The End of Intravitreal Injections?

12 08 2012

Intravitreal injection of drugs is now the standard of care for treatment of many forms of diabetic macular edema and “wet” macular degeneration. The results after injection can be amazing, yet every time I inject a patient, I am acutely aware that there is a small risk of developing endophthalmitis with the injection – 1/1000 per injection. And as is often the case, the organisms causing infection with intravitreal injections often lead to loss of vision. It’s a small risk, but a very real one. I wish there were some way to deliver the drugs without having to inject them into the vitreous.

So, I was excited to read of the study by Patel et al from the Georgia Institute of Technology that was published in the July 2012 issue of Ophthalmology and Visual Science. They describe using a 750 micron needle to inject drugs into the suprachoroidal space (SCS) of rabbit eyes. This technique delivers drugs to the chorioretinal area, through the sclera, and avoids entering the vitreous cavity. Presumably, the risk of endophthalmitis would be greatly reduced, if not eliminated, with this technique.

750 micron needle used for suprachoroidal injections (Courtesy of Medscape)

The study showed that drug concentrations after such injections were higher in the posterior segment of the eye than the anterior segment, and that they stayed around longer than with intravitreal injections.

Of course, the efficacy of this technique would need to be proven in humans, and I look forward such studies. I’m thrilled that such work is being done as it could help reduce the main concern of intravitreal injection, endophthalmitis.

If you are Medscape member, you can read more about the study here.  Microneedle Developed for Back-of-the-Eye Injections


CATT Results Released: Avastin and Lucentis Equivalent for AMD

29 04 2011

Well, it’s the big news today. The one-year results of the NIH sponsored CATT trial, comparing Avastin (bevacizumab) and Lucentis (ranibizumab) in a head-to-head comparison for the treatment of AMD were released early. The results at one year show equal efficacy in terms of visual acuity. Interestingly, the study also show that prn dosing results in equal visual acuities when compared to monthly dosing.

Here are the key paragraphs from the NIH press release. The full article is available online at the NEJM site, if you have access to that site.

“NEI launched CATT in 2008 to compare Lucentis and Avastin for treatment of wet AMD. The study has now reported results for 1,185 patients treated at 43 clinical centers in the United States. Patients were randomly assigned and treated with one of four regimens for a year. They received Lucentis monthly or PRN, or Avastin monthly or PRN. Enrollment criteria required that study participants had active disease.

“Patients in the monthly dosing groups received an initial treatment and then had an injection every 28 days. Patients in the PRN groups received an initial treatment and were then examined every 28 days to determine medical need for additional treatment. PRN groups received subsequent treatment when there were signs of disease activity, such as fluid in the retina. Ophthalmologists involved in patient care did not know which study drug a patient was getting, to make sure that the data was not affected by how anyone felt about the treatment.

“Change in visual acuity served as the primary outcome measure for CATT. Thus far, visual acuity improvement was virtually identical (within one letter difference on an eye chart) for either drug when given monthly. In addition, no difference was found in the percentage of patients who had an important gain or loss in visual function. Also, when each drug was given on a PRN schedule, there also was no difference (within one letter) between drugs. PRN dosing required four to five fewer injections per year than monthly treatment. Visual gains were about two letters less with PRN than with monthly treatment but overall visual results were still excellent.

“‘In addition to the primary finding of equivalence between Lucentis and Avastin for visual acuity, CATT also demonstrates that PRN dosing is a viable treatment option for either of these drugs,’ said Daniel F. Martin M.D., study chair for CATT and chairman of the Cole Eye Institute at the Cleveland Clinic. ‘Substantial visual acuity gains may be accomplished with a lower treatment burden.’

“Adverse events indicate development or worsening of a medical condition. They may or may not be causally associated with the clinical trial treatment, but they are always monitored and reported in any clinical trial. The median age of patients in CATT was over 80 years, and a high rate of hospitalizations might be anticipated as a result of chronic or acute medical conditions more common to older populations.

“Serious adverse events (primarily hospitalizations) occurred at a 24 percent rate for patients receiving Avastin and a 19 percent rate for patients receiving Lucentis. These events were distributed across many different conditions, most of which were not associated with Avastin in cancer clinical trials where the drug was administered at 500 times the dose used for AMD. The number of deaths, heart attacks, and strokes were low and similar for both drugs during the study. CATT was not capable of determining whether there is an association between a particular adverse event and treatment. Differences in serious adverse event rates require further study.

“Investigators in the CATT study will continue to follow patients through a second year of treatment. These additional data will provide information on longer-term effects of the drugs on vision and safety.

Rifampin for Central Serous Chorioretinopathy

30 08 2010

The first full day of the American Society of Retina Specialists Annual Meeting has just ended here in Vancouver. It’s been a productive meeting so far, with a few clinical pearls that will change the way I practice. Yesterday’s case conference session had a memorable patient presented by Kirk Packo’s group in Chicago that will change the way I approach central serous retinopathy (CSR).

They presented a patient who had widespread chorioretinal lesions and a serous macular detachment. The thinking was that it could be due either to tuberculosis or to chronic CSR. The patient was placed on standard multi-drug therapy for tuberculosis and while on the therapy, the macular edema resolved. However, when the drugs were stopped, the fluid recurred. The patient was placed again on the TB drugs, the fluid again resolved, and when the drugs were stopped, the fluid again re-accumulated. At this point, they began to consider that the patient may not have TB, but may rather have chronic CSR and that one of the TB medications was having a positive independent effect on the serous retinopathy.

The reviewed the list of TB drugs and realized that rifampin has the ability to decrease endogenous steroid production, and that perhaps this was resulting in resolution of the patients serous retinopathy. They gave the patient a trial of rifampin alone (600 mg per day) and voila, the fluid resolved again.

Since that time, they had used rifampin in several other patients with CSR and have seen resolution of the fluid within 1-4 weeks. The case presentation doesn’t prove anything, and further study is required, but it does raise the possibility that rifampin may help CSR, particularly chronic cases, and I will now consider it as an option to offer select patients with central serous chorioretinopathy.

Lucentis vs Avastin: Who will win the race?

9 08 2010

As we await the results of the NIH sponsored CATT trial, a head-to-head comparison of ranibizumab (Lucentis) and bevacizumab (Avastin), other smaller studies are trickling in, pointing to what many of us suspect from clinical experience — their efficacy is pretty much the same. Fong et al. published a study in the February issue of Ophthalmology (Intravitreal bevacizumab and renibizumab for age-related macular degeneration) which compared the results from 342 pateints treated with bevacizumab and 128 with ranibiumab. It was a non-randomized trial and the bevacizumab patients were younger. The study showed that both groups had fairly equal results at 12 months. About 23% of bevacizumab patients and 25% of ranibizumab patients had visual acuity of 20/40 or better. More patients with bevacizumab (27%) had improved vision compared to ranibizumab (20%). So, I think we can call it close to even, and remember that about 25% will have improvement of vision and the same percentage with get to 20/40 or better.

The one point that stood out was that in a year, the bevacizumab patients needed about 4 injections, whereas the ranibizumab patients needed 6 injections. This is in keeping with the thought that bevacizumab, being a larger molecule, stays around longer and has a longer duration of action.

Which brings us to the road ahead. The fact that the CATT study has not been stopped, points to the fact that bevacizumab is likely not either significantly less effective or more dangerous than its $2000 per dose cousin, ranibizumab. Both drugs are made by Genentech, bevacizumab being the off-label one. It’s tough for a company to have two of its drugs competing against one another, especially if the off-label one wins or comes close to being even. I believe that Genentech’s approach in the coming years will be to make ranibizumab easier to use. The concern with both these drugs is the ongoing burden of repeat injections. We have a history of the development of sustained-release intraocular drugs in ophthalmology, and the next step is to get one of these drugs into a sustained release form to decrease the number of injections required. Genentech will develop ranibizumab in a sustained release form, leaving bevacizumab to be injected repeatedly. So the issue of equal efficacy will become moot.

Let Them Eat Viagra: No link to central serous retinopathy

20 03 2010

In the February 2010 issue of Retina there is an important article by French and Margo titled “Central Serous Chorioretionopathy and Phosphodiesterase-5 Inhibitors.”  It’s a well done surveillance study that shows no association between the use of these erectile disfunction drugs and CSR.  You’ll recall that in 2008 Fraunfelder and Fraunfelder reported 11 men with CSR who were taking sildenafil (Viagra).  Although the association was not proven, the report raised the issue and added an extra question to the work-up of patients with CSR, “Are you taking Viagra?”.

French and Margo performed a surveillance study on the Veteran’s Healthcare Administration’s (VHA) database of 5.2 million veterans, 90% of whom are male.  They used the ICD9 codes for central serous retinopathy to find such patients, and as controls they used codes for the diagnosis of acute conjunctivitis and of macular pucker.  In order to minimize the possibility of including misdiagnoses of macular degeneration, they excluded patients age 60 and over.  They limited their search to the fiscal year 2004-05, and they also excluded anyone who had a diagnosis of CSR prior to these dates.  They also excluded individuals who had sources of exogenous or endogenous glucocorticoids, because of their association with CSR.

Their search identified 577 men with newly diagnosed CSR during the study period.  For statistical purposes they identified 1154 age-matched men in the two control groups, acute conjunctivitis and macular pucker.  Then they asked, how many men in each group were prescribed PDE-5 inhibitors prior to their diagnosis?  In the CSR group, 19.2% had been prescribed PDE-5 inhibitors.  The numbers were 18.5% and 21.5% in the two control groups.  Since these numbers are all about the same, the authors conclude that there is no statistically significant association between PDE-5 use and CSR.

The authors do note the limitations of their study, particularly that it was a surveillance study, with no access to individual patient charts.  Therefore, there is the possibility of misdiagnosis.  Also, although the patients were prescribed PDE-5 inhibitors prior to the onset of CSR, there is no way to know if they were actually taking the drug at the time of the diagnosis.  They also point out that some cases of CSR may be causally related to PDE-5 use, but that in this case-controlled study there is no statistically significant association.

Preparing for Anaphylaxis from Fluorescein Angiography

3 02 2010

I was interviewed for an article that appeared in the January issue of Retinal Physician magazine. The article was titled, “Putting ‘Management’ Into Risk Management: Proactive policies for keeping patients safe.” My discussion regarding developing a plan for performing fluorescein angiography were rather comprehensive, so Retinal Physician turned it into a sidebar in the article. In my comments I mentioned that the protocol we developed for our office is available here at The Retina Blog. For some reason it seems to have disappeared. If you would like a copy, leave a comment (it will show your email address to me, but no one else can see it) and I’d be happy to send the protocol to you. I only ask that you let me know how useful you find the protocol, and if you have any suggestions for improvements.

Here is my discussion in the January 2010 issue of Retinal Physician:


The biggest risk we need to be prepared for in retina practices is anaphylaxis from fluorescein angiography. A publication this past summer showed the incidence of anaphylaxis is one per 350 fluoresceins.2 Because anaphylaxis can be life threatening, and because it is treatable, we must train our staff to recognize it — and our retina practices must be prepared to deal with it.

Anaphylaxis is a multisystem allergic reaction. The severity of the reaction is difficult to predict at its outset, and the internal medicine literature advocates treating it early with subcutaneous or intramuscular epinephrine (1:1000). Anaphylaxis can involve four major organ systems: respiratory, cardiovascular, gastrointestinal and cutaneous. Involvement of any two of these organ systems meets the definition of anaphylaxis. So if a patient has itching and shortness of breath, the internal medicine literature advocates treating with epinephrine. I think many of us are hesitant to give epinephrine in our practices because of the potential risks in our patients with diabetes and cardiac disease. But it’s certainly worth having epinephrine on hand. It’s also important to have a protocol in place.

In developing protocols for our practice, I consulted with other retina specialists to see what protocols they had in place. At hospital-based practices, the most common protocol was “call the code blue team.” Most private practices I spoke to don’t have written protocols. In developing ours, I sought the advice of my colleagues in internal medicine, emergency medicine and anesthesia. Among these specialties, they advocated a wide range of reasonable approaches. If you have easy access to a 911 team, then your protocol can be minimal — administer diphenhydramine, maybe epinephrine, and monitor blood pressure. If, however, you want to be more complete, because of a slower anticipated response time, your protocol could include administering oxygen or IV fluids, monitoring oxygen saturation and a cardiac tracing, and having available an airway, bag and an automatic external defibrillator.

Our protocol is primarily a checklist that includes several sections: preparation, which includes checking the blood pressure and pulse, reviewing drug allergies, pregnancy, and any history of prior cardiac or respiratory problems. Most of these data are already on the chart, so this portion of the checklist is a sort of a “time out” to review things before injecting the fluorescein.

The second portion of the checklist addresses things that must be available prior to injecting: Personnel —which is an MD in the office; Equipment — which, in our case, is a cardiac monitor, and bag and mask; Supplies — such as IV fluid, needle, tape and gauze; Drugs — we stock oral and IV diphenhydramine, and IM/SQ epinephrine (1:1000); and Paperwork — the consent form, the symptoms checklist, and treatment flowsheets.

The protocol in our practice is symptom-based. For example, if there is nausea or vomiting, the protocol directs staff to provide an emesis basin, support, and monitor the patient for 30 minutes. In the event of mild hives or itching, the protocol calls for notification of the MD, the administration of oral diphenhydramine 25-50 mg, and monitoring until symptoms improve. For more severe hives or itching, we administer the drug IV.

In the event of the onset of respiratory symptoms, our protocol calls for us to record oxygen saturation and blood pressure, to call 911, to prepare the epinephrine for possible administration, and to get the doctor in the emergency room on the phone for guidance while awaiting arrival of the emergency team. Our entire protocol is available online at

The truth is that severe symptoms are rare, and fortunately, death from fluorescein angiography is reported to be only 1/220,000. But it is prudent to be prepared to deal with the more severe symptoms in the unlikely event they arise.

In terms of training, I think we physicians have to periodically review the management of anaphylaxis, and an excellent review is available in the article, “Office Approach to Anaphylaxis: Sooner Better than Later” by Stephen F. Kemp, which appeared in the Amercian Journal of Medicine (2007) 120, 664-668.

I do think it’s reasonable for nonphysicians to administer fluorescein, under the supervision of a physician.* In our case, it means that they have shown their ability to administer FA successfully 10 times under direct physician observation. After that, a physician must be present in the office and available to deal with any emergencies during administration. In our office, all staff (including doctors) maintain current CPR certification.

* OMIC warns that allowing unlicensed staff to inject may be illegal. See their risk management recommendations on FAs.

Bilateral same-day intravitreal injections? Yes!

8 01 2010

One of the growing issues in treating age-related macular degeneration (AMD) is to reduce the burden of treatment for the patients, many of whom are elderly, and may require injections of VEGF inhibitors every 4-8 weeks. A long-standing tradition in ophthalmology has been to avoid bilateral same-day surgical intervention. The rationale has been that if you operate on two eyes on the same day, any infection that develops in one eye may spread to the other eye, and that if there is any contamination of instruments or compounds used during the surgery, operating on both eyes on the same day increases the likelihood that both eyes will be adversely affected.

Intravitreal injections, which have become standard of care for treatment of AMD are fairly innocuous in terms of surgical insult. The “wound” is a 31 gauge needle entry site. Nevertheless, endophthalmitis may develop. For example, in the VISION study of intravitreal pegaptanib, the incidence of endophthalmitis was 0.16%. In the MARINA study of intravitreal ranibizumab, it was 0.05%. And in the ANCHOR study of ranibizumab, it was 0.05% (3 cases our of 5,921 injections). A study by Pilli et all reported an incidence of 0.029% (3 cases in 10,254 injections). These incidences were for unilateral injections.

So, is there an increased risk for bilateral same-day injections? Lima et al (Yannuzzi’s group) out of New York published their retrospective analysis of bilateral same-day intravitreal injections of VEGF inhibitors in the October 2009 issue of Retina. They report that of 1,534 bilateral injections (3,068 injections total), the incidence of culture-proven endophthalmitis was 0.065%, and the incidence of acute intraocular inflmaation was 0.033%. None of those cases were bilateral. There were no cases of retinal breaks. They conclude that it appears that there is no increased risk for same-day bilateral injections of VEGF inhibitors, as the complication rates are similar.

It’s noteworthy that all patients in their study were done in the office, and received surface disinfection with 5% providone-iodine solution, followed by 2 days of either polytrim or ofloxacin qid.

The study is very useful and gives comfort to those of us considering bilateral same-day injections. In any disease process where the incidence of an occurence is very low, large numbers of patients are needed to determine whether or not there is a difference in incidence between groups, and so, it is fair to say that there is no obvious increase in the risk of complications from bilateral VEGF inhibitor injections, while recognizing the limitation of the comparison.

Bevacizumab helpful in BRVO

7 12 2009

The SCORE study showed that intravitreal triamcinolone was no more helpful than standard therapy (macular laser) in improving vision in patients with branch retinal vein occlusions (BRVO), and thus the SCORE study concluded to continue using standard therapy to treat BRVO.  The question is, might another agent, say a VEGF inhibitor like bevacizumab (Avastin) be helpful in treating BRVO?

We get some indirect answers in a study published in the November/December issue of Retina.  The study (Comparison of two doses of intravitereal bevacizumab as primary treatment for macular edema secondary to branch retinal vein occlusion:  results of the pan american collaborative retina study group at 24 months) by Wu et al and the Pan-American Collaborative Retina Study Group was a retrtospective multicenter study looking at results from 63 eyes treated with two different doses of bevacizumab:  1.25 mg and 2.5 mg.  Other studies had shown benefit in visual acuity and central macular thickness (CMT) with bevacizumab in BRVO, and the authors wondered if an increased dose may improve the visual outcome or decrease the frequency of injections.  Patients with macular edema were given bevacizumab at one of the two doses (depending on the center they visited) and were examined on a monthly basis.  OCT’s were performed at months 1, 3, 6, 12 and 24.  If at any time the OCT measured CMT > 250 microns, they were classified as “recurrent” or “persistent” macular edema, and were re-injected.  Re-injections were also given if visual acuity decreased by more than 5 letters (1 line).

Their study found that there was no additional benefit from the higher dose, and that both groups experienced improved vision and decreased CMT.  The results showed that on average, CMT improved from about 450 mic at the first visit, to 250 mic at 3 months, and ended up at the 240 mic range at 24 months.  Furthermore, the mean time for repeat injections was in the 10-13 week range.

But is this better than laser?  Well, the study doesn’t directly look at this question, but it does state that among its small sample size of 63 eyes, of those receiving 1.25 mg, 68% experienced improvement in vision by 3 or more lines of vision (72% in the 2.5 mg group).  That’s pretty significant percentage of improved vision.  You’ll recall that the SCORE Study patients had improved vision in the 26-29% range for IVTA and laser.  So bevacizumab seems to give better results than laser for BRVO.  I realize I’m mixing studies — not really a meta-analysis (maybe a “mini”-analysis), and clearly this needs to be studied further.  But the authors seem to lay out a reasonable approach for treating BRVO with bevacizumab, and their results seem to be significantly better than best results we have available with laser or IVTA.  Check vision monthly, and if worse by 5 letters, treat with bevacizumab;  check OCT at 3, 6 12 and 24 mo, and if CMT > 250 treat with bevacizumab.  Expect about 70% to improve 3 or more lines of vision.

Early Treatment for Retinopathy of Prematurity (ETROP)

2 12 2009

I have a patient in the NICU that I’ve been following for the past couple of weeks with ROP, so it’s given me the opportunity to review the ETROP results, the final results of which were published in 2004 by William V. Good on behalf of the ETROP Cooperative Group.

The CRYO-ROP trial had given us the “threshold” of treatment for ROP.  Threshold was defined by the CRYO-ROP as at least five clock hours of contiguous, or eight cumulative clock hours of stage 3 ROP in zone I or II in the presence of plus disease.  When I was a resident, a quick trick we were taught was to look at the posterior pole for plus disease.  If there was no plus disease, no matter how extensive the ROP might be, you weren’t at threshold.  And I still do this now, looking first at the posterior pole for plus disease, before heading off into the periphery to determine zones and stages.

The ETROP study came about because of a sense that treating some patients sooner than these threshold definitions could be beneficial, and in fact the ETROP proved it to be so, effectively changing the threshold for treatment.  The most important change is that treatment can now be considered sooner without regard to the number of clock hours involved.  The new criteria defined by ETROP are:

1.  Zone I ROP with plus disease (doesn’t matter what stage)

2.  Zone I ROP with stage 3 disease (don’t need plus disease)

3.  Zone II ROP, stage 2 or 3, with plus disease.

These three criteria are defined as “Type I” ROP.

Treating these eyes instead of waiting for CRYO-ROP threshold criteria resulted in reduction in unfavorable visual acuity outcomes from 19.8% to 14.3%, and a reduction in unfavorable structural outcomes from 15.6% to 9.0% (both at 9 months).  Bear in mind that these results are not comparing “treated” to “untreated”, but are comparing “treating earlier” to “waiting for CRYO-ROP threshold”.  Treating these Type I eyes resulted in treatment on average 2 weeks earlier than waiting for CRYO-ROP threshold.

Looking at these numbers, there is a definite clinical benefit to treating earlier, but you can spin the numbers in a couple of ways.  You can say that unfavorable visual acuity outcomes were reduced by 25% (from 19.8 down to 14.3%), but I think it’s more meaningful to look at the absolute percentage reduction, which is about 5%.  A 5% decrease means that for every 20 eyes you treat early, you see a benefit in one of them.  Not great, but given that you’re only changing the timing of the treatment in most cases, (66% of control eyes went on to need treatment by CRYO-ROP criteria anyway), it’s pretty good.  The question though is, what risk was involved in treating these infants two weeks earlier?  The study showed that there was a definite increase in morbidity in the early treatment group (but not mortality) — bradycardia, re-intubation, etc.  How often did this occur?  If you treat two patients early, one had a potentially serious systemic side effect.  So, the decision regarding the timing of treatment is not so cut and dried as the study numbers might suggest.  On the other hand, it does support treating ROP at “pre-threshold” levels, and that should be taken into account.  Furthermore it points to the need to follow pre-threshold eyes more closely, and to be prepared to treat within 48 hours of reaching threshold.

Is Avastin the new laser for diabetic macular edema?

19 11 2009

In the June 2009 issue of Ophthalmology Soheilian et al, out of Tehran, published an important paper comparing intravitreal bevacizumab (IVB) alone or in combination with triamcinolone (+IVTA) to macular photocoagulation for treatment of diabetic macular edema.  They randomized a total of 150 eyes with clinically significant macular edema (CSME) into three treatment groups, and followed them for up to 36 weeks (which is about 9 months).   If CSME persisted at 12 week (3 mo) intervals, the eye were retreated with the same intervention.  Their study showed that intravitreal bevacizumab yielded better visual acuity results at 36 weeks than did laser treatment.  Specifically, vision improved by > 2 Snellen lines of acuity in 37% of the IVB group, 25% of the IVB+IVTA group, and 15% of the laser group.  On the other hand, retreatment was required in 28%, 20% and 6% of the three groups respectively.  IVB improves vision more, but needs to be used more often.

One of the things that the study did not do, was differentiate the patients with CSME into those with cystic edema and those with non-cystic edema.  My hunch is that we would find that in the group with cystic edema, the difference between the effectiveness in bevacizumab and laser would be even greater.  Will bevacizumab become a replacement for laser in treating CSME?  It’s too soon to say.  My hope is that the Sohelian et al will continue to follow these patients and report on results as they reach the 1 year and 2 year marks.

In clinical practice, this paper seems to support what many of us are currently doing:  using bevacizumab as first line treatment for diabetic macular edema to decrease the edema and improve the vision, and following it up with laser treatment for long term control.  I’d like to see a study that looks at this situation.  I use the sequential approach with greatest success in patients with massive or cystic macular edema.

Here is the abstract:

Ophthalmology. 2009 Jun;116(6):1142-50. Epub 2009 Apr 19.

Randomized trial of intravitreal bevacizumab alone or combined with triamcinolone versus macular photocoagulation in diabetic macular edema.

Soheilian MRamezani AObudi ABijanzadeh BSalehipour MYaseri MAhmadieh HDehghan MHAzarmina MMoradian SPeyman GA.

Ophthalmology Department, Labbafinejad Medical Center, Shaheed Beheshti Medical University, Tehran, Iran.

PURPOSE: To compare the results of intravitreal bevacizumab (IVB) injection alone or in combination with intravitreal triamcinolone acetonide (IVT) versus macular laser photocoagulation (MPC) as a primary treatment of diabetic macular edema (DME). DESIGN: Randomized 3-arm clinical trial. PARTICIPANTS: A total of 150 eyes of 129 patients with clinically significant DME and no previous treatment.

METHODS: The eyes were randomly assigned to 1 of the 3 study arms: the IVB group, patients who received 1.25 mg IVB (50 eyes); the IVB/IVT group, patients who received 1.25 mg of IVB and 2 mg of IVT (50 eyes); and the MPC group, patients who underwent focal or modified grid laser (50 eyes). Retreatment was performed at 12-week intervals whenever indicated. MAIN OUTCOME MEASURES: Change in best-corrected visual acuity (VA) at week 24.

RESULTS: VA changes among the groups were statistically significant at 6 (P<0.001) and 24 (P = 0.012) weeks. The significant treatment effect was demonstrated in the IVB group at all follow-up visits and in the IVB/IVT group at 6 and 12 weeks. VA changes +/- standard deviation at 36 weeks were -0.28+/-0.25, -0.04+/-0.33, and +0.01+/-0.27 logarithm of minimum angle of resolution in the IVB, IVB/IVT, and MPC groups, respectively (P = 0.053). Significant central macular thickness (CMT) reduction was observed in all groups only up to 6 weeks; however, CMT changes were not significant among the groups in all visits. Overall, retreatment was required for 27 eyes up to 36 weeks (14 in the IVB group, 10 in the IVB/IVT group, and 3 in the MPC group). In the IVB group, in which a greater VA improvement was observed, only 1 injection was required in 72% of the cases. VA improvement >2 Snellen lines at 36 weeks was detected in 37%, 25%, and 14.8% of patients in the IVB, IVB/IVT, and MPC groups, respectively.

CONCLUSIONS: Intravitreal bevacizumab injection in patients with DME yielded a better visual outcome at 24 weeks compared with macular photocoagulation. A change in CMT beyond the 6-week time point that corresponded to the vision change was not detected. No adjunctive effect of IVT was demonstrated.