Pars plana vitrectomy in uveitis of diverse etiologies: indications and surgical outcomes Authors: Nilufer Zorlutuna Kaymak, Hatice Selen Kanar, Guzide Akcay, Ezgi Tanyeri Kılınç, Ilknur Sungu, Murat Oklar, Burak Tanyıldız Published: 2026 ABSTRACT AbstractBackgroundTo evaluate the indications, surgical outcomes, and diagnostic role of pars plana vitrectomy (PPV) in uveitis of diverse etiologies.MethodsThis retrospective study included 32 eyes (30 patients) undergoing 23-gauge PPV during 2019–2025. Uveitis etiology, surgical indications, and pre- and postoperative best-corrected visual acuity (BCVA) were analyzed. The primary outcome was change in BCVA, while secondary outcomes included anatomical success in retinal detachment (RD) cases, clinically meaningful visual improvement, diagnostic yield, and postoperative complications.ResultsThe most common etiologies were idiopathic uveitis (16.7%) and suspected primary vitreoretinal lymphoma (16.7%). Main surgical indications included tractional RD (28.1%), rhegmatogenous RD (21.9%), and diagnostic vitrectomy (18.8%). Final anatomical success was achieved in 93.8% of RD cases. BCVA improvement remained statistically significant at 1, 6, and 12 months after correction for multiple comparisons, while a gain of ≥ 2 Snellen lines was achieved in 46.2% of eyes at the final follow-up visit. No severe postoperative inflammation occurred.ConclusionsPPV is a safe and effective approach for managing vitreoretinal complications and diagnostic evaluation in uveitis, with favorable anatomical and functional outcomes.Keywords: Pars plana vitrectomy, Uveitis, Surgical outcomes, Diagnostic vitrectomy, Vitreoretinal complications Background Uveitis is a major cause of visual morbidity and may lead to irreversible structural damage and permanent visual impairment if not adequately controlled, as the eye has limited tolerance for inflammation due to the delicate structure of intraocular tissues [1–3]. The first report describing the role of pars plana vitrectomy (PPV) in uveitis was published in 1978 by Diamond and Kaplan, demonstrating favorable visual outcomes following combined lensectomy and vitrectomy for complicated cataract secondary to uveitis [4]. Since then, PPV has evolved as both a therapeutic and diagnostic tool in the management of uveitis and its complications, including removal of persistent vitreous inflammation and management of vision-threatening complications such as epiretinal membranes, macular holes, and retinal detachment (RD) [5–8]. However, most previous studies have focused on specific etiologies or isolated surgical indications, and data reflecting real-world clinical practice with heterogeneous etiologies and surgical indications remain limited [9,10]. Few studies have evaluated both the diagnostic and therapeutic roles of PPV across heterogeneous cohorts including both infectious and non-infectious uveitis in a real-world clinical setting. In this study, we aimed to evaluate the indications, surgical outcomes, and diagnostic yield of PPV in patients with uveitis of diverse etiologies. Methods This retrospective study included 32 eyes of 30 patients with uveitis whose initial presentations to our uveitis center occurred between January 2015 and October 2025. The PPV procedures included in the study were performed between May 2019 and July 2025. Patients with a postoperative follow-up of < 6 months were excluded. Follow-up duration varied among patients, and outcome measures at each time point were analyzed based on available follow-up data. Indications for PPV included removal of vitreous opacities and inflammatory debris, management of uveitis-related complications, and diagnostic vitreous sampling. Collected data included demographic characteristics, laterality, follow-up duration, interval between uveitis diagnosis and PPV, lens status, anatomical classification and etiology of uveitis, disease course, uveitis-related complications, laboratory findings, medical treatments, surgical indications, and intraoperative and postoperative complications and outcomes. Surgical indications and complication rates were analyzed on an eye basis (32 eyes of 30 patients), whereas demographic and etiological data were evaluated on a patient basis. All patients underwent standard ophthalmologic evaluation at each visit, including assessment of best-corrected visual acuity (BCVA) using Snellen charts, slit-lamp and fundus examination, intraocular pressure (IOP) measurement with Goldmann applanation tonometry, and optical coherence tomography (OCT). Fluorescein angiography (FA), indocyanine green angiography (ICG), and ocular ultrasonography were performed when indicated. For statistical analysis, BCVA values were converted to logarithm of the minimum angle of resolution (logMAR) units. Anterior chamber cell and flare counts were determined according to the Standardization of Uveitis Nomenclature (SUN) Working Group grading scheme [11]. Vitreous haze was graded based on standard photographs developed by Nussenblatt et al. [12]. The diagnosis of uveitis was established based on medical history, clinical examination, relevant laboratory investigations, and radiographic findings. Cataract was defined as any lens opacity and elevated IOP as > 21 mmHg. All surgeries were performed using 23-gauge PPV by experienced vitreoretinal surgeons, under local or general anesthesia. Combined cataract surgery was performed in cases with coexisting cataract, and vitreous samples were collected at the beginning of the procedure for diagnostic analysis when indicated. ILM peeling was not routinely performed but was undertaken selectively in eyes with an epiretinal membrane or proliferative vitreoretinopathy, according to the underlying macular pathology and intraoperative assessment of the macular surface. The choice of intraocular tamponade was individualized according to the underlying retinal pathology, the presence and severity of proliferative vitreoretinopathy, the extent and complexity of retinal detachment, the need for short- or long-term tamponade, and the surgeon’s intraoperative assessment. Silicone oil was generally preferred in complex RD cases, particularly those associated with extensive traction or proliferative vitreoretinopathy, whereas SF6 or C3F8 gas was used when temporary gas tamponade was considered sufficient. Air was used in selected cases requiring only short-term tamponade. The primary outcome measure was the change in BCVA from baseline, evaluated at 1, 3, 6, and 12 months and at the final follow-up. Secondary outcome measures included anatomical success in eyes with RD, defined as retinal reattachment; clinically meaningful visual improvement, defined as a gain of ≥ 2 Snellen lines at the final follow-up; the diagnostic yield of diagnostic PPV; and postoperative complications and additional surgical interventions during follow-up. For eyes with RD, primary anatomical success was defined as retinal reattachment after the initial PPV, whereas final anatomical success was defined as retinal reattachment at the last follow-up after any required reoperation. Statistical analyses were performed using SPSS Statistics (version 25.0) with p < 0.05 considered statistically significant. Given the limited sample size and small numbers within several clinical subgroups, subgroup comparisons were considered exploratory and hypothesis-generating rather than confirmatory. Preoperative and final follow-up IOP values were compared using the Wilcoxon signed-rank test because the data were not normally distributed and included extreme values. Changes in BCVA from baseline to each postoperative time point were evaluated using the Wilcoxon signed-rank test. To account for multiple comparisons across postoperative time points, p values were adjusted using the Holm–Bonferroni method. For very low visual acuity levels (< 0.01), counting fingers (CF) and hand motion (HM) at two feet were converted to Snellen equivalents according to the method described by Holladay, corresponding to 0.01 (+ 2.0 logMAR) and 0.001 (+ 3.0 logMAR), respectively [13]. This study adhered to the Declaration of Helsinki and was approved by the Local Ethics Committee of our hospital (No: 2025/010.99/19/35) on 27.08.2025. Standard surgical informed consent was obtained from all patients prior to surgery. Results A total of 30 patients were included in the study. Regarding ocular involvement, 15 patients (50%) had bilateral uveitis, while the remaining 15 patients (50%) had unilateral involvement. Pars plana vitrectomy was performed in 32 eyes of these 30 patients; 28 patients underwent unilateral PPV, whereas two patients had bilateral PPV. Two patients diagnosed with primary vitreoretinal lymphoma (PVRL) after PPV died during follow-up due to systemic disease, one before the 6-month visit and the other before the 12-month visit. For these two patients, the BCVA recorded at their last available postoperative visit before death was used as the final follow-up BCVA value. Missing BCVA data at the scheduled postoperative time points mainly reflected follow-up visits occurring outside the predefined assessment windows rather than exclusion from the study. A flowchart illustrating study inclusion and the availability of BCVA data at each follow-up assessment is presented in Fig. 1. The baseline demographic and clinical characteristics of the study population are summarized in Table 1. Flowchart of eyes included in the BCVA analysis at each follow-up time point. The number of eyes with available BCVA data decreased at intermediate visits mainly because some postoperative examinations occurred outside the predefined assessment windows; final follow-up values reflect the last available visit for each eye Table 1Baseline demographic and clinical characteristics of the study populationParameterValueNumber of patients / eyes30/32Age, mean ± SD (range), years50.7 ± 14.8 (21–76)Sex, n (%) Male16 (53.3) Female14 (46.7)PPV performed, n (%) Unilateral surgery28 (93.3) Bilateral surgery2 (6.7)Median follow-up duration from initial presentation, months (IQR, range)36 (41, 6–120)Median time from uveitis diagnosis to PPV, months (IQR, range)4.8 (19.38, 0.07–75.23)Anatomical classification of uveitis Panuveitis21 (70.0) Posterior uveitis5 (16.7) Intermediate uveitis3 (10.0) Anterior uveitis1 (3.3)Abbreviations: SD, standard deviation; IQR, interquartile range; PPV, pars plana vitrectomy Baseline demographic and clinical characteristics of the study population Abbreviations: SD, standard deviation; IQR, interquartile range; PPV, pars plana vitrectomy The most frequent underlying diagnoses leading to PPV were idiopathic uveitis and suspected PVRL (5 patients each, 16.7%), followed by sarcoidosis and acute retinal necrosis (ARN) (4 patients each, 13.3%). Behçet’s disease and syphilitic uveitis were identified in 3 patients each (10.0%), while toxoplasma uveitis was detected in 2 patients (6.7%). Pars planitis, lupus, Fuchs uveitis, and tuberculosis-associated vasculitis were the least frequent etiologies, occurring in 1 patient each (3.3%). The indications for PPV and their distribution according to infectious and non-infectious uveitis are summarized in Table 2. Tractional RD was the most frequent indication for surgery, followed by rhegmatogenous RD and diagnostic vitrectomy. Bilateral surgeries were performed in two patients (6.7%). All 16 RD cases were considered related to uveitis or its sequelae based on their clinical context; no eye had high myopia or pre-existing peripheral retinal degeneration unrelated to uveitis. Table 2Indications for pars plana vitrectomy according to uveitis etiology (n = 32 eyes)Indications for PPVTotal n (%)Infectious n (%)Non-infectious n (%)Tractional RD9 (28.1)5 (55.6)4 (44.4)Rhegmatogenous RD7 (21.9)4 (57.1)3 (42.9)Diagnostic vitrectomy6 (18.8)0 (0.0)6 (100.0)Vitreous hemorrhage4 (12.5)1 (25.0)3 (75.0)Epiretinal membrane4 (12.5)0 (0.0)4 (100.0)Vitreous opacity1 (3.1)0 (0.0)1 (100.0)PVR1 (3.1)1 (100.0)0 (0.0)Abbreviations: PPV, pars plana vitrectomy; RD, retinal detachment; PVR, proliferative vitreoretinopathy Indications for pars plana vitrectomy according to uveitis etiology (n = 32 eyes) Abbreviations: PPV, pars plana vitrectomy; RD, retinal detachment; PVR, proliferative vitreoretinopathy At baseline, 13 eyes (40.6%) were phakic without cataract, 12 eyes (37.5%) were phakic with cataract, and seven eyes (21.9%) were pseudophakic. Combined phacoemulsification and intraocular lens (IOL) implantation with PPV was performed in 16 eyes (50.0%), whereas the remaining eyes underwent PPV alone. Intraocular tamponade included silicone oil in 16 eyes (50.0%), balanced salt solution in 8 eyes (25.0%), SF6 gas in 3 eyes (9.4%), air in 3 eyes (9.4%), and C3F8 gas in 2 eyes (6.3%). Membrane peeling was performed in 14 eyes (43.8%), including internal limiting membrane (ILM) peeling in 8 eyes (25.0%), epiretinal membrane (ERM) peeling in 4 eyes (12.5%), and combined ERM-ILM peeling in 2 eyes (6.3%). Diagnostic PPV was performed in six eyes (18.8%) of five patients, with PVRL being the presumed diagnosis in all six eyes. No aqueous humor sampling or intraocular PCR analysis had been performed before PPV; therefore, PPV represented the initial invasive intraocular diagnostic procedure rather than an investigation performed after inconclusive prior intraocular testing. Retinal biopsy was performed in four eyes. Histopathological examination established a definitive diagnosis of PVRL in three eyes, leading to the initiation of lymphoma-directed treatment. Retinal biopsy yielded a negative result in the remaining eye, and a definitive diagnosis could not be established at that time. This patient subsequently developed neurological manifestations, and lymphoma was later confirmed by brain biopsy, after which lymphoma-directed treatment was initiated. In the remaining two eyes, which belonged to the same patient, cytological examination of vitreous samples revealed atypical B lymphocytes. However, retinal biopsy could not be performed because no identifiable retinal lesion suitable for biopsy was present. Although these cytological findings strengthened the clinical suspicion of PVRL, they were insufficient to establish a definitive histopathological diagnosis, and lymphoma-directed treatment could not be initiated. The overall definitive diagnostic yield of diagnostic PPV at the time of surgery was 50.0% (3/6 eyes), while retinal biopsy established a definitive diagnosis in 75.0% of the eyes in which it was performed (3/4 eyes). The mean preoperative IOP was 14.59 ± 3.51 mmHg (range, 10–23), whereas the mean IOP at the final follow-up visit was 16.25 ± 5.45 mmHg (range, 7–38). Because the IOP data were not normally distributed and included extreme values, the comparison was performed using the Wilcoxon signed-rank test. No statistically significant difference was observed between preoperative and final IOP values (Z = − 1.645, p = 0.100). During follow-up, 22 eyes (68.8%) required no additional surgical intervention. Primary anatomical success after the initial PPV was achieved in 13 of 16 eyes with RD (81.3%), while final anatomical success after repeat surgery, when required, was achieved in 15 of 16 eyes (93.8%). Among eyes with RD, recurrent RD associated with proliferative vitreoretinopathy occurred in three eyes and necessitated repeat PPV. These included one eye with HIV-associated syphilitic uveitis that initially underwent PPV for tractional RD, one eye with idiopathic uveitis that initially underwent PPV for rhegmatogenous RD, and one eye with acute retinal necrosis that initially underwent PPV for rhegmatogenous RD. Following reoperation, retinal reattachment was achieved in the eyes with HIV-associated syphilitic uveitis and acute retinal necrosis, whereas anatomical success could not be achieved in the eye with idiopathic uveitis because of persistent proliferative vitreoretinopathy. No major intraoperative complications were observed. Silicone oil removal was performed in five eyes (15.6%). During follow-up, angle-closure glaucoma secondary to pupillary block developed in one eye with idiopathic uveitis that had initially undergone PPV for rhegmatogenous RD and was successfully managed with laser iridotomy. Trabeculectomy was required because of uncontrolled IOP in two eyes: one eye with Fuchs uveitis that had undergone PPV for vitreous opacities and one eye with idiopathic uveitis that had undergone PPV for rhegmatogenous RD. IOP was controlled following trabeculectomy in both eyes. In addition, IOP elevation following an intravitreal injection occurred in one eye with HIV-associated syphilitic uveitis that had initially undergone PPV for tractional RD and was successfully controlled with medical treatment. The highest final IOP value (38 mmHg) was observed in the eye with Fuchs uveitis that required trabeculectomy, whereas the lowest value (7 mmHg) occurred in an eye with RD and reflected mild hypotony without clinically significant sequelae. Two eyes (6.3%) underwent intravitreal injection for choroidal neovascularization, and one eye (3.1%) received additional retinal laser treatment. At the last postoperative visit, normal macular anatomy without edema was observed in 21 eyes (65.6%), whereas macular edema, macular atrophy, and macular hole were detected in 6 eyes (18.8%), 4 eyes (12.5%), and 1 eye (3.1%), respectively. Secondary cataract developed in two eyes (6.3%): one eye with advanced sarcoid uveitis that had initially undergone PPV for rhegmatogenous RD and one eye with idiopathic uveitis that had undergone PPV for epiretinal membrane and vitreomacular traction. Both eyes subsequently underwent cataract surgery with favorable outcomes. Seven eyes (21.9%) underwent intravitreal dexamethasone implant injection during follow-up. Among these, two eyes (6.3%) required additional sub-Tenon triamcinolone injection due to recalcitrant cystoid macular edema. The mean preoperative BCVA was 1.72 ± 1.20 logMAR (range, 0.00–3.00; n = 32). Postoperatively, the mean BCVA was 0.98 ± 0.83 logMAR at 1 month (n = 27), 1.00 ± 0.85 logMAR at 3 months (n = 22), 0.79 ± 0.62 logMAR at 6 months (n = 21), and 0.73 ± 0.72 logMAR at 12 months (n = 19). The mean BCVA at the final follow-up visit, corresponding to the last available postoperative assessment for each eye, was 1.26 ± 1.12 logMAR (n = 26). After Holm–Bonferroni correction for multiple comparisons, BCVA improvement compared with baseline remained statistically significant at 1 month (adjusted p = 0.035), 6 months (adjusted p = 0.036), and 12 months (adjusted p = 0.036). However, the differences at 3 months and the final follow-up visit were not statistically significant after adjustment (adjusted p = 0.058 for both). At the final available follow-up visit (n = 26), visual acuity improved in 13 eyes (50.0%), remained stable in 9 eyes (34.6%), and worsened in 4 eyes (15.4%). Thirteen eyes (50.0%) achieved a BCVA of 20/200 or better, including 7 eyes (26.9%) with 20/40 or better visual acuity at the final follow-up visit. Changes in mean BCVA over time are shown in Fig. 2. Mean best-corrected visual acuity (BCVA) in logMAR units at baseline and during postoperative follow-up. Error bars represent standard deviations. The numbers of eyes with available BCVA data were 32 at baseline, 27 at 1 month, 22 at 3 months, 21 at 6 months, 19 at 12 months, and 26 at the final follow-up visit. Statistical comparisons were performed against baseline using the Wilcoxon signed-rank test with Holm–Bonferroni correction for multiple comparisons. *Adjusted p < 0.05; ns, not statistically significant At the time of surgery, 17 eyes (53.1%) had active inflammation, and 15 eyes (46.9%) were in a quiescent stage. Of the 17 eyes with active inflammation, 11 eyes (64.7%) had non-infectious uveitis, whereas 6 eyes (35.3%) had an infectious etiology. Among the 15 eyes in the quiescent stage, 10 eyes (66.7%) had non-infectious uveitis, whereas 5 eyes (33.3%) had an infectious etiology. A visual gain of ≥ 2 Snellen lines was observed in 12 eyes (46.2%) at the final follow-up visit. In exploratory subgroup analyses, a ≥ 2 Snellen-line visual gain was observed in 6 of 15 eyes (40.0%) with active inflammation and in 6 of 11 eyes (54.5%) with quiescent inflammation, with no statistically significant difference detected between the groups (p = 0.462). Similarly, a ≥ 2 Snellen-line visual gain was observed in 53.8% of non-RD eyes and 38.5% of RD eyes, with no statistically significant difference detected between the groups (p = 0.431). Given the limited subgroup sizes, these findings should be interpreted cautiously. At the time of surgery, 10 of 30 patients (33.3%) were receiving systemic immunosuppressive therapy, including biologic treatment in five (16.7%) patients. All patients with infectious uveitis were receiving appropriate antimicrobial therapy at the time of surgery. No severe postoperative inflammatory exacerbation or hypotony resulting in clinically significant sequelae was observed during follow-up. Discussion Pars plana vitrectomy plays an important role in the management of uveitis-related vitreoretinal complications as well as in establishing the diagnosis in selected cases [5–9]. In the present study, we evaluated the indications and outcomes of PPV in a heterogeneous cohort of patients with uveitis. Our results demonstrated that PPV was performed for a wide range of diagnostic and therapeutic indications and was associated with functional visual improvement in nearly half of the operated eyes. In addition, high anatomical success was achieved in eyes undergoing surgery for RD, and no severe postoperative inflammatory exacerbation was observed. Unlike many previous studies that focused on specific etiologies or isolated surgical indications, our study reflects real-world clinical practice by including patients with heterogeneous uveitis etiologies and a wide spectrum of surgical indications. Importantly, our cohort encompassed both infectious and non-infectious uveitis, allowing a more comprehensive evaluation of the role of PPV in both diagnostic and therapeutic settings. This approach may better represent daily clinical practice and highlights the broad applicability of PPV in diverse clinical scenarios. In our cohort, the most common indication for PPV was RD, followed by diagnostic vitrectomy and ERM. Previous reports have emphasized the role of PPV in the management of vitreoretinal complications in uveitis patients, including RD, ERM, and diagnostic indications [14]. However, it is now well established that PPV serves both therapeutic and diagnostic purposes in the management of uveitis [15]. In line with this evolving role, RD was the leading indication in our series, while diagnostic vitrectomy represented the second most common indication, reflecting the increasing recognition of diagnostic vitrectomy as an important tool in contemporary clinical practice [16]. All 16 RD cases were considered related to uveitis or its sequelae based on their clinical context. The nine eyes with tractional RD developed in the context of infectious retinitis/vasculitis or inflammatory retinal vasculitis, including syphilitic/HIV-associated retinitis (bilateral in one patient), acute retinal necrosis, Behçet’s disease, pars planitis, and idiopathic retinal vasculitis. These inflammatory and proliferative processes were considered to have contributed to tractional membrane formation. Among the seven eyes with rhegmatogenous RD, retinal breaks were attributed to necrotizing retinitis in two eyes, toxoplasma retinochoroiditis in two eyes, and sarcoidosis-related retinal vasculitis in one eye. The remaining two eyes had idiopathic uveitis, in which a specific inflammatory mechanism could not be further characterized. No eye had high myopia or pre-existing peripheral retinal degeneration unrelated to uveitis. The wide IQR relative to the median interval between uveitis diagnosis and PPV reflects the heterogeneous clinical course and surgical indications in this cohort. Some eyes required early intervention for acute or vision-threatening complications, whereas others underwent surgery later because of chronic or progressive vitreoretinal complications. Retinal detachment constituted a substantial proportion of surgical indications in our series. In these eyes, PPV achieved favorable anatomical outcomes, with a final retinal reattachment rate of 93.8%. This success rate is comparable to previously reported outcomes in uveitic RD, which is known to have variable anatomical and functional results and remains a challenging complication of intraocular inflammation, largely due to proliferative vitreoretinopathy and inflammation-related tractional changes [7]. These favorable outcomes may be related to advances in surgical techniques and improved case selection in recent years. Visual outcomes after surgery showed improvement in a considerable proportion of eyes. A clinically meaningful visual gain of ≥ 2 Snellen lines was achieved in 46.2% of eyes at the final follow-up visit. Although the degree of visual improvement varied depending on baseline visual acuity and underlying pathology, BCVA improvement remained statistically significant at 1, 6, and 12 months after correction for multiple comparisons. Collectively, these findings suggest a potential functional benefit of PPV in selected patients with uveitis. The apparently worse mean BCVA at the final follow-up compared with the 6- and 12-month values should not be interpreted as a uniform longitudinal decline in visual acuity. The final follow-up represented the last available postoperative visit for each eye rather than a fixed postoperative time point, and the number and composition of eyes contributing data differed across assessments. In addition, several eyes included in the final analysis had poor visual outcomes associated with structural macular abnormalities, including macular edema, macular atrophy, and macular hole, which may have contributed to the higher mean logMAR BCVA at the final visit. However, visual recovery in uveitic eyes may be limited by pre-existing structural damage to the retina or macula. Chronic or recurrent inflammation may result in cumulative ocular damage, including macular edema, macular atrophy, epiretinal membrane formation, or retinal detachment, all of which can compromise the potential for visual improvement despite successful anatomical outcomes [17]. This may partly explain the variability in visual outcomes observed in our cohort. Although surgery is traditionally recommended during quiescent phases, more than half of the eyes in our cohort had active inflammation at the time of surgery, and no severe postoperative inflammatory exacerbation was observed. The optimal timing of PPV in uveitis remains controversial. While adequate control of intraocular inflammation prior to surgery has classically been recommended, delayed intervention may result in irreversible structural damage, particularly in cases with chronic macular involvement. Therefore, surgical timing should be individualized based on disease activity, underlying pathology, and the risk of vision-threatening complications [18]. Our findings are consistent with previous reports indicating that PPV can be safely performed in selected cases with active inflammation when clinically indicated [18,19]. The beneficial effects of vitrectomy in uveitis are likely multifactorial. Removal of the vitreous gel may reduce the intraocular inflammatory load by eliminating inflammatory mediators, immune complexes, and inflammatory cells. In addition, vitrectomy may facilitate the effectiveness and intraocular distribution of therapeutic agents, thereby contributing to improved inflammation control and visual stabilization [18]. In addition to its therapeutic role, PPV contributed to the diagnostic evaluation of selected cases with clinically suspected PVRL. Diagnostic PPV enabled histopathological confirmation of PVRL in three eyes and directly influenced subsequent management by allowing the initiation of lymphoma-directed treatment. However, in one eye with a negative retinal biopsy, lymphoma was subsequently confirmed by brain biopsy after the development of neurological manifestations, illustrating the potential limitations of ocular tissue sampling in cases with persistent clinical suspicion. In the remaining two eyes, vitreous cytology demonstrated atypical B lymphocytes but did not provide definitive histopathological confirmation, further illustrating the diagnostic challenges associated with PVRL. Collectively, these findings support the clinical value of intraocular tissue sampling in suspected PVRL while emphasizing the importance of continued clinical and systemic evaluation when ocular diagnostic findings are inconclusive [20]. This study has several limitations. First, its retrospective, single-center design and relatively small sample size may limit the generalizability of the findings. The heterogeneity of uveitis etiologies and surgical indications represents an important limitation of this study. Because the included conditions differed substantially in their prognosis and surgical objectives, pooling these clinically diverse cases may have obscured meaningful differences in anatomical and functional outcomes between subgroups. Further stratification according to individual uveitis etiologies or surgical indications would have resulted in very small subgroups and was therefore not performed, as such analyses would not support meaningful statistical inference and could be misleading. The small number of eyes within several clinical subgroups limited statistical power and increased the possibility of type II error; therefore, non-significant findings should not be interpreted as evidence of the absence of an association. Subgroup comparisons were considered exploratory and hypothesis-generating rather than confirmatory and should be interpreted cautiously. Furthermore, the limited sample size and low frequency of several clinical variables precluded a reliable multivariable analysis. Second, incomplete availability of preoperative macular structural data limited the evaluation of baseline prognostic factors. Conclusions Pars plana vitrectomy represents an important surgical option in the management of uveitis, offering both therapeutic and diagnostic benefits. In a real-world heterogeneous cohort, PPV provided favorable anatomical outcomes and meaningful visual improvement without inducing severe postoperative inflammatory exacerbation, supporting its role in contemporary clinical practice.