Transcript 16980: Non-Invasive Keratograph Tear Break

Non-Invasive Keratograph Tear Break-Up Time versus Invasive
Fluorescein-Associated Tear Stability Test: Experimental Validation
Study
Nizar S Abdelfattah, M.D., SriniVas R Sadda, M.D., Olivia L Lee, M.D.
Doheny Image Reading Center, Doheny Eye Institute and the Department of Ophthalmology,
David Geffen School of Medicine at UCLA, Los Angeles, CA
Financial Disclosures
Dr. Lee
Dr. Sadda
Dr. Abdelfattah
Consultant: Allergan
Research Grant Support: Allergan
Consultant: Allergan, Genentech, Regeneron, Optos, Carl Zeiss Meditec
Research Grant Support: Optos, Carl Zeiss Meditec, Allergan, Genentech
None
2
Purpose
• To test the validity of the automated noninvasive tear break-up time
(NIKBUT) measurement tool offered by the Oculus Keratograph 5M.
3
Patients & Methods
• Seventy participants (94 eyes) were recruited prospectively over 9 months and divided into
two groups.
• The ocular surface diseases (OSD) group included 63 eyes (74% females, 63.5 ± 14.6
years) diagnosed clinically with dry eye and/or Meibomian gland dysfunction.
• The control group included 31 eyes of healthy volunteers (53 % females, 34.8 ± 10.5
years).
4
Methods
• All eyes were imaged non-invasively with the Oculus K5M (Wetzlar, Germany) using the NIKBUT tool.
• Based on infrared imaging of the tear film over the corneal surface, automated analysis of first break up time (NIKBUT-F) and
average value for break up times (NIKBUT-A) was performed by the instrument’s software.
• Manual assessment of tear break up (NIKBUT-M) was also analyzed, guided by distortion of the Oculus mires reflection on
tear film.
• For comparison, fluorescein was used to stain the tear film and tear break-up time (TBUT) was manually graded based on
videos captured on the Oculus K5M using cobalt blue light.
• All subjects completed the Ocular Surface Disease Index (OSDI) survey.
5
Results
Before Break up
I. First Breakup Time (NIKBUT-F):
This represents the time at which the first breakup of tears takes
place.
II. Average Breakup Time (NIKBUT-A):
This represents the average time at which the all breakup fields
of tears take place during period of video recording.
III. Manual Breakup Time (NIKBUT-M):
This represents the time at which device mires undergo fixed
distortion. This method is not offered automatically by the device
but manually determined by the gradrer.
After Break up
Results
• As expected, strong negative correlation was detected between standard TBUT and
OSDI score (r = 0.8, p < 0.001).
• Poor correlation was detected between all NIKBUT readings and OSDI survey
scores (r = 0.8, p < 0.05)
• Poor correlation was found between non-invasive measurements (NIKBUT-F &
NIKBUT-A) and standard fluorescein TBUT (r = 0.17, p < 0.1 and r = 0.14, p < 0.01
respectively).
7
Results
NIKBUT-F in Controls vs. MGD
9
NIKBUT-A in Controls vs. MGD
11
P = 0.54
8
P = 0.73
10.8
7
10.6
6
10.4
5
MGD
4
Controls
3
10.2
MGD
10
Controls
9.8
2
9.6
1
0
9.4
NIKBUT-First
NIKBUT-Average
There is no statistically significant difference between NIKBUT readings in eyes with MGD vs. controls.
Results
TBUT in Controls vs. MGD
6.8
6.6
6.4
R = 0.70
6.2
6
P < 0.05
MGD
5.8
Controls
5.6
5.4
5.2
5
TBUT
There is a statistically significant difference between TBUT readings in eyes with MGD vs. controls.
And strong negative correlation between OSDI and TBUT.
Results
25
18
16
r = 0.8
p < 0.05
14
r = 0.63
p < 0.01
20
12
15
NIKBUT- 10
First
8
NIKBUTAverage
10
6
4
5
2
0
0
0
5
10
15
NIKBUT-Manual
20
25
0
5
10
15
NIKBUT-First
20
25
Results
Statistically significant strong positive correlation was appreciated
between:
►NIKBUT-F and NIKBUT-M (r = 0.8, p < 0.05)
►NIKBUT-F and NIKBUT-A (r = 0.63, p < 0.01)
►NIKBUT-A and NIKBUT-M (r = 0.74, p < 0.05)
11
Conclusions
• The Oculus Keratograph 5M is able to provide automated tear break up time values without the use of
fluorescein and cobalt blue light.
• However, the automated NIKBUT values are not strongly correlated to either standard fluorescein
TBUT or the OSDI questionnaire.
• The automated measurements of tear break up time do not demonstrate statistically significant
differences between ocular surface disease and normal controls, while the traditional fluorescein
method does.
• Nevertheless, the presence of a correlative pattern among NIKBUT subtypes suggests a potential
differentiating trait amongst ocular surface disease states that can be further explored.
12