Clinical macular edema following uneventful cataract surgery in eyes without pre-existing fundus diseases: risk factors and treatment prognosis

Abstract

Background

To investigate the risk factors and prognosis of clinical pseudophakic cystoid macular edema (PCME) after uneventful phacoemulsification surgery in patients without pre-existing fundus diseases.

Methods

This was a retrospective case-control study. Medical records between August 2020 and August 2023 were reviewed for patients who had no previous fundus diseases and developed clinical PCME. A control group was randomly chosen and the risk factors for PCME was analyzed by binary logistic regression. Structure and visual prognosis of the PCME cohort were observed and compared among subgroups undergoing different treatment measures.

Results

Forty-seven eyes of 47 patients with PCME were included. The development of PCME was associated with higher systolic blood pressure (OR, 1.048; 95%CI 1.002, 1.097; P = .042), no posterior vitreous detachment (OR, 0.215; 95%CI: 0.553, 0.887; P = .032) and shorter axial lengths (OR, 0.401; 95%CI 0.161, 0.997; P = .049) compared to controls. During a mean follow-up of 8.26 months, 36 eyes (76.6%) showed visual improvement with decreased macular thickness. Different treatment modalities, including observation, topical NSAIDs, and intervention therapy, have no significant differences on the visual prognosis (P = 1.000). However, the intervention group had a shorter recovery time compared to the observation group (28.6 vs. 45.9 days, P = .037).

Conclusion

PCME remains an encountered morbidity in patients without pre-existing fundus diseases. Shorter axial lengths, absence of posterior vitreous detachment, and higher systolic blood pressure are risk factors of PCME. Active intervention failed to improve the prognosis of PCME but could shorten the recovery time.

Background

Phacoemulsification cataract surgery is currently the most popular and safe effective intervention for cataracts [1]. A normal structure and function of the retina and macula are essential for achieving favorable visual outcomes following surgery. However, some patients with normal fundus before surgery may develop pseudophakic cystoid macular edema (PCME), manifesting as early postoperative visual function improvement followed by visual acuity decline and metamorphopsia approximately 4–6 weeks later [2]. This phenomenon is also known as Irvine-Gass syndrome, first described by Irvine in 1953 [3], and further studied by Gass and Norton in 1966 using fluorescein fundus angiography (FFA) [4]. Although the incidence rate of clinical PCME has once been reported to be 0.1–2.35% [5, 6], asymptomatic PCME as detected by optical coherence tomography (OCT) or FFA far exceeds this proportion [7,8,9]. In the United States, the annual total ophthalmic payments were 47% ($1,092) higher for those who developed PCME than those who did not [10]. PCME remains a commonly encountered morbidity.

The pathogenesis of PCME is thought to be multifactorial. Increased intraocular inflammatory mediators after surgical manipulation appears to the major etiology. Inflammation breaks down the blood-retinal barrier, leading to increased capillary permeability and resulting in the accumulation of fluid in the outer plexiform and inner nuclear layers of the retina to create cystic spaces [11]. Surgical complications appear to be the risk factors of PCME, which include eyes with posterior capsule rupture and vitreous loss, vitreous traction at incision sites, iris trauma, and intraocular lens dislocation [12]. In addition, PCME more likely develops in those with a prior history of fundus diseases, such as diabetic retinopathy and diabetic macular edema, uveitis, retinal vein occlusion, epiretinal membrane, and retinal detachment repair [13]. The epidemiology and risk factors of PCME in populations without a history of surgical complications and fundus diseases are not yet clear.

There is currently no standardized treatment or prophylactic protocol for PCME. Although most patients with clinical PCME will experience spontaneous revolve within 3 to 6 months [9], there are still some patients who progress to chronic CME and ultimately affect vision. Furthermore, it remains unclear whether active intervention, such as injection of anti-vascular endothelial growth factor (VEGF) or steroids, can bring additional benefits to patients with PCME.

In this study, we reviewed patients who developed PCME after uneventful phacoemulsification cataract surgery without a history of fundus diseases and evaluated its impact on postoperative visual function upon different treatment measures. The potential risk factors of PCME were also investigated using a control group randomly selected from the same hospital patients of the same period.

Materials and methods

Subjects

This retrospective case-control study conformed to the tenets of the Declaration of Helsinki. The study was conducted in compliance with a suitable accredited institutional review board (IRB) from the Ethics Committee of Joint Shantou International Eye Center (JSIEC). The study was deemed exempt from informed consent due to its retrospective design. Medical records for patients who underwent phacoemulsification cataract surgery between August 2020 and August 2023 were reviewed. Those who met the following criteria were included in the PCME group: (1) underwent uneventful phacoemulsification combined with intraocular lens implantation; (2) reported a blurred vision or metamorphopsia during the follow-up period; (3) diagnosed with macular edema confirmed by OCT and/or FFA; (4) displayed no signs of macular edema or other macular pathologies before the cataract surgery. Exclusion criteria included: (1) had a history of retinopathy or other diseases that might cause macular edema, including but not limited to diabetic retinopathy, retinal vein occlusion and other retinal vascular diseases, uveitis, retinitis pigmentosa, neovascular macular degeneration, epiretinal membrane, and a post-vitrectomy setting; (2) Loss to follow-up. In addition, patients who underwent phacoemulsification cataract surgery were randomly selected as the control group using a computer randomization program. The selection process involved excluding patients meeting the aforementioned exclusion criteria.

Data collection

Medical records were reviewed for the following content: (1) general information including gender, age, occupation, education level, medical history, systemic diseases, etc.; (2) ocular examination findings conducted within one week prior to the cataract surgery, including best-corrected visual acuity (BCVA), intraocular pressure (IOP), axial length, anterior chamber depth, central corneal thickness, lens thickness, endothelial cell density, iris width white to white, ocular b-scan ultrasound, and OCT; (3) systemic examination conducted within one week prior to the cataract surgery, including blood pressure measurement, complete blood count, blood biochemistry, fasting blood glucose, and others; (4) Surgery records, such as surgeon seniority, pupil size, phaco parameters, operational illumination time, bottle height of the irrigation fluid; (5) follow-up findings including slit-lamp examination, BCVA assessment and OCT imaging on the 1st postoperative day, 1st postoperative week, 1st and 3rd postoperative months, and the last visit. During the follow-up period, improvement of BCVA by equal to or greater than 1 Snellen line accompanied by a reduction in CRT was defined as improvement of PCME. Decrease in BCVA by more than 1 Snellen line with an increase in CRT compared to pre-treatment were defined as worsening of PCME, while the remaining cases were defined as no change in PCME severity. Patients with incomplete preoperative systemic examination results were included only in the analysis of treatment and prognosis, while being excluded from the analysis of risk factors.

Management options

This was a retrospective study, and the choice of treatment may have depended on multiple factors, including the ophthalmologist’s preference, the patient’s needs, and the microstructure and extent of macular edema. Some patients were managed with observation without any intervention. Some received ocular eyedrops, including NSAIDs with or without artificial tears. Some patients underwent periocular steroid injections (triamcinolone acetonide, 20 mg), while others received intravitreal injections, including anti-VEGF agents (ranibizumab, 0.5 mg/mL) or steroids (triamcinolone acetonide, 4 mg, or dexamethasone implant, 0.7 mg). One patient with PCME underwent vitrectomy combined with posterior vitreous detachment induction and internal limiting membrane peeling due to the presence of posterior vitreous cortex adhesion and macular traction.

Statistical analyses

Nominal data were presented as frequencies and percentages and numerical data as mean with standard deviation (SD). The chi-square (χ²) test with Yates’s correction for continuity or Fisher’s exact tests were used for the comparison of categorical variables between the PCME group and the control group, including analysis for risk factor and treatment outcomes. Unpaired t-test was used for continuous variables and Mann–Whitney U-test for nonparametric variables. Binary logistic regression analysis was used to confirm risk factors and establish risk assessment models. One way analysis of variance (ANOVA) and Dunn’s multiple pairwise comparison was used to evaluate the difference in outcomes between the treatment groups and observation group. A p-value of ≤ 0.05 indicated a significant difference. All statistical analyses were performed using IBM SPSS ver. 25 (IBM Corp., Armonk, NY, USA).

Results

Demographics

A total of 47 eyes from 47 patients diagnosed with PCME were enrolled in this study. Among them, 21 were male, with a mean age of 67.9 ± 9.1 (range: 44 ~ 86) years. The average time between cataract surgery and PCME onset was 39.8 ± 19.0 (range: 7 ~ 76) days. Baseline data were presented in Table 1.

Table 1 Demographics of patients with PCME

Risk factors

Detailed clinical and experimental data from 29 PCME cases were used for risk factor analysis. Meanwhile, 36 eyes without PCME as confirmed by OCT were randomly selected as controls. Ophthalmic examination revealed no significant differences in preoperative BCVA and IOP between the PCME group and the control group. One-month following cataract surgery, both groups exhibited significant BCVA improvements, whereas the PCME group demonstrated a significant decrease in BCVA compared than control group (0.43 ± 0.19LogMAR vs. 0.08 ± 0.08LogMAR, P < .001) (Fig. 1A). No significant difference in postoperative IOP were detected between the two groups (Fig. 1B). In addition, the occurrence of PCME was associated with lower incidence of posterior vitreous detachment (PVD) detected by B-scan ultrasonography (44.8% vs. 83.3%, χ = 6.157, P = .013) and a shorter axial length (23.1 ± 0.7 mm vs. 23.9 ± 1.6 mm, P = .028) compared to the control group. Systemic and laboratory examinations showed that the PCME group had higher platelet count (261.3 ± 68.8 × 10⁹/L vs. 222.5 ± 46.4 × 10⁹/L, P = .009) and higher fasting blood glucose levels (7.2 ± 2.5mmol/L vs. 6.2 ± 1.8mmol/L, P = .017), as compared to the control group. No significant differences were observed between the two groups in other factors, such as surgeon seniority, surgical duration, and phaco energy (Tables 2 and 3).

Fig. 1

Comparison of perioperative best-corrected visual acuity (A) and intraocular pressure (B) between patients with PCME and non-PCME. Data were presented with mean ± standard deviation. P values were generated by Mann-Whitney U test. Statistically significant at P < .05. (PCME = pseudophakic cystoid macular edema)

Table 2 Comparison of nominal characteristic in patients with PCME and without PCME

Table 3 Comparison of numerical characteristic in patients with PCME and without PCME

Binary logistic regression analysis was further conducted to assess the potential influencing factors for PCME. Two systemic factors, systolic blood pressure (P = .052) and triglyceride levels (P = .051), were also included in the analysis. The results indicated that the development of PCME was associated with elevated systolic blood pressure (OR, 1.048; 95%CI 1.002, 1.097; P = .042). Conversely, the PCME eyes presented with lower incidence of PVD (OR, 0.215; 95%CI: 0.553, 0.887; P = .032) and shorter axial length (OR, 0.401; 95%CI 0.161, 0.997; P = .049) compared to controls. The area under the receiver operating characteristic curve (AUC) from the logistic model was 0.8579, indicating the collective significance of all predictor features (Fig. 2).

Fig. 2

Logistic regression analysis of risk factors for PCME occurrence. (A) Odds ratio of risk factors in forest plot; (B) ROC curve of the logistics model. (AL = axial length, PVD = posterior vitreous detachment, SBP = systolic blood pressure, PLT = platelet count, TG = triglyceride, FBG = fasting blood glucose. OR = Odds ratio, CI = confidence interval, ROC = receiver operating characteristic)

Treatments and follow-ups

BCVA and OCT-based macular edema was evaluated with a mean follow-up of 8.26 ± 2.55 months. As shown in Fig. 3A, the BCVA significantly decreased at about one month postoperatively due to the presence of PCME. After that, the BCVA showed gradual improvement and could recover to the early postoperative levels at six months postoperatively. Specifically, among the 47 cases, 36 (76.6%) showed visual improvement during the follow-up period, 10 (21.3%) remained unchanged, and 1 (2.1%) experienced visual deterioration. The average CMT increased significantly at the onset of PCME compared to pre-surgical levels (444.7 ± 178.9 μm vs. 182.9 ± 21.8 μm, P < .001). After a follow-up of 34.5 ± 16.7 days, the CMT significantly decreased to 277.3 ± 156.8 μm (P < .001) (Fig. 3B).

Fig. 3

Follow-up of patients with PCME for central retinal thickness (A) and best-corrected visual acuity (B). Data were presented with mean ± standard deviation. P values were generated by student t test. Statistically significant at P < .05. (PCME = pseudophakic cystoid macular edema, d = day, wk = week, mo = month)

Among the patients in this group, 17 cases (36.2%) underwent intervention treatments, which included intravitreal anti-VEGF injection (n = 3), intravitreal or periocular injection of steroid drugs (n = 11), and surgical treatment (n = 1). Fifteen cases (31.9%) received topical nonsteroidal anti-inflammatory drugs (NSAIDs). The other 15 cases (31.9%) did not undergo any treatment measures and were only under observation. There were no significant differences in the rate of treatment success during the follow-up period among the three subgroups (P = .918) (Table 4). However, further analysis of patients showing improvement in BCVA and CRT revealed significant differences in recovery time among the three groups (P = .032). Multiple pairwise comparisons indicated that the intervention group had a shorter recovery time compared to the observation group (28.6 ± 16.7 days vs. 45.9 ± 16.3 days, P = .037) (Table 5). A further subgroup analysis of the interventions revealed that among the four groups (observation, topical NSAIDs, periocular injection, and intravitreal injection), there was no significant difference in visual prognosis (χ = 0.738, P = 1.000), although all management options led to a significant improvement in visual outcomes (Fig. 4).

Table 4 Treatments and visual prognosis in patients with PCME

Table 5 Recovery time of PCME in different treatment groups

Fig. 4

Changes in best-corrected visual acuity at presentation of PCME and the final follow-ups among different management options. Four management options (observation, topical NSAIDs, periocular injection, and intravitreal injection) led to a significant improvement in visual outcomes. *P < .05, **P < .01. (PCME = pseudophakic cystoid macular edema)

Discussion

In this study, we included 47 eyes from 47 patients diagnosed with clinical PCME. During the same period, the hospital conducted approximately 11,000 cataract surgeries, resulting in a PCME incidence rate of 0.43%, which was comparable to previous research reporting the rates ranging from 0.1 to 2.35% [5, 6]. However, this incidence rate should be considered a minimal estimate, as there may be cases of patients who were either followed up at external hospitals or did not follow at all. In addition, clinicians may attribute a mild postoperative vision decline to factors like refractive shifts or posterior capsule opacification. Indeed, the incidence of subclinical PCME based on FFA or OCT evidence has been reported to be 11%~30% [7, 8] and even exceeding 40% [9]. Careful consideration should therefore be given to the presence of PCME, especially in patients manifesting unexpected vision loss approximately one month after surgery.

Several risk factors for PCME have been previously described [14]. Surgical complications, such as posterior capsule rupture and vitreous loss, vitreous traction at incision sites, and intraocular lens dislocation, are thought to be the high-risk factors [15]. Ocular parameters may also associate with the development of PCME. In this study, the PCME group had a significantly lower incidence of PVD compared to the control group, which is consistent with previous research [14]. Tight adhesion of the posterior vitreous cortex to the neuroretina in the posterior pole results in a mechanical force exerted by vitreomacular traction [16], leading to alterations in the blood-retina barrier. Therefore, the presence of PVD acts as a protective factor against PCME, and preoperative assessment of the vitreoretinal interface using B-scan ultrasonography or high-resolution OCT is advisable. In addition, a small but statistically significant difference in axial length was detected between the PCME and the control group. Specifically, the PCME group exhibited shorter axial lengths, with 44.7% of eyes having axial lengths less than 23 mm. One possible explanation is that elongation of the globe causes stretching and thinning of vessels, which reduces blood flow and lowers capillary pressure in the macular area [17, 18], thus decreasing vascular dilation and fluid leakage.

Another factor influencing retinal blood flow is the blood pressure. In this study, although a history of hypertension was not the risk factor for PCME, subsequent analysis revealed a higher preoperative systolic blood pressure in the PCME group than controls (141.9 mmHg vs. 134.1 mmHg). Elevated systolic blood pressure may predispose to PCME by impairing autoregulation of retinal capillary blood flow, although there is no direct evidence supporting this notion. Furthermore, pre-existing fundus diseases, especially chorioretinal vasculopathy such as diabetic retinopathy and diabetic macular edema [19, 20], retinal vein occlusion [5], and uveitis [7], have been reported to correlate with PCME. It should be noted, however, that cataract surgery was thought to accelerate progression of these diseases [21], and distinguishing between PCME and macular edema secondary to these diseases remains a challenge. Therefore, we did not include patients with preoperative fundus diseases in the analysis.

In this study, we also found that patients in the PCME group had higher platelet counts than in the control group (261.3 × 10⁹/L vs. 222.5 × 10⁹/L), which was consistent with the findings of another study [22]. Platelets serve as the primary site for the production of platelet-derived growth factor (PDGF), an angiogenic factor that promotes cell proliferation and migration and plays a role in inflammation and tissue repair processes. Overexpression of PDGF can lead to increased vascular permeability and extravasation of vascular contents [23]. However, this is a bold speculation and prospective controlled studies are needed to elucidate the role of PDGF in the pathophysiology of PCME. We also analyzed other factors of interest that might be associated with the development of PCME, including the surgeon’s seniority, the patient’s education level, and some surgical parameters. It was reported that the incidence of PCME was higher among trainee surgeons compared to experienced surgeons, and increased with prolonged surgical duration [24]; however, these correlations were not found in this cohort. Additionally, we focused on intraoperative factors such as microscope illumination, phaco energy, and infusion pressure, which might affect the macular structure and microcirculation, but found no significant differences.

There is currently no standardized prevention or treatment protocols for PCME due to a lack of strong randomized clinical trials and comparative effectiveness studies. Topical NSAIDs are the mainstay of perioperative PCME prophylaxis [25]. Corticosteroids via topical administration, periocular injection and intravitreal injections [26], as well as anti-VEGF agents [27], carbonic anhydrase inhibitors [11], immunosuppressants [28], and surgical interventions [29], have been used in the management of PCME. Although these treatments were often beneficial, some patients with chronic PCME required repeated treatment procedures or were even unresponsive [30]. In this study, no significant difference in visual outcomes was found among the three treatment modalities: observation, topical NSAIDs, and intervention treatments. Nevertheless, intervention treatments could significantly shorten the recovery time compared to observation (28.6 days vs. 45.9 days). This suggests that conservative treatment, including observation or topical NSAIDs, can be considered in most cases as they are cost-effective and can prevent patients from undergoing the discomfort of a second operation, whereas for patients anxious about visual impairment or having a need for rapid recovery, intervention treatments such as intravitreal injections of corticosteroids or anti-VEGF agents can be considered. For prognosis, we found 23.4% (11/47) of patients did not achieve complete or partial recovery in BCVA and CRT during the average 8.26 months of follow-up. It was noted that intervention treatments did not increase the proportion of patients achieving visual recovery, even though the majority of them have acute PCME that remains untreated.

This study is limited by its retrospective design, which restrict the uniformity in the timing of examination and follow-up intervals, and the selection of treatment or no treatment was not randomized. Furthermore, the small sample size reduced the power of statistical analysis, especially in comparing the effects of risk factors and different treatment modalities. The potential roles of NSAIDs, steroids, anti-VEGF agents, and other drugs in the prevention and treatment of PCME require further exploration. Finally, the mechanism of PCME may involve changes in vascular permeability due to inflammation. Analyzing OCT angiography in future prospective studies could enhance understanding of the vascular changes and their impact on PCME.

Conclusions

PCME remains an encountered morbidity even in patients without pre-existing fundus pathologies. Postoperative macula assessment using OCT is necessary, especially in patients with risk factors like posterior vitreoretinal adhesion, shorter axial length, and higher systolic blood pressure. Although most cases of acute PCME spontaneously resolve, active intervention can shorten the recovery time and should be considered for those experiencing anxiety or requiring prompt visual recovery. Lastly, some cases may develop chronic PCME and visual impairment, which highlights the importance of preoperative assessment of risk factors and adequate patient counseling.