Ophthalmology Management

Environmental factors may not be at the forefront of one's mind when contemplating glaucoma's pathophysiology, but an interesting presentation by Harvard's Louis Pasquale, MD provides evidence of an association between geographic latitude and exfoliation syndrome. Citing prior studies, Dr. Pasquale observed that "in countries closest to the equator the prevalence is around 1%, and as you go closer to the arctic circle the prevalence ends up around 20%," although there are a few notable exceptions among Eskimos and some regions of China.

To ascertain the breakdown within the continental US, Dr. Pasquale reviewed data from 120,000 enrollees (79,000 female, 41,000 male) in the Nurses Health Study and the Health Professionals Follow-up Study, two trials that proved ideal for this purpose given that the cohorts are large and subjects are widely dispersed throughout the US. Dr. Pasquale and his team identified 214 cases of exfoliation syndrome from that population.

They stratified the US into a northern tier (above 41-42° latitude), a middle tier (37-41° latitude) and a lower tier (below 37° latitude) and mapped the childhood residences of the study subjects; any who shifted tiers during the first 15 years of life were excluded. After controlling for risk factors such as age, race and gender, his group found that the middle and southern tiers were associated with a reduced risk of exfoliation syndrome.

Differences in temperature and UV exposure by latitude are the two most likely factors at play, Dr. Pasquale suggested; these associations will be addressed in follow-up studies. He concluded that the data could have implications for screening methodologies and may aid development of primary prevention strategies.

Tags:

That was the question posed by Robert L. Stamper, MD in his Surgery Day talk.

Dr. Stamper, a professor of clinical ophthalmology at UC San Francisco, began with a little perspective on the goals of therapy: first, to prevent "disabling" visual loss, not just "any visual loss." At least some visual degradation is inevitable in glaucoma, and is indeed one of its defining characteristics. The clinician should also strive to maintain the patient's health and quality of life, and the treatment shouldn't be worse than the condition itself.

With those goals in mind, which is more appropriate as first-line therapy in newly diagnosed open-angle glaucoma, medications or laser trabeculoplasty? Medical therapy is of course easily modified if the initial result is suboptimal, and despite concerns about compliance "the fact is that the vast majority of patients do just fine with medical therapy for very long periods of time," Dr. Stamper said. LTP, by contrast, has a limited lifespan of effectiveness and the long-term benefit of retreatment, he believes, has not been well established. In LTP's favor: it substantially lowers IOP in the short term and obviates the cost, convenience, compliance and side effect concerns of medical therapy. Patients also prefer the "gee whiz" aspect of a high-tech laser procedure over the more mundane eye drop.

Dr. Stamper then cited data from the Glaucoma Laser Trial (a multicenter, prospective randomized clinical trial comparing ALT to timolol) that found the mean IOP was slightly lower in patients receiving ALT as initial therapy, at both the two- and seven-year marks. By the end of the original two-year study period, about half of all ALT patients required a pressure-lowering medication and about 70% of those who had started on timolol required additional medical therapy. He pointed out, however, that some of the newer medications introduced in the years since the GLT began have since shown a general equivalence in efficacy to that of LTP in the first year.

As for the LTP options, "there is no solid evidence to suggest superiority for any one of the laser modalities over another," he said, although "we do know that 360 degree treatment is more effective than 180, but only slightly, and SLT may be slightly more cost effective than medical therapy in newly-diagnosed open-angle glaucoma."

Offering his own opinion on how to answer the titular question of his talk, Dr. Stamper finished up by saying that medical therapy is probably appropriate in most newly-diagnosed patients. Circumstances that would argue in favor of beginning instead with LTP are:

• a known compliance problem (i.e., frequently missed exam appointments)

• the patient is African American

• exfoliative glaucoma

• The patient is very elderly and would stand to benefit more from the immediate impact and convenience of LTP rather than a protracted course of medical therapy.

Tags:

Sixteen years after it began, the Ocular Hypertension Treatment Trial is still generating fresh insights into one of the most vexing questions in glaucoma care: does treating ocular hypertension delay or perhaps even prevent glaucoma? About 4-8% of people in the US over the age of 40 (4.8 to 9.5 million individuals) have ocular hypertension. When questioning standards of care in a group of that size, any conclusive data will necessarily have profound implications for the economics of healthcare delivery and, of course, the lives of millions.

Attendees of this year's AGS were given an early look at the OHTS phase 2 data by Michael A. Kass, MD, principal investigator of the long-running study, which dates back to Bill Clinton's first term in office. In his 40-minute presentation, Dr. Kass reviewed the trial's phase 1 data, published in 2002, and provided highlights of the recently-completed second phase, on which a full report appears in the March 2010 Archives of Ophthalmology.

OHTS's goals were twofold, Dr. Kass said: (1) to answer once and for all the question of whether or not early treatment confers a protective effect and (2) to identify the baseline demographic and clinical factors that would predict which patients are likely to get glaucoma. The study enrolled 1636 subjects with baseline pressures of 24 to 32 mm Hg in one eye and 21 to 32 in the fellow eye, but normal visual fields at initial presentation. Subjects were randomized to either observation or a regimen of medical therapy sufficient to reduce IOP by at least 20% (the agent was chosen at the discretion of the investigator).

At the five-year mark, the incidence of POAG was nearly 60% lower in the medication group than the observation group. OHTS phase 1, Dr. Kass said, provided proof of concept; namely, that medication reduces the incidence of POAG. It did not, however, indicate when medical therapy should begin or whether or not all ocular hypertensives should begin treatment. Those were addressed in the second phase, by splitting the study group into early-treatment and late-treatment groups. After 7.5 years of observation, patients in the phase 1 observation group were offered the option to begin therapy; those in the original treatment group simply continued therapy. The early-treatment group continued therapy for 13 years while the late-treatment group has thus far logged 5.5 years of data in phase 2. Of the original 1636 individuals enrolled in OHTS, 1159 remained in the study for the duration of phase 2. The two groups differed in medication use, owing to the course of drug development during the study period. Most of those in the early-treatment group had begun the study on beta-blockers while the late-treatment group benefited from more recent medication launches and predominantly used prostaglandin analogs.

Charting the incidence of POAG in the two groups over the 13-year span of the study shows curves that initially diverge (i.e., a higher incidence in the observation group) but then become essentially parallel as the previously-untreated patients begin therapy.  "The incidence of glaucoma was not different between the observation and medication groups once patients began treatment," Dr. Kass said. "This effect appears to be relatively rapid." The median time to develop glaucoma was 8.7 years in the early-treatment group and six years in the late-treatment group. As would be expected, African American patients developed glaucoma more readily than other ethnicities; however, race is not a significant predictor in a multivariate model that accounts for central corneal thickness and baseline c/d ratio. "If we'd had 10,000 African Americans instead of 400, race may have stayed in the model," Dr. Kass said, "but this is the limits of what we can do with the data that we have." Dr. Kass's group at Washington University in St. Louis have produced a risk calculator based on OTHS data that is available here. 

Using the parameters at the basis of that calculator to stratify subjects into three groups by baseline risk (low, middle and high) reveals some notable findings that have implications for care. "If you look at the difference between the medication group and the observation group, there's very little absolute reduction in the occurrence of glaucoma" among low-risk patients, Dr. Kass said. Those at low risk in the observation (i.e., late-treatment) group had a cumulative 13-year incidence of glaucoma of 8%; for the early-treatment group, the incidence was 7%. By contrast, in the high-risk stratum "there is a substantial absolute reduction in the occurrence of glaucoma in this high-risk group," he said. The high-risk, late-treatment group's cumulative incidence was 40% vs. 28% for the high-risk, early-treatment group.

There are likely to be a wide range of opinions on exactly how to incorporate these new findings into clinical practice, Dr. Kass said. "Some clinicians may elect to follow all ocular hypertensives without treatment," he observed, adding "I have no objection to that" provided the clinician is sophisticated in glaucoma diagnosis (as AGS members no doubt are) and aware of the need for timely visits and appropriate testing.

He finished up the 20th AGS lecture with the following observations and conclusions from the OTHS:

• Early medical treatment reduces the cumulative incidence of POAG; delaying treatment increases it.

• The absolute risk of glaucoma is greatest in high-risk individuals.

• Most ocular hypertensives are at low risk for glaucoma, and most low-risk patients can be followed without medication.

• Delaying treatment for 7.5 years resulted in only a small absolute increase in POAG in low-risk patients.

• Starting treatment of POAG at diagnosis has no major negative effect on prognosis over five years.

• High-risk patients may benefit from more frequent exams and early treatment, after taking into consideration factors such as patient age, health status, life expectancy and personal preference.

• For ocular hypertensives, the risk of developing glaucoma continues over at least a 15-year follow-up period.

• African Americans develop glaucoma at a higher rate despite similar treatment protocols and IOP levels. The higher incidence is related to their baseline risk factors.

• Individualized assessment of risk is useful to both patients and clinicians.

Tags:

UCLA's Anne Coleman, MD, began the Surgery Day session by reporting on trends in the volume of various surgical procedures from 1994 to 2008 based on her review of Medicare data from that period. By far the biggest mover was laser trabeculoplasty, which declined in volume sharply from 1994 to 2001, and then — just as sharply — rebounded after the introduction of SLT in that year.

The trends were less dramatic for other glaucoma surgical procedures. Peripheral laser iridotomy, aqueous drainage devices and cyclophotocoagulation all experienced slight increases in the past decade, with the latter becoming dominated by endoscopic procedures. Trabeculectomy volume, however, declined gradually but consistently over the 14-year period, a trend that continues today.

Each uptick in a procedure’s volume — most notably for SLT, ECP and drainage devices — correlated with advances in technology. “As we get more innovation, we tend to do more surgery,” Dr. Coleman pointed out. This highlights the need for advances in surgical technology or technique to translate into better outcomes for patients if the additional procedure volume is to be justified, she said.

Tags:

Karim F. Damji, MD, of the University of Ottawa presented data from a five-year-long randomized clinical trial comparing ALT with SLT in patients who had failed medical therapy. In the study, success was defined as pressure lowering of at least 20% with no need for additional laser or medical intervention. The combined success rate for both procedures was about 63-65% at the one-year mark, which then dropped off by 9-10% in each subsequent year. At the conclusion of the five-year study, the success rate was about 25%, Dr. Damji said, and the median survival period was two years.

Dr. Damji pointed out that the two modalities were essentially equivalent in efficacy but that many patients in the study did ultimately require additional medical, laser or surgical interventions over time. He also reminded attendees of research showing that all forms of LTP are capable of reducing diurnal IOP fluctuations to some extent, and that SLT can reduce visit-to-visit fluctuation in pressure.

Predictors of response to ALT in the study included older age, more trabecular meshwork pigmentation, and certain subtypes of glaucoma such as pseudoexfoliative, whereas in SLT the main predictor seemed to be the baseline pressure. The higher the baseline IOP, the greater the pressure-lowering effect. Extent of pigmentation was not correlated with response to SLT. "In fact, if you have too much pigmentation, such as in long-standing pigmentary glaucoma, you can actually tip the function of the meshwork and cause an irreversible spike in pressure," Dr. Damji said.

ALT and SLT also differ in repeatability. When retreatment is defined as an initial procedure with a full 360-degree application of laser pulses and a second procedure that is equally extensive (i.e., rather than two 180-degree procedures), ALT has "quite diminishing returns and in fact quite a high-risk of pressure spikes," he said. SLT, by contrast, is more effective as a repeat procedure in eyes with prior LTP (be it ALT or SLT).

Dr. Damji concluded with the following observations about the suitability of the two procedures depending on clinical circumstances: in primary open-angle glaucoma, ALT and SLT are essentially equivalent as an initial laser procedure; in exfoliative and pigmentary glaucoma, the comparative evidence is as-yet inconclusive (but more data will be forthcoming at this year's AAO meeting); and in eyes with prior ALT, SLT seems more efficacious than repeating ALT.

Tags:

Jay Katz, MD, of Philadelphia's Wills Eye Institute began his talk about laser trabeculoplasty's long-term results and clinical implications by quoting a famous George W. Bush malapropism: once, in reacting to criticism, he claimed that the press h ad "misunderestimated" him. Can we say the same about LTP? Possibly. "Laser trabeculoplasty has certainly been misunderstood," Dr. Katz said, "and we probably have also underestimated its value."

He reminded attendees that the glaucoma management model has always been maximal medical therapy followed by LTP and lastly filtration surgery, but observed that the AGIS trial deviated somewhat from that by including a treatment arm that employed trabeculectomy prior to trabeculoplasty. After seven years follow-up, one conclusion drawn from the AGIS was that there was a lower failure rate among African-American patients in the arm starting with ALT rather than filtering surgery. Visual field loss rate was better in that group as well.

Dr. Katz commented that the first LTP paper in 1981 showed an impressive 12.7 mm drop in IOP after one year, but that subsequent studies (notably those by Brad Shingleton in 1987 and George Spaeth in 1992) found that the first-year failure rate is about 20%, and there's a continued attrition of 8% to 10% per year thereafter. The studies in question continued for over five years, he said.

When comparing medications to LTP, Dr. Katz said the four factors upon which to base your evaluation should be efficacy, compliance, safety and cost.

He mentioned that in the Glaucoma Laser Trial, which compared ALT to timolol as first-line therapy and reported results at two and seven years postop, ALT's success rate was found superior to medical therapy at both time points. Newer trials comparing SLT to prostaglandin analogs have found LTP similar in efficacy to prostaglandin therapy, Dr. Katz said.

On the topic of safety, the major concern in LTP is the potential for an IOP spike. With medical therapy, safety concerns involve hyperemia, ocular surface allergic reaction and systemic effects of beta-blockers.

Adherence to therapy is an ongoing battle in medication use, Dr. Katz said, and pointed to studies estimating compliance rates of 50% to 75% at best with medication regimens. LTP, naturally, obviates this concern.

Several cost analyses done in the US, Canada and the UK found LTP to be more cost effective, Dr. Katz mentioned.

He concluded his presentation by stating that LTP (whether primary or adjunctive) achieves IOP reduction comparable to that of medication regimens, with more favorable safety and cost profiles. However, clinicians should be mindful of LTP's long-term gradual attrition rate.

Tags:

Irked by the latest round of SGR-triggered Medicare fee cuts and dismayed by the dim prospects for a long-term solution, physicians are acutely aware that medical care does not occur inside a bubble, shielded from financial considerations. With this sobering fact in mind, Joshua Stein, MD, of the University of Michigan gave a talk on cost effectiveness (CE) studies of glaucoma management options. "With the aging of the baby boomer population and the need for more eyecare services in the years to come, it's becoming increasingly important to not only determine whether our interventions are more efficacious but also whether they confer value and are more cost effective relative to others."

In 2008, he said, the US spent $2.4 trillion on healthcare — nearly twice per capita what other industrialized nations spent. Innovations certainly benefit patients, but burden the healthcare economy in the process by inflating costs. Do the incremental gains in outcomes justify a larger price tag? Unfortunately, studies seeking to provide guidance are often vulnerable to skepticism by dint of their industry influence. Dr. Stein noted that a review of 19 CE studies specific to glaucoma found that 18 were sponsored by industry, and all concluded that the sponsor's product or category was more cost effective than the alternate. The lone independent study, by the Canadian government, found timolol the most cost effective glaucoma medication, followed by the prostaglandin analogs.

Dr. Stein conducted his own CE analysis, comparing trabeculectomy with drainage implant surgery. He designed a model based on the frequency of complications found in the one-year TVT (tube vs. trab) results and calculated the likely costs if 100,000 patients had undergone these two procedures, using the 2010 Medicare physician and facility reimbursements as a cost basis. The result: drainage device procedures would cost approximately $84 million more per 100,000 patients than trabeculectomy, his model estimated. He then incorporated the increase in quality-adjusted life years (QALY) generated by each procedure into the results. For every additional QALY gained, glaucoma drainage device surgery cost an additional $93,500.

When assessing medical, laser or incisional procedures for glaucoma, Dr. Stein concluded, after studies have proven their efficacy "we need to next perform cost effectiveness analyses to ascertain which of these therapies deliver the best bang for the buck."

A future report will incorporate the three-year TVT data into the model.

Tags:

Although it's long been known that IOP is positively correlated with glaucoma risk, and that pressure-lowering interventions slow the progression of glaucomatous optic nerve damage, the model lacks precision. Diurnal fluctuations make IOP a rather capricious target to begin with, and the pressure measurements themselves are vulnerable to changes in corneal thickness and elasticity.

Stanford's Kuldev Singh, MD, gave an interesting talk on the clinical implications by delving into the relative merits of three parameters used to assess pressure: mean IOP, peak IOP, and fluctuation/variability of IOP. "We still don't have all the answers on which of these parameters are most important as risk factors for glaucoma progression," he began. Mean IOP has historically been the focus of both research and clinical practice, but the other two factors have recently gained in prominence. Dr. Singh reviewed several notable but somewhat contradictory investigations comparing them.

• One study that recorded diurnal curves and visual fields every two months for two years found that the group with the greatest IOP fluctuation was also most likely to show progressive field loss.

• Another study found that the amount of fluctuation increases as IOP itself increases: every additional 1 mm of mean IOP adds another 0.17 mm of fluctuation.

• The AGIS study found that patients who maintained pressure under 18 mm showed no net progression, arguing in favor of an influence attributable to peak IOP.

• Another analysis of AGIS data found that the group with the largest standard deviation (≥3) was more likely to show progression than those with more stable IOP. Critics, however, pointed out that intervention in high-IOP patients reduced pressure and thus decreased variability in IOP — and in so doing made it difficult to attribute causation to any one factor. The authors answered with a new data analysis indicating that long-term variability was only important when mean IOP was low.

• In the EMGT, which didn't include a treatment arm and thus did not introduce treatment-induced variability in IOP, mean IOP was strongly predictive of progression while IOP variability was not found to be an independent risk factor.

• The DIGS study data showed that mean IOP but not long-term variability predicted conversion from ocular hypertension to glaucoma.

Surveying this swath of conflicting evidence, Dr. Singh observed that "to say that this is a controversial issue is indeed an understatement."

He proposed a hypothesis that he arrived at by comparing findings of the EMGT and the CIGTS. In the EMGT, every millimeter of IOP reduction led to a 10% reduced risk of glaucomatous progression. In CIGTS, there was no difference in visual preservation at five years between the medicine-first and surgery-first groups, even though the surgery group did achieve an early advantage in IOP reduction. "Even though they're different analyses, one could make the argument that there are diminishing returns," Dr. Singh said. "At some point, further IOP lowering doesn't give you as much as it did initially."

If that theory can be borne out, it would suggest that the effects of variability and peak IOP are masked when considering mean IOP alone. Consider two patients, he suggested: one with IOP consistently at 13 and another whose results vary from 10 to 16. Both have a mean IOP of 13. In the second patient, however, the risk while at a pressure of 16 is greater than the benefit he stands to gain from being as low as 10.

Dr. Singh summarized his presentation by concluding that peak IOP, short-term fluctuation, and long-term IOP variability are not proven independent risk factors for glaucoma progression. IOP variability should not be given undue importance, he said, and average IOP over multiple visits should not be discounted as an indicator of control.

Tags:

As a companion to Donald Budenz's presentation the day before about efficacy of the Ahmed and Baerveldt drainage devices, on March 5th Keith Barton of Moorefields Eye Hospital presented one-year data on complication rates found in the Ahmed Baerveldt Comparison (ABC) study, which will continue to the five-year mark.

Mean visual acuity scores were roughly equal between the groups. At one year, roughly 30% of patients in each group lost about two lines of vision, similar to the VA trends in the TVT (tube vs. trabeculectomy) study findings. However, "the vast maj ority were not due to the surgery" but rather glaucoma progression, maculopathy or other disease-based VA loss. Diplopia rates were similar in both groups (10% Ahmed, 8% Baerveldt).

There were relatively few intraoperative complications in either group (mostly hyphema), but more postoperative interventions were required in the Baerveldt group. However, Dr. Barton said the difference was of borderline significance.

The data do show a statistically significantly higher rate of complications (mainly tube occlusions and corneal edema) during the first three months for the Baerveldt patients. After the three-month mark, there were no significant differences between the groups on complication rates. The number of serious complications — defined as either those requiring reoperation or those that involved two or more lines of VA loss — were also significantly more common in the Baerveldt group than the Ahmed group.

Did surgical experience influence outcomes? Dr. Barton (one of the principal investigators) said they attempted to recruit surgeons skilled in the use of both devices but that was a bit of a challenge, as surgeons typically favor one implant and use it predominantly. "Interestingly, lack of experience with one implant or the other didn't seem to make any difference to the outcome," Dr. Barton said, because all surgeons in the study are well-versed in tube procedures. They did find that tube-corneal touch and vitreous hemorrhage occurred more often in surgeons lacking experience with Baerveldt implantation, but not at a level reaching statistical significance.

Correlating the outcomes and complications data, Dr. Barton concluded that the Baerveldt implant lowered IOP slightly more than the Ahmed group, but this was achieved at a higher cost in terms of complications and reoperations.

Future analyses of their data will look at individual substrata of the sample based on glaucoma type and perhaps other factors at initial presentation. 

Tags:

Two studies presented at AGS shed new light on the relative performance of two popular tube shunts. Although previous studies have compared the two, they're typically retrospective and suffer from selection bias. The Ahmed Baerveldt Comparison (or ABC) study, by contrast, is a randomized, unmasked prospective trial with five-year follow-up planned. The ambitious project involves 276 patients at 16 investigational sites spread across four continents. Additional details on the study can be found here .

At AGS, Donald Budenz of the University of Miami presented one-year efficacy data on Thursday, March 4th, and his coinvestigator Keith Barton of Moorefields Eye Hospital presented data on complications the following day.

Baseline IOP averaged 31.2 mm Hg in the Ahmed group (n =143) and 31.8 mm Hg in the Baerveldt group (n=133). At one-year postop, 132 Ahmed patients and 117 Baerveldt maintained their enrollment and were seen for evaluation.

At one day and one week postop, the Ahmed group achieved a statistically significant 7-8 mm greater IOP reduction than the Baerveldt patients. "This might be expected," Dr. Budenz said in his presentation, "since the Baerveldt group had their tubes tied off completely." By one month, however, the Baerveldt group's IOP readings were 2-3 mm lower than Ahmed's and this persisted to the one-year mark. At that visit, average IOP was 15.4mm for the Ahmed group and 13.2mm for Baerveldt.

Failure rates vary based on the criteria used. Probability of failure at one year, as defined by the study's primary criteria, was 16.4% in the Ahmed group and 14% in the Baerveldt group, although the difference was not statistically significant. Using WGA criteria for alternative drainage methods (IOP greater than 17 mm, et al.), the Ahmed group's failure rate was 22% and Baerveldt's was 16%; this difference was also not statistically significant. When using a second, more stringent WGA criterion for failure  — IOP greater than 14 mm — 38.6% of Ahmed and 24% Baerveldt cases would be deemed failures, and that difference was statistically significant. Looking at reasons for failure, Dr. Budenz noted that reoperations were more common in the Ahmed group.

(A recap of Dr. Barton's presentation on complications will be posted later today.)

Tags: