- Email: [email protected]
Comparison of a digital and a handheld infrared pupillometer for determining scotopic pupil diameter
Comparison of a digital and a handheld infrared pupillometer for determining scotopic pupil diameter Thomas Kohnen, MD, Evdoxia Terzi, MD, Jens Bu¨hren, MD, Eva-Maria Kohnen, MD Purpose: To compare a digital infrared pupillometer with a handheld infrared pupillometer for determining scotopic pupil size. Setting: Department of Ophthalmology, Johann Wolfgang Goethe-University, Frankfurt am Main, Germany. Methods: In 100 eyes of 50 healthy individuals (mean age 38.8 years ⫾ 10.7 [SD]), the scotopic pupil size was measured with a digital (Procyon) and a handheld (Colvard) infrared pupillometer. After dark adaptation of 1 minute, measurements were performed with both devices by 2 examiners (E1 and E2). Agreement and repeatability were analyzed using a comparison method described by Bland and Altman. Results: The mean scotopic pupil diameter was 5.90 ⫾ 0.97 mm (range 3.24 to 7.91 mm) with the Procyon and 5.78 ⫾ 0.98 mm (range 3.00 to 7.30 mm) with the Colvard pupillometer. The mean difference between the 2 devices was ⫺0.01 mm (E1) and ⫺0.24 mm (E2). The limits of agreement ranged from 2.84 (E1) to 3.24 (E2). The coefficients of repeatability were 0.64 (Procyon) and 1.16 (Colvard). The mean difference between E1 and E2 was ⫺0.10 for the Procyon and 0.13 for the Colvard pupillometer. The limits of agreement ranged from 1.28 (Procyon) to 2.32 (Colvard). Conclusions: The digital infrared pupillometer (Procyon) demonstrated better repeatability and agreement in measuring scotopic pupil size than a handheld device (Colvard). J Cataract Refract Surg 2003; 29:112–117 © 2003 ASCRS and ESCRS
P
upil diameter is a parameter of great interest in modern refractive surgery. Some of the postoperative complaints of patients having refractive corneal or lens surgery are related to visual quality and are associated with large scotopic pupils since optical aberrations become more dominant when the pupil dilates. If the pupil size exceeds the optical zone size, a marked increase in higher-order aberrations could be observed. Patients might complain about poor contrast sensitivity, increased glare disability, or halos. Precise measurement of Accepted for publication September 26, 2002. Reprint requests to Thomas Kohnen, MD, Johann Wolfgang GoetheUniversity, Department of Ophthalmology, Theodor-Stern-Kai 7, 60590 Frankfurt am Main, Germany. E-mail: [email protected]. © 2003 ASCRS and ESCRS Published by Elsevier Science Inc.
the scotopic pupil diameter has therefore become an essential step in the preoperative evaluation of patients suitable for refractive surgery and is an important inclusion or exclusion criterion. Precise pupil diameter measurement can be performed only if the patient’s pupillary behavior in general is taken into consideration.1 The parameters that describe pupillary behavior are pupillary unrest (“hippus,” PU) and anisocoria. Pupillary unrest occurs at all levels of illumination.1 Ignoring it by taking a single measurement may lead to false determination of pupil size. Anisocoria can be determined accurately by binocular simultaneous measurements. The Procyon digital infrared pupillometer provides dynamic pupillometry and thus should give quantitative information about the behavior of the individual pupil. 0886-3350/03/$–see front matter doi:10.1016/S0886-3350(02)01898-9
PROCYON VERSUS COLVARD INFRARED PUPILLOMETER
This study compared the Procyon digital infrared pupillometer and the handheld infrared Colvard pupillometer in determining scotopic pupil size.
Subjects and Methods Procyon Pupillometer The Procyon pupillometer (P2000 SA pupillometer, Procyon Instruments Ltd.) is a digital infrared device for binocular simultaneous measurements of pupil diameter at 3 illuminance levels (scotopic [0.07 lux], low mesopic [0.88 lux], and high mesopic [6.61 lux]). Infrared light-emitting diodes (IR-LEDs) illuminate the eyes with long-wave light that does not affect pupil size but is recognized by the charge-coupled device camera in the system. At each illuminance level, a sequence of 10 images is acquired within 2 seconds and stored in a portable computer. The patient looks into the device and fixates on a black spot at a virtual distance of 10 m. Rubber eyecups provide tight occlusion around the patient’s eyes so pupil width is not influenced by ambient light. The acquired images can be reviewed by the examiner immediately after the measurement. For each level of illuminance, a diagram showing PU and anisocoria is presented. The mean, range, and standard deviation of scotopic, mesopic, and high-mesopic pupil diameters are assessed by the software and displayed as a diagram.1
Colvard Pupillometer The handheld Colvard pupillometer (Oasis Medical) has been described.2 Briefly, the device uses light amplification technology. The examiner is able to focus the iris and pupil by moving the pupillometer slightly forward and back. The patient is asked to fixate on an IR-LED, which emits red light at very low levels. A millimeter ruler is superimposed by a reticle in the device. In this study, an attempt was made to measure the largest horizontal scotopic pupil diameter. Examiners tried to estimate pupil size with a precision of 0.1 mm.
with a luxmeter (illuminance meter, Minolta). Measurements with the Colvard device under scotopic illumination (0.28 lux) were performed first and then Procyon measurements at the scotopic illuminance level of 0.07 lux. Although this difference might influence readings, the intention was to compare Procyon measurements with Colvard measurements taken under the usual conditions in our institution2 and those recommended by the manufacturer. Therefore, a slight background illumination was allowed for Colvard measurements, resulting in the mentioned difference. Colvard measurements were done first so the examiner was not influenced by the objective measurement of the Procyon pupillometer. Before the measurements, the objective refraction was determined.
Statistical Analysis The method described by Bland and Altman3,4 incorporated in a statistical analysis software (BiAS 7.06) was used to analyze the data. This method compares 2 methods by plotting their means against their differences. The limits of agreement and interrater reliability of the 2 examiners were also assessed. The limits of agreement were defined as the mean differences in measurements using the 2 techniques ⫾ 2 SDs. The limits of agreement were determined for measurements taken by 1 examiner with the 2 devices, as well as for measurements performed by both examiners with 1 pupillometer. For each pupillometry technique, the coefficients of interrater repeatability were defined as twice the standard deviation of the differences between both examiners’ measurements. Thus, lower values indicate higher repeatability.
Results Pupil Diameter The mean scotopic pupil diameters with the 2 pupillometers are shown in Table 1. Values varied more Table 1. Scotopic pupil measurements with the Procyon and Colvard pupillometers.
Subjects One hundred eyes of 50 healthy individuals with a mean age of 38.8 years ⫾ 10.7 (SD) (range 20 to 59 years) were included in the study. Subjects were excluded if they had a history of eye disease, eye surgery, eye trauma, use of topical eyedrops, or a morphological eye abnormality. In addition to measuring the pupil diameter, sphere and cylinder (autorefractor, RM-A 2300, Topcon) were determined. The mean sphere was ⫺0.88 ⫾ 2.42 diopters (D) (range ⫺9.50 to ⫹3.50 D) and the mean cylinder, ⫺0.68 ⫾ 0.59 D (range ⫺3.25 to 0.00 D).
Device/Examiner
Scotopic Pupil Measurement (mm)
Procyon E1
5.85 ⫾ 1.01
E2
5.95 ⫾ 0.93
Mean
5.90 ⫾ 0.97
Colvard E1
5.84 ⫾ 1.03
Measurements
E2
5.71 ⫾ 0.92
Examinations were performed by 2 independent examiners (E1 and E2). A 1-minute dark adaptation was achieved for all measurements. Background illuminance was measured
Mean
5.78 ⫾ 0.98
All values are mean ⫾ SD. E1 ⫽ examiner 1; E2 ⫽ examiner 2
J CATARACT REFRACT SURG—VOL 29, JANUARY 2003
113
PROCYON VERSUS COLVARD INFRARED PUPILLOMETER
with the Colvard pupillometer than with the Procyon pupillometer (Figure 1, A and B). The measurements taken by the 2 examiners with each pupillometer showed better agreement for the Procyon than for the Colvard (Table 2; Figure 2, C and D). Scotopic pupil size measured by 1 examiner with the 2 pupillometers showed a similar correlation for both examiners. Measurements taken by E1 showed less bias than those taken by E2 (Table 3; Figure 1, C and D). Agreement The mean differences between the Procyon and the Colvard pupillometers are shown in Table 3. Examiner 1 showed a smaller range of error within the limits of agreement than E2. There was a smaller but similar mean difference for the Procyon than for the Colvard (Table 2), but the standard deviation and subsequently the limits of agreement were higher for the Colvard device.
Table 2. Limits of agreement between E1 and E2 measurements. Mean Difference ⴞ SD (mm)
Comparison
Limits of Agreement (mm)*
Range (mm)
E1–E2 (Procyon)
⫺0.10 ⫾ 0.32
⫺0.74, 0.54
1.28
E1–E2 (Colvard)
0.13 ⫾ 0.58
⫺1.03, 1.29
2.32
*Mean difference ⫾ 2 SD of the difference between Colvard and Procyon pupillometer measurements
Repeatability The coefficient of interrater repeatability was smaller for the Procyon pupillometer (0.64) than for the Colvard pupillometer (1.16). This indicates better repeatability for the Procyon device. Figure 2, C and D, show that measurements made with the Procyon pupillometer fell into a smaller range of ⫾ 2 SD around the assessed mean difference than measurements made with the Colvard device by the 2 examiners.
Figure 1. (Kohnen) A: Pupil size measured with the Procyon pupillometer (E1 versus E2). B: Pupil size measured with the Colvard pupillometer (E1 versus E2). C: Pupil size measured by E1 using the Procyon and Colvard pupillometers. D: Pupil size measured by E2 using the Procyon and Colvard pupillometers. All measurements are in millimeters.
114
J CATARACT REFRACT SURG—VOL 29, JANUARY 2003
PROCYON VERSUS COLVARD INFRARED PUPILLOMETER
Figure 2. (Kohnen) A: Agreement of the Procyon and Colvard pupillometers for measurements obtained by E1. B: Agreement of the Procyon and Colvard pupillometers for measurements obtained by E2. C: Interrater repeatability for the Procyon pupillometer. D: Interrater repeatability for the Colvard pupillometer. All measurements are in millimeters.
Discussion Refractive corneal and lens surgery is becoming more popular for the correction of refractive errors. Several parameters determine the limits of various procedures; these include refraction and aberrations, age, corneal thickness, endothelial cells, anterior chamber depth, lens status, and pupil size. Because at night the dilated pupil allows unwanted light phenomena, scotopic pupil size has become an important factor in refractive surgery. The largest possible scotopic pupil size has not been established; only a combination of all parameters will determine the inclusion or exclusion criteria for refractive surgery. Patients with large scotopic pupils often report glare phenomena, halos, or poor contrast sensitivity under mesopic light conditions. Night vision, and especially driving at night, can be a serious problem for some patients, and their quality of life in general can be strongly
affected by such complaints. These phenomena are considered to be the result of a large scotopic pupil.5–10 To solve these problems for all refractive procedures, an accurate, objective, standardized, and reliable method to measure pupil size in mesopic light conditions must be established.11 Pupil size is a dynamic value that is affected by external factors such as illumination and drugs and by intrinsic factors such as psychological state, age, and refraction.1,12–14 In addition, UP, the so-called hippus, makes a sole measurement of pupil diameter inaccurate, so taking more than 1 measure and assessing the mean value would result in more precise measurements.1 The mean differences in the pupil diameters measured by the Procyon and Colvard devices were surprisingly low and similar. The tendency of smaller pupil size measurements with the Colvard pupillometer might be due to the slightly higher ambiance illumination or an effect of accommodative miosis when subjects fixated on
J CATARACT REFRACT SURG—VOL 29, JANUARY 2003
115
PROCYON VERSUS COLVARD INFRARED PUPILLOMETER
Table 3. Limits of agreement between Colvard and Procyon measurements. Mean Difference ⴞ SD (mm)
Limits of Agreement (mm)*
Range (mm)
Procyon–Colvard (E1)
⫺0.01 ⫾ 0.71
⫺1.43, 1.41
2.84
Procyon–Colvard (E2)
⫺0.24 ⫾ 0.81
⫺1.86, 1.38
3.24
Comparison
*Mean difference ⫾ 2 SD of the difference between Colvard and Procyon pupillometer measurements
the IR-LED in this device. From the present study, we conclude that the digital infrared device shows less variation in scotopic pupil diameter and has better interrater repeatability than the handheld infrared device. Digital pupillometers such as the Procyon device allow examination of the dynamic pupil process and objectively determine the value using a computer. Further advantages are a better resolution than handheld infrared devices and well-defined illuminance conditions. The digital device also offers high comparability and standardization for research purposes. This is an important aspect in studies with different settings and measurements under different luminance levels. Handheld infrared pupillometers such as the Colvard unit can also follow the dynamic pupil process if an experienced examiner is using the device. Measurements appear to be possible in the range of 0.1 mm steps.
However, considerable clustering of readings at integer values (eg, 5.0, 5.5, 6.0, 6.5 mm) were measured. This is seen in the straight horizontal and vertical lines of data points. These phenomena were not observed with the Procyon pupillometer. This underlines the fact that measurements using the handheld device are performed on a subjective basis. If examiners have less probability of reading between millimeter values, information is lost. The advantages of the handheld infrared device are a lower cost and portability. Since we have used the Colvard pupillometer for patient evaluation on a permanent basis, problems with severe glare induced by small myopic laser ablation zones have disappeared. We currently use optical zones (OZs) that are larger than the scotopic pupil diameter. If this OZ size is not achievable, eg, in highly myopic eyes or thinner corneas, we recommend not treating the patient until better modalities are available. Preventing problems in refractive surgery is better than solving postoperative complications, and we believe that a digital pupillometer such as the Procyon can provide more precise measurements than the handheld Colvard pupillometer. Pupil measurements are of interest to anterior segment surgeons. Table 4 shows the results of studies examining pupil size with various devices. The variation in scotopic pupil sizes emphasizes the need for precise measurements of each case before refractive surgery. Mea-
Table 4. Scotopic pupil diameter measured with different devices in adults. Illumination/ Measurement
Device Photography (Nikon)15 Rosenbaum chart
Iowa infrared pupillometer
Colvard infrared pupillometer
17
VIVA (Video Vision Analyzer)2 Colvard infrared pupillometer
2
Procyon infrared pupillometer Colvard infrared pupillometer
1
18
Age (Y)
Scotopic Pupil Diameter
242
20–70
4.9 ⫾ 0.8* (2.7–6.6)
0.42, 25.55, 344.33 cd/m
14
20–70
5.4 ⫾ 1.1 (—)
IR/OBS
0.42, 25.55, 344.33 cd/m
2
14
28–42
4.95 ⫾ 1.08 (—)
IR/OBS
15, 3 lux
200
18–50
6.2 (3.0–9.0)
IR/OBS
0.5–0.6 lux
66
19–55
6.24 ⫾ 1.28 (3.5–9.0)
IR/OBS
0.5–0.6 lux
66
19–55
6.08 ⫾ 1.16 (3.2–8.4)
IR/DIG
0.02, 0.15, 10.6 lux
WL/OBS 16
Eyes
2
WL/OBS
16
Luminance/Illuminance 0, 20, 80, 320, 1000 ft-c
116
Mean 46 ⫾ 11.7
6.61 (—) 6.1 ⫾ 0.9 (4.0–8.0)
IR/OBS
0.7, 2–5 lux
58
—
Procyon infrared pupillometer (current data)
IR/DIG
0.07, 0.88, 6.61 lux
100
20–59
5.90 ⫾ 0.97 (3.24–7.91)
Colvard infrared pupillometer (current data)
IR/OBS
0.28 lux
100
20–59
5.78 ⫾ 0.98 (3.00–7.30)
*Measured pupil size reduced 14% to account for corneal magnification WL ⫽ white light illumination; IR ⫽ infrared illumination; OBS ⫽ measurement by observer; DIG ⫽ digital measurement; (—) data not provided
116
J CATARACT REFRACT SURG—VOL 29, JANUARY 2003
PROCYON VERSUS COLVARD INFRARED PUPILLOMETER
surements that are too small would accept inappropriate patients for treatment, and measurements that are too large would reject appropriate patients. In summary, we believe that infrared measurements should be state-of-the-art technology to obtain scotopic pupil sizes before refractive surgery. Although the differences between the readings of the Procyon and Colvard devices appear to be small, digital pupillometers will currently provide the most accurate measurement of the scotopic pupil.
9. 10.
11. 12.
References 1. Rosen ES, Gore CL, Taylor D, et al. Use of a digital infrared pupillometer to assess patient suitability for refractive surgery. J Cataract Refract Surg 2002; 28:1433– 1438 2. 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 3. Bland JM, Altman DG. Statistical methods for assessing agreement between two methods of clinical measurement. Lancet 1986; 1:307–310 4. Altman DG, Bland JM. Measurements in medicine: the analysis of method comparison studies. Statistician 1983; 32:307–318 5. Applegate RA, Gansel KA. The importance of pupil size in optical quality measurements following radial keratotomy. Refract Corneal Surg 1990; 6:47–54 6. Holladay JT, Lynn MJ, Waring GO III, et al. The relationship of visual acuity, refractive error, and pupil size after radial keratotomy. Arch Ophthalmol 1991; 109: 70 –76 7. Roberts CW, Koester CJ. Optical zone diameters for photorefractive corneal surgery. Invest Ophthalmol Vis Sci 1993; 34:2275–2281 8. O’Brart DPS, Corbett MC, Lohmann CP, et al. The effects of ablation diameter on the outcome of excimer laser photorefractive keratectomy; a prospective, ran-
13.
14.
15.
16.
17.
18.
domized, double-blind study. Arch Ophthalmol 1995; 113:438 –443 Nixon WS. Pupil size in refractive surgery [letter]. J Cataract Refract Surg 1997; 23:1435–1436 Schnitzler E-M, Kohnen T, Baumeister M, Steinkamp GWK. Korrelation von Pupillenweite, Ablationszone und Blendungsempfindlichkeit nach Laser in situ Keratomileusis (LASIK) mit dem Planoscan Excimer-Lasersystem [abstract]. Ophthalmologe 1999; 96(suppl):S16 Kohnen T. Measuring vision in refractive surgery [editorial]. J Cataract Refract Surg 2001; 27:1897–1898 Borgmann H. Grundlagen fu¨ r eine klinische Pupillographie: III: Die Abha¨ngigkeit der Parameter der Lichtreaktion vom Lebensalter. Albrecht von Graefes Arch Klin Exp Ophthalmol 1972; 185:1–10 Winn B, Whitaker D, Elliott DB, Phillips NJ. Factors affecting light-adapted pupil size in normal human subjects. Invest Ophthalmol Vis Sci 1994; 35:1132–1137 Loewenfeld IE. The Pupil; Anatomy, Physiology, and Clinical Applications. Ames, IA, Iowa State University Press, 1993 Koch DD, Samuelson SW, Haft EA, Merin LM. Pupillary size and responsiveness; implications for selection of a bifocal intraocular lens. Ophthalmology 1991; 98:1030– 1035 Boxer Wachler BS, Krueger RR. Agreement and repeatability of infrared pupillometry and comparison method. Ophthalmology 1999; 106:319 –323 Colvard M. Preoperative measurement of scotopic pupil dilation using an office pupillometer. J Cataract Refract Surg 1998; 24:1594 –1597 Pop M, Payette Y, Santoriello E. Comparison of the pupil card and pupillometer in measuring pupil size. J Cataract Refract Surg 2002; 28:283–288
From the Department of Ophthalmology, Johann Wolfgang GoetheUniversity, Frankfurt am Main, Germany. None of the authors has a proprietary or financial interest in any product mentioned. Hanns Ackermann, PhD, Institute for Statistics, Johann Wolfgang Goethe-University, Frankfurt am Main, Germany, helped with the statistical evaluation.
J CATARACT REFRACT SURG—VOL 29, JANUARY 2003
117
P
upil diameter is a parameter of great interest in modern refractive surgery. Some of the postoperative complaints of patients having refractive corneal or lens surgery are related to visual quality and are associated with large scotopic pupils since optical aberrations become more dominant when the pupil dilates. If the pupil size exceeds the optical zone size, a marked increase in higher-order aberrations could be observed. Patients might complain about poor contrast sensitivity, increased glare disability, or halos. Precise measurement of Accepted for publication September 26, 2002. Reprint requests to Thomas Kohnen, MD, Johann Wolfgang GoetheUniversity, Department of Ophthalmology, Theodor-Stern-Kai 7, 60590 Frankfurt am Main, Germany. E-mail: [email protected]. © 2003 ASCRS and ESCRS Published by Elsevier Science Inc.
the scotopic pupil diameter has therefore become an essential step in the preoperative evaluation of patients suitable for refractive surgery and is an important inclusion or exclusion criterion. Precise pupil diameter measurement can be performed only if the patient’s pupillary behavior in general is taken into consideration.1 The parameters that describe pupillary behavior are pupillary unrest (“hippus,” PU) and anisocoria. Pupillary unrest occurs at all levels of illumination.1 Ignoring it by taking a single measurement may lead to false determination of pupil size. Anisocoria can be determined accurately by binocular simultaneous measurements. The Procyon digital infrared pupillometer provides dynamic pupillometry and thus should give quantitative information about the behavior of the individual pupil. 0886-3350/03/$–see front matter doi:10.1016/S0886-3350(02)01898-9
PROCYON VERSUS COLVARD INFRARED PUPILLOMETER
This study compared the Procyon digital infrared pupillometer and the handheld infrared Colvard pupillometer in determining scotopic pupil size.
Subjects and Methods Procyon Pupillometer The Procyon pupillometer (P2000 SA pupillometer, Procyon Instruments Ltd.) is a digital infrared device for binocular simultaneous measurements of pupil diameter at 3 illuminance levels (scotopic [0.07 lux], low mesopic [0.88 lux], and high mesopic [6.61 lux]). Infrared light-emitting diodes (IR-LEDs) illuminate the eyes with long-wave light that does not affect pupil size but is recognized by the charge-coupled device camera in the system. At each illuminance level, a sequence of 10 images is acquired within 2 seconds and stored in a portable computer. The patient looks into the device and fixates on a black spot at a virtual distance of 10 m. Rubber eyecups provide tight occlusion around the patient’s eyes so pupil width is not influenced by ambient light. The acquired images can be reviewed by the examiner immediately after the measurement. For each level of illuminance, a diagram showing PU and anisocoria is presented. The mean, range, and standard deviation of scotopic, mesopic, and high-mesopic pupil diameters are assessed by the software and displayed as a diagram.1
Colvard Pupillometer The handheld Colvard pupillometer (Oasis Medical) has been described.2 Briefly, the device uses light amplification technology. The examiner is able to focus the iris and pupil by moving the pupillometer slightly forward and back. The patient is asked to fixate on an IR-LED, which emits red light at very low levels. A millimeter ruler is superimposed by a reticle in the device. In this study, an attempt was made to measure the largest horizontal scotopic pupil diameter. Examiners tried to estimate pupil size with a precision of 0.1 mm.
with a luxmeter (illuminance meter, Minolta). Measurements with the Colvard device under scotopic illumination (0.28 lux) were performed first and then Procyon measurements at the scotopic illuminance level of 0.07 lux. Although this difference might influence readings, the intention was to compare Procyon measurements with Colvard measurements taken under the usual conditions in our institution2 and those recommended by the manufacturer. Therefore, a slight background illumination was allowed for Colvard measurements, resulting in the mentioned difference. Colvard measurements were done first so the examiner was not influenced by the objective measurement of the Procyon pupillometer. Before the measurements, the objective refraction was determined.
Statistical Analysis The method described by Bland and Altman3,4 incorporated in a statistical analysis software (BiAS 7.06) was used to analyze the data. This method compares 2 methods by plotting their means against their differences. The limits of agreement and interrater reliability of the 2 examiners were also assessed. The limits of agreement were defined as the mean differences in measurements using the 2 techniques ⫾ 2 SDs. The limits of agreement were determined for measurements taken by 1 examiner with the 2 devices, as well as for measurements performed by both examiners with 1 pupillometer. For each pupillometry technique, the coefficients of interrater repeatability were defined as twice the standard deviation of the differences between both examiners’ measurements. Thus, lower values indicate higher repeatability.
Results Pupil Diameter The mean scotopic pupil diameters with the 2 pupillometers are shown in Table 1. Values varied more Table 1. Scotopic pupil measurements with the Procyon and Colvard pupillometers.
Subjects One hundred eyes of 50 healthy individuals with a mean age of 38.8 years ⫾ 10.7 (SD) (range 20 to 59 years) were included in the study. Subjects were excluded if they had a history of eye disease, eye surgery, eye trauma, use of topical eyedrops, or a morphological eye abnormality. In addition to measuring the pupil diameter, sphere and cylinder (autorefractor, RM-A 2300, Topcon) were determined. The mean sphere was ⫺0.88 ⫾ 2.42 diopters (D) (range ⫺9.50 to ⫹3.50 D) and the mean cylinder, ⫺0.68 ⫾ 0.59 D (range ⫺3.25 to 0.00 D).
Device/Examiner
Scotopic Pupil Measurement (mm)
Procyon E1
5.85 ⫾ 1.01
E2
5.95 ⫾ 0.93
Mean
5.90 ⫾ 0.97
Colvard E1
5.84 ⫾ 1.03
Measurements
E2
5.71 ⫾ 0.92
Examinations were performed by 2 independent examiners (E1 and E2). A 1-minute dark adaptation was achieved for all measurements. Background illuminance was measured
Mean
5.78 ⫾ 0.98
All values are mean ⫾ SD. E1 ⫽ examiner 1; E2 ⫽ examiner 2
J CATARACT REFRACT SURG—VOL 29, JANUARY 2003
113
PROCYON VERSUS COLVARD INFRARED PUPILLOMETER
with the Colvard pupillometer than with the Procyon pupillometer (Figure 1, A and B). The measurements taken by the 2 examiners with each pupillometer showed better agreement for the Procyon than for the Colvard (Table 2; Figure 2, C and D). Scotopic pupil size measured by 1 examiner with the 2 pupillometers showed a similar correlation for both examiners. Measurements taken by E1 showed less bias than those taken by E2 (Table 3; Figure 1, C and D). Agreement The mean differences between the Procyon and the Colvard pupillometers are shown in Table 3. Examiner 1 showed a smaller range of error within the limits of agreement than E2. There was a smaller but similar mean difference for the Procyon than for the Colvard (Table 2), but the standard deviation and subsequently the limits of agreement were higher for the Colvard device.
Table 2. Limits of agreement between E1 and E2 measurements. Mean Difference ⴞ SD (mm)
Comparison
Limits of Agreement (mm)*
Range (mm)
E1–E2 (Procyon)
⫺0.10 ⫾ 0.32
⫺0.74, 0.54
1.28
E1–E2 (Colvard)
0.13 ⫾ 0.58
⫺1.03, 1.29
2.32
*Mean difference ⫾ 2 SD of the difference between Colvard and Procyon pupillometer measurements
Repeatability The coefficient of interrater repeatability was smaller for the Procyon pupillometer (0.64) than for the Colvard pupillometer (1.16). This indicates better repeatability for the Procyon device. Figure 2, C and D, show that measurements made with the Procyon pupillometer fell into a smaller range of ⫾ 2 SD around the assessed mean difference than measurements made with the Colvard device by the 2 examiners.
Figure 1. (Kohnen) A: Pupil size measured with the Procyon pupillometer (E1 versus E2). B: Pupil size measured with the Colvard pupillometer (E1 versus E2). C: Pupil size measured by E1 using the Procyon and Colvard pupillometers. D: Pupil size measured by E2 using the Procyon and Colvard pupillometers. All measurements are in millimeters.
114
J CATARACT REFRACT SURG—VOL 29, JANUARY 2003
PROCYON VERSUS COLVARD INFRARED PUPILLOMETER
Figure 2. (Kohnen) A: Agreement of the Procyon and Colvard pupillometers for measurements obtained by E1. B: Agreement of the Procyon and Colvard pupillometers for measurements obtained by E2. C: Interrater repeatability for the Procyon pupillometer. D: Interrater repeatability for the Colvard pupillometer. All measurements are in millimeters.
Discussion Refractive corneal and lens surgery is becoming more popular for the correction of refractive errors. Several parameters determine the limits of various procedures; these include refraction and aberrations, age, corneal thickness, endothelial cells, anterior chamber depth, lens status, and pupil size. Because at night the dilated pupil allows unwanted light phenomena, scotopic pupil size has become an important factor in refractive surgery. The largest possible scotopic pupil size has not been established; only a combination of all parameters will determine the inclusion or exclusion criteria for refractive surgery. Patients with large scotopic pupils often report glare phenomena, halos, or poor contrast sensitivity under mesopic light conditions. Night vision, and especially driving at night, can be a serious problem for some patients, and their quality of life in general can be strongly
affected by such complaints. These phenomena are considered to be the result of a large scotopic pupil.5–10 To solve these problems for all refractive procedures, an accurate, objective, standardized, and reliable method to measure pupil size in mesopic light conditions must be established.11 Pupil size is a dynamic value that is affected by external factors such as illumination and drugs and by intrinsic factors such as psychological state, age, and refraction.1,12–14 In addition, UP, the so-called hippus, makes a sole measurement of pupil diameter inaccurate, so taking more than 1 measure and assessing the mean value would result in more precise measurements.1 The mean differences in the pupil diameters measured by the Procyon and Colvard devices were surprisingly low and similar. The tendency of smaller pupil size measurements with the Colvard pupillometer might be due to the slightly higher ambiance illumination or an effect of accommodative miosis when subjects fixated on
J CATARACT REFRACT SURG—VOL 29, JANUARY 2003
115
PROCYON VERSUS COLVARD INFRARED PUPILLOMETER
Table 3. Limits of agreement between Colvard and Procyon measurements. Mean Difference ⴞ SD (mm)
Limits of Agreement (mm)*
Range (mm)
Procyon–Colvard (E1)
⫺0.01 ⫾ 0.71
⫺1.43, 1.41
2.84
Procyon–Colvard (E2)
⫺0.24 ⫾ 0.81
⫺1.86, 1.38
3.24
Comparison
*Mean difference ⫾ 2 SD of the difference between Colvard and Procyon pupillometer measurements
the IR-LED in this device. From the present study, we conclude that the digital infrared device shows less variation in scotopic pupil diameter and has better interrater repeatability than the handheld infrared device. Digital pupillometers such as the Procyon device allow examination of the dynamic pupil process and objectively determine the value using a computer. Further advantages are a better resolution than handheld infrared devices and well-defined illuminance conditions. The digital device also offers high comparability and standardization for research purposes. This is an important aspect in studies with different settings and measurements under different luminance levels. Handheld infrared pupillometers such as the Colvard unit can also follow the dynamic pupil process if an experienced examiner is using the device. Measurements appear to be possible in the range of 0.1 mm steps.
However, considerable clustering of readings at integer values (eg, 5.0, 5.5, 6.0, 6.5 mm) were measured. This is seen in the straight horizontal and vertical lines of data points. These phenomena were not observed with the Procyon pupillometer. This underlines the fact that measurements using the handheld device are performed on a subjective basis. If examiners have less probability of reading between millimeter values, information is lost. The advantages of the handheld infrared device are a lower cost and portability. Since we have used the Colvard pupillometer for patient evaluation on a permanent basis, problems with severe glare induced by small myopic laser ablation zones have disappeared. We currently use optical zones (OZs) that are larger than the scotopic pupil diameter. If this OZ size is not achievable, eg, in highly myopic eyes or thinner corneas, we recommend not treating the patient until better modalities are available. Preventing problems in refractive surgery is better than solving postoperative complications, and we believe that a digital pupillometer such as the Procyon can provide more precise measurements than the handheld Colvard pupillometer. Pupil measurements are of interest to anterior segment surgeons. Table 4 shows the results of studies examining pupil size with various devices. The variation in scotopic pupil sizes emphasizes the need for precise measurements of each case before refractive surgery. Mea-
Table 4. Scotopic pupil diameter measured with different devices in adults. Illumination/ Measurement
Device Photography (Nikon)15 Rosenbaum chart
Iowa infrared pupillometer
Colvard infrared pupillometer
17
VIVA (Video Vision Analyzer)2 Colvard infrared pupillometer
2
Procyon infrared pupillometer Colvard infrared pupillometer
1
18
Age (Y)
Scotopic Pupil Diameter
242
20–70
4.9 ⫾ 0.8* (2.7–6.6)
0.42, 25.55, 344.33 cd/m
14
20–70
5.4 ⫾ 1.1 (—)
IR/OBS
0.42, 25.55, 344.33 cd/m
2
14
28–42
4.95 ⫾ 1.08 (—)
IR/OBS
15, 3 lux
200
18–50
6.2 (3.0–9.0)
IR/OBS
0.5–0.6 lux
66
19–55
6.24 ⫾ 1.28 (3.5–9.0)
IR/OBS
0.5–0.6 lux
66
19–55
6.08 ⫾ 1.16 (3.2–8.4)
IR/DIG
0.02, 0.15, 10.6 lux
WL/OBS 16
Eyes
2
WL/OBS
16
Luminance/Illuminance 0, 20, 80, 320, 1000 ft-c
116
Mean 46 ⫾ 11.7
6.61 (—) 6.1 ⫾ 0.9 (4.0–8.0)
IR/OBS
0.7, 2–5 lux
58
—
Procyon infrared pupillometer (current data)
IR/DIG
0.07, 0.88, 6.61 lux
100
20–59
5.90 ⫾ 0.97 (3.24–7.91)
Colvard infrared pupillometer (current data)
IR/OBS
0.28 lux
100
20–59
5.78 ⫾ 0.98 (3.00–7.30)
*Measured pupil size reduced 14% to account for corneal magnification WL ⫽ white light illumination; IR ⫽ infrared illumination; OBS ⫽ measurement by observer; DIG ⫽ digital measurement; (—) data not provided
116
J CATARACT REFRACT SURG—VOL 29, JANUARY 2003
PROCYON VERSUS COLVARD INFRARED PUPILLOMETER
surements that are too small would accept inappropriate patients for treatment, and measurements that are too large would reject appropriate patients. In summary, we believe that infrared measurements should be state-of-the-art technology to obtain scotopic pupil sizes before refractive surgery. Although the differences between the readings of the Procyon and Colvard devices appear to be small, digital pupillometers will currently provide the most accurate measurement of the scotopic pupil.
9. 10.
11. 12.
References 1. Rosen ES, Gore CL, Taylor D, et al. Use of a digital infrared pupillometer to assess patient suitability for refractive surgery. J Cataract Refract Surg 2002; 28:1433– 1438 2. 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 3. Bland JM, Altman DG. Statistical methods for assessing agreement between two methods of clinical measurement. Lancet 1986; 1:307–310 4. Altman DG, Bland JM. Measurements in medicine: the analysis of method comparison studies. Statistician 1983; 32:307–318 5. Applegate RA, Gansel KA. The importance of pupil size in optical quality measurements following radial keratotomy. Refract Corneal Surg 1990; 6:47–54 6. Holladay JT, Lynn MJ, Waring GO III, et al. The relationship of visual acuity, refractive error, and pupil size after radial keratotomy. Arch Ophthalmol 1991; 109: 70 –76 7. Roberts CW, Koester CJ. Optical zone diameters for photorefractive corneal surgery. Invest Ophthalmol Vis Sci 1993; 34:2275–2281 8. O’Brart DPS, Corbett MC, Lohmann CP, et al. The effects of ablation diameter on the outcome of excimer laser photorefractive keratectomy; a prospective, ran-
13.
14.
15.
16.
17.
18.
domized, double-blind study. Arch Ophthalmol 1995; 113:438 –443 Nixon WS. Pupil size in refractive surgery [letter]. J Cataract Refract Surg 1997; 23:1435–1436 Schnitzler E-M, Kohnen T, Baumeister M, Steinkamp GWK. Korrelation von Pupillenweite, Ablationszone und Blendungsempfindlichkeit nach Laser in situ Keratomileusis (LASIK) mit dem Planoscan Excimer-Lasersystem [abstract]. Ophthalmologe 1999; 96(suppl):S16 Kohnen T. Measuring vision in refractive surgery [editorial]. J Cataract Refract Surg 2001; 27:1897–1898 Borgmann H. Grundlagen fu¨ r eine klinische Pupillographie: III: Die Abha¨ngigkeit der Parameter der Lichtreaktion vom Lebensalter. Albrecht von Graefes Arch Klin Exp Ophthalmol 1972; 185:1–10 Winn B, Whitaker D, Elliott DB, Phillips NJ. Factors affecting light-adapted pupil size in normal human subjects. Invest Ophthalmol Vis Sci 1994; 35:1132–1137 Loewenfeld IE. The Pupil; Anatomy, Physiology, and Clinical Applications. Ames, IA, Iowa State University Press, 1993 Koch DD, Samuelson SW, Haft EA, Merin LM. Pupillary size and responsiveness; implications for selection of a bifocal intraocular lens. Ophthalmology 1991; 98:1030– 1035 Boxer Wachler BS, Krueger RR. Agreement and repeatability of infrared pupillometry and comparison method. Ophthalmology 1999; 106:319 –323 Colvard M. Preoperative measurement of scotopic pupil dilation using an office pupillometer. J Cataract Refract Surg 1998; 24:1594 –1597 Pop M, Payette Y, Santoriello E. Comparison of the pupil card and pupillometer in measuring pupil size. J Cataract Refract Surg 2002; 28:283–288
From the Department of Ophthalmology, Johann Wolfgang GoetheUniversity, Frankfurt am Main, Germany. None of the authors has a proprietary or financial interest in any product mentioned. Hanns Ackermann, PhD, Institute for Statistics, Johann Wolfgang Goethe-University, Frankfurt am Main, Germany, helped with the statistical evaluation.
J CATARACT REFRACT SURG—VOL 29, JANUARY 2003
117