = 303]; group 2 = pseudophakic lens and PPV alone [n = 118]; group 3 = combined cataract extraction with intraocular lens and PPV, or phacovitrectomy [n = 198]). Outcomes included postoperative hemorrhage and postoperative complications within 90 days of surgery. Logistic regression was performed to evaluate the association between study group and outcomes, adjusting for relevant confounders. Results: A total of 619 eyes of 562 patients with PDR and an indication for PPV (nonclearing vitreous hemorrhage or tractional retinal detachment [TRD]) were included. Univariate analysis revealed that age, hypertension, perioperative bevacizumab use, TRD, intraoperative tamponade type, and operative status were associated with greater odds of experiencing postoperative complications and hemorrhage. After adjusting for confounders, multivariate analyses showed that group 3 patients (combined phacovitrectomy) had lower odds of postoperative complications (odds ratio [OR], 0.646 [95% CI, 0.439-0.952]; P = .027) and hemorrhage (OR, 0.640 [95% CI, 0.433-0.943]; P = .036) compared with group 1 patients (phakic vitrectomy), but group 2 patients (pseudophakic vitrectomy) did not have significantly decreased odds of either outcome. Conclusions: Patients undergoing combined phacovitrectomy have decreased odds of experiencing postoperative complications and hemorrhage within 90 days of surgery, compared with patients with phakic lens undergoing PPV alone. Lens removal at the time of vitrectomy for patients with PDR may reduce postoperative complication risk.