- Email: [email protected]
Clinical performance of a handheld digital infrared monocular pupillometer for measurement of the dark-adapted pupil diameter
ARTICLE
Clinical performance of a handheld digital infrared monocular pupillometer for measurement of the dark-adapted pupil diameter Jay C. Bradley, MD, Karl C. Bentley, MD, Aleem I. Mughal, MD, Sandra M. Brown, MD
PURPOSE: To compare the accuracy of a handheld infrared digital pupillometer and digital infrared photography for measurement of the dark-adapted pupil diameter. SETTING: Department of Ophthalmology and Visual Sciences, Texas Tech University Health Sciences Center, Lubbock, Texas, USA. METHODS: The right horizontal pupil diameter in healthy volunteers was measured using a NeurOptics PLR-200 pupillometer and then videographed using the infrared function of a CyberShot video camera after 2 minutes and 5 minutes dark adaptation at 1 lux ambient illumination. The best still image was extracted from the video file, and the horizontal pupil diameter was determined by comparison against an internal photographic length standard using digital image software. Accommodation and alertness were controlled during testing. RESULTS: The mean horizontal pupil diameter by infrared photography after 2 minutes of dark adaptation by subject age was 7.71 mm for ages 20 to 29 years, 6.80 mm for ages 30 to 39 years, 6.53 mm for ages 40 to 49 years, 5.94 mm for ages 50 to 59 years, and 6.01 mm for ages 60 to 69 years. The mean difference (infrared photography minus pupillometer) was C0.09 mm (range C0.30 to 0.14 mm) at 2 minutes of adaptation and C0.07 mm (range C0.25 to 0.13 mm) at 5 minutes. CONCLUSIONS: The pupillometer accurately measured the horizontal pupil diameter at 1 lux, with no measurement more than 0.3 mm different from infrared photography measurements. The pupillometer had a slight negative bias that is unlikely to introduce an error greater than 0.5 mm in clinical measurements. Financial Disclosure: No author has a financial or proprietary interest in any material or method mentioned. J Cataract Refract Surg 2010; 36:277–281 Q 2010 ASCRS and ESCRS
Surgeons performing corneal refractive surgery or implanting multifocal intraocular lenses (IOLs) are interested in the patient’s dark-adapted pupil diameter, which can influence the choice of ablation optical zone diameter or IOL design. Corneal refractive surgery is the most common elective operation in the United States, and cataract surgery is one of the most common operations in the elderly. Therefore, a compact and portable pupillometer of modest cost is valuable. Several pupillometers are commercially available, with the Colvard pupillometer (Oasis Medical, Inc.) probably the most widely used in the U.S. Although several studies1–7 have compared pupillometers, only the Colvard pupillometer has been compared with the laboratory method of digital infrared photography.8 Q 2010 ASCRS and ESCRS Published by Elsevier Inc.
The purpose of this study was to compare the clinical performance of a handheld digital infrared monocular pupillometer and the laboratory standard of digital infrared photography. SUBJECTS AND METHODS This observational cohort study was approved by an institutional review board. All subjects provided informed consent before participating. Volunteers with no history of eye disease, injury, or surgery were recruited from friends and colleagues of the investigators, medical students, and clinical and administrative staff of the Department of Ophthalmology and Visual Sciences, Texas Tech University, Lubbock, Texas, USA. For preliminary retinal adaptation, subjects donned wraparound ‘‘cataract’’ sunglasses for a minimum of 5 minutes before testing. After the subject entered the testing room, 0886-3350/10/$dsee front matter doi:10.1016/j.jcrs.2009.09.025
277
278
CLINICAL PERFORMANCE OF HANDHELD INFRARED PUPILLOMETER
dark adaptation progressed for 2 minutes at 1 lux ambient illumination, which was monitored with an industrial light meter (model CA814, AMEC Instruments) placed near the subject’s chair. Volitional accommodation during measurement was controlled by instructing the subject to gaze at the far end of the examination room (15 feet), which contained no point light sources. At 2 minutes (measured by a stopwatch), an alertness test was administered by asking the subject to identify the number of odd numbers in a sequence of 10 numbers. Infrared digital videography of the right pupil was performed using a CyberShot DCRTRV740 digital Handycam video camera (Sony); a small strip of printed ruler was taped to the right lower lid as close to the eyelash line as possible to provide an internal photographic width standard (Figure 1). A NeurOptics PLR-200 pupillometer (NeurOptics, Inc.) was then used to measure the horizontal pupil diameter according to the manufacturer’s instructions. Additional dark adaptation was performed for 5 minutes, and the measurements were repeated. All clinical testing was performed by the same ophthalmologist (K.C.B.) in the same standardized testing suite. The videography files were reviewed and the clearest still images extracted. Although blinded to the pupillometry data, the same ophthalmologist (J.C.B.) measured the horizontal pupil diameter with digital image software, using the ruler as an internal length standard. Additional details of the infrared photography and digital image analysis process have been published.9,10 This method has sufficient precision for the current study, with a mean interobserver difference of 0.03 mm.9 The handheld infrared pupillometer in this study uses monocular occlusion, with the tested eye occluded by a rubber cup. The pupillometer is self-calibrating and focuses automatically. It acquires 14 images of the pupil over 3 seconds, discards outliers to account for hippus, and provides an average value. The software measures the best circular fit of the pupil margin and automatically adjusts for vertex distance. During measurement, subjects were instructed not to look at the examiner’s face but rather to maintain distance fixation. The pupillometry data were compared with the infrared photography data using Bland-Altman graphs. A difference of G0.5 mm was considered clinically significant (ie, an error that could affect surgical planning).
RESULTS Of the 58 subjects recruited and tested, 8 were excluded due to excessive ruler slant (2), video decentration (1),
Submitted: July 2, 2009. Final revision submitted: August 26, 2009. Accepted: September 8, 2009. From the Department of Ophthalmology and Visual Sciences (Bradley, Bentley, Mughal), Texas Tech University Health Sciences Center, Lubbock, Texas, and a private eye center (Brown), Concord, North Carolina, USA.. Corresponding author: Sandra M. Brown, MD, 201 LePhillip Court Northeast, Concord, North Carolina 28025, USA. E-mail: [email protected].
Figure 1. Example of an infrared photograph extracted from a digital video file showing the internal length standard. The distance between the white hash marks is 5.0 mm. The ‘‘3’’ is the subject number.
pupil corectopia (1), and excessive hippos or accommodative miosis (4). The mean age of the final cohort of 50 subjects (28 men, 22 women) was 35 years (median 30.5; range 19 to 62 years). Table 1 shows the results of testing by age decade. Figure 2 shows the distribution of horizontal darkadapted pupil diameter measured by infrared photography after 2 minutes of dark adaptation as a function of age. Figure 3 shows a comparison between pupillometry and infrared photography after 2 minutes and after 5 minutes of dark adaptation results using Bland-Altman analysis. The mean difference was C0.094 mm (range 0.14 to C0.30 mm) after 2 minutes and C0.071 mm (range 0.13 to C0.25 mm) after 5 minutes. Figure 4 shows the distribution of the difference between pupillometry and infrared photography. Table 1 shows the difference in dark-adapted pupil diameter between 2 minutes and 5 minutes of dark adaptation as a function of age; the difference is plotted in Figure 5. DISCUSSION Until the advent of corneal laser refractive surgery, there was little interest in sophisticated office measurement of a patient’s pupil diameter under low-light conditions. The Rosenbaum pupil card sufficed because anisocoria was of greater diagnostic interest than the absolute value of the pupil diameter. The first lightweight, handheld infrared pupillometer was the Colvard, which was marketed in 1998.11 In our laboratory, we found this device to have a steep learning curve and excessive user dependence,8 making it inaccurate, especially in unknowledgeable hands. The Procyon pupillometer1–6,12–15 (Procyon, Ltd.) was released in approximately 2002. This tabletop device
J CATARACT REFRACT SURG - VOL 36, FEBRUARY 2010
279
CLINICAL PERFORMANCE OF HANDHELD INFRARED PUPILLOMETER
Table 1. Effects of age and duration of dark adaptation on horizontal pupil diameter. Horizontal Pupil Diameter (mm) Difference 5 Minutes 2 Minutes
2 Minutes Adaptation Infrared Photography Age (y) 0–19 20–29 30–39 40–49 50–59 60–69 All
Pupillometry
Infrared Photography
Eyes (n)
Mean
Range
Mean
Range
Mean
Range
1 22 10 7 7 3 50
6.83 7.71 6.80 6.53 5.94 6.01 7.00
d 5.97 to 8.80 5.16 to 8.78 4.79 to 7.24 4.31 to 7.33 5.67 to 6.25 4.31 to 8.80
6.80 7.60 6.68 6.47 5.87 5.98 6.90
d 6.00 to 8.60 5.30 to 8.70 4.80 to 7.20 4.40 to 7.20 5.80 to 6.10 4.40 to 8.70
0.03 0.045 0.064 0.016 0.000 0.005 0.039
d 0.229 to 0.140 0.318 to 0.103 0.260 to 0.162 0.081 to 0.074 0.005 to 0.027 0.229 to 0.162
was significantly more expensive than the Colvard pupillometer but had the stated advantages of minimal user dependence and hardcopy output. Despite a high degree of measurement precision (2 decimal places), the machine was clinically inaccurate16 and puzzlingly user dependent, with an interobserver difference at 0.4 lux of 1.8 to C2.4 mm.17 The Procyon pupillometer also had a profound negative bias, with a mean difference compared with the pupillometer we used in our study (Procyon minus NeurOptics) of 0.76 mm at 0.04 lux and 2.10 mm at 0.40 lux.17 Some authors use a wavefront analyzer or other ‘‘chin on the paper, forehead against the strap’’ machines to measure the pupil diameter.2,3,15,18–20 This method does not allow distance fixation and may induce accommodation, even in myopic patients, through psychogenic awareness of near. Because measuring pupil diameter is important, it would be preferable to have a dedicated device to perform the test.21,22 We found the NeurOptics PLR-200 pupillometer to have clinically satisfactory accuracy when compared with the laboratory standard of infrared photography. We defined an unsatisfactory difference as G0.5 mm
Figure 2. Horizontal dark-adapted pupil diameter measured by infrared photography after 2 minutes of dark-adaptation at 1 lux as a function of age (DAPD Z dark-adapted pupil diameter).
for any given subject. No pupillometer reading fell outside this range. After 2 minutes of dark adaptation, 41 (82%) of 50 of readings were less than G0.2 mm different from the infrared photography reading and the maximum difference (infrared photography minus pupillometry) was C0.30 mm. After 5 minutes, 47 readings (94%) were less than G0.2 mm different and the maximum difference was C0.25 mm. Photographic measurements were more often slightly larger
Figure 3. Bland-Altman plot of the horizontal dark-adapted pupil diameter measured by infrared photography and by the pupillometer after 2 minutes and 5 minutes of dark adaptation at 1 lux (DAPD Z dark-adapted pupil diameter; IRP Z infrared photography).
J CATARACT REFRACT SURG - VOL 36, FEBRUARY 2010
280
CLINICAL PERFORMANCE OF HANDHELD INFRARED PUPILLOMETER
Figure 4. Distribution of the difference between the horizontal darkadapted pupil diameters measured by infrared photography and by the pupillometer (DAPD Z dark-adapted pupil diameter; IRP Z infrared photography).
Figure 5. Difference between the horizontal dark-adapted pupil diameters measured with infrared photography after 2 minutes and 5 minutes of dark adaptation at 1 lux as a function of age (20 Z 2 minutes; 50 Z 5 minutes; DAPD Z dark-adapted pupil diameter; IRP Z infrared photography).
(2 minutes adaptation: n Z 36, 72%; 5 minutes adaptation: n Z 38, 76%) than pupillometry measurements. The clinician’s goal is to measure the stable darkadapted pupil diameter at a predetermined level of low ambient illumination. Note that dark has no specific definition in the literature.23 Also, there is no such thing as scotopic illumination because the term scotopic refers to a state of retinal adaptation (in which cone photoreceptors are entirely quiescent) and the level of illumination that corresponds to scotopic adaptation varies between individuals.24 Room illumination, the duration of dark adaptation, accommodation, and alertness must all be controlled if the measured pupil diameter is to represent the dark-adapted pupil diameter. No device to measure pupil diameter can overcome inadequate methodology.25 We used 1 lux ambient illumination in our study because it is typical of night-driving conditions on residential side streets or unlighted highways in the U.S.10 A camera (photographic) light meter cannot be used to determine room illumination because its purpose is to measure reflected light and the readout is not in lux.26 To achieve room illumination at or near 1 lux requires that all fluorescent fixtures be turned off and incandescent lights be dimmed as low as possible. Practically speaking, turning off all lights except the reading lamp on the slitlamp stand, turning this light as low as possible, and aiming it toward the intersection of the wall and the ceiling behind the patient can achieve diffuse ambient illumination close to this low level. If you can read a medical journal in your examination room, you have to find other ways to further reduce the light level. It is also important to eliminate focal light sources from the pupil testing room. Hallway light entering under the door can have a marked effect on the stability of the pupil diameter and must be blocked in some fashion. We placed a blanket across the bottom of the
door; more decorative solutions, such as a door draftstopper, are available. In this study, we found that 2 minutes of dark adaptation was sufficient to achieve stable retinal adaptation to 1 lux ambient illumination. No horizontal pupil diameter was more than 0.25 mm larger after longer adaptation. However, we used a preliminary adaptation in which the subject wore dark wraparound cataract glasses for a minimum of 5 minutes. This was necessary because many of our subjects continued with regular work activities under fluorescent lights before testing. In a customary clinical setting, examination rooms are often dimly lit (although much brighter than 1 lux). Having the subject spend 2 to 3 minutes in the examination room before further reducing the light level should be sufficient to obtain stable retinal adaptation after 2 minutes at 1 lux. It is crucial that tests that cause photoreceptor bleaching, such as corneal topography, pupil reactivity, or slitlamp examination, be performed after the dark-adapted pupil diameter is measured. Even testing visual acuity can cause rod bleaching from the white background of the projected chart. Surgeons must consider these issues in the context of their customary clinical flow and establish a clear stepwise protocol for the examination process, especially if it is delegated in large part to technicians. Because there is a strong correlation between age and dark-adapted pupil diameter,22,23 an outlier measurement (eg, a 5.5 mm pupil diameter in a 20-year-old patient) should be repeated. An additional question is whether monocular occlusion (as with the NeurOptics and Colvard pupillometers) changes the dark-adapted pupil diameter compared with that obtained under natural binocular viewing conditions.24 In theory, total monocular occlusion should increase the dark-adapted pupil diameter because photoreceptor stimulation is halved. That the nonoccluded infrared photography measurements
J CATARACT REFRACT SURG - VOL 36, FEBRUARY 2010
CLINICAL PERFORMANCE OF HANDHELD INFRARED PUPILLOMETER
were more frequently larger than the pupillometry measurements in our study suggests that this effect is subclinical at 1 lux illumination or that it was lost in the mild negative bias of the pupillometer relative to infrared photography. It may be clinically detectable at higher light levels. Alternatively, monocular occlusion may induce involuntary accommodative convergence through awareness of the nearness of the examiner, even when subjects are instructed to maintain distance fixation. We found the NeurOptics PLR-200 pupillometer to be an easy-to-use, accurate, portable, and relatively inexpensive device for measurement of horizontal pupil diameter under reduced ambient illumination. It requires almost no user intervention (eg, focusing) or user image interpretation (eg, reading a reticule). Hard copy output is possible through wireless transmission to a printer. Compared with the laboratory standard of infrared photography, it may have a slight negative bias on average that is not clinically significant. Surgeons can be confident that if a correct darkadaptation protocol at 1 lux is followed, the pupillometer measurement will be within G0.5 mm of the photographed dark-adapted pupil diameter and will likely represent the patient’s true pupil diameter under night-driving conditions. REFERENCES 1. Bootsma S, Tahzib N, Eggink F, de Brabander J, Nuijts R. Comparison of two pupillometers in determining pupil size for refractive surgery. Acta Ophthalmol Scand 2007; 85:324–328 2. McDonnell C, Rolincova M, Venter J. Comparison of measurement of pupil sizes among the Colvard pupillometer, Procyon pupillometer, and NIDEK OPD-scan. J Refract Surg 2006; 22:S1027–S1030 3. Schmitz S, Krummenauer F, Henn S, Dick HB. Comparison of three different technologies for pupil diameter measurement. Graefes Arch Clin Exp Ophthalmol 2003; 241:472–477 4. Kohnen T, Terzi E, Bu¨hren J, Kohnen E-M. Comparison of a digital and a handheld infrared pupillometer for determining scotopic pupil diameter. J Cataract Refract Surg 2003; 29:112–117 5. Kohnen T, Terzi E, Kasper T, Kohnen E-M, Bu¨hren J. Correlation of infrared pupillometers and CCD-camera imaging from aberrometry and videokeratography for determining scotopic pupil size. J Cataract Refract Surg 2004; 30:2116–2123 6. Michel AW, Kronberg BP, Narva´ez J, Zimmerman G. Comparison of 2 multiple-measurement infrared pupillometers to determine scotopic pupil diameter. J Cataract Refract Surg 2006; 32:1926–1931 7. Schnitzler E-M, Baumeister M, Kohnen T. Scotopic measurement of normal pupils: Colvard versus Video Vision Analyzer infrared pupillometer. J Cataract Refract Surg 2000; 26:859–866 8. Bradley JC, Anderson JE, Xu KT, Brown SM. Comparison of Colvard pupillometer and infrared digital photography for measurement of the dark-adapted pupil diameter. J Cataract Refract Surg 2005; 31:2129–2132
281
9. Brown SM, Khanani AM, Xu KT. Day to day variability of the dark-adapted pupil diameter. J Cataract Refract Surg 2004; 30:639–644 10. Khanani AM, Brown SM, Xu KT. Six-month variability of the dark-adapted pupil diameter. J Cataract Refract Surg 2005; 31:987–990 11. Colvard M. Preoperative measurement of scotopic pupil dilation using an office pupillometer. J Cataract Refract Surg 1998; 24:1594–1597 12. Cheng ACK, Rao SK, Cheng LL, Lam DSC. Assessment of pupil size under different light intensities using the Procyon pupillometer. J Cataract Refract Surg 2006; 32:1015–1017 13. Kurz S, Krummenaer F, Pfeiffer N, Dick HB. Monocular versus binocular pupillometry. J Cataract Refract Surg 2004; 30:2551–2556 14. Rosen ES, Gore CL, Taylor D, Chitkara D, Howes F, Kowalewski E. Use of a digital infrared pupillometer to assess patient suitability for refractive surgery. J Cataract Refract Surg 2002; 28:1433–1438 15. Wickremasinghe SS, Smith GT, Stevens JD. Comparison of dynamic digital pupillometry and static measurements of pupil size in determining scotopic pupil size before refractive surgery. J Cataract Refract Surg 2005; 31:1171–1176 16. Brown SM, Bradley JC. Pupil size in refractive surgery candidates [letter]. J Refract Surg 2005; 21:303 17. Schallenberg M, Bangre V, Steuhl K-P, Kremmer S, Selbach JM. Comparison of the Colvard, Procyon, and Neuroptics pupillometers for measuring pupil diameter under low ambient illumination. J Refract Surg 2009; In press 18. Hsieh Y-T, Hu F-R. The correlation of pupil size measured by Colvard pupillometer and Orbscan II. J Refract Surg 2007; 23:789–795 19. Mantry S, Banerjee S, Naroo S, Shah S. Scotopic measurement of normal pupil size with the Colvard pupillometer and the Nidek auto-refractor. Contact Lens Anterior Eye 2005; 28:53–56 20. Periman LM, Ambrosio R Jr, Harrison DA, Wilson SE. Correlation of pupil sizes measured with a mesopic infrared pupillometer and a photopic topographer. J Refract Surg 2003; 19:555–559 21. Brown SM. Comparison of the Colvard pupillometer and the Zywave for measuring scotopic pupil diameter [letter]. J Refract Surg 2005; 21:92; reply by ACK Cheng, DSC Lam, 92–93 22. Khanani AM, Brown SM. Determining scotopic pupil size [letter]. J Cataract Refract Surg 2005; 31:1266–1267; reply by T Kohnen, E Terzi, T Kasper, E-M Kohnen, J Bu¨hren, 1267 23. Brown SM, Bradley JC, Khanani AM. Dark adaptation is critical for accurate pupil measurement [letter]. Arch Ophthalmol 2008; 126:584 24. Brown SM. Monocular versus binocular pupillometry [letter]. J Cataract Refract Surg 2006; 32:374–375 25. Brown SM, Khanani AM. Night vision complaints after LASIK [letter]. Ophthalmology 2004; 111:1619–1620; reply by M Pop 1920 26. Brown SM, Khanani AM. Measuring room illuminance [letter]. J Cataract Refract Surg 2006; 32:702; reply by SS Wickremasinghe, GT Smith, JD Stevens, 702
J CATARACT REFRACT SURG - VOL 36, FEBRUARY 2010
First author: Jay C. Bradley, MD Department of Ophthalmology and Visual Sciences, Texas Tech University Health Sciences Center, Lubbock, Texas, USA
Clinical performance of a handheld digital infrared monocular pupillometer for measurement of the dark-adapted pupil diameter Jay C. Bradley, MD, Karl C. Bentley, MD, Aleem I. Mughal, MD, Sandra M. Brown, MD
PURPOSE: To compare the accuracy of a handheld infrared digital pupillometer and digital infrared photography for measurement of the dark-adapted pupil diameter. SETTING: Department of Ophthalmology and Visual Sciences, Texas Tech University Health Sciences Center, Lubbock, Texas, USA. METHODS: The right horizontal pupil diameter in healthy volunteers was measured using a NeurOptics PLR-200 pupillometer and then videographed using the infrared function of a CyberShot video camera after 2 minutes and 5 minutes dark adaptation at 1 lux ambient illumination. The best still image was extracted from the video file, and the horizontal pupil diameter was determined by comparison against an internal photographic length standard using digital image software. Accommodation and alertness were controlled during testing. RESULTS: The mean horizontal pupil diameter by infrared photography after 2 minutes of dark adaptation by subject age was 7.71 mm for ages 20 to 29 years, 6.80 mm for ages 30 to 39 years, 6.53 mm for ages 40 to 49 years, 5.94 mm for ages 50 to 59 years, and 6.01 mm for ages 60 to 69 years. The mean difference (infrared photography minus pupillometer) was C0.09 mm (range C0.30 to 0.14 mm) at 2 minutes of adaptation and C0.07 mm (range C0.25 to 0.13 mm) at 5 minutes. CONCLUSIONS: The pupillometer accurately measured the horizontal pupil diameter at 1 lux, with no measurement more than 0.3 mm different from infrared photography measurements. The pupillometer had a slight negative bias that is unlikely to introduce an error greater than 0.5 mm in clinical measurements. Financial Disclosure: No author has a financial or proprietary interest in any material or method mentioned. J Cataract Refract Surg 2010; 36:277–281 Q 2010 ASCRS and ESCRS
Surgeons performing corneal refractive surgery or implanting multifocal intraocular lenses (IOLs) are interested in the patient’s dark-adapted pupil diameter, which can influence the choice of ablation optical zone diameter or IOL design. Corneal refractive surgery is the most common elective operation in the United States, and cataract surgery is one of the most common operations in the elderly. Therefore, a compact and portable pupillometer of modest cost is valuable. Several pupillometers are commercially available, with the Colvard pupillometer (Oasis Medical, Inc.) probably the most widely used in the U.S. Although several studies1–7 have compared pupillometers, only the Colvard pupillometer has been compared with the laboratory method of digital infrared photography.8 Q 2010 ASCRS and ESCRS Published by Elsevier Inc.
The purpose of this study was to compare the clinical performance of a handheld digital infrared monocular pupillometer and the laboratory standard of digital infrared photography. SUBJECTS AND METHODS This observational cohort study was approved by an institutional review board. All subjects provided informed consent before participating. Volunteers with no history of eye disease, injury, or surgery were recruited from friends and colleagues of the investigators, medical students, and clinical and administrative staff of the Department of Ophthalmology and Visual Sciences, Texas Tech University, Lubbock, Texas, USA. For preliminary retinal adaptation, subjects donned wraparound ‘‘cataract’’ sunglasses for a minimum of 5 minutes before testing. After the subject entered the testing room, 0886-3350/10/$dsee front matter doi:10.1016/j.jcrs.2009.09.025
277
278
CLINICAL PERFORMANCE OF HANDHELD INFRARED PUPILLOMETER
dark adaptation progressed for 2 minutes at 1 lux ambient illumination, which was monitored with an industrial light meter (model CA814, AMEC Instruments) placed near the subject’s chair. Volitional accommodation during measurement was controlled by instructing the subject to gaze at the far end of the examination room (15 feet), which contained no point light sources. At 2 minutes (measured by a stopwatch), an alertness test was administered by asking the subject to identify the number of odd numbers in a sequence of 10 numbers. Infrared digital videography of the right pupil was performed using a CyberShot DCRTRV740 digital Handycam video camera (Sony); a small strip of printed ruler was taped to the right lower lid as close to the eyelash line as possible to provide an internal photographic width standard (Figure 1). A NeurOptics PLR-200 pupillometer (NeurOptics, Inc.) was then used to measure the horizontal pupil diameter according to the manufacturer’s instructions. Additional dark adaptation was performed for 5 minutes, and the measurements were repeated. All clinical testing was performed by the same ophthalmologist (K.C.B.) in the same standardized testing suite. The videography files were reviewed and the clearest still images extracted. Although blinded to the pupillometry data, the same ophthalmologist (J.C.B.) measured the horizontal pupil diameter with digital image software, using the ruler as an internal length standard. Additional details of the infrared photography and digital image analysis process have been published.9,10 This method has sufficient precision for the current study, with a mean interobserver difference of 0.03 mm.9 The handheld infrared pupillometer in this study uses monocular occlusion, with the tested eye occluded by a rubber cup. The pupillometer is self-calibrating and focuses automatically. It acquires 14 images of the pupil over 3 seconds, discards outliers to account for hippus, and provides an average value. The software measures the best circular fit of the pupil margin and automatically adjusts for vertex distance. During measurement, subjects were instructed not to look at the examiner’s face but rather to maintain distance fixation. The pupillometry data were compared with the infrared photography data using Bland-Altman graphs. A difference of G0.5 mm was considered clinically significant (ie, an error that could affect surgical planning).
RESULTS Of the 58 subjects recruited and tested, 8 were excluded due to excessive ruler slant (2), video decentration (1),
Submitted: July 2, 2009. Final revision submitted: August 26, 2009. Accepted: September 8, 2009. From the Department of Ophthalmology and Visual Sciences (Bradley, Bentley, Mughal), Texas Tech University Health Sciences Center, Lubbock, Texas, and a private eye center (Brown), Concord, North Carolina, USA.. Corresponding author: Sandra M. Brown, MD, 201 LePhillip Court Northeast, Concord, North Carolina 28025, USA. E-mail: [email protected].
Figure 1. Example of an infrared photograph extracted from a digital video file showing the internal length standard. The distance between the white hash marks is 5.0 mm. The ‘‘3’’ is the subject number.
pupil corectopia (1), and excessive hippos or accommodative miosis (4). The mean age of the final cohort of 50 subjects (28 men, 22 women) was 35 years (median 30.5; range 19 to 62 years). Table 1 shows the results of testing by age decade. Figure 2 shows the distribution of horizontal darkadapted pupil diameter measured by infrared photography after 2 minutes of dark adaptation as a function of age. Figure 3 shows a comparison between pupillometry and infrared photography after 2 minutes and after 5 minutes of dark adaptation results using Bland-Altman analysis. The mean difference was C0.094 mm (range 0.14 to C0.30 mm) after 2 minutes and C0.071 mm (range 0.13 to C0.25 mm) after 5 minutes. Figure 4 shows the distribution of the difference between pupillometry and infrared photography. Table 1 shows the difference in dark-adapted pupil diameter between 2 minutes and 5 minutes of dark adaptation as a function of age; the difference is plotted in Figure 5. DISCUSSION Until the advent of corneal laser refractive surgery, there was little interest in sophisticated office measurement of a patient’s pupil diameter under low-light conditions. The Rosenbaum pupil card sufficed because anisocoria was of greater diagnostic interest than the absolute value of the pupil diameter. The first lightweight, handheld infrared pupillometer was the Colvard, which was marketed in 1998.11 In our laboratory, we found this device to have a steep learning curve and excessive user dependence,8 making it inaccurate, especially in unknowledgeable hands. The Procyon pupillometer1–6,12–15 (Procyon, Ltd.) was released in approximately 2002. This tabletop device
J CATARACT REFRACT SURG - VOL 36, FEBRUARY 2010
279
CLINICAL PERFORMANCE OF HANDHELD INFRARED PUPILLOMETER
Table 1. Effects of age and duration of dark adaptation on horizontal pupil diameter. Horizontal Pupil Diameter (mm) Difference 5 Minutes 2 Minutes
2 Minutes Adaptation Infrared Photography Age (y) 0–19 20–29 30–39 40–49 50–59 60–69 All
Pupillometry
Infrared Photography
Eyes (n)
Mean
Range
Mean
Range
Mean
Range
1 22 10 7 7 3 50
6.83 7.71 6.80 6.53 5.94 6.01 7.00
d 5.97 to 8.80 5.16 to 8.78 4.79 to 7.24 4.31 to 7.33 5.67 to 6.25 4.31 to 8.80
6.80 7.60 6.68 6.47 5.87 5.98 6.90
d 6.00 to 8.60 5.30 to 8.70 4.80 to 7.20 4.40 to 7.20 5.80 to 6.10 4.40 to 8.70
0.03 0.045 0.064 0.016 0.000 0.005 0.039
d 0.229 to 0.140 0.318 to 0.103 0.260 to 0.162 0.081 to 0.074 0.005 to 0.027 0.229 to 0.162
was significantly more expensive than the Colvard pupillometer but had the stated advantages of minimal user dependence and hardcopy output. Despite a high degree of measurement precision (2 decimal places), the machine was clinically inaccurate16 and puzzlingly user dependent, with an interobserver difference at 0.4 lux of 1.8 to C2.4 mm.17 The Procyon pupillometer also had a profound negative bias, with a mean difference compared with the pupillometer we used in our study (Procyon minus NeurOptics) of 0.76 mm at 0.04 lux and 2.10 mm at 0.40 lux.17 Some authors use a wavefront analyzer or other ‘‘chin on the paper, forehead against the strap’’ machines to measure the pupil diameter.2,3,15,18–20 This method does not allow distance fixation and may induce accommodation, even in myopic patients, through psychogenic awareness of near. Because measuring pupil diameter is important, it would be preferable to have a dedicated device to perform the test.21,22 We found the NeurOptics PLR-200 pupillometer to have clinically satisfactory accuracy when compared with the laboratory standard of infrared photography. We defined an unsatisfactory difference as G0.5 mm
Figure 2. Horizontal dark-adapted pupil diameter measured by infrared photography after 2 minutes of dark-adaptation at 1 lux as a function of age (DAPD Z dark-adapted pupil diameter).
for any given subject. No pupillometer reading fell outside this range. After 2 minutes of dark adaptation, 41 (82%) of 50 of readings were less than G0.2 mm different from the infrared photography reading and the maximum difference (infrared photography minus pupillometry) was C0.30 mm. After 5 minutes, 47 readings (94%) were less than G0.2 mm different and the maximum difference was C0.25 mm. Photographic measurements were more often slightly larger
Figure 3. Bland-Altman plot of the horizontal dark-adapted pupil diameter measured by infrared photography and by the pupillometer after 2 minutes and 5 minutes of dark adaptation at 1 lux (DAPD Z dark-adapted pupil diameter; IRP Z infrared photography).
J CATARACT REFRACT SURG - VOL 36, FEBRUARY 2010
280
CLINICAL PERFORMANCE OF HANDHELD INFRARED PUPILLOMETER
Figure 4. Distribution of the difference between the horizontal darkadapted pupil diameters measured by infrared photography and by the pupillometer (DAPD Z dark-adapted pupil diameter; IRP Z infrared photography).
Figure 5. Difference between the horizontal dark-adapted pupil diameters measured with infrared photography after 2 minutes and 5 minutes of dark adaptation at 1 lux as a function of age (20 Z 2 minutes; 50 Z 5 minutes; DAPD Z dark-adapted pupil diameter; IRP Z infrared photography).
(2 minutes adaptation: n Z 36, 72%; 5 minutes adaptation: n Z 38, 76%) than pupillometry measurements. The clinician’s goal is to measure the stable darkadapted pupil diameter at a predetermined level of low ambient illumination. Note that dark has no specific definition in the literature.23 Also, there is no such thing as scotopic illumination because the term scotopic refers to a state of retinal adaptation (in which cone photoreceptors are entirely quiescent) and the level of illumination that corresponds to scotopic adaptation varies between individuals.24 Room illumination, the duration of dark adaptation, accommodation, and alertness must all be controlled if the measured pupil diameter is to represent the dark-adapted pupil diameter. No device to measure pupil diameter can overcome inadequate methodology.25 We used 1 lux ambient illumination in our study because it is typical of night-driving conditions on residential side streets or unlighted highways in the U.S.10 A camera (photographic) light meter cannot be used to determine room illumination because its purpose is to measure reflected light and the readout is not in lux.26 To achieve room illumination at or near 1 lux requires that all fluorescent fixtures be turned off and incandescent lights be dimmed as low as possible. Practically speaking, turning off all lights except the reading lamp on the slitlamp stand, turning this light as low as possible, and aiming it toward the intersection of the wall and the ceiling behind the patient can achieve diffuse ambient illumination close to this low level. If you can read a medical journal in your examination room, you have to find other ways to further reduce the light level. It is also important to eliminate focal light sources from the pupil testing room. Hallway light entering under the door can have a marked effect on the stability of the pupil diameter and must be blocked in some fashion. We placed a blanket across the bottom of the
door; more decorative solutions, such as a door draftstopper, are available. In this study, we found that 2 minutes of dark adaptation was sufficient to achieve stable retinal adaptation to 1 lux ambient illumination. No horizontal pupil diameter was more than 0.25 mm larger after longer adaptation. However, we used a preliminary adaptation in which the subject wore dark wraparound cataract glasses for a minimum of 5 minutes. This was necessary because many of our subjects continued with regular work activities under fluorescent lights before testing. In a customary clinical setting, examination rooms are often dimly lit (although much brighter than 1 lux). Having the subject spend 2 to 3 minutes in the examination room before further reducing the light level should be sufficient to obtain stable retinal adaptation after 2 minutes at 1 lux. It is crucial that tests that cause photoreceptor bleaching, such as corneal topography, pupil reactivity, or slitlamp examination, be performed after the dark-adapted pupil diameter is measured. Even testing visual acuity can cause rod bleaching from the white background of the projected chart. Surgeons must consider these issues in the context of their customary clinical flow and establish a clear stepwise protocol for the examination process, especially if it is delegated in large part to technicians. Because there is a strong correlation between age and dark-adapted pupil diameter,22,23 an outlier measurement (eg, a 5.5 mm pupil diameter in a 20-year-old patient) should be repeated. An additional question is whether monocular occlusion (as with the NeurOptics and Colvard pupillometers) changes the dark-adapted pupil diameter compared with that obtained under natural binocular viewing conditions.24 In theory, total monocular occlusion should increase the dark-adapted pupil diameter because photoreceptor stimulation is halved. That the nonoccluded infrared photography measurements
J CATARACT REFRACT SURG - VOL 36, FEBRUARY 2010
CLINICAL PERFORMANCE OF HANDHELD INFRARED PUPILLOMETER
were more frequently larger than the pupillometry measurements in our study suggests that this effect is subclinical at 1 lux illumination or that it was lost in the mild negative bias of the pupillometer relative to infrared photography. It may be clinically detectable at higher light levels. Alternatively, monocular occlusion may induce involuntary accommodative convergence through awareness of the nearness of the examiner, even when subjects are instructed to maintain distance fixation. We found the NeurOptics PLR-200 pupillometer to be an easy-to-use, accurate, portable, and relatively inexpensive device for measurement of horizontal pupil diameter under reduced ambient illumination. It requires almost no user intervention (eg, focusing) or user image interpretation (eg, reading a reticule). Hard copy output is possible through wireless transmission to a printer. Compared with the laboratory standard of infrared photography, it may have a slight negative bias on average that is not clinically significant. Surgeons can be confident that if a correct darkadaptation protocol at 1 lux is followed, the pupillometer measurement will be within G0.5 mm of the photographed dark-adapted pupil diameter and will likely represent the patient’s true pupil diameter under night-driving conditions. REFERENCES 1. Bootsma S, Tahzib N, Eggink F, de Brabander J, Nuijts R. Comparison of two pupillometers in determining pupil size for refractive surgery. Acta Ophthalmol Scand 2007; 85:324–328 2. McDonnell C, Rolincova M, Venter J. Comparison of measurement of pupil sizes among the Colvard pupillometer, Procyon pupillometer, and NIDEK OPD-scan. J Refract Surg 2006; 22:S1027–S1030 3. Schmitz S, Krummenauer F, Henn S, Dick HB. Comparison of three different technologies for pupil diameter measurement. Graefes Arch Clin Exp Ophthalmol 2003; 241:472–477 4. Kohnen T, Terzi E, Bu¨hren J, Kohnen E-M. Comparison of a digital and a handheld infrared pupillometer for determining scotopic pupil diameter. J Cataract Refract Surg 2003; 29:112–117 5. Kohnen T, Terzi E, Kasper T, Kohnen E-M, Bu¨hren J. Correlation of infrared pupillometers and CCD-camera imaging from aberrometry and videokeratography for determining scotopic pupil size. J Cataract Refract Surg 2004; 30:2116–2123 6. Michel AW, Kronberg BP, Narva´ez J, Zimmerman G. Comparison of 2 multiple-measurement infrared pupillometers to determine scotopic pupil diameter. J Cataract Refract Surg 2006; 32:1926–1931 7. Schnitzler E-M, Baumeister M, Kohnen T. Scotopic measurement of normal pupils: Colvard versus Video Vision Analyzer infrared pupillometer. J Cataract Refract Surg 2000; 26:859–866 8. Bradley JC, Anderson JE, Xu KT, Brown SM. Comparison of Colvard pupillometer and infrared digital photography for measurement of the dark-adapted pupil diameter. J Cataract Refract Surg 2005; 31:2129–2132
281
9. Brown SM, Khanani AM, Xu KT. Day to day variability of the dark-adapted pupil diameter. J Cataract Refract Surg 2004; 30:639–644 10. Khanani AM, Brown SM, Xu KT. Six-month variability of the dark-adapted pupil diameter. J Cataract Refract Surg 2005; 31:987–990 11. Colvard M. Preoperative measurement of scotopic pupil dilation using an office pupillometer. J Cataract Refract Surg 1998; 24:1594–1597 12. Cheng ACK, Rao SK, Cheng LL, Lam DSC. Assessment of pupil size under different light intensities using the Procyon pupillometer. J Cataract Refract Surg 2006; 32:1015–1017 13. Kurz S, Krummenaer F, Pfeiffer N, Dick HB. Monocular versus binocular pupillometry. J Cataract Refract Surg 2004; 30:2551–2556 14. Rosen ES, Gore CL, Taylor D, Chitkara D, Howes F, Kowalewski E. Use of a digital infrared pupillometer to assess patient suitability for refractive surgery. J Cataract Refract Surg 2002; 28:1433–1438 15. Wickremasinghe SS, Smith GT, Stevens JD. Comparison of dynamic digital pupillometry and static measurements of pupil size in determining scotopic pupil size before refractive surgery. J Cataract Refract Surg 2005; 31:1171–1176 16. Brown SM, Bradley JC. Pupil size in refractive surgery candidates [letter]. J Refract Surg 2005; 21:303 17. Schallenberg M, Bangre V, Steuhl K-P, Kremmer S, Selbach JM. Comparison of the Colvard, Procyon, and Neuroptics pupillometers for measuring pupil diameter under low ambient illumination. J Refract Surg 2009; In press 18. Hsieh Y-T, Hu F-R. The correlation of pupil size measured by Colvard pupillometer and Orbscan II. J Refract Surg 2007; 23:789–795 19. Mantry S, Banerjee S, Naroo S, Shah S. Scotopic measurement of normal pupil size with the Colvard pupillometer and the Nidek auto-refractor. Contact Lens Anterior Eye 2005; 28:53–56 20. Periman LM, Ambrosio R Jr, Harrison DA, Wilson SE. Correlation of pupil sizes measured with a mesopic infrared pupillometer and a photopic topographer. J Refract Surg 2003; 19:555–559 21. Brown SM. Comparison of the Colvard pupillometer and the Zywave for measuring scotopic pupil diameter [letter]. J Refract Surg 2005; 21:92; reply by ACK Cheng, DSC Lam, 92–93 22. Khanani AM, Brown SM. Determining scotopic pupil size [letter]. J Cataract Refract Surg 2005; 31:1266–1267; reply by T Kohnen, E Terzi, T Kasper, E-M Kohnen, J Bu¨hren, 1267 23. Brown SM, Bradley JC, Khanani AM. Dark adaptation is critical for accurate pupil measurement [letter]. Arch Ophthalmol 2008; 126:584 24. Brown SM. Monocular versus binocular pupillometry [letter]. J Cataract Refract Surg 2006; 32:374–375 25. Brown SM, Khanani AM. Night vision complaints after LASIK [letter]. Ophthalmology 2004; 111:1619–1620; reply by M Pop 1920 26. Brown SM, Khanani AM. Measuring room illuminance [letter]. J Cataract Refract Surg 2006; 32:702; reply by SS Wickremasinghe, GT Smith, JD Stevens, 702
J CATARACT REFRACT SURG - VOL 36, FEBRUARY 2010
First author: Jay C. Bradley, MD Department of Ophthalmology and Visual Sciences, Texas Tech University Health Sciences Center, Lubbock, Texas, USA