1. COLLABORATIVE NORMAL-TENSION GLAUCOMA STUDY GROUP
Glaucoma conference
COLLABORATIVE NORMAL-TENSION GLAUCOMA STUDY GROUP
R2 황규연/R3 김찬영/Pf.안명덕

2. Glaucomatous Progression
Comparison of
Glaucomatous Progression
Between Untreated Patients With Normal-Tension Glaucoma
and Patients With Therapeutically Reduced Intraocular Pressures
Am J Ophthalmol 1998;126;

3. Purpose IOP is still a contributing factor to optic nerve damage
asymmetric normal-tension glaucoma is often associated with asymmetric intraocular pressure suggests that the level of intraocular pressure, even when statistically normal
it appears that risk factors other than IOP must be present
there is an inverse correlation between intraocular pressure and neural rim area
To ascertain the influence of intraocular pressure level on the course of normal-tension glaucoma
Am J Ophthalmol 1998;126;

4. Purpose prospective multicenter study compared
1) an untreated group of normal-tension glaucoma patients
2) with patients in whom intraocular pressure was lowered by 30%
untreated control group
the aim to learn more about the untreated natural history of the disease
control group : followed up from the baseline state at the time of randomization,
treated group : followed up from a new baseline established as soon as the 30% intraocular pressure reduction was stable (219 ±158 days after randomization)
Am J Ophthalmol 1998;126;

5. PATIENTS AND METHOD
unilateral or bilateral normal-tension glaucoma with optic disk abnormalities and visual field defects
no recorded IOP over 24 mm Hg in either eye
older than 20 years and younger than 90 years
10 baseline IOP readings(After a 4-week wash out of any existing medication)
8:00 AM and 6:00 PM on one day: on other days= 6:4
The median of the 10 readings :20 mm Hg or less
no reading above 24 mm Hg
no more than one reading of 23 or 24 mm Hg
Am J Ophthalmol 1998;126;

6. PATIENTS AND METHOD 3 good baseline visual fields
performed within 1 month
with the Octopus program 32 or the Humphrey Visual Field Analyzer 30-2 (Humphrey- Zeiss, San Francisco, California) full threshold program, including the point of fixation
A pupil diameter: 2.5 mm or greater
reliable field criteria
false positive rate 15% or less
false negative rate 30% or less
fixation loss less than 15%
If the fixation loss <10%,->a false negative rate :30~50%
Minimal visual field defect
A cluster of 3 adjacent points depressed by at least 5 dB from normal age values, with one of these points depressed by at least 10 dB from normal values for age
At least 3 points of such a cluster, including the 10-dB depressed point, were required to be on one side of the horizontal meridian.
There had to be other points elsewhere that were at least 10 dB higher than the densest point in the scotoma.
Am J Ophthalmol 1998;126;

7. PATIENTS AND METHOD Exclusion
taking systemic beta-blockers or clonidine
were unable to perform reliable fields or who had a nonglaucomatous condition that might later affect the visual field.
previous laser treatment, previous ocular surgery(except strabismus surgery), or cyclodestructive procedures
eyes with field defects attributable to nonglaucomatous conditions
narrow anterior chamber angles, corneal abnormalities
BCVA< 20/30
visual fields too damaged to detect further progression reliably
Am J Ophthalmol 1998;126;

8. PATIENTS AND METHOD Am J Ophthalmol 1998;126;487-497
unilateral normal-tension glaucoma,
the eligible eye was followed up until progression occurred and was then randomized
unless fixation was threatened from the outset in which case randomization was performed after baselines were established.
In patients with bilateral disease
In whom fixation was threatened in only one eye
the nonthreatened eye was selected as the study eye and randomized when it met the criteria for randomization.
If fixation was threatened in both eyes
the less affected eye was randomized at the outset.
The less affected eye was determined initially if there was a difference of 3 dB or more between the eyes in the mean of the three baseline mean defects (MDs). If the MD difference was smaller than 3 dB, the eye with the lower median IOP (a difference of 3 mm Hg or greater) was selected.
If the eyes were within these limits of MD and intraocular pressure, then the eye to be included in the study was randomly selected
In patients in whom both eyes were eligible and neither threatened fixation,
both were followed up untreated in the study.
The first eye to progress (as defined by the protocol) was randomized.
If both eyes showed progression simultaneously, the less affected eye, using the criteria outlined, was randomized.
Am J Ophthalmol 1998;126;

9. PATIENTS AND METHOD
Group of IOP reduced 30% from the mean of the last three prerandomization pressure
undergoing filtration surgery, a 20% intraocular pressure reduction was accepted without requiring the patient to undergo a second procedure
neither eye could be treated with beta-adrenergic blockers or adrenergic agonists
Systemic carbonic anhydrase inhibitors could be used only
The goal was to achieve the 30% pressure lowering within 6 months, but in fact, it often took longer.
All randomly assigned patients were monitored for occurrence of visual field progression, change in degree of glaucomatous optic disk damage, or both.
Am J Ophthalmol 1998;126;

10. PATIENTS AND METHOD
The protocol definition of visual field progression
ensured identification of minimal field alterations to minimize any risk to eyes in the untreated control arm of the study.
These criteria
deepening of an existing scotoma
the expansion of an existing scotoma
a new or expanded threat to fixation
a fresh scotoma in a previously normal part of the visual field.
two adjacent points in a baseline defect needed to have declined 10 dB from their initial average of the three baseline values.
The decline must also have been at least three times the short term fluctuation
threshold sensitivity had to have been worse than any value at that location in any of the three baseline visual fields.
A new defect : a cluster of at least 3 points meeting the criteria for a visual field defect occurring in a previously normal part of the field.
A new threat to fixation : one paracentral point that was normal in all three baseline visual fields became sufficiently abnormal to meet the definition of a threat to fixation, or became ncluded within a defect that constituted a pre-existing threat to fixation, thus expanding its boundary.
Am J Ophthalmol 1998;126;

11. PATIENTS AND METHOD optic disk changes occurred, end point was reached
photographs were submitted so that the change could be confirmed by the reading committee.
Both members of the reading committee (D.R.A., S.M.D.) were independently presented with masked sets of stereo disk photographs.
both had to identify correctly the baseline photographs.
end point was reached
by virtue of disk progression or visual field loss (as defined),
All therapeutic constraints were also lifted, and the patients were treated according to the individual clinician’s judgment.
Am J Ophthalmol 1998;126;

12. PATIENTS AND METHOD
to identify an end point in a follow-up visual field relative to the three baseline fields according to the following four-of-five criteria
2 or more points that had changed by at least 10 dB, relative to the average baseline values for these points.
These 2 progressing points
adjacent to each other
both could not be peripheral
not crossing the nasal meridian
the sensitivity at each deteriorating point had to be less than the minimum of the values of this point in each of the three baseline visual fields.
progression : at least 1 of the innermost 4 points showed at least 10-dB deterioration relative to its average value at baseline
Progression was considered to be confirmed when four of five consecutive follow-up fields showed progression relative to baseline fields, with at least one nonperipheral progressing point (or the one central point) being common to all four fields.
Am J Ophthalmol 1998;126;

13. PATIENTS AND METHOD Kaplan-Meier survival analysis
Cox regression analysis
log-transformed intervisit time intervals in the two study groups by means of nested analyses of variance for different years of follow-up and overall
thinned the data in the group with the significantly higher visit frequency by a random mechanism, matching the frequencies, the mean intervisit times, and correspondingvariances
new baselines for the control group with a delay after randomization that was matched to the Corresponding delays in the treated group
To correct for the potential biasing effect caused by the time lag between randomization and stabilization in the treated group
a sensitivity analysis by censoring the data from patients in both groups at the time of the cataract diagnosis and reanalyzing the adjusted data for survival.
Am J Ophthalmol 1998;126;

14. RESULT 230 eye enrolled 140 eyes group No Tx. 79(56%) Tx. 61(44%)
Threat to fixation
50(63%)
42(64%)
Protocol end point
28(35%)
7(12%)
4 of 5 end point
24(30%)
11(18%)
Am J Ophthalmol 1998;126;

15. RESULT Am J Ophthalmol 1998;126;487-497 11 7 2049±129 2688±123 22
1047±139 28
1695±143 11
2255±18 21
1837±168
Am J Ophthalmol 1998;126;

16. RESULT 230 eye enrolled 34 cataract group No Tx. 11(14%) Tx. 23(38%)
surgical
16(26%)
medical
7(11%)
Am J Ophthalmol 1998;126;

17. Discussion Ethical problems
only one eye per patient, enroll the patient’s better eye, to exclude advanced disease
randomize eyes with threats to fixation immediately, rather than to wait for them to show progression
visual field progression to end point was defined to be exceedingly small
when IOP is lowered by 30%, the disease subsequently shows a slower rate of visual field progression than in eyes in which no effort was made to lower intraocular pressure
By topical drug, laser trabeculoplasty, or both
The disease is also often either slow or nonprogressive.
Am J Ophthalmol 1998;126;

18. of Intraocular Pressure Reduction in the Treatment of
The Effectiveness
of Intraocular Pressure Reduction
in the Treatment of
Normal-Tension Glaucoma
Am J Ophthalmol 1998;126;

19. Purpose
how easily and how often such intraocular pressure–lowering can be achieved, as well as the side effects of drugs and complications of surgical intervention required to obtain such a pressure reduction
intent-to-treat analysis of the study data to determine the effectiveness of pressure reduction
Am J Ophthalmol 1998;126;

20. Result
Am J Ophthalmol 1998;126;

21. Discussion discrepancy in the results of the two analyses
primarily attributed to the fact that the baseline visual field threshold values of the treated group were higher at the time of randomization (the baseline used in the present analysis) than the baseline values at the time of pressure stabilization used in the analysis reported in the companion paper
Am J Ophthalmol 1998;126;

22. Discussion
This is particularly important, as assertive measures that may have adverse effects are often needed to lower the intraocular pressure successfully.
Those patients destined to be nonprogressive or only slowly progressive would derive no benefit from treatment but would have been exposed to the risks.
To guide our treatment of patients with normal-tension glaucoma, it would therefore be helpful to find those clinical features that may enable us to distinguish the potential progressors from those whose disease is currently stable.
It should also be noted that although fewer patients progressed in the treated group, there were nevertheless some who continued to progress after a 30% intraocular pressure reduction.
Perhaps, in them, greater intraocular pressure reduction might have been beneficial, but our study interestingly showed no relation between outcome and the pressure levels during the period of observation.
Am J Ophthalmol 1998;126;

23. Normal-Tension Glaucoma
Natural History of
Normal-Tension Glaucoma
Ophthalmology 2001;108:

24. Purpose
an analysis of the natural course of NTG during the time eyes were not receiving therapy, either in the time interval awaiting randomization or after being randomly assigned not to receive treatment to lower the IOP

25. Method
The untreated study group in this report therefore consisted of the following subgroups: (total: 160eyes)
Those never randomly assigned, whose natural course was followed without reaching the protocol criterion for randomization (group NR, “never randomized,” n=56).
Those, not immediately randomly assigned, whose natural course was followed until an event that caused them to be randomly assigned with assignment to receive therapy (group DRT, ultimately “randomized to treatment” but with a delay before randomization, which produced an untreated interval that could be studied, n=24). The untreated course from enrollment to randomization was analyzed.
Those, not immediately randomly assigned, whose natural course included an event that later caused them to be randomly assigned to remain untreated (group DRC, “randomized to control” with the delay of an untreated interval, N=31). Their entire untreated course, both before and after randomization, was included in the analysis.
Those randomly assigned immediately after enrollment on the basis of fixation threat and who were assigned not to receive therapy (group IRC, “immediately randomized” to control, n=49).
Ophthalmology 2001;108:

26. Method
The first three of these groups were not immediately randomized (NIR) and are thus distinguished from the fourth group, immediately randomized (group IRC) because of a fixation threat.
Because the NIR and IRC groups potentially have different natural courses (because of more advanced disease or because the location of visual change is critically close to the center), they were analyzed separately as well as together.

27. Result
Ophthalmology 2001;108:

28. Result
By Kaplan-Meier analysis of all untreated subjects combined, approximately one third showed localized progression within 3 years and about half within 5 to 7 years. 52
2046±103
Ophthalmology 2001;108:

29. Discussion
Some cases of NTG progress more rapidly than others. Although approximately half of cases showed a confirmed localized visual field deterioration by 7 years, the change is typically small and slow.
Ophthalmology 2001;108: