প্রতিলিপি 🔗
WEBVTT
00:00:00.000 --> 00:00:03.839
Once glaucoma starts progressing, how hard is it to stop?
00:00:04.080 --> 00:00:06.559
A new 12-year study gives an answer.
00:00:06.799 --> 00:00:23.120
A first confirmed visual field progression event is a warning, not a guarantee that vision will keep worsening, but a large, long-running study from Bascombe Palmer Eye Institute suggests that a meaningful share of eyes do progress again, even while receiving routine care.
00:00:23.359 --> 00:00:37.439
Published online in JAMA ophthalmology on October 22, 2026, the study found that recurrence was more likely when vision had been declining quickly before the first event and when average eye pressure remained higher afterward.
00:00:37.600 --> 00:00:43.200
More pressure reduction was associated with less recurrence, but this was an observational study.
00:00:43.359 --> 00:00:47.039
It cannot prove that treatment escalation caused the difference.
00:00:48.479 --> 00:01:00.240
What the Bascombe Palmer study measured, researchers reviewed records from 7,457 eyes belonging to 5,086 people with open-angle glaucoma or suspected glaucoma.
00:01:00.640 --> 00:01:03.920
The records spanned 1998 to 2025.
00:01:04.079 --> 00:01:12.480
Across the full group, average follow-up was 12.4 years, with an average of 12.3 visual field tests per eye.
00:01:12.640 --> 00:01:17.439
A visual field test measures how well a person sees in different parts of their field of vision.
00:01:17.599 --> 00:01:22.239
The first progression analysis included all 7,457 eyes.
00:01:22.480 --> 00:01:24.480
The recurrence analysis was narrower.
00:01:24.640 --> 00:01:36.239
1,951 eyes had an initial likely progression event, but only 635 had enough later visual field tests and eye pressure records for the researchers to assess recurrence.
00:01:36.480 --> 00:01:43.200
The recurrence results therefore describe this smaller, well-monitored group, not every eye in the original registry.
00:01:43.439 --> 00:01:46.079
How the study defined a progression event.
00:01:46.319 --> 00:01:52.719
The researchers used guided progression analysis, a tool for comparing visual field test results over time.
00:01:52.879 --> 00:02:04.400
It compares later tests with the average of the first two reliable tests and looks for worsening at the same test locations beyond the amount expected from normal test-to-test variation.
00:02:04.640 --> 00:02:13.120
An eye met the study's definition of likely progression when three or more of the same locations worsened on three consecutive examinations.
00:02:13.360 --> 00:02:16.800
The event was dated to the first of those three examinations.
00:02:17.039 --> 00:02:28.400
This is a repeatable, computer-based signal, not a guarantee that every flag change represents irreversible glaucoma damage without any possible test or eye health influence.
00:02:28.960 --> 00:02:32.639
To look for a second event, the researchers reset the comparison.
00:02:32.879 --> 00:02:37.520
The first progressing visual field and the next examination became the new baseline tests.
00:02:37.680 --> 00:02:41.120
They then reapplied the same progression rule to the remaining tests.
00:02:41.280 --> 00:02:47.199
The recurrence analysis required at least four visual field tests after the first progressing examination.
00:02:47.439 --> 00:02:49.439
How often did progression recur?
00:02:49.680 --> 00:02:59.439
Across the full cohort, 1,951 of 7,457 eyes, 26.2% had an initial likely progression event.
00:02:59.599 --> 00:03:11.199
Among the 635 eyes with enough subsequent testing and pressure data, 167-26.3% had a recurrent event during the follow-up available to the study.
00:03:11.360 --> 00:03:17.039
The researchers also used a time-to-event method that accounts for eyes having different lengths of follow-up.
00:03:17.280 --> 00:03:23.919
Its estimated cumulative recurrence was, for the full table, please open this article on visualfieldtest.com.
00:03:24.159 --> 00:03:29.919
At the three-year mark, the 95% confidence interval was 11.6% to 17.1%.
00:03:30.479 --> 00:03:37.759
At five years, it was 22.1 to 29.5%, and at 10 years it was 35.8% to 52.1%.
00:03:38.159 --> 00:03:42.479
The number at risk fell from 635 at the start to 30 at 10 years.
00:03:42.639 --> 00:03:50.080
It was 578 at 2 years, 382 at 4 years, 151 at 6 years, and 45 at 8 years.
00:03:50.240 --> 00:03:54.639
The 10-year estimate is therefore much less precise than the earlier estimates.
00:03:55.120 --> 00:03:57.120
These numbers are not contradictory.
00:03:57.280 --> 00:04:03.599
26.3% is the share of the 635 eyes with a recurrence recorded during their available follow-up.
00:04:04.159 --> 00:04:11.520
The 43.5% figure is the estimated cumulative chance by 10 years, accounting for differences in follow-up time.
00:04:11.759 --> 00:04:17.199
It does not mean that 43.5% of all 7,457 eyes recurred.
00:04:17.360 --> 00:04:34.160
The paper also notes that recurrence was recorded more often in the 635 eye analysis than in the broader group of eyes with initial progression, 26.3% versus 15.2%, in part because eyes without enough later tests could not register a recurrence.
00:04:34.399 --> 00:04:36.720
Which eyes were more likely to progress again?
00:04:36.959 --> 00:04:41.279
The strongest signal was how quickly the visual field had already been worsening.
00:04:41.519 --> 00:04:54.240
In the study's adjusted analysis, each 1 decibel per year faster loss in mean deviation, a summary measure of visual field damage, was associated with a 73% higher hazard of recurrence.
00:04:54.480 --> 00:04:58.079
Hazard describes the relative rate of a new event over time.
00:04:58.240 --> 00:05:02.319
It is not the same as an individual's absolute probability of recurrence.
00:05:02.560 --> 00:05:14.399
Other adjusted associations included higher average eye pressure after the first event, each additional 1 mm of mercury, was associated with a 9% higher recurrence hazard.
00:05:14.639 -->
Load full transcript
00:05:23.360
Greater pressure reduction, each additional 10% of reduction from the pre-event average, was associated with an 11% lower hazard.
00:05:23.600 --> 00:05:28.879
The pressure level and percentage reduction were analyzed in separate models because they were related.
00:05:29.040 --> 00:05:32.959
They should not be added together as if they were independent effects.
00:05:33.279 --> 00:05:40.959
Older age, each additional five years, was associated with roughly an 8 to 11% higher hazard, depending on the model.
00:05:41.360 --> 00:05:51.120
Visual field damage at the first event, eyes with a less negative mean deviation score, generally less advanced loss, had a higher measured recurrence hazard.
00:05:51.279 --> 00:05:59.920
The authors cautioned that progression can be harder to detect in advanced disease because damaged test locations have less remaining range to worsen.
00:06:00.480 --> 00:06:06.959
The study's combined prediction model had moderate, not high, ability to distinguish which eyes would recur.
00:06:07.439 --> 00:06:12.639
It is a useful risk signal, not a validated calculator for an individual patient.
00:06:13.040 --> 00:06:15.199
Fast progressors stood out.
00:06:15.600 --> 00:06:27.120
Among 96 eyes losing more than 1 decibel per year before the first event, 42.7% had recurrent progression, compared with 23.4% of slower progressing eyes.
00:06:27.360 --> 00:06:30.399
The difference was about 19 percentage points.
00:06:30.720 --> 00:06:37.199
This does not mean that every fast progressing eye will worsen again or that every slower progressing eye is safe.
00:06:37.360 --> 00:06:41.360
It does mean that the earlier rate of change contains important information.
00:06:41.680 --> 00:06:47.759
A first progression event in an eye that was already losing field quickly deserves particular attention.
00:06:48.000 --> 00:06:51.439
What happened to fast progressing eyes after the first event?
00:06:51.600 --> 00:06:58.879
The study compared the change in mean deviation slope for fast progressing eyes that did and did not have a recorded recurrence.
00:06:59.120 --> 00:07:08.399
Fast progressors without a recorded recurrence slowed from about minus 1.71 to minor.39 decibels per year after the first event.
00:07:08.639 --> 00:07:15.920
Fast progressors with recurrence slowed less, from about 1.70 to minus 1.31 decibels per year.
00:07:16.160 --> 00:07:19.360
The group's pre-event slopes were nearly identical.
00:07:19.600 --> 00:07:34.319
Afterward, the non-recurrence group had a lower average pressure, 14.0 versus 16.2 mm of mercury, and a pressure reduction greater than 20% was more common among them, 36.4% versus 12.2%.
00:07:34.959 --> 00:07:40.240
These are striking observations, but they do not establish why the group's courses diverged.
00:07:40.480 --> 00:07:44.480
Does this show that stronger pressure lowering changed the disease course?
00:07:44.720 --> 00:07:48.800
It is consistent with that possibility, but it is not
proof.
00:07:49.120 --> 00:07:55.120
Pressure lowering is already supported by randomized trials as a way to reduce glaucoma progression.
00:07:55.360 --> 00:08:10.399
In the Baskom-Palmer study, recurrence was also associated with higher post-event pressure, and eyes with a larger pressure reduction tended to fare better, but the study did not randomly assign people to different pressure targets or treatment plans.
00:08:11.040 --> 00:08:13.360
Several other explanations could contribute.
00:08:13.600 --> 00:08:25.759
Regression to the mean, an eye selected because it had a particularly steep prior decline or a triggering progression event may show a less extreme slope later, even without an effective treatment change.
00:08:26.000 --> 00:08:31.839
This is especially relevant when comparing a very fast pre-event period with the period afterward.
00:08:32.159 --> 00:08:33.840
Treatment selection bias.
00:08:34.080 --> 00:08:38.559
Clinicians may intensify treatment for eyes thought to be at greatest risk.
00:08:38.799 --> 00:08:46.720
In observational data, treatment and disease severity are intertwined, and the direction of the resulting bias is not predictable.
00:08:47.039 --> 00:08:48.879
Unknown treatment and adherence.
00:08:49.039 --> 00:08:56.559
The investigators did not have records of medication changes, laser treatment, surgery, or how consistently treatment was used.
00:08:56.799 --> 00:09:04.000
A lower recorded pressure could reflect any mix of treatment changes, improved adherence, or ordinary variation.
00:09:04.320 --> 00:09:05.600
Selection and follow-up.
00:09:06.000 --> 00:09:14.320
Only 635 of the 1,951 eyes with initial progression had the post-event tests and pressure data needed for this analysis.
00:09:14.480 --> 00:09:21.279
Eyes that remain in care and receive more tests also have more opportunity for another event to be detected.
00:09:21.759 --> 00:09:26.159
Long-term estimates depend on assumptions about eyes that leave follow-up.
00:09:26.559 --> 00:09:29.519
So the study supports a clinically plausible message.
00:09:29.759 --> 00:09:38.799
Pressure reduction may help after progression is detected, but it does not show that a particular escalation caused the slowing scene in one subgroup.
00:09:38.960 --> 00:09:41.679
What does the pressure after the event tell us?
00:09:41.919 --> 00:09:48.399
The proportion of eyes with a recorded recurrence rose across the study's categories of average post-event pressure.
00:09:48.639 --> 00:09:53.120
For the full table, please open this article on VisualFieldTest.com.
00:09:53.279 --> 00:10:02.080
These are crude recurrence proportions by pressure category, not proof of a safe cutoff or a guarantee that a particular pressure prevents damage.
00:10:02.320 --> 00:10:11.200
The analysis found a graded association, but some eyes still progressed even when their recorded average was 12 mm of mercury or lower.
00:10:11.440 --> 00:10:20.320
A low clinic average does not necessarily
capture pressure peaks, pressure throughout the day and night, or whether treatment was consistently used.
00:10:20.639 --> 00:10:27.840
Nor does this study establish that all ongoing damage at a low measured pressure is truly independent of pressure.
00:10:28.080 --> 00:10:29.919
Other factors may matter too.
00:10:30.080 --> 00:10:34.240
Blood pressure patterns and blood flow to the optic nerve have been studied.
00:10:34.480 --> 00:10:45.759
For example, one prospective study in people with normal tension glaucoma found an association between prolonged nighttime blood pressure dips and later visual field loss.
00:10:46.000 --> 00:10:55.840
But those factors were not measured in the Bascom Palmer recurrence study, and such findings do not prove that changing blood pressure prevents glaucoma progression.
00:10:56.080 --> 00:10:59.519
Patients should not change blood pressure medicines on their own.
00:10:59.679 --> 00:11:04.879
Concerns belong in a discussion with their eye doctor and prescribing clinician.
00:11:06.000 --> 00:11:11.039
Pressure-independent injury and neuroprotection remain important research questions.
00:11:11.360 --> 00:11:20.000
The European Glaucoma Society's current guidance says there is not enough evidence to recommend neuroprotective agents as glaucoma treatment.
00:11:20.240 --> 00:11:27.200
Pressure lowering remains the established treatment approach, even though it may not stop progression in every eye.
00:11:28.159 --> 00:11:30.639
What do the major glaucoma trials add?
00:11:30.799 --> 00:11:37.679
The Bascombe Palmer study asks a specific question: what happens after an eye has already shown likely progression?
00:11:37.919 --> 00:11:47.120
The major trials help assess the broader claim that lowering eye pressure can slow glaucoma, but most studied initial treatment rather than recurrence after a first event.
00:11:47.440 --> 00:11:49.519
Early manifest glaucoma trial.
00:11:49.679 --> 00:11:54.080
In people with early glaucoma, treatment lowered average pressure by about 25%.
00:11:54.720 --> 00:12:02.399
Over a median of six years, progression occurred in 45% of treated participants and 62% of controls.
00:12:03.200 --> 00:12:11.440
This is randomized evidence that treatment can delay progression, but it was not a trial of a particular pressure reduction after a first progression event.
00:12:11.679 --> 00:12:16.399
Early manifest glaucoma trial, United Kingdom glaucoma treatment study.
00:12:16.559 --> 00:12:27.840
In a randomized placebo-controlled study of open-angle glaucoma, average pressure reduction at 24 months was 3.8 mm of mercury with latanoprost and 0.9 with placebo.
00:12:28.080 --> 00:12:32.480
The hazard of visual field deterioration was lower in the latanoprost group.
00:12:32.639 --> 00:12:37.519
Again, this supports pressure lowering, not a precise post-progression target.
00:12:38.240 --> 00:12:43.120
United Kingdom Glaucoma Treatment Study, Advanced Glaucoma Intervention Study.
00:12:43.360 --> 00:12:57.200
In people with advanced glaucoma after surgery, eyes with pressure below 18 mm of mercury at all visits had little average visual field worsening, while less consistent pressure control was associated with more worsening.
00:12:58.000 --> 00:13:05.759
Pressure levels were not randomly assigned, so this supports an association, not a universal below 18 is safe rule.
00:13:06.080 --> 00:13:08.399
Advanced glaucoma intervention study.
00:13:08.799 --> 00:13:11.600
Collaborative initial glaucoma treatment study.
00:13:11.759 --> 00:13:17.440
This trial compared initial medicine with surgery and used eye-specific pressure targets.
00:13:17.600 --> 00:13:25.360
Even with aggressive treatment, substantial worsening of mean deviation was observed in more than one-fifth of participants by eight years.
00:13:25.600 --> 00:13:30.399
Treatment effects varied with factors including the amount of damage at the start.
00:13:30.720 --> 00:13:33.519
Collaborative initial glaucoma treatment study.
00:13:33.759 --> 00:13:36.399
Collaborative normal tension glaucoma study.
00:13:36.799 --> 00:13:44.960
This trial is especially relevant because participants were selected for documented progression, high-risk defects, or other warning signs.
00:13:45.200 --> 00:13:48.159
A 30% pressure reduction goal was studied.
00:13:48.399 --> 00:13:58.480
The results favored lowering pressure, but interpretation of visual field outcomes was complicated by cataracts, which were more common after some surgical treatments.
00:13:58.799 --> 00:14:01.759
Collaborative Normal Tension Glaucoma Study.
00:14:02.000 --> 00:14:13.919
Together, these studies support lower pressure as a way to reduce risk while showing that it does not guarantee stability and that treatment targets cannot be reduced to one number for everyone.
00:14:14.159 --> 00:14:17.919
Should the target be based on the person's own rate of progression?
00:14:18.159 --> 00:14:18.799
Yes.
00:14:19.039 --> 00:14:24.240
Once a reliable personal rate is available, it should carry substantial weight.
00:14:24.480 --> 00:14:28.559
A percentage reduction or pressure formula can be a useful starting point.
00:14:28.720 --> 00:14:33.200
It cannot tell whether that pressure is adequate for a particular optic nerve.
00:14:33.519 --> 00:14:40.960
The American Academy of Ophthalmology says target pressure should be individualized and adjusted as the disease course becomes clear.
00:14:41.120 --> 00:14:50.799
The European Glaucoma Society likewise emphasizes that there is no single target appropriate for every patient and includes progression rate among the factors to consider.
00:14:51.200 --> 00:14:53.919
American Academy of Ophthalmology Guidance.
00:14:54.240 --> 00:14:56.559
European Glaucoma Society Guidelines.
00:14:56.799 --> 00:14:58.639
The practical distinction is important.
00:14:58.799 --> 00:15:03.120
A target is a working estimate, not a promise that damage will stop.
00:15:03.440 --> 00:15:08.159
If
repeated, reliable testing shows continued loss at that target.
00:15:08.399 --> 00:15:10.399
The estimate may need to be reconsidered.
00:15:10.559 --> 00:15:18.639
The rate of loss, current damage, pressure history, life expectancy, the other eye, and the risks and burdens of further treatment all matter.
00:15:18.799 --> 00:15:21.440
A simple model of different prior rates.
00:15:21.840 --> 00:15:29.279
The study reported a 1.73-fold higher recurrence hazard for each 1 decibel per year faster prior loss.
00:15:29.519 --> 00:15:36.879
The table below applies that model mathematically, comparing the three example rates while holding other measured factors constant.
00:15:37.200 --> 00:15:41.360
For the full table, please open this article on VisualFieldTest.com.
00:15:41.519 --> 00:15:49.600
Under this model, the eye losing 0.7 decibels per year has about a 1.3 times hazard compared with the eye losing 0.2.
00:15:49.919 --> 00:15:53.360
The eye losing 1.5 has about twice the hazard.
00:15:53.600 --> 00:16:02.159
These are relative model estimates, not absolute probabilities, validated personal forecasts, or proof that a specific treatment will offset the risk.
00:16:02.399 --> 00:16:06.320
The study's overall prediction model had only moderate discrimination.
00:16:07.039 --> 00:16:10.159
How much additional pressure reduction might be considered?
00:16:10.320 --> 00:16:16.799
There is no randomized trial establishing one universal extra pressure reduction after a confirmed event.
00:16:17.039 --> 00:16:25.279
In the Bascom Palmer study, mean pressure was 16.4 mm of mercury before the first event and 14.5 afterward.
00:16:25.440 --> 00:16:29.919
The average percentage reduction, calculated eye by eye, was 8.7%.
00:16:30.879 --> 00:16:36.399
Each additional 10% of reduction was associated with an 11% lower recurrence hazard.
00:16:36.559 --> 00:16:40.720
That is evidence of an association, not a guaranteed treatment effect.
00:16:41.279 --> 00:16:51.120
As a population level illustration only, the study's model would associate a 10% point greater total reduction with a hazard ratio of 0.89.
00:16:51.519 --> 00:16:57.039
Extending the same model mathematically to 20% points gives about 0.79.
00:16:57.519 --> 00:17:02.080
Those calculations assume the observed relationship continues in a straight line.
00:17:02.240 --> 00:17:07.200
The study did not randomize those reductions, so they should not be read as causal promises.
00:17:07.519 --> 00:17:16.640
Earlier European guidance described a further 20% reduction as one possible response when progression threatened quality of life despite a target being reached.
00:17:16.880 --> 00:17:22.400
The current guidance emphasizes individualized targets rather than a universal increment.
00:17:22.640 --> 00:17:33.759
At the study's average post-event pressure of 14.5, a further 10% to 20% would equal roughly 1.5 to 2.9 mm of mercury arithmetically.
00:17:34.720 --> 00:17:37.680
That is not a recommended target for any individual.
00:17:37.839 --> 00:17:46.720
The appropriate amount, if any, depends on the eye's measured course, pressure history, existing damage, treatment tolerance, and clinical judgment.
00:17:46.960 --> 00:17:51.519
If your glaucoma has already progressed once, what matters most next?
00:17:51.839 --> 00:17:59.440
For predicting another detected progression event in this study, the factors can be ranked practically, with an important caveat.
00:17:59.680 --> 00:18:06.240
The model was not designed to prove that the items can be ranked precisely against one another because their units differ.
00:18:06.480 --> 00:18:08.319
Previous progression speed.
00:18:08.559 --> 00:18:10.480
This was the clearest warning signal.
00:18:10.720 --> 00:18:17.519
Each 1 decibel per year faster prior loss was associated with a 73% higher recurrence hazard.
00:18:17.839 --> 00:18:22.240
Fast progressors had recurrence recorded more often than slower progressors.
00:18:22.559 --> 00:18:24.319
Current average eye pressure.
00:18:24.559 --> 00:18:31.039
Each additional 1 mm of mercury after the first event was associated with a 9% higher hazard.
00:18:31.599 --> 00:18:33.519
Percentage pressure reduction.
00:18:33.759 --> 00:18:39.119
Each additional 10% of reduction was associated with an 11% lower hazard.
00:18:39.359 --> 00:18:45.839
This was assessed in a separate model from current pressure, so it is not an independent extra benefit to add on top.
00:18:46.079 --> 00:18:46.720
Age.
00:18:47.119 --> 00:18:52.559
Older age was independently associated with recurrence, although the increase per five years was modest.
00:18:52.799 --> 00:18:55.119
Damage severity at the first event.
00:18:55.440 --> 00:19:05.759
In this study, less advanced damage was associated with more recorded recurrence, likely in part because further change is harder to detect in very damaged test locations.
00:19:06.240 --> 00:19:09.279
That does not mean advanced glaucoma is less serious.
00:19:09.599 --> 00:19:13.440
Advanced damage can make any additional loss more consequential.
00:19:13.680 --> 00:19:18.559
The most useful predictor in this study was the eye's previous rate of visual field loss.
00:19:18.799 --> 00:19:22.720
Current pressure and the amount of pressure reduction also mattered.
00:19:23.039 --> 00:19:30.400
The central lesson is not that every eye should be driven to the same pressure, or that progression can always be stopped.
00:19:30.799 --> 00:19:40.640
It is that after a reliable progression event, the eye's own history should help determine how closely it is watched and whether its treatment target still makes sense.
00:19:41.359 --> 00:20:11.359
Sources Moloch and Colleagues, JAMA Ophthalmology, October 22, 2026, Rialini, Invited Commentary in JAMA Ophthalmology Early Manifest Glaucoma Trial, United Kingdom Glaucoma Treatment Study, Advanced Glaucoma Intervention Study, Collaborative Initial Glaucoma Treatment Study, Collaborative Normal Tension Glaucoma Study, American Academy of Ophthalmology, Primary Open Angle Glaucoma, Preferred Practice Pattern, European Glaucoma Society, 6th Edition Guidelines.
00:20:11.599 --> 00:20:15.440
All links to sources are available in the text version of this article.
00:20:15.680 --> 00:20:19.599
You can find the full article at Visual Fieldtest.com.
00:20:19.839 --> 00:20:20.720
Thanks for listening.
00:20:20.880 --> 00:20:26.799
To check your visual field, click the link at the bottom of this article or visit Visualfieldtest.com.