Transcript Standard Refractive Outcomes

Quality Control in
Refractive Surgery
Stefan Pieger*, M.Sc.
Wendelstein, Germany
* Ingenieurbüro Pieger GmbH
Nidek Germany Office
Introduction



Personal experience as application specialist
for refractive excimer lasers since 1987.
(Meditec, Schwind, Nidek)
Progress in PRK&LASIK was usually based
on outcomes analysis.
Excimer Laser Surgery and refractive surgery
in general well suited for a systematic
approach on quality control.
Why Quality Control?






Increase confidence level about refractive
procedures offered in your center.
Verify current nomogram settings and make
adjustments if necessary.
Reduce enhancement rate.
Use for marketing and advertisement.
Discover trends and technical problems in
order to react more rapidly.
Fulfill requirements of ophthalmic societies.
How to collect your data?

Patients files

Excel Spread Sheet

Database Software (Access; Filemaker; etc.)

Outcomes Analysis Software
(Datagraph; ASSORT; Refr. Consultant; etc.)
How to analyze refractive data?

Standard Refractive Outcomes


Additional Outcome Parameters



Stability / Safety / Predictability / Efficacy
Astigmatism Outcomes: Surgically Induced change in
Cylinder (SIA); Double Angle Scatter Plot
Defocus Equivalent / Contrast Sensitivity in mesopic
conditions / pre OP BSCVA vs. post OP UCVA
Wavefront Based Outcomes


Defocus + Cyl (‘aberrometer refraction’)
Higher Order RMS / Spherical Aberration (Z12) / Coma /
Trefoil
Standard Refractive Outcomes:
Safety
100%
90%
2 or more lines lost
3 % at 1 m post OP
80%
70%
57%
54%
60%
50%
40%
30%
22%
20%
1 (770)
3 (653)
13%
10%
0%
month (eyes)
25%
19%
1% 0%
lost > 2
2% 3%
2% 0%
lost 2
lost 1
Number of eyes
per Follow up
visit.
unchanged
gained 1
2. SAFETY: Change in BSCVA - Percent
gained 2
0% 0%
gained > 2
Follow up time
interval
Standard Refractive Outcomes:
Efficacy
100%
90%
72% 20/20 or better
3 month post Op
80%
70%
60%
50%
46%
40%
40%
month (eyes)
30%
26%
20%
10%
1 m (691)
18%
3 m (562)
14%
14%
10%
7%
4% 6%
3% 4%
5% 3%
0%
20/12 or
better
20/15
20/20
20/25
Number of eyes
per Follow up
visit.
20/30
4. EFFICACY: UCVA - Percent
20/40
20/50 or
worse
Follow up time
interval
Standard Refractive Outcomes:
Stability (SEQ)
1,00
0,00
pre OP
-1,00
-0,15
-0,20
-0,18
1m
3m
6m
-0,22
12 m
-2,00
± 1 StDev
Follow up time
interval
-3,00
-4,70
-4,00
Mean value of
SEQ
-5,00
-6,00
-7,00
1189
-8,00
744
662
42
1. STABILITY: Achieved Change in Refr. over Time
25
Number of eyes
per Follow up
visit.
Standard Refractive Outcomes:
Predictability (SEQ)
Achieved [D]
PREDICTABILITY: Attempted vs Achieved (Scatter)
744 eyes
15
Number of eyes at follow up
14
overcorrected
13
12
11
10
±1 D ‘happiness’ Zone
9
8
7
Trend line
6
5
y = -0.00x2 + 0.94x + 0.21
4
3
undercorrected
2
Datagraph
1
0
0
1
2
3
4
5
6
7
8
9
10
11 12
13 14
15
Attempted delta SR equiv. [D]
y = -0.00x2 + 0.94x + 0.21
Regression Formula: ‘Achieved
= 0.94*Attempted’
(~6% undercorrection)
Astigmatism Outcomes:
SIA (based on Vector Analysis)
Achieved [D]
Attempted Cyl vs SIA (Scatter)
3 month postOP 490 eyes
-2/-1@180°
7
overcorrected
6
0/-0.5@90°
(-0.5/+0.5@180°)
0.5 D Cyl
0.5 D Cyl
Undercorrection! Overcorrection!
0/-0.5@180°
5
4
3
2
y = -0.00x + 0.90x + 0.04
2
y = 0.90 * x
(~10% undercorrection)
Datagraph
1
undercorrected
0
0
1
2
3
4
5
6
Attempted Cyl [D]
7
Astigmatism Outcomes:
Double Angle Scatter Plot
PreOP Cyl & Axis
Double Angle Scatter Plot
PostOP Cyl & Axis
918 eyes
Double Angle Scatter Plot 3 m postOP 491 eyes
45°
45°
-5
-5
-4
-4
-3
-3
Mean Cyl 0,46 @ 179,6°
Mean Cyl 0,07 @ 14,4°
-2
-2
-1
-1
-0
90°
-5
-4
-3
-2
-1
-0
-1
-2
-3
-4
-5
0°
-5
-4
-3
-2
-1
-0
-1
-1
-2
-2
-3
-3
-4
-4
-5
-5
Datagraph
-0
90°
135°
Datagraph
135°
-1
-2
-3
-4
-5
0°
Additional Outcomes:
pre OP BSCVA vs. post OP UCVA
month (eyes)
3 m (397)
100%
96%
preSCVA (703)
100%
100%
99%
99%
98% 95%
98%
91%
90%
79%
80%
70%
60%
50%
39%
34%
40%
30%
20%
7%
10%
1%
0%
'20/10'
20/12 or
better
1%0%
20/15 or
better
20/20 or
better
20/25 or
better
20/30 or
better
20/40 or
better
preOP BSCVA vs. postOP UCVA - Percent
20/50 or
better
20/60 or
worse
Additional Outcomes:
Defocus Equivalent
99%99%
100%
100%
100%
100%
100%
100%
100%
SEQ
=
SPH + ½ CYL
86%88%
90%
80%
70%
60%
64%
60%
DEQ
=
|SEQ|+|½ CYL|
50%
40%
30%
month (eyes)
20%
1 m (615)
3 m (554)
10%
0%
<=0,5D
<=1D
<=2D
<=3D
<=4D
DEFOCUS EQUIVALENT - Percent
<=5D
Defocus Equivalent vs.
Refractive Outcome
99% 99%
100%100%
-5.0/+10.0@90°
SEQ  Plano
DEQ  +5
92% 92%
100%
90%
80%
76%
72%
70%
60%
99% 99%
50%
100%
40%
90%
30%
80%
20%
70%
10%
60%
0%
+- 0,5
+- 1,0
100% 100%
86% 88%
60%
1 m (615)
3 m (554)
64%
50%
+- 2,0
40%
+- 3,0
Refractive outcome - % within | Attempted |
1 m (744)
3 m (662)
30%
20%
10%
DEQ ~ ‚Blur Circle‘
0%
<=0,5D
<=1D
<=2D
DEFOCUS EQUIVALENT - Percent
<=3D
Additional Outcomes:
Mesopic Contrast Sensitivity
Contrast Sensitivity Mesopic
2,20
2,00
month (eyes)
1,92
1,88
pre op (112)
1,80
1,60
1 m (72)
1,68
1,63
1,61
1,57
1,40
1,22
1,21
1,20
1,00
0,80
0,60
0,40
A (3cpd)
B (6cpd)
C (12cpd)
D (18cpd)
Spatial Frequency [cycles/degree]
VA in LogMar Scale
Normal
Population Range
(Vector Vision
CSV 1000)
Wavefront based Outcomes –
2ndOrder / Coma / Trefoil



„Work in progress“ !
Using Aberrometer Refraction rather
than Manifest Refraction?
Presenting horizontal and vertical
Coma individually?  Vector
calculation to present magnitude and
axis in [D]! (0.5 D Coma @ 230°)
Trefoil: Axis? Present only
magnitude?
f

f
Wavefront based Outcomes –
Higher Order RMS [µm ±StDev]
0.90
Pupil Diameter: 6.0 mm
0.80
0.70
0.53
0.60
0.51
0.54
29
11
3m
6m
0.54
0.50
0.40
0.31
0.30
0.20
0.10
52
0.00
pre OP
26
1m
RMS higher order over time
3
9m
Wavefront based Outcomes –
Spherical Aberration [µm] or [D]
0,10
3,00
Pupil Diameter: 6.0 mm
0,05
2,50 -0,03
0,00
pre OP
-0,05
2,00
1m
3m
9m
1,52
1,36
1,31
-0,10
1,42
1,50
-0,15
-0,23
-0,20
1,00
-0,24
Datagraph
-0,25
-0,25
0,41
0,50
-0,30
-0,35 18
0,00
pre 18
OP
-0,40
13
11
7
11
13
3m
6m
Spherical
Aberration
(Z12)
DEQ
over
time
[D]
Spherical Aberration (Z12) over time [µm]
1m
6
6
9m
Making Outcome-based
Nomogram Adjustments



Comparison of Laser Settings vs. Achieved
change in refraction (and not Attempted vs.
Achieved).
Reduce random errors as far as possible as
nomograms can only compensate systematic
errors!
Must be specific for major laser parameters
like OZ, TZ, ablation profile type as well as
for refraction types.
Nomograms for
Individual Patient
Groups





Data must be filtered on certain parameters:
Refraction Type
(Myp/MyoAsti/Hyp/HypAsti…)
Surgery Type (PRK; LASIK; LASEK;
Custom…)
Optical Zone Diameter
Others (age, laser software version;
humidity…)
Nomogram Improvements –
Laser Settings vs. Achieved
Laser Settings vs Achieved
Achieved [D]
1166 eyes
15
14
overcorrected
13
12
11
10
9
y = -0,01x2 + 1,22x - 0,08
8
7
6
y = -0,01x 2 + 1,22x - 0,08
5
reduce attempted
SEQ by 22%!
4
3
undercorrected
2
Datagraph
1
0
0
1
2
3
4
5
6
7
8
9
10
11 12
13 14
Laser setting [D]
15
Identify and exclude outliers
Laser Settings vs Achieved
Achieved [D]
1166 eyes
15
14
overcorrected
13
12
11
10
9
8
7
6
y = -0,01x 2 + 1,22x - 0,08
5
4
3
undercorrected
2
Datagraph
1
0
0
1
2
3
4
5
6
7
8
9
10
11 12
13 14
Laser setting [D]
15
Laser Setting CYL vs. Surgical
Induced change in Astigmatism
Laser Setting (cyl) vs SIA
Achieved [D]
1480 eyes
7
overcorrected
6
5
4
3
y = 0.03x 2 + 0.60x + 0.26
2
undercorrected
Datagraph
1
0
0
1
2
3
4
5
6
LaserSetCyl [D]
7
1. High Scatter! (further
analysis necessary!)
2. 40 % systematic
undercorrection
Nomograms:
General Comments
1.
2.
3.
4.
5.
Reduce Scatter by Standardized Surgery and OR
Environment
Exclude Outliers from Data Analysis
Exclude Enhancements
Choose appropriate follow up interval (≥ 3 m)
Create Formula („-10%“) / Lookup Table or use
Nomogram Software
Summary




Improving the results of refractive surgery
procedures must be based on an individual
quality control system.
Nomograms can compensate for systematic
errors, but not for random errors.
Modern outcomes analysis software allows
constant monitoring of your results.
Conventional Outcomes will be extended by
HO-RMS, Spherical Aberration and Coma.
Thank You!
www.datagraph-med.com