- Email: [email protected]
Normal pupillary size in fluorescent and bright light
OPHTHALMOLOGY/BRIEF RESEARCH REPORT
Normal Pupillary Size in Fluorescent and Bright Light
Michael D. Witting, MD Deepi Goyal, MD From the Division of Emergency Medicine, University of Maryland School of Medicine, Baltimore, MD. Dr. Goyal is currently affiliated with the Mayo Clinic, Rochester, MN.
Study objective: Despite its common clinical use, the range of normal pupillary size has been described only crudely. The objective of this report is to describe the distribution of normal pupillary sizes in 2 light conditions that are available in clinical settings. Methods: Pupillary size measurements were taken from healthy patients by the principal investigator using a modified Haab scale. Measurements were obtained in areas with fluorescent lighting with an intensity of between 2,700 and 5,400 lux and by using bright handheld light sources producing a light intensity of greater than 54,000 lux. The effect of varying the type of handheld device (otoscope, ophthalmoscope, or penlight) on mean pupillary size was analyzed on the basis of intervals calculated from the t distribution. Results: One hundred twenty-eight patients were enrolled, with a mean age of 35±9 years. The mean pupillary size in fluorescent light was 3.6±0.7 mm, and the mean size in bright light was 2.6±0.5 mm. Extreme values in fluorescent light were 2.6 mm (5th percentile) and 5.0 mm (96th percentile). Extreme values in bright light were 1.9 mm (3rd percentile) and 3.6 mm (96th percentile). The type of bright light source had no effect on pupillary size measurement. Conclusion: Pupillary sizes of greater than 5.0 mm or less than 2.6 mm are rare (<10%) in normal individuals in fluorescent lighting (2,700 to 5,400 lux), and sizes of greater than 3.6 mm or less than 1.9 mm are rare (<10%) in bright light. [Ann Emerg Med. 2003;41:247-250.]
Copyright © 2003 by the American College of Emergency Physicians. 0196-0644/2003/$30.00 + 0 doi:10.1067/mem.2003.8
FEBRUARY 2003
41:2
ANNALS OF EMERGENCY MEDICINE
2 4 7
NORMAL PUPILLARY SIZE Witting & Goyal
INTRODUCTION
Abnormal pupillary size, when detected, can dramatically alter a patient’s management. It might indicate a toxic ingestion, ocular pathology, or intracranial pathology.1,2 Yet unlike other physiologic parameters, such as vital signs, the range of normal pupillary size is known only crudely. Pupillary size depends on a balance of sympathetic and parasympathetic tone and on one’s level of alertness.3 Perhaps most importantly, it varies under different lighting conditions.3,4 Just as light intensity can vary in 2 examining rooms, so can pupillary size. Perhaps because of this variation with light intensity, the range of normal pupillary size has defied description, and we were unable to find any study designed to define the range of normal pupillary size. The purpose of this report is to describe the distribution of normal pupillary sizes in light intensities produced by 2 sources available in clinical settings, fluorescent light and handheld light sources. M AT E R I A L S A N D M E T H O D S
Patients were healthy volunteers and were generally emergency department workers. Those with baseline anisocoria exceeding 1 mm, prior iridectomy, or a history of glaucoma were excluded. The study design was approved by our institutional review board. Light intensity was measured with a portable incident light meter (Acquamarine, Saginaw, MI) that has a range of 0 to 54,000 lux (0 to 5,000 foot-candles). Light intensity was measured in 2 fluorescent-lit EDs, one in a community hospital and the other in an academic tertiary care center. The light intensity varied between 2,700 and 5,400 lux in all treatment rooms, irrespective of whether the gauge was aimed directly at a light source or generally toward the ceiling. Therefore, fluorescent light was defined as any fluorescent light intensity of between 2,700 and 5,400 lux. Several handheld bright light sources were tested. Sources exceeding the upper range of our light meter (54,000 lux) were defined as bright. All 5 penlights obtained from various pharmaceutical representatives,
2 4 8
wall-mounted otoscopes, and ophthalmoscopes (model 11078, Welch Allyn Medical Products, Skaneateles Falls, NY), as well as a battery-powered otoscope (model 5578, Welch Allyn Medical Products), all produced bright light. Because the light meter only measured to 54,000 lux, bright light sources might have had variable intensity of greater than 54,000 lux. As part of a larger study, each patient received, according to randomized, double-blinded assignment, a mydriatic agent (phenylephrine) in one eye and placebo in the other. This report describes measurements taken from placebo-exposed eyes by the principal investigator using a gauge with a modified Haab scale.5 The gauge featured circles of increasing diameter corresponding to 0.2-mm gradations between 1.6 and 5 mm and 0.5-mm gradations between 6.0 and 9.0 mm. For fluorescent lighting, patients were asked to gaze toward the ceiling to avoid the near reflex. They were asked to stare at a spot between 2 light sources because light intensity here would be less sensitive to eye movement. For bright lighting, the principal investigator measured the pupil at its smallest size when exposed to bright light, and no attempt was made to avoid the near reflex. The bright light sources were varied throughout the study, and the type of light source used was recorded for each patient. All light sources were tested periodically to ensure they continued to produce bright light. After all data were collected, the principal investigator noted that, in photocopying, he had inadvertently enlarged the entire scale by 10%. Because this same enlarged scale was used for all measurements taken in the study, we adjusted the measured values by 10% and report only the adjusted values. Statistical analysis was done with Epicalc software (Epicalc 2000, Brixton Books, London, United Kingdom). Confidence intervals (CIs) for mean values and the paired difference between fluorescent and bright measured values were calculated from the t distribution. R E S U LT S
Pupils were measured in 128 patients, with a mean age of 35±9 years. Fifty-eight percent of patients had brown
ANNALS OF EMERGENCY MEDICINE
41:2
FEBRUARY 2003
NORMAL PUPILLARY SIZE Witting & Goyal
eyes. Fluorescent light measurements were taken in 27 different locations, and bright light measurements were taken with a variety of otoscopes, ophthalmoscopes, and penlights. The mean pupillary size in fluorescent light was 3.6±0.7 mm, and the mean pupillary size in bright light was 2.6±0.5 mm, with a mean difference of 0.97 mm (95% CI 0.85 to 1.07 mm; Figure). Pupillary size in fluorescent light ranged from 2.1 to 5.9 mm, with 2.6 mm and 5.0 mm corresponding to the 5th and 96th percentiles, respectively. Pupillary size in bright light ranged from 1.8 to 3.8 mm, with 1.9 and 3.6 mm corresponding to the 3rd and 96th percentiles, respectively. Patients measured with a penlight, otoscope, or ophthalmoscope had similar mean pupillary sizes (Table). DISCUSSION
It is common knowledge that the pupil’s size varies with light intensity, but no prior study has attempted to de-
Figure.
Distribution of pupillary sizes in fluorescent and bright light (n=128). Sizes are offset slightly to display duplicate entries.
6
Pupil size (mm)
5
4
scribe the range of normal pupillary size at a given light intensity. Without a precise range, detection of miosis or mydriasis might be inaccurate. Our data suggest logical limits of normal pupillary size in light conditions that are available in clinical settings. In the most extensive work to date on pupillary size, Loewenfeld6 studied pupillary size in darkness and the amplitude of the pupillary light reflex in various age groups. She did not measure pupillary size in light intensities found in clinical settings. In addition to pupillary size, the briskness of the pupillary light reflex and its amplitude might also indicate pupillary dysfunction. Nevertheless, these other parameters are difficult to measure, and detection of abnormality depends on operator expertise. Only size is recorded regularly in patient records. For clinicians to use the ranges of normal pupillary sizes we report, similar light intensity must be provided. This is accomplished more easily with bright light than with fluorescent light, and bright light sources can be available in any setting. The amount of light hitting a retina in room light depends on whether the person gazes directly at the light or at a darker area of the room. This effect is enhanced with incandescent light sources, which produce a spotlight effect. Although fluorescent lighting provides a diffuse light with little spotlight effect, the intensity of fluorescent light decreases as the bulb ages.7 Even if another setting also has fluorescent lighting, it might have a different watt density than that in our study setting. In many settings and gaze directions, light intensity might fall outside the 2,700- to 5,400-lux range. By contrast, a variety of available portable handheld sources supply light intensity exceeding 54,000 lux.
3
Table.
Effect of bright light source on pupillary size. 2 Light Source 1 Fluorescent light
FEBRUARY 2003
41:2
Bright light
ANNALS OF EMERGENCY MEDICINE
Penlight Otoscope Ophthalmoscope
Mean Pupillary Size, mm (95% CI)
No.
2.7 (2.5–2.8) 2.5 (2.4–2.7) 2.6 (2.5–2.8)
70 28 30
2 4 9
NORMAL PUPILLARY SIZE Witting & Goyal
When light is sufficiently bright to saturate both rod and cone light receptors, differences in light intensity become less important in producing a physiologic response.8 One would expect, then, that bright light measurement would be less sensitive to differences in light intensity than fluorescent light measurement. Indeed, there was no difference in pupillary size when measured with penlights, otoscopes, or ophthalmoscopes. These devices might have differed markedly in their light intensities because our meter measured only to 54,000 lux. Bright light pupillary size might be less affected by aging than fluorescent light pupillary size. In her study of pupillary size, Loewenfeld6 noted that pupillary size in darkness varies with age, with a peak in adolescence and a progressive decline with aging. She also noted that the light reflex amplitude declines with aging. With aging, then, the changes that affect pupillary size in bright light trend in opposite directions, tending to maintain a constant size. We expect pupillary size in fluorescent light to decrease with aging, as it does in darkness. Our data set did not have enough elderly patients to allow analysis of this theory. There are also a few practical advantages of fluorescent light measurement: it requires no equipment, and it is more comfortable for awake photophobic patients. Nevertheless, the balance of advantages favors bright light measurement. One limitation of this study is that a single person created the scale and took all measurements, allowing for human error. The postmeasurement adjustment, on the basis of comparison with the study’s scale and commercially available scales, limited potential bias associated with the scale itself. Also, because of a research interest in the pupil, the principal investigator might have systematically measured its size differently than most health care providers, such as avoiding the near reflex. Some interobserver reliability data from this study will be presented in a later report. An additional limitation is the age range of the population. The data might not be generalizable to all age groups. Further study should define the ranges of normal pupillary size in pediatric and geriatric groups.
2 5 0
In summary, because pupillary size depends on light intensity, description of the range of normal size requires a constant light intensity. Our data suggest ranges of pupillary size from 1.9 to 3.6 mm in bright light (>54,000 lux) and 2.6 mm to 5.0 mm in fluorescent light (2,700 to 5,400 lux). Author contributions: MDW designed the study, and collected and analyzed study data. DG contributed to the design of the study and collected pilot data. MDW drafted the manuscript, and DG substantially contributed to its revision. MDW takes responsibility for the paper as a whole. Received for publication April 1, 2002. Revision received July 16, 2002. Accepted for publication August 30, 2002. Presented at the 4th Annual Mid-Atlantic Regional Society for Academic Emergency Medicine Research Conference, Charlotte, NC, March 2001. Address for reprints: Michael D. Witting, MD, Division of Emergency Medicine, University of Maryland Medical School of Medicine, 419 West Redwood Street, Suite 280, Baltimore, MD 21201; E-mail [email protected]. We thank all of our patients for their trust and effort in participating in our study. We also thank Linda Kesselring, MS, ELS, for her editorial support.
REFERENCES 1. Plum F, Posner JB. The Diagnosis of Stupor and Coma. 3rd ed. Philadelphia, PA: Davis; 1980. 2. Smilkstein M. Ophthalmic principles. In: Goldfrank LR, ed. Goldfrank’s Toxicologic Emergencies. 6th ed. Stamford, CT: Appleton & Lange; 1998:447-456. 3. Burde R, Savino P, Trobe J. Clinical Decisions in Neuro-ophthalmology. St. Louis, MO: Mosby-Year Book; 1992. 4.
Walsh T. Neuroophthalmology. 4th ed. Baltimore, MD: Williams & Wilkins; 1997.
5.
Alexandridis E. The Pupil. New York, NY: Springer-Lange; 1985.
6. Loewenfeld I. The Pupil: Anatomy, Physiology, and Clinical Applications. Woburn, MA: Butterworth-Heinemann; 1999. 7. Schremp G, N S, eds. Reader’s Digest New Complete Do-It-Yourself Manual. Pleasantville, NY: The Reader’s Digest Association; 1995. 8.
Davson H. Physiology of the Eye. 5th ed. New York, NY: Pergamon Press; 1990.
ANNALS OF EMERGENCY MEDICINE
41:2
FEBRUARY 2003
Normal Pupillary Size in Fluorescent and Bright Light
Michael D. Witting, MD Deepi Goyal, MD From the Division of Emergency Medicine, University of Maryland School of Medicine, Baltimore, MD. Dr. Goyal is currently affiliated with the Mayo Clinic, Rochester, MN.
Study objective: Despite its common clinical use, the range of normal pupillary size has been described only crudely. The objective of this report is to describe the distribution of normal pupillary sizes in 2 light conditions that are available in clinical settings. Methods: Pupillary size measurements were taken from healthy patients by the principal investigator using a modified Haab scale. Measurements were obtained in areas with fluorescent lighting with an intensity of between 2,700 and 5,400 lux and by using bright handheld light sources producing a light intensity of greater than 54,000 lux. The effect of varying the type of handheld device (otoscope, ophthalmoscope, or penlight) on mean pupillary size was analyzed on the basis of intervals calculated from the t distribution. Results: One hundred twenty-eight patients were enrolled, with a mean age of 35±9 years. The mean pupillary size in fluorescent light was 3.6±0.7 mm, and the mean size in bright light was 2.6±0.5 mm. Extreme values in fluorescent light were 2.6 mm (5th percentile) and 5.0 mm (96th percentile). Extreme values in bright light were 1.9 mm (3rd percentile) and 3.6 mm (96th percentile). The type of bright light source had no effect on pupillary size measurement. Conclusion: Pupillary sizes of greater than 5.0 mm or less than 2.6 mm are rare (<10%) in normal individuals in fluorescent lighting (2,700 to 5,400 lux), and sizes of greater than 3.6 mm or less than 1.9 mm are rare (<10%) in bright light. [Ann Emerg Med. 2003;41:247-250.]
Copyright © 2003 by the American College of Emergency Physicians. 0196-0644/2003/$30.00 + 0 doi:10.1067/mem.2003.8
FEBRUARY 2003
41:2
ANNALS OF EMERGENCY MEDICINE
2 4 7
NORMAL PUPILLARY SIZE Witting & Goyal
INTRODUCTION
Abnormal pupillary size, when detected, can dramatically alter a patient’s management. It might indicate a toxic ingestion, ocular pathology, or intracranial pathology.1,2 Yet unlike other physiologic parameters, such as vital signs, the range of normal pupillary size is known only crudely. Pupillary size depends on a balance of sympathetic and parasympathetic tone and on one’s level of alertness.3 Perhaps most importantly, it varies under different lighting conditions.3,4 Just as light intensity can vary in 2 examining rooms, so can pupillary size. Perhaps because of this variation with light intensity, the range of normal pupillary size has defied description, and we were unable to find any study designed to define the range of normal pupillary size. The purpose of this report is to describe the distribution of normal pupillary sizes in light intensities produced by 2 sources available in clinical settings, fluorescent light and handheld light sources. M AT E R I A L S A N D M E T H O D S
Patients were healthy volunteers and were generally emergency department workers. Those with baseline anisocoria exceeding 1 mm, prior iridectomy, or a history of glaucoma were excluded. The study design was approved by our institutional review board. Light intensity was measured with a portable incident light meter (Acquamarine, Saginaw, MI) that has a range of 0 to 54,000 lux (0 to 5,000 foot-candles). Light intensity was measured in 2 fluorescent-lit EDs, one in a community hospital and the other in an academic tertiary care center. The light intensity varied between 2,700 and 5,400 lux in all treatment rooms, irrespective of whether the gauge was aimed directly at a light source or generally toward the ceiling. Therefore, fluorescent light was defined as any fluorescent light intensity of between 2,700 and 5,400 lux. Several handheld bright light sources were tested. Sources exceeding the upper range of our light meter (54,000 lux) were defined as bright. All 5 penlights obtained from various pharmaceutical representatives,
2 4 8
wall-mounted otoscopes, and ophthalmoscopes (model 11078, Welch Allyn Medical Products, Skaneateles Falls, NY), as well as a battery-powered otoscope (model 5578, Welch Allyn Medical Products), all produced bright light. Because the light meter only measured to 54,000 lux, bright light sources might have had variable intensity of greater than 54,000 lux. As part of a larger study, each patient received, according to randomized, double-blinded assignment, a mydriatic agent (phenylephrine) in one eye and placebo in the other. This report describes measurements taken from placebo-exposed eyes by the principal investigator using a gauge with a modified Haab scale.5 The gauge featured circles of increasing diameter corresponding to 0.2-mm gradations between 1.6 and 5 mm and 0.5-mm gradations between 6.0 and 9.0 mm. For fluorescent lighting, patients were asked to gaze toward the ceiling to avoid the near reflex. They were asked to stare at a spot between 2 light sources because light intensity here would be less sensitive to eye movement. For bright lighting, the principal investigator measured the pupil at its smallest size when exposed to bright light, and no attempt was made to avoid the near reflex. The bright light sources were varied throughout the study, and the type of light source used was recorded for each patient. All light sources were tested periodically to ensure they continued to produce bright light. After all data were collected, the principal investigator noted that, in photocopying, he had inadvertently enlarged the entire scale by 10%. Because this same enlarged scale was used for all measurements taken in the study, we adjusted the measured values by 10% and report only the adjusted values. Statistical analysis was done with Epicalc software (Epicalc 2000, Brixton Books, London, United Kingdom). Confidence intervals (CIs) for mean values and the paired difference between fluorescent and bright measured values were calculated from the t distribution. R E S U LT S
Pupils were measured in 128 patients, with a mean age of 35±9 years. Fifty-eight percent of patients had brown
ANNALS OF EMERGENCY MEDICINE
41:2
FEBRUARY 2003
NORMAL PUPILLARY SIZE Witting & Goyal
eyes. Fluorescent light measurements were taken in 27 different locations, and bright light measurements were taken with a variety of otoscopes, ophthalmoscopes, and penlights. The mean pupillary size in fluorescent light was 3.6±0.7 mm, and the mean pupillary size in bright light was 2.6±0.5 mm, with a mean difference of 0.97 mm (95% CI 0.85 to 1.07 mm; Figure). Pupillary size in fluorescent light ranged from 2.1 to 5.9 mm, with 2.6 mm and 5.0 mm corresponding to the 5th and 96th percentiles, respectively. Pupillary size in bright light ranged from 1.8 to 3.8 mm, with 1.9 and 3.6 mm corresponding to the 3rd and 96th percentiles, respectively. Patients measured with a penlight, otoscope, or ophthalmoscope had similar mean pupillary sizes (Table). DISCUSSION
It is common knowledge that the pupil’s size varies with light intensity, but no prior study has attempted to de-
Figure.
Distribution of pupillary sizes in fluorescent and bright light (n=128). Sizes are offset slightly to display duplicate entries.
6
Pupil size (mm)
5
4
scribe the range of normal pupillary size at a given light intensity. Without a precise range, detection of miosis or mydriasis might be inaccurate. Our data suggest logical limits of normal pupillary size in light conditions that are available in clinical settings. In the most extensive work to date on pupillary size, Loewenfeld6 studied pupillary size in darkness and the amplitude of the pupillary light reflex in various age groups. She did not measure pupillary size in light intensities found in clinical settings. In addition to pupillary size, the briskness of the pupillary light reflex and its amplitude might also indicate pupillary dysfunction. Nevertheless, these other parameters are difficult to measure, and detection of abnormality depends on operator expertise. Only size is recorded regularly in patient records. For clinicians to use the ranges of normal pupillary sizes we report, similar light intensity must be provided. This is accomplished more easily with bright light than with fluorescent light, and bright light sources can be available in any setting. The amount of light hitting a retina in room light depends on whether the person gazes directly at the light or at a darker area of the room. This effect is enhanced with incandescent light sources, which produce a spotlight effect. Although fluorescent lighting provides a diffuse light with little spotlight effect, the intensity of fluorescent light decreases as the bulb ages.7 Even if another setting also has fluorescent lighting, it might have a different watt density than that in our study setting. In many settings and gaze directions, light intensity might fall outside the 2,700- to 5,400-lux range. By contrast, a variety of available portable handheld sources supply light intensity exceeding 54,000 lux.
3
Table.
Effect of bright light source on pupillary size. 2 Light Source 1 Fluorescent light
FEBRUARY 2003
41:2
Bright light
ANNALS OF EMERGENCY MEDICINE
Penlight Otoscope Ophthalmoscope
Mean Pupillary Size, mm (95% CI)
No.
2.7 (2.5–2.8) 2.5 (2.4–2.7) 2.6 (2.5–2.8)
70 28 30
2 4 9
NORMAL PUPILLARY SIZE Witting & Goyal
When light is sufficiently bright to saturate both rod and cone light receptors, differences in light intensity become less important in producing a physiologic response.8 One would expect, then, that bright light measurement would be less sensitive to differences in light intensity than fluorescent light measurement. Indeed, there was no difference in pupillary size when measured with penlights, otoscopes, or ophthalmoscopes. These devices might have differed markedly in their light intensities because our meter measured only to 54,000 lux. Bright light pupillary size might be less affected by aging than fluorescent light pupillary size. In her study of pupillary size, Loewenfeld6 noted that pupillary size in darkness varies with age, with a peak in adolescence and a progressive decline with aging. She also noted that the light reflex amplitude declines with aging. With aging, then, the changes that affect pupillary size in bright light trend in opposite directions, tending to maintain a constant size. We expect pupillary size in fluorescent light to decrease with aging, as it does in darkness. Our data set did not have enough elderly patients to allow analysis of this theory. There are also a few practical advantages of fluorescent light measurement: it requires no equipment, and it is more comfortable for awake photophobic patients. Nevertheless, the balance of advantages favors bright light measurement. One limitation of this study is that a single person created the scale and took all measurements, allowing for human error. The postmeasurement adjustment, on the basis of comparison with the study’s scale and commercially available scales, limited potential bias associated with the scale itself. Also, because of a research interest in the pupil, the principal investigator might have systematically measured its size differently than most health care providers, such as avoiding the near reflex. Some interobserver reliability data from this study will be presented in a later report. An additional limitation is the age range of the population. The data might not be generalizable to all age groups. Further study should define the ranges of normal pupillary size in pediatric and geriatric groups.
2 5 0
In summary, because pupillary size depends on light intensity, description of the range of normal size requires a constant light intensity. Our data suggest ranges of pupillary size from 1.9 to 3.6 mm in bright light (>54,000 lux) and 2.6 mm to 5.0 mm in fluorescent light (2,700 to 5,400 lux). Author contributions: MDW designed the study, and collected and analyzed study data. DG contributed to the design of the study and collected pilot data. MDW drafted the manuscript, and DG substantially contributed to its revision. MDW takes responsibility for the paper as a whole. Received for publication April 1, 2002. Revision received July 16, 2002. Accepted for publication August 30, 2002. Presented at the 4th Annual Mid-Atlantic Regional Society for Academic Emergency Medicine Research Conference, Charlotte, NC, March 2001. Address for reprints: Michael D. Witting, MD, Division of Emergency Medicine, University of Maryland Medical School of Medicine, 419 West Redwood Street, Suite 280, Baltimore, MD 21201; E-mail [email protected]. We thank all of our patients for their trust and effort in participating in our study. We also thank Linda Kesselring, MS, ELS, for her editorial support.
REFERENCES 1. Plum F, Posner JB. The Diagnosis of Stupor and Coma. 3rd ed. Philadelphia, PA: Davis; 1980. 2. Smilkstein M. Ophthalmic principles. In: Goldfrank LR, ed. Goldfrank’s Toxicologic Emergencies. 6th ed. Stamford, CT: Appleton & Lange; 1998:447-456. 3. Burde R, Savino P, Trobe J. Clinical Decisions in Neuro-ophthalmology. St. Louis, MO: Mosby-Year Book; 1992. 4.
Walsh T. Neuroophthalmology. 4th ed. Baltimore, MD: Williams & Wilkins; 1997.
5.
Alexandridis E. The Pupil. New York, NY: Springer-Lange; 1985.
6. Loewenfeld I. The Pupil: Anatomy, Physiology, and Clinical Applications. Woburn, MA: Butterworth-Heinemann; 1999. 7. Schremp G, N S, eds. Reader’s Digest New Complete Do-It-Yourself Manual. Pleasantville, NY: The Reader’s Digest Association; 1995. 8.
Davson H. Physiology of the Eye. 5th ed. New York, NY: Pergamon Press; 1990.
ANNALS OF EMERGENCY MEDICINE
41:2
FEBRUARY 2003