November 2012




Complicated cataract cases

Cataract surgery and diabetes

by Thomas A. Oetting, M.D.


Kevin Miller, M.D.

Diabetes mellitus is one of the most common medical conditions in the western world. A Saturday Night Live skit on the iPhone 5 jokingly suggested that diabetes is the United States' number one export ( Saturday-night-live/ video/tech-talk-iphone-5/1420759).

Primary concerns with patients who have diabetes mellitus include the development of post-op cystoid macular edema, worsening diabetic macular edema, progression to proliferative diabetic retinopathy, and the development of rubeosis iridis.

Patients with well-controlled type 2 diabetes mellitus can often be treated as normal individuals.

Those with retinopathy and macular edema require a thoughtful approach. Whenever possible, the retinopathy and macular edema should be treated before cataract surgery. This is not always possible, however.

Poor vision may lead to poor diabetic control because patients cannot read their glucose monitors or measure out the correct amount of insulin for injection. In such patients, it may be necessary to remove visually significant cataracts before their blood sugars can be brought under control.

There are occasional patients who have such dense cataracts that the extent of their diabetic retinopathy cannot be ascertained pre-op. The cataract surgeon must be prepared to inject a VEGF inhibitor or perform panretinal laser photocoagulation through an indirect ophthalmoscope on the operating table or shortly thereafter.

Surgeons, anesthesiologists, and hospital or ambulatory surgery centers often have varying recommendations for diabetic patients on the morning of surgery. At the Jules Stein Eye Institute we instruct diabetics to withhold their oral hypoglycemics and insulin after midnight. We do not want their blood sugars to bottom out in the morning hours. We check them on arrival to the pre-op suite and administer short-acting insulin if needed.

In this month's column, Thomas A. Oetting, M.D., discusses his approach to patients with diabetes mellitus. He reviews the use of nonsteroidal anti-inflammatory agents, intraocular lens selection, and pre-op consultation with a retina specialist. Diabetes mellitus is a common medical condition, and all ophthalmologists should be comfortable managing diabetic patients pre-op, intraoperatively, and through the post-op recovery period.

Kevin Miller, M.D., Complicated cataract cases editor

Patient with early cataract

Figure 1. Patient with early cataract following vitrectomy with lens following at the start of phacoemulsification through pre-existing tear in the posterior capsule

Silicone oil bubbles Figure 2. Silicone oil bubbles in anterior chamber following uncomplicated phacoemulsification in patient with history of complex retinal detachment repair Source (all): Thomas Oetting, M.D.

NSAIDs treated table Table 1

Ocular coherence tomography (OCT) and vascular endothelial growth factor (VEGF) inhibition have changed the management of cataract surgery in our patients with diabetes mellitus (DM). Patients with DM cannot be ignored. The National Eye Institute estimates that over 26% of our patients over 65 have DM.1 Most of these patients with DM are treated like all of our other cataract patients; however, some patients with poor control or retinopathy will need extra attention before, during, and after surgery.2

Prevention of cystoid macula edema (CME) and limiting the acceleration of diabetic macular edema (DME) following cataract surgery are the most important issues. While the use of nonsteroidal anti-inflammatory drugs (NSAIDs) in routine patients is controversial, most agree that NSAIDs are important in high risk patients, such as those with DM. I typically use a nonsteroidal for a week prior to surgery. I usually stop the NSAIDs for a week following surgery while the epithelium heals and then start them back up. Guidance for the duration of treatment is not clear. I usually use NSAIDs for a month for lower risk patients (e.g., DM with no retinopathy) and for 2-3 months for higher risk patients (e.g., DM with pre-op macular edema). Patients with pre-op retinal edema should be considered for consultation with a retina specialist for possible injection of VEGF inhibitors or intravitreal or sub-Tenon's corticosteroid injection.3-8 OCT is critical to the perioperative management of patients with DM as a sensitive test for the development of CME and progression of DME. IOL selection is also an important issue for patients with DM.2 Patients with macular edema or those at high risk for macular edema are not good candidates for multifocal IOLs, which can limit contrast sensitivity. I think that multifocal IOLs are reasonable in type 2 DM patients with great control of their blood sugars and no existing retinopathy. However, I personally avoid multifocal IOLs in type 2 DM patients with poor control or all patients with type 1 DM even if they have no retinopathy as they are at increased risk for retinopathy in the future. Vitreous surgeons can be bothered by silicone IOLs. Silicone IOLs can cloud with air fluid exchange and really cloud with silicone oil, which adheres to the silicone IOL. In deference to our retinal surgeons, I tend to use acrylic IOLs in patients with DM who are at risk for vitrectomy in the future. While I typically will use a single-piece acrylic lens (SPA) for DM patients, I often use a large optic three-piece monofocal acrylic IOL for patients at very high risk for vitrectomy (e.g., neovascular disease). Patients with type 2 DM with excellent control and no pre-op retinopathy are not likely to need vitrectomy in the future. In these patients, it seems reasonable to use a silicone IOL, however I still typically use an acrylic IOL just in case they progress in the future. I think that multifocal IOLs are also reasonable in this set of patients. Even this set of patients with type 2 DM and excellent control of their blood sugars are at increased risk for cystoid macular edema post-op and should be treated with pre- and post-op NSAIDs (Table 1).

Patients with type 1 DM with good control and no retinopathy are similar in many ways to the type 2 DM. However, one might assume that over time they would be more likely to develop retinopathy, so I would most likely avoid multifocal IOLs and would not use a silicone IOL in these patients (Table 1). Patients with poor control of their blood sugars, whether type 1 or type 2, present some difficult issues. One strategy would be to withhold the surgery until they had developed a strategy with their endocrinologist for better control. However, sometimes this is not possible as better vision is necessary for the management of their diabetes. In these patients, I would consider a monofocal acrylic IOL as their poor control makes macular edema and future vitrectomy more likely (Table 1).

In patients with poor control who have background diabetic retinopathy, I am more concerned about the possibility of worsening retinopathy following the surgery and, specifically, of diabetic macular edema. Typically for these patients, pre-op I will do an OCT to rule out subtle diabetic macular edema and if this is present, will refer them to a retina colleague for possible treatment. I will also use a monofocal acrylic lens in these patients, as macular function will be an issue in the future and they are at risk for pars plana vitrectomy. I use the usual regimen with a nonsteroidal pre-op with a break around the time of surgery for a week and then for 1-2 months following surgery. In patients with poor control who have existing diabetic macular edema, it is important to consult with your retina colleagues for pretreatment in these patients. The use of OCT and possibly fluorescein angiography is indicated in these patients to assess the level of diabetic macular edema. Focal laser and more recently VEGF inhibitors in this set of patients prior to surgery can limit the possibility of worsening macular edema. I use my usual NSAID regimen starting 1 week prior to surgery, 1 week rest around the immediate post-op period, and then continue the NSAIDs for 1-2 months following surgery. These patients should be followed with your retina colleagues after the surgery, typically at 1 month after the surgery. Patients with poor control with pre-existing neovascularization of the retina, or even worse, neovascularization of the iris, require close attention.3 In these patients, a monofocal acrylic lens is the best choice, and I typically will use a three-piece intraocular lens or a capsular tension ring (CTR) with a SPA, as these patients are at risk for future pars plana vitrectomy and possible weakening of the zonular apparatus. This set of patients need a retinal consult with most likely an OCT and fluorescein angiography and are often treated with panretinal photocoagulation (PRP) and VEGF inhibitors prior to the surgery. Operatively, these patients will have a small pupil, particularly if they have had previous PRP.2,3 These patients do well with either iris hooks or devices such as the Malyugin ring. I typically will add a suture to the incision as they may require a contact lens for further laser immediately following the surgery. The post-op period includes close collaboration with retinal colleagues typically at 1 week and 4 weeks following surgery (Table 1). Patients who have had vitrectomy in the past for their diabetic retinopathy can present problems for the cataract surgeon especially if silicone oil was used.5 These patients should get a monofocal acrylic lens and may need a CTR. The surgeon should be prepared for sulcus placement as always with a large acrylic optic. Usually these patients have already had a pre-op retina evaluation, which will most likely include OCT to evaluate for pre-existing macular edema. Patients following a vitrectomy often have a small pupil, particularly if they have had significant PRP in the past. Rarely, particularly in patients with an early cataract following vitrectomy, the posterior capsule can be damaged from the vitrectomy. In these patients, the surgery can be quite difficult. Patients with an early cataract following vitrectomy should be dealt with similar to a patient with a posterior polar cataract where the assumption is that there is a hole in the posterior capsule and one should avoid hydrodissection (Figure 1). Rarely these patients will have silicone oil in place following extensive retinal surgery. Silicone oil can be quite difficult and presents issues both for IOL power estimation and for surgery. If one can use the IOLMaster (Carl Zeiss Meditec, Dublin, Calif.), and if one is going to remove the silicone oil, then these patients can be relatively routine. However, if the silicone oil is going to remain in place, it presents issues due to the difference in refractive index of the silicone oil and how it interfaces with the intraocular lens. See the website of Warren Hill, M.D., for more information on how to calculate intraocular lens power in this situation.9 Silicone oil can leak into the anterior chamber during surgery, which must be removed by the end of the case (Figure 2). Post-op, these patients should have continued retinal consultation, typically at the 1 week and 1 month visit (Table 1). Cataract surgery in patients with diabetes can range from being quite routine to being quite complex depending on the level of retinopathy and the amount of prior procedures for the retinopathy. Diabetes is extremely common, particularly in the set of patients that we are dealing with for cataract surgery, and it is important that cataract surgeons are comfortable managing these patients.


1. National Diabetes Clearing House. Accessed on September 30, 2012. 2. Ostri C, Lund-Andersen H, Sander B, La Cour M. Phacoemulsification cataract surgery in a large cohort of diabetes patients: visual acuity outcomes and prognostic factors. J Cataract Refract Surg. 2011;37(11):2006-2012. Epub 2011 Sep 1.

3. Suto C, Kitano S, Hori S. Optimal timing of cataract surgery and panretinal photocoagulation for diabetic retinopathy. Diabetes Care. 2011;34(7):e123.

4. Elman MJ, Bressler NM, Qin H, et al. Diabetic Retinopathy Clinical Research Network. Expanded 2-year follow-up of ranibizumab plus prompt or deferred laser or triamcinolone plus prompt laser for diabetic macular edema. Ophthalmology. 2011;118(4):609-614.

5. Horozoglu F, Yanyali A, Aytug B, et al. Macular thickness changes after phacoemulsification in previously vitrectomized eyes for diabetic macular edema. Retina. 2011;31(6):1095-1100.

6. Akinci A, Muftuoglu O, Altınsoy A, Ozkılıc E. Phacoemulsification with intravitreal bevacizumab and triamcinolone acetonide injection in diabetic patients with clinically significant macular edema and cataract. Retina. 2011;31(4):755-758.

7. Szaflik JP, Szaflik J. Prevention of vision loss after cataract surgery in diabetic macular edema with intravitreal bevacizumab: a pilot study. Retina. 2010;30(8):1328-1329.

8. Ahmadabadi HF, Mohammadi M, Beheshtnejad H, Mirshahi A. Effect of intravitreal triamcinolone acetonide injection on central macular thickness in diabetic patients having phacoemulsification. J Cataract Refract Surg. 2010;36(6):917-922.

9. Hill, Warren. Accessed September 30, 2012.

Editors' note: Dr. Oetting is professor of clinical ophthalmology, and director, ophthalmology residency program, University of Iowa, Iowa City. He has no financial interests related to this article.

Contact information


Cataract surgery and diabetes Cataract surgery and diabetes
Ophthalmology News - EyeWorld Magazine
283 110
216 149
True, 11