Increased Levels of Inflammatory Immune Mediators in Vitreous From Pseudophakic Eyes

Abstract

Purpose.: To determine if pseudophakic eyes have an increased and sustained level of inflammatory immune mediators in the vitreous compared to phakic eyes.

Methods.: Vitreous fluid samples were obtained from 73 patients undergoing elective pars plana vitrectomy (PPV) as a result of a macular hole, epiretinal membrane, vitreous macular traction, or vitreous floaters. Forty eyes were pseudophakic and had previously undergone uncomplicated cataract surgery, ranging from a few months to several years prior to PPV. The vitreous samples were analyzed for 29 different inflammatory immune mediators using multiplex bead immunoassays.

Results.: A total of 14 cytokines (eotaxin, interferon-γ–induced protein-10 [IP-10], monocyte chemotactic protein-1 [MCP-1], macrophage derived chemokine [MDC], macrophage inflammatory protein [MIP]-1α, MIP-1β, thymus activation regulated chemokine [TARC], IL-12p40, IL-15, IL-16, IL-7, VEGF, IL-6, and IL-8) were detected in the vitreous of both study groups. Using multiple linear regression analysis, pseudophakia was significantly correlated with higher levels of vitreous immune mediators compared to phakia. Elevated vitreous levels were estimated to decrease over time for IL-6, IL-8, IL-15, IL-16, and VEGF, though they remained elevated for many months and even years compared to the levels detected in phakic eyes.

Conclusions.: This is the first study to demonstrate that cataract surgery and pseudophakia can induce increased vitreous levels of a substantial range of inflammatory immune mediators. The elevated levels seem to be maintained for a long period of time. These increased levels of cytokines may be involved in inflammatory processes leading to several complications to cataract surgery, both early and late.

Cataract surgery with implantation of an intraocular lens prosthesis (IOL) is one of the most common and successful medical procedures performed. Nevertheless, even if surgery is initially uncomplicated, some patients will experience future complications. In the short term there may be prolonged postoperative inflammation and/or cystoid macular edema (CME), whereas posterior capsular opacification (PCO), pseudophakic retinal detachment (RD), or late spontaneous dislocation of the IOL may occur in the long term. The detailed mechanism underlying these events is not fully understood. The question arises whether the pseudophakic condition itself is responsible for later unexpected pathological reactions. 

It is known that vitreous proteome alterations occur in the pseudophakic eye.¹ Due to the development of highly sensitive immunoassay techniques, it is possible to analyze a large number of immune mediators using extremely small quantities of vitreous fluid. A number of cytokines have been detected in vitreous of patients with various ocular diseases, such as diabetic retinopathy, retinal vein occlusion, and retinal detachment.^2,3 In several of these studies, vitreous of patients with less aggressive vitreoretinal disorders like macular hole or macular pucker has been used as control, revealing low but detectable levels of various cytokines. Recent studies of aqueous humor in pseudophakic glaucoma eyes demonstrated increased content of inflammatory mediators compared to phakic glaucoma eyes.⁴ The present study, using a multiplex bead immunoassay for a range of cytokines, chemokines, and proinflammatory factors, is the first to demonstrate prolonged elevation of inflammatory molecules in the posterior segment of pseudophakic eyes compared to phakic eyes. 

Patients and Methods

Study Population

A total of 73 patients were consecutively enrolled in this prospective observational cohort study. All patients were subjected to elective pars plana vitrectomy (PPV) because of macular hole (MH), epiretinal membrane (ERM), vitreous macular traction (VMT), or idiopathic vitreous floaters (VF). The patients were divided into two groups—phakic (n = 33) and pseudophakic (n = 40)—and the pseudophakic group was in turn divided into early pseudophakic (n = 17; cataract surgery performed within 6 months prior to PPV) and late pseudophakic (n = 23; cataract surgery performed 6 months or more prior to PPV). Exclusion criteria were previous complicated cataract surgery, for example, capsular rupture, previous intraocular surgery including intravitreal injections, history of uveitis, current vitreous hemorrhage, any degree of diabetic retinopathy (DRP), ongoing treatment with topical steroids or NSAID (nonsteroid anti-inflammatory drug eye drops, and age below 30 years. The Regional Ethics Committee in Gothenburg, Västra Götaland County, Sweden, approved the study. Informed consent was provided by the participating patients, and the procedures were in compliance with the Declaration of Helsinki. 

Collection of Human Vitreous Samples

Samples were collected during the period from November 1, 2013, to June 30, 2014, at the Department of Ophthalmology, Sahlgrenska University Hospital, Mölndal, Sweden. All surgeries were performed using small-gauge (23 or 25 gauge) nonsutured incisions, and undiluted vitreous fluid (0.2–0.3 mL) was obtained by a vitreous cutter (Alcon, Fort Worth, TX, USA) from the central vitreous compartment. Samples were collected in an exchangeable tube connected to the aspiration line of the vitrectomy machine (Alcon) and immediately transferred to cool storage and long-term storage at −80°C until final simultaneous analysis of all samples. 

Biochemical Analyses

Vitreous samples were studied using a multiplex immunoassay system (MSD Multi-Spot Assay System; Meso Scale Discovery, Gaithersburg, MD, USA). In total, 29 different inflammatory mediators were analyzed for each vitreous sample. The Chemokine Panel 1 (human) kit was used to measure eotaxin, eotaxin-3, IL-8, interferon-γ–induced protein-10 (IP-10), monocyte chemotactic protein-1 (MCP-1), MCP-4, macrophage derived chemokine (MDC), macrophage inflammatory protein [MIP]–1α, MIP-1β, and thymus activation regulated chemokine (TARC). Cytokine Panel 1 (human) was used to measure granulocyte-macrophage colony–stimulating factor (GM-CSF) IL-1α, IL-12/IL-23p40, IL-15, IL-16, IL-17A, IL-5, IL-7, TNF-β, and VEGF. Proinflammatory Panel 1 (human) was used to measure IFN-γ, IL-1β, IL-10, IL-12p70, IL-13, IL-2, IL-4, IL-6, IL-8, and TNF-α. 

Cryopreserved vitreous samples as well as all kit components were brought to room temperature before analysis. Reverse pipetting was applied in all pipetting steps to avoid bubbles. Samples were spun at 2000g for 20 minutes to remove cell debris and aggregates and diluted 2-fold in sample diluent. The analyses were performed according to the instructions from the manufacturer (Meso Scale Discovery). Briefly, duplicates of diluted calibrator and samples were loaded on the plate and the plate was incubated on a shaker for 2 hours. After washing, labeled detection antibodies were pipetted in the wells and the plate was incubated for another 2 hours. After incubation, the plate was washed and read buffer was added to the plate just before reading in the MESO QuickPlex SQ 120 instrument using the MSD DISCOVERY WORKBENCH analysis software. Curve fitting was done with the same software using a four-parameter logistic model with a 1/Y² weighting according to the manufacturer's instructions, and concentrations were determined from the standard curves. The lower limit of detection (LLOD) was the calculated concentration corresponding to the average signal 2.5 standard deviations above the background (zero calibrator) for each analysis. All measurements were performed using one batch of reagents by board-certified laboratory technicians who were blinded to clinical data. Only inflammatory mediators for which at least 90% of the duplicate samples were in detection range were included in the statistical evaluation. A total of 14 out of 29 analyzed factors fulfilled this criterion. A list of the inflammatory immune mediators that were in detection range and those that were not is presented in Table 1. 

Table 1

 

Studied Inflammatory Immune Mediators

 

Statistical Analyses

For some immune mediators, a small number of samples were detectable but below curve fit of the standards. In order to avoid bias by excluding those patients, these samples were set to the lowest value that was in detection range for that specific immune modulator. As a consequence, nonparametric measures (median and interquartile range, IQR) and statistical analyses (Mann-Whitney U test for independent groups and Spearman's rank correlation) were used to analyze differences in cytokine levels with age or between patient groups. The difference in age and sex distribution between the phakic and pseudophakic groups was calculated with Fisher's exact test and Student's t-test. A P value < 0.05 was considered statistically significant. A multiple linear regression analysis was performed with immune mediators as dependent variable after logarithmation of data to compensate for skewed distribution. Some covariates, that is, age, sex, phakia/pseudophakia, and diagnosis (ERM and MH only), were entered in a stepwise manner, including those with a P value < 0.150. Independent variables with a P value > 0.05 were excluded from the final model. Statistical analysis was performed using SPSS for Macintosh software (version 22.0; IBM SPSS Statistics, Armonk, NY, USA). For multiple linear regression analysis, SAS, version 9.2 (SAS Institute, Inc., Cary, NC, USA), was used. 

Results

Demographic analysis (Table 2) showed no difference in age or sex between pseudophakic and phakic patients. Vitreous levels for the detected inflammatory immune mediators in pseudophakic and phakic eyes are summarized in Table 3. Pseudophakic eyes exhibited significantly higher levels of immune mediators in the vitreous except for VEGF (Mann-Whitney U test). The median level of different immune mediators varied from less than 2 pg/mL to more than 700 pg/mL. Highest levels were found for MCP-1, IP-10, and MDC. These variations were similar with a parallel distribution in both groups. The increase in median vitreous levels between phakic and pseudophakic eyes ranged from 20% to 165%, median value 39% (VEGF excluded). 

Table 2

 

Patient Characteristics

 

Table 3

 

Vitreous Levels for the Main Detected Immune Mediators Comparing Pseudophakic and Phakic Eyes

 

For the vast majority of immune mediators, no significant correlation was found with age, either for the whole study group or when phakic and pseudophakic eyes were analyzed separately (Table 4) using Spearman's rank correlation. 

Table 4

 

Correlation of Age and Vitreous Levels of Immune Mediators for Pseudophakic and Phakic Eyes

 

Correlation of vitreous levels in pseudophakic patients with short IOL duration (<6 months) compared to longer duration (≥6 months) showed no significant difference, whereas both pseudophakic subgroups had significantly higher levels compared with phakic eyes (Table 5). 

Table 5

 

Correlation (P Values) Between Different Subgroups of IOL Duration or Phakia and Vitreous Levels of Immune Mediators

 

The duration of pseudophakia and the vitreous level for each substance were analyzed using Spearman's rank correlation test. Four immune mediators (Fig. 1) had a statistically significant trend of decreasing concentration over time: IL-6 (R = −0.518, P = 0.001), IL-8 (R = −0.392, P = 0.01), IL-15 (R = −0.487, P = 0.002), and IL-16 (R = −0.355, P = 0.026). The other inflammatory immune mediators showed no statistically significant equalization over time to the levels of phakic eyes, although the correlation coefficient was negative in all cases, indicating a declining trend. The increased levels of immune mediators in pseudophakic eyes compared to phakic eyes were maintained for a long period of time, and a representation of these immune mediators is shown in Figure 2. 

Figure 1

 

Correlation of vitreous levels of IL-6, IL-8, IL-15, and IL-16 and duration of pseudophakia. All immune mediators exhibiting a significant decrease over time are included in the figure. Each dot represents a single pseudophakic eye. x-axis: Duration of pseudophakia in months, log.scale. y-axis: pg/mL, log.scale. Horizontal line: Median level (pg/mL) for all phakic cases. Dashed line: Linear relation between declining level of the detected immune mediator and increasing duration of pseudophakia. Spearman's rank correlation (R) and P values calculated for each immune mediator: IL-6: R = −0.518, P = 0.001; IL-8: R = −0.392, P = 0.01; IL-15: R = −0.487, P = 0.002; IL-16: R = −0.355, P = 0.026.

Figure 2

 

Correlation of vitreous levels of eotaxin, IP-10, MCP-1, MDC, MIP-1α, and MIP-1β and duration of pseudophakia. Six representative immune mediators with sustained levels over time are shown (VEGF, TARC, IL-12p40, and IL-7 are not included in the figure). Each dot represents a single pseudophakic eye. x-axis: Duration of pseudophakia in months, log.scale. y-axis: pg/mL, log.scale. Horizontal line: Median level (pg/mL) for all phakic cases. Dashed line: Linear relation between declining level of the detected immune mediator and increasing duration of pseudophakia. Spearman's rank correlation (R) and P values calculated for each immune mediator: eotaxin: R = −0.163, P = 0.314; IP-10: R = −0.053, P = 0.745; MCP-1: R = −0.189, P = 0.242; MDC: R = −0.240, P = 0.136; MIP-1α: R = −0.074, P = 0.652; MIP-1β: R = −0.020, P = 0.902.

Using multiple linear regression analysis with age, sex, phakia/pseudophakia, and diagnosis (only patients with ERM or MH were included) as covariates, the increased levels of immune mediators in pseudophakic eyes were confirmed (Table 6). All immune mediators except VEGF were significantly elevated in the vitreous of pseudophakic as opposed to phakic eyes. In the majority of immune mediators, pseudophakia was the only covariate that showed significant correlation with concentration. However, for MCP-1 and IL-7, there were significant correlations with age (lower and higher age, respectively); and for IL-16, IL-7, IL-6, and IL-8, higher concentrations of immune mediators were significantly associated with diagnosis of ERM as compared to MH. There was no correlation between sex and various cytokine levels. 

Table 6

 

Multiple Linear Regression Analysis of Immune Mediators in Vitreous of Phakic and Pseudophakic Eyes

 

Discussion

The vitreous of the eye is not only a transparent gel composed of hyaluronic acid, collagen, and water. It also contains a substantial variety of bioactive molecules, and most of these reflect physiological or pathological activities in the ocular tissues.⁵ The aim of this study was to examine whether cataract surgery and pseudophakia can induce elevated levels of vitreous inflammatory immune mediators and if so, whether the increased levels are sustained. 

In 1992, Aiello et al.⁶ measured VEGF in ocular fluids using an enzyme-linked immunosorbent assay (ELISA) technique and established that VEGF is a major contributor to the neovascularization in retinal ischemic eyes; and other studies confirmed the presence of cytokines in the vitreous of patients with various ocular diseases.^7–9 The limitation of using the conventional ELISA technique is that only a minor part of all immune mediators can be analyzed per sample. The development of highly sensitive techniques with multiplex bead immunoassays¹⁰ has enabled detection of a substantial number of immune mediators, and several studies have revealed a large variety of bioactive molecules present in fluid of eyes with various ocular disorders.^2,3,11,12 

Uncomplicated pseudophakia is usually not considered a pathological condition. However, in a study from 2005, Neal et al.¹ compared the vitreous humor proteome and viscosity in phakic and pseudophakic donor eyes. The protein composition was altered in the pseudophakic eyes, and there was also a reversed profile of the vitreous viscosity. 

Pseudophakia has been shown by Inoue et al.⁴ to induce an increased level of cytokine MCP-1 in the aqueous humor of glaucomatous pseudophakic eyes compared to phakic glaucoma eyes. The elevated cytokine level that was revealed was sustained, and the clinical implication suspected was a lower success rate in pseudophakic eyes following filtering surgery due to increased risk of tissue fibrosis. Findings revealed a low but significant inflammatory activity in the anterior segment of glaucomatous pseudophakic eyes. 

During cataract surgery with phacoemulsification, the barrier between the anterior and posterior segment is affected by intraocular pressure fluctuations and immense fluid perfusion through the anterior chamber. This can cause cortical debris to leak into the vitreous even without capsular damage, which, together with mechanical forces strong enough to arouse ocular globe deformation, may lead to vitreous destabilization and inflammatory reactions in the vitreoretinal tissues.¹³ Our study demonstrates a significant increase in various immune mediators in the vitreous, which was maintained for an extensive period of time. These bioactive molecules may be primarily due to the inflammation caused by the surgery. It is known that the surgical trauma in small-incision cataract surgery causes prolonged blood–aqueous barrier dysregulation, which can be detected in the anterior chamber using a laser flare cell meter,¹⁴ and this can contribute to elevated intraocular levels of immune mediators. Because the vitreous is capable of eliminating potentially damaging substances through a protein turnover mechanism,¹⁵ sustained levels of cytokines are probably the result of continuous production by immunoactive cells in the eye itself. One possible source of cytokines to the vitreous is continuous proliferation of lens epithelial cells, especially when the cells involve the posterior lens capsule.¹⁶ The vitreoretinal compartment is an immune-privileged environment, separated by the blood–retina barrier from circulating blood cells; hence the retinal conditions in the studied population may themselves be preceded by inflammatory mechanisms such as ERM formation.¹⁷ Nonetheless, the difference in cytokine levels remained when other diagnoses like MH or VMT were excluded. Multiple regression analysis revealed that pseudophakia was the strongest predictor for increased vitreous levels for all immune mediators except VEGF, while diagnosis (ERM versus MH) was a predictor in only four out of 14 increased immune mediator levels. The levels of VEGF were equally low in phakic and pseudophakic eyes, precluding angiogenesis or increased vascular permeability as an explanation for the increased vitreous levels of other cytokines.¹⁸ Another study also concluded that soluble factors detected in vitreous are mainly recruited from ocular tissues.² 

In a recent study, Bromberg-White et al.³ detected MCP-1, IP-10, and MDC in the vitreous of both study patients and controls when analyzing cytokines in vitreous of patients with proliferative DRP. In our study, these three immune mediators—MCP-1, IP-10, and MDC—showed the same pattern with relatively high vitreous levels. Monocyte chemotactic protein-1 is expressed by endothelial and inflammatory cells and is upregulated after tissue injury leading to recruitment of inflammatory cells. Interferon-γ–induced protein-10 has both antiangiogenic and anti-inflammatory properties,¹⁹ and MDC exert chemotactic activity. However, cytokines, chemokines, and proinflammatory factors often have pleiotropic properties. Exactly how the interactive mechanisms are mediated in the immune response in the eye is poorly understood. 

Although there was a declining trend over time, the elevated levels of immune mediators were maintained for several months and even years. In the study by Inoue et al.⁴ on aqueous humor from pseudophakic glaucoma eyes, elevated levels of MCP-1 and IL-8 compared to phakic glaucoma eyes were found in cases with a mean duration of pseudophakia of over 7 years. In our study, cytokine levels eventually decreased to the levels obtained with phakic eyes (Fig. 1), and this effect lasted an extended period of time. However, for the majority of inflammatory immune mediators, no significant equalization to the levels of phakic eyes could be detected with Spearman's rank correlation test. This implies that the sustained elevated levels of immune mediators are related to the pseudophakic status itself, and not only as residuals of an inflammatory reaction induced by the previous cataract surgical procedure. 

Most patients in our study were subjected to PPV due to ERM, and this condition can theoretically cause higher vitreous levels of immune molecules due to the presence of posterior vitreous detachment and increased mobility of fluids in the vitreoretinal compartment compared, for example, to MH, where the vitreous is often still attached. Using multiple linear regression analysis, the cytokine concentrations of IL-16, IL-7, IL-6, and IL-8 were significantly higher in vitreous samples from patients with ERM compared to MH. Another study found no correlation between cytokine levels in vitreous samples from ERM and MH, respectively.³ 

Limitations of this study are the lack of data concerning the type of IOL implanted and also the lack of information relating to previous Nd:YAG laser capsulotomy. However, we believe that the majority of the pseudophakic eyes had sharp-edged hydrophobic acrylic IOLs, which are the most common type of IOL used in Sweden. Moreover, the cataract surgery was performed less than 24 months prior to PPV in the majority (75%) of cases, which suggests a low PCO prevalence.²⁰ 

The clinical relevance for sustained elevated vitreous levels of inflammatory immune mediators in pseudophakia is yet unclear. In the majority of pseudophakic cases this probably has no clinical implication. However, for those patients who develop complications from cataract surgery, for example, CME, postoperative chronic uveitis, PCO, progression of DRP, pseudophakic RD, or late spontaneous IOL dislocation, a low but sustained increase in immunological activity due to cataract surgery and pseudophakia may play a crucial role in the pathophysiological processes leading to these conditions. Inflammatory reactions are the main cause of both CME and postoperative uveitis and also for progression of DRP.³ Pseudophakic RD develops due to vitreous collapse with formation of retinal tears,²¹ and it has been suggested that cytokines are involved in the pathogenesis.²² Late IOL dislocation can occur when the zonular fibers are gradually weakened. Both these latter conditions are preceded by normal aging processes, which can be accelerated by altered immunological vitreous activity. A low grade of increased inflammatory reaction can be described as parainflammation and might contribute to age-related pathological alterations in the eye.^23,24 

Cystoid macular edema is mediated by an inflammatory postoperative reaction following cataract surgery. It is clinically evident only in 2% of patients experiencing impaired visual acuity. If the diagnosis of CME is based on other methods like fluorescein angiography, the prevalence of CME can be established in up to 20% of cases,^25,26 implying a more frequent postoperative intraocular inflammation also extending to the posterior segment. In a recent systematic review of clinical trials for treatment of pseudophakic CME, the authors concluded that treatment with topical NSAID is effective in preventing inflammation and reducing the incidence of CME.²⁷ We believe our data support the need for effective and perhaps prolonged treatment with local anti-inflammatory topical drugs in order to control postoperative intraocular inflammation. 

In conclusion, the present study is the first to demonstrate increased levels of inflammatory immune mediators in the vitreous of pseudophakic eyes. Even if this does not imply a clinical, chronic intraocular inflammation, it indicates an amplified immunological response sustained for an extended time with a potential to influence pathophysiological processes involved in postoperative complications following cataract surgery. Although several studies indicate the central importance of inflammatory components, the significance of increased levels of immune mediators to influence the mechanisms for these processes is not known in detail. However, in view of these findings, the belief that the difference between a phakic and a pseudophakic eye is merely an exchange of the lens properties must be dismissed as a myth. 

Acknowledgments

Supported by grants from the Swedish Research Council (#2011-3132), Swedish government (“Agreement concerning research and education of doctors”; ALF-GBG-145921), Göteborg Medical Society, Marianne and Marcus Wallenberg Foundation, Dr Reinhard Marcuses Foundation, Konung Gustaf V:s och Drottning Victorias Frimurarestiftelse, Hjalmar Svensson Foundation, Greta Andersson Foundation, Herman Svensson Foundation, Ögonfonden, De Blindas Vänner and Kronprinsessan Margaretas Arbetsnämnd för Synskadade. 

None of the authors has any financial interest in any of the techniques or devices described. 

Disclosure: G. Jakobsson, None; K. Sundelin, None; H. Zetterberg, None; M. Zetterberg, None