Risks After Cataract, Glaucoma Surgery Tied to Higher Hemoglobin A1c Levels

NEW ORLEANS — Higher preoperative hemoglobin A1c (A1c) levels significantly increased the risk of cystoid macular edema (CME) after cataract surgery, a large retrospective review showed.

The incidence of CME increased from 4.2% in patients with preoperative A1c <6.3% to 6.9% for patients with an A1c ≥8% before surgery. Preoperative A1c ≥8% conferred a greater risk of CME as compared with a diabetes diagnosis. A multivariable analysis identified preoperative diabetic retinopathy as the strongest predictor of CME, but A1c also was a significant predictor and additive to retinopathy.

“The incidence of cystoid macular edema in patients with diabetes was 5.3%, higher than the general population, but it grew to 6.9% in patients with a hemoglobin A1c above 8,” said Daniel Fraser, MD, of the University of Minnesota in Minneapolis, at the American Academy of Ophthalmology (AAO) annual meeting. “The rate just keeps going. Some of this has already been elucidated in the literature, but it was good to see that our results were matching that.”

“I would like to see an interventional study,” he added. “I want to see, in those patients in the higher quartiles of A1c, whether having an intervention, bringing them down to the lower quartiles, maybe under 6.3%, benefits them with a lower rate of cystoid macular edema after cataract surgery. Is there a sweet spot?”

Though statistically significant, the odds ratio for CME associated with higher A1c was “only” 1.06 per 1% increase, noted session moderator Jeff Pettey, MD, of the University of Utah Moran Eye Center in Salt Lake City. “I’m curious what you feel you could comfortably and confidently draw from the study in terms of clinical relevance. Is there a level at which you would alter treatment, delay surgery, et cetera.”

“That’s the biggest point from this study, deciding when we’re going to do more in-depth counseling or maybe more specific counseling with patients preoperatively and seeing that their risk levels are starting to jump with each 1% of hemoglobin A1c,” said Fraser. “Somebody with an A1c of 6% is way different than 10%, and it’s going to keep compounding with that variable. I think this gets the ball rolling for preoperative counseling.”

An unidentified member of the audience pointed to the PREMED study from Europe as a source of potential guidance on the issue. In that trial, patients with preoperative diabetic retinopathy, irrespective of severity, were randomized to no additional treatment or to subconjunctival triamcinolone acetonide alone or plus intravitreal bevacizumab (Avastin). No patients who received the corticosteroid developed CME, and the anti-VEGF therapy had no additive effect.

A separate study reported during a glaucoma session at AAO showed that higher preoperative A1c correlated with an increased risk of blebitis or endophthalmitis after tube shunt surgery. That almost doubled after 5 years for patients with an A1c >8.4% as compared with a normal preoperative A1c, reported Mohsen Adelpour, MD, of the University of Arkansas for Medical Sciences in Little Rock.

Macular Edema After Cataract Surgery

By way of introduction to the study, Fraser noted that pseudophakic CME is the most common vision-threatening complication of cataract surgery, occurring in up to 4% of surgically treated eyes. Several small, single-center studies have suggested an association between preoperative A1c and pseudophakic CME.

To examine the issue in a large population, investigators analyzed data from the VA Health System for the years 2010-2022. They identified 6,808 patients who had postoperative macular edema within 6 months of cataract surgery and compared them with 121,164 patients who did not develop CME after surgery.

Overall, pseudophakic CME occurred in 5.3% of the cataract surgeries. Patients who developed CME had a slightly higher preoperative A1c (7.6% vs 7.3%). After separating preoperative A1c levels into quartiles, Fraser and colleagues saw a direct correlation between A1c and CME incidence, increasing from 4.2% in patients with a preoperative A1c <6.3% to an incidence of 6.9% in patients with an A1c ≥8.0%.

A multivariable analysis showed that preoperative diabetic retinopathy tripled the odds ratio for postoperative CME (OR 3.28, 95% CI 3.10-3.47, P<0.001) and the odds increased with retinopathy severity. By comparison, each 1% increase in preoperative A1c increased the risk of CME by 6% (95% CI 1.05-1.08, P<0.001).

Fraser acknowledged the limitations inherent to a retrospective analysis, as well as use of VA data, which is heavily skewed toward older white males.

Eye Infections and A1c

In contrast to data on the association between A1c and CME, few studies have examined the impact of preoperative A1c on blebitis/endophthalmitis after shunt surgery for glaucoma, said Adelpour. To investigate the issue, his group analyzed data from the TriNetX clinical research network, identifying adults who had records of A1c prior to shunt surgery. They excluded patients with a history of endophthalmitis or active intraocular infection prior to surgery.

After propensity matching, data analysis included 4,182 patients who had preoperative A1c <7.0% (defined as normal), 1,166 patients with A1c values of 7.0% to 8.4%, and 777 patients with A1c levels >8.4% (defined as high). Comparison of the A1c-defined groups showed that patients with high preoperative A1c had a 5-year hazard for blebitis/endophthalmitis of 1.93 versus the group with normal A1c values (95% CI 1.04-3.58). The intermediate group did not have a significantly higher risk versus the normal-value group (HR 1.11, 95% CI 0.59-2.08).

A multivariable analysis showed that pre-existing myopia “massively” elevated the infection risk, separate from A1c levels (HR 3.58, 95% CI 1.57-8.15, P<0.01). Additionally, a higher comorbidity burden had an additive impact to poorly controlled diabetes (HR 2.35, 95% CI 1.14-4.36).

“Severe hyperglycemia serves as an independent predictor of long-term infection complications after tube shunt surgery, potentially due to impaired immune defenses and suboptimal wound healing,” said Adelpour. “Our myopia findings align with [previously reported observations] linking myopia to thinned scleral thickness profiles in filtering tracks, structurally predisposing eyes to tissue breakdown, exposure, and intraocular migration.”