FormalPara Key Summary Points

Why carry out this study?

It is unclear if patients with non-ocular autoimmune disease are at higher risk for postoperative complications or persistent anterior uveitis (PAU) following phacoemulsification cataract surgery.

This study reviewed charts and evaluated the outcomes of patients with and without non-ocular autoimmune disease who underwent phacoemulsification cataract surgery.

We hypothesized that patients with non-ocular autoimmune disease would not be at higher risk for postoperative complications or PAU compared to patients without autoimmune disease.

What was learned from the study?

Patients with non-ocular autoimmune disease, regardless of immunosuppression status, were not at higher risk for postoperative complications or PAU following phacoemulsification cataract surgery.

These patients do not appear to require additional preoperative counseling or a modified perioperative medication regimen.

Introduction

Cataract extraction is one of the most common surgical procedures performed in the USA and around the world. Cataract surgery with modern phacoemulsification techniques is highly successful in the general population, with relatively high rates of improved postoperative best-corrected visual acuity (BCVA) [1], and low rates of complications such as intraocular pressure (IOP) elevation and persistent anterior uveitis (PAU) [2]. However, patients with pre-existing ocular inflammation or uveitis have higher rates of complications following cataract surgery [3]. Therefore, patients with uveitis are generally recommended to have ocular inflammation controlled for at least 1–3 months prior to surgery [4], and are often given additional perioperative corticosteroids to reduce the risk of chronic or recurrent inflammation following surgery [5,6,7].

For patients with autoimmune disease without ocular involvement, however, there is no existing data to guide clinicians on the potential need for additional preoperative counseling of higher complication rates, or for modification of perioperative medication regimens. In this study, we evaluate the outcomes of this patient population following phacoemulsification cataract extraction surgery.

Methods

A retrospective cohort study was done via records from the Cataracts Outcome Registry at the University of Colorado School of Medicine Department of Ophthalmology. The study received approval from the Colorado Multiple Institutional Review Board and all research conformed to the tenets of the Helsinki Declaration of 1964 and its later amendments. The registry was used to identify the patients included in this study. Professional research assistants, trained specifically in the abstraction of information related to cataract surgery, review the information from medical charts of patients who undergo cataract surgery at our institution and enter data into REDCap (Research Electronic Data Capture), a secure, web-based application. The registry includes data on demographic information, medical history, preoperative medication history, intraoperative and postoperative complications, and preoperative and postoperative eye examinations.

This study included patients who underwent phacoemulsification cataract surgery with intraocular lens implantation between January 1, 2014 and December 31, 2019 at our institution. Exclusion criteria were history of ocular inflammation or cataract surgery combined with another intraocular surgery. Charts of patients who were noted in the registry to have a history of autoimmune disease were further evaluated. Patients were only included as having autoimmune disease if the diagnosis was confirmed by a specialist at the University of Colorado Hospital. Patients who were identified as having autoimmune disease in the registry but were not confirmed by a specialist were excluded from both the autoimmune and control groups. The following autoimmune diseases were characterized as systemic: rheumatoid arthritis, multiple sclerosis, systemic lupus erythematosus, inflammatory bowel disease, psoriasis, Sjögren disease, polymyalgia rheumatica, giant cell arteritis, sarcoidosis, ANCA-associated vasculitis, systemic sclerosis, scleroderma, dermatomyositis, polymyositis, ankylosing spondylitis, reactive arthritis, and Behçet’s disease; whereas the following were characterized as organ-specific: myasthenia gravis, Hashimoto thyroiditis, autoimmune hepatitis, Graves’ disease, autoimmune encephalitis, and pemphigoid.

Cataract surgeries were performed using phacoemulsification with clear corneal incisions. Brand moxifloxacin was either injected intracamerally at the end of surgery or placed topically via a soaked collagen shield. Following surgery, patients were typically prescribed a topical postoperative eyedrop regimen consisting of a fluoroquinolone antibiotic, non-steroidal anti-inflammatory drug, and corticosteroid. This regimen did not differ between patients with or without a history of autoimmune disease. The antibiotic was typically discontinued at the week 1 postoperative appointment. Depending on surgeon preference, the steroid was either prescribed as one drop two times daily for 2 weeks (using brand difluprednate 0.05%) or four times daily for 1 week followed by a weekly taper for 3 weeks (using generic prednisolone acetate 1%). However, if there was persistent intraocular inflammation, topical steroids were continued after the month 1 postoperative visit.

The following data were collected: age, self-reported race/ethnicity, sex, laterality of surgery, identification as complex surgery, length of surgery, cumulative dissipated energy (CDE) in seconds used during phacoemulsification, as well as measures of BCVA in logMAR, IOP in mmHg, grading of anterior chamber (AC) cell, presence of cystoid macular edema (CME) on macular optical coherence tomography (OCT), and steroid usage at 3, 6, and 12 months postoperatively. PAU was defined as at least 0.5 + anterior chamber cell and/or continuation of corticosteroid therapy at or beyond 3 months postoperatively with no other etiology found other than postoperative state. If no specific diagnosis was indicated from patient history, patients with presumed PAU generally underwent serological testing for syphilis and a chest radiograph. Patients with autoimmune disease were compared to patients without autoimmune disease during the study time period. However, the control group for PAU analysis was obtained from our prior paper and included eyes from 2014 to 2016 [2]. Postoperative IOP spike was defined as IOP greater than 30 mmHg or 10 points higher than preoperative IOP at the postoperative day 1 visit. Systemic immunomodulatory therapy (IMT) and corticosteroid use and dosage at the time of cataract surgery were also recorded among patients with autoimmune disease. Patients were considered to be on immunosuppression if they were being treated with IMT or at least 7.5 mg of oral prednisone-equivalent daily.

Statistical Analysis

Basic frequencies and percentages were calculated for categorical variables to compare patient eyes with autoimmune disease and those without. Continuous variables were summarized with mean and standard deviation for patient age, and means, standard deviations and medians for IOP, BCVA, and CDE measures. The two groups were compared with logistic regression modeling with generalized estimating equations utilizing an unstructured correlation to account for many patients having two eyes included in the study. Patient eyes on IMT were also compared to no IMT with the same statistical methods.

Results

There were 762 eyes from 524 patients identified as having an autoimmune disease without ocular involvement within the Cataract Outcome Registry for the specified time period. After further chart review of these 524 patients, 422 eyes from 248 patients were found to have a confirmed autoimmune diagnosis by a specialist at the University of Colorado Hospital. The most common autoimmune diseases included rheumatoid arthritis (148 eyes), multiple sclerosis (52 eyes), systemic lupus erythematosus (39 eyes), and inflammatory bowel disease (36 eyes) (Table 1). There were 10,201 eyes among patients without a history of autoimmune disease or ocular inflammation included within the control group. For PAU specifically, there were 3013 eyes included in the control group.

Table 1 Autoimmune disease entities
Full size table

The autoimmune group was more likely to be female (79.6% versus 56.4%) and the two groups were similar with regards to race/ethnicity (Table 2). The autoimmune group and control group had no statistically significant differences with regard to preoperative IOP, surgery length, postoperative BCVA, rates of postoperative day 1 IOP spike, postoperative IOP, and PAU, whereas the control group had worse preoperative BCVA and required higher CDE during surgery (Table 3).

Table 2 Baseline demographics for autoimmune patients and controls
Full size table
Table 3 Clinical outcomes for all autoimmune cases and control groups
Full size table

Within the autoimmune group, 378 of the 422 eyes were included in the systemic autoimmune group, whereas 44 eyes were defined as having organ-specific autoimmune disease. Comparison of the systemic autoimmune group and control group had similar findings as the overall autoimmune and control groups, with no statistically significant differences with regard to preoperative IOP, surgery length, postoperative BCVA, rates of postoperative day 1 IOP spike, postoperative IOP, postoperative IOP, and PAU. Again, the control group had worse preoperative BCVA and required higher CDE during surgery (Table 4).

Table 4 Clinical outcomes for systemic autoimmune cases and control groups
Full size table

Patients within the systemic autoimmune group were further stratified into those who were and were not on systemic IMT at the time of cataract surgery. There were 253 eyes included in the immunosuppressed group and 124 eyes who were not immunosuppressed. The non-immunosuppressed group had more African-Americans compared to the immunosuppressed group (p = 0.013). There were no statistically significant differences between these two groups with regards to preoperative IOP and BCVA, surgery length, CDE, postoperative BCVA and IOP, and rates of postoperative day 1 IOP spike. The non-immunosuppressed group did have higher rates of PAU (5.6% versus 1.6%), but this did not reach statistical significance (p = 0.065) (Table 5).

Table 5 Clinical outcomes of autoimmune patient eyes by immunosuppression status for patients with systemic autoimmune disease
Full size table

Discussion

Whereas there is a known higher risk of complications and persistent inflammation following cataract surgery in patients with uveitis or ocular inflammation, necessitating additional preoperative counseling and modified perioperative medications [3], there is no similar data or guidance for patients with non-ocular autoimmune disease undergoing cataract surgery. In this study, patients with confirmed non-ocular autoimmune disease, diagnosed by specialists at our hospital, were not at higher risk for postoperative complications or PAU compared to patients without autoimmune disease. Importantly, the autoimmune and control groups both had similar percentages of African-American patients. Previous studies from our group [2] and others [8] have found that patients who self-identified as African-American have higher rates of PAU following cataract surgery compared to Caucasians.

In further analysis, patients with organ-specific autoimmune disease, such as myasthenia gravis and Hashimoto thyroiditis, which tend not to cause inflammation outside the relevant anatomical area or organ system, were excluded and only patients with systemic autoimmune disease, such as rheumatoid arthritis and systemic lupus erythematosus, were compared to the control group and, again, not found to be at higher risk for postoperative complications. Patients with systemic autoimmune disease were then stratified into an immunosuppressed group, who were on systemic IMT or at least 7.5 mg of oral prednisone-equivalent daily at the time of cataract surgery, and non-immunosuppressed group. Patients who were not immunosuppressed were also not more likely to have postoperative complications compared to the immunosuppressed group, although the non-immunosuppressed group was found to have an increased likelihood of PAU that trended towards statistical significance. This result, however, may have been confounded by the higher number of African-American patients within the non-immunosuppressed group. Taken together, these findings suggest that patients with non-ocular autoimmune disease, regardless of immunosuppression state, do not appear to require additional preoperative counseling or a modified perioperative corticosteroid regimen.

Other findings from this study include that patients with autoimmune disease in this cohort were more likely to be female than patients in the control group, which is consistent with the known demographics of autoimmune disease [9]. The autoimmune group also required less CDE during surgery than the control group. This could be related to increased systemic corticosteroid exposure in the autoimmune group, leading to higher rates of posterior subcapsular versus nuclear sclerosis cataract.

The primary limitation of this study is its retrospective nature. However, this study design allowed for the large sample sizes of both autoimmune disease and control patients. We also attempted to minimize the misclassification of patients by only including patients as having an autoimmune disease if the diagnosis was confirmed by a relevant specialist at our hospital. This also allowed us to confirm systemic medication use at the time of surgery.

Conclusion

Patients with non-ocular autoimmune disease do not appear to be at higher risk for postoperative complications, including worse BCVA or increased rates of IOP elevation and PAU, following phacoemulsification cataract surgery. These patients likely do not require modification of the typical perioperative medication regimen, unlike patients with ocular autoimmune disease.