1040-5488/14/9111-1372/0 VOL. 91, NO. 11, PP. 1372Y1376 OPTOMETRY AND VISION SCIENCE Copyright * 2014 American Academy of Optometry
ORIGINAL ARTICLE
Color Vision Deficiency in Zahedan, Iran: Lower than Expected Hamed Momeni-Moghaddam*, Jason S. Ng†, Hassan Robabi*, and Farshid Yaghubi‡
ABSTRACT Purpose. To estimate the prevalence of congenital red-green color vision defects in the elementary school students of Zahedan in 2012. Methods. In this cross-sectional study, 1000 students with a mean (TSD) age of 9.0 (T1.4) years were selected randomly from a large primary school population. Color vision was evaluated using the Ishihara pseudoisochromatic color plates (38-plate edition). A daylight fluorescent tube was used as an illuminant C equivalent (i.e., 860 lux, color rendering index greater than 92, and color temperature = 6500 K). Having more than three misreadings on the test was considered a failing criterion. Data were analyzed in SPSS version 17 software using W2 tests. Results. Nine students (0.9%) made more than three errors on the Ishihara test. Based on this criterion, the prevalence of red-green color vision deficiency in girls and boys was 0.2 and 1.6% (p = 0.02), respectively. Conclusions. The prevalence of red-green color vision deficiency was found to be significantly lower in Zahedan than comparable reports in the literature. (Optom Vis Sci 2014;91:1372Y1376) Key Words: color vision deficiency, color blindness, prevalence, Iran
C
olor vision can be defined as a characteristic of visual sense that enables a person to discriminate between light of different wavelengths.1 This is attributed to the presence of three types of cone photoreceptors in the retina in patients with normal color vision.2 When normal trichromatic vision does not exist, it is called anomalous color vision, color vision deficiency (CVD), or most commonly the somewhat erroneous term color blindness.3 Color vision deficiency can be classified by etiology as acquired or congenital. Clinically, these conditions are further divided into partial CVD (red-green and blue-yellow) and total color blindness (achromatopsia). The most common form overall is congenital red-green CVD, which is an X-linked recessive (XLR) trait and, by definition, it is more common in male than in female subjects.4 This type of CVD has two subtypes, protan and deuteran, each
*MSc † OD, PhD ‡ BSc Department of Optometry, School of Paramedical Sciences, Mashhad University of Medical Sciences, Mashhad, Iran (HM-M); Southern California College of Optometry, Marshall B. Ketchum University, Fullerton, California (JSN); and School of Nursing (HR) and School of Rehabilitation (FY), Zahedan University of Medical Sciences, Zahedan, Iran.
of which is further subdivided into dichromacy (protanopia and deuteranopia) and anomalous trichromacy (protanomaly and deuteranomaly).2 The dichromatic state is characterized by missing one type of cone photopigment (usually M- and, to a lesser extent, L-cones), and in anomalous trichromacy, there is a shift in cone peak sensitivity toward the lower or higher wavelengths owing to a change in one of these pigments.5 Generally, M-cone (green) photopigment anomalies are the cause of the redgreen CVD in 75% of color-deficient men.6 X-linked recessive anomalies of color vision are prevalent throughout the world and more so in men because they have only one X chromosome and thus would more commonly inherit an XLR trait. Women have two X chromosomes and would need to inherit defective genes on both X chromosomes. Standard references cite that up to 8% of the European male population and 0.4% of the female population have anomalous trichromatic color vision or dichromacy.6 These percentages are somewhat different in other areas and among other ethnicities.7 Some previous studies reported CVD prevalences of 0.9 to 9.6% using the Ishihara isochromatic color plates.8Y22 Although isolated familial marriages may pose little risk to increasing the general prevalence of recessive genetic disorders, repeated consanguineous marriages within a group are more problematic. After repeated generations of familial marriage, the actual
Optometry and Vision Science, Vol. 91, No. 11, November 2014
Copyright © American Academy of Optometry. Unauthorized reproduction of this article is prohibited.
Color Vision Deficiency in IranVMomeni-Moghaddam et al.
genetic relationship between two people is closer than the most immediate relationship would suggest. Recessive genetic disorders can affect any family; however, marriages between close relatives significantly increase the probability that children will be affected by a recessive genetic disorder, including XLR disorders such as red-green color deficiency. Alleles that are rare in large populations can randomly increase to high frequency in small groups within a few generations because of the founder effect and accelerated genetic drift in a breeding pool of restricted size. Genetic drift explains how disadvantageous alleles can become common in small populations.23 The higher frequency of familial marriage in Zahedan compared with other regions of Iran persuaded us to design this study to estimate the prevalence of CVD among elementary schoolYaged students in southeastern Iran. A higher prevalence than normally reported was expected owing to higher rates of familial marriage. The selection of elementary schoolYaged students as the lowest age level of students in Iran will assist in prevention of some problems that children may encounter in the following years such as failure in academic achievement or job selection. It would also help a child and his or her educators to know the status of a child’s color vision. Also, although a range of diseases can produce acquired CVDs, in the studied population (elementary school students), acquired CVDs are less common than hereditary CVDs. In addition, management of a child with CVD through patient education and possibly using tinted lenses could improve his or her quality of life.
METHODS In this cross-sectional study, the target population was primary school students (grades 1 to 5). The period of elementary school in Iran is 5 years. In total, 1000 students were selected as the sample out of a population of 63,187 students using a cluster random sampling strategy. The city of Zahedan has two educational areas with 100 schools in each area. The schools in each area were considered as a cluster. A total of 50 clusters were randomly selected based on a list of schools obtained from the Zahedan education department. The students to be screened were randomly selected based on a random number table, so that 20 students were in each cluster. The nature, purpose, and methods of this research work were clearly explained to the school administrators and parents of the children in advance in their native language, and informed consent was obtained from all of them. Also, because of the potential existence of a social stigma for CVD, they were assured that the children’s information would be kept confidential in accordance with the tenets of the Declaration of Helsinki. All participants were transferred to a clinic for an extensive eye examination. Those who met inclusion criteria were entered into the study. It was determined that the enumerated students who did not attend the examination process be contacted again, and if they were not able to participate in the study because of personal problems, or they were excluded from the study, other students would be randomly selected based on the random number table to replace them. Although we planned for the possibility of needing to exclude subjects that could not be found or recruited the first time, all subjects originally recruited were tested. That is, no student declined to participate in the study and our random replacement procedure was never used.
1373
Inclusion criteria were absence of anterior and posterior segment pathologies diagnosed with slit lamp biomicroscopy and direct ophthalmoscopy, no history of eye trauma, and no use of ocular or systemic drugs at the time of evaluation. For each student, refractive error was determined with retinoscopy and subjective refraction. The mean and SD of the bestcorrected decimal visual acuity in the right and left eyes of the students with normal and abnormal color vision status was 0.96 T 0.12 (95% confidence interval [CI], 0.95 to 0.98), 0.95 T 0.13 (95% CI, 0.95 to 0.96) and 0.95 T 0.08 (95% CI, 0.95 to 0.96), 0.97 T 0.60 (95% CI, 0.94 to 1.00), respectively. The mean visual acuity was not different between the students with normal and abnormal color vision status (p = 0.9 for the right eye and p = 0.6 for the left eye). Anterior and posterior segment evaluation was done using slit lamp biomicroscopy and direct ophthalmoscopy to rule out any organic causes of color deficiency. Evaluation of color vision was performed using the Ishihara isochromatic color plates (38 plates) under binocular viewing conditions, because inherited defects are bilateral conditions and resultantly an inherited color vision defect is generally not missed owing to binocular testing. A daylight fluorescent tube with a color rendering index of greater than 92 and a color temperature of 6500 K was used for this purpose because this source has illumination characteristics close enough (80 fc or 860 lux) to the Macbeth Illuminant C to make it a reasonable substitute.5,23 Illumination was measured with a lux meter (Hagner Photometer, model EC1, Sweden) at the level of test surface, a distance about 75 cm from the subject. With the best correction in a trial frame and while the head was stable, the color vision plates were held about 75 cm from the subject by another examiner, parallel to the subject’s face and perpendicular to their line of sight. The sequence of pages was demonstration (page 1), transformation (pages 2 to 9), vanishing (pages 10 to 17), hidden digits (pages 18 to 21), and diagnostic plates (pages 22 to 25). Each plate was presented for 3 to 5 seconds. If there were three or less misreadings on plates 2 to 21 of the 38-plate test version, we considered the child to have passed the color vision test because a failing criterion of more than three errors has a sensitivity of 97.7% in the detection of CVD.24 Also, Cole25 has cited in a review study that three or four errors will indicate a probable CVD and more than five errors will show a certain CVD. After data collection, the data were analyzed by W2 tests in the software program Statistical Package for the Social Sciences (SPSS version 17, SPSS Inc, Chicago, IL). In all tests, the significance level was considered to be 0.05.
RESULTS Of the 1000 students under study from grade 1 to grade 5, 502 (50.2%) were girls. The age range was 7 to 11 years. The mean (TSD) age was 9.0 (T1.4) years, with no significant difference in mean age between the two sexes (p = 0.60). The number of misreadings on plates 2 to 21 of the 38-plate test version in the two sexes is displayed in Table 1. Based on a failing criterion, more than two and three misreadings, the sex distribution of color vision status is shown in Table 2.
Optometry and Vision Science, Vol. 91, No. 11, November 2014
Copyright © American Academy of Optometry. Unauthorized reproduction of this article is prohibited.
1374 Color Vision Deficiency in IranVMomeni-Moghaddam et al. TABLE 1.
Frequency distribution of the number of misreadings separately in the two sexes No. misreadings Sex
One
Two
Three
More than three
Boys Girls Total
13 7 20
11 4 15
9 3 12
8 1 9
As can be seen in Table 2, based on a failing criterion of more than three errors, 0.9% (95% CI, 0.4 to 1.7) of students likely had CVD, and the prevalence in girls and boys was 0.2% (95% CI, 0.0 to 1.1) and 1.6% (95% CI, 0.8 to 3.1), respectively. With attention to a failing criterion of more than two errors, the prevalence of CVD in all students and separately in girls and boys was 1.2% (95% CI, 0.7 to 2.0), 0.6% (95% CI, 0.2 to 1.7), and 1.8% (95% CI, 0.9 to 3.4), respectively. The difference between the two sexes was found to be statistically significant (p = 0.02, Fisher exact test). The mean (TSD) age in students with and without CVD was 8.8 (T1.3) and 9.0 (T1.4) years, respectively, which was not found to be significantly different (p = 0.70, independent samples t test).
DISCUSSION Based on our findings, the prevalence of CVD in Zahedan province appears to be two times lower in girls (0.2 vs. 0.4%) and five times lower in boys (1.6 vs. 8.0%), compared with their European counterparts based on a failing criterion of more than three misreadings. There have been very few large-scale studies addressing the prevalence of inherited CVD in Iran and none was conducted in Zahedan to our knowledge. A recent study by Birch22 examined the prevalence of red-green CVD worldwide. In that study, Iran is grouped into the ‘‘European’’ group, one of the five anthropologically defined groups as the author describes. The Iranian people (also known as Iranic people) are a diverse IndoEuropean ethno-linguistic group that comprises the speakers of Iranian languages.26 This grouping is somewhat reasonable given that Modarres et al.21 found prevalences of red-green CVD of 8.18 and 0.43% in male and female subjects, respectively, in the secondary school population of Tehran, Iran. They stated that their study was the ‘‘first study to determine prevalence of congenital color blindness in Iran,’’ but in fact,
several other studies were conducted previously. Zarrabi and Sadighian16 determined the prevalence of CVD in primary school children in Tehran in the 1970s and found prevalences of 2.04 and 0.53% in male and female subjects, respectively. Similarly, Dargahi et al.15 and Farokhfar11 found male prevalence of CVD around 2%. All other previous studies, except for Modarres et al.,21 found similar prevalences to this present study. The discrepancy in prevalences between Zarrabi and Sadighian16 and Modarres et al.,21 at least for Tehran, may be due to the fact that migration into and possibly the diversity of the gene pool in Tehran have increased according to the Statistical Center of Iran.27 Such increases in migration and settlement of people with an increased diversity in backgrounds push the population more toward Hardy-Weinberg equilibrium as Birch suggests.22 A large, randomly mating population in Hardy-Weinberg equilibrium would show stable congenital CVD prevalences of about 8 and 0.4% for male and female subjects, respectively. In Zahedan, the population is somewhat smaller than that in Tehran and relatively more isolated (i.e., less migration). The familial marriage rates are higher than those in Tehran28 and the hypothesis was that a higher prevalence rate of CVD would be found,29 but unexpectedly the prevalence in this study was significantly lower than that reported by Modarres et al.21 and significantly lower in other ethnic groups such as Asians and Africans.22 The reason for this finding is unclear. Although familial marriage rates are relatively high in Zahedan, it is not a small isolated ethnic group. Although we used random sampling in this study and the population we sampled from was rather large, the data appear to indicate that we have identified a particular extreme in a local population. The female prevalence of CVD in this study is not necessarily significantly different from European populations when the CI is considered. However, the male prevalence of CVD in this study is significantly lower compared with European populations and even in the aforementioned studies. This is not easily explainable. With a point estimate of 1.6% for male prevalence, one would expect based on theory alone that the female point estimate of prevalence would be 0.03% (i.e., 1.6% squared), but the estimate found in this study was a tenth of that. It is possible that the sample size was insufficient to reliably estimate female prevalence given that it is already expected to be relatively low. Another aspect to consider is whether the lower prevalence found in this study could be attributed to our methodology. We recruited at random from a large population of schoolchildren
TABLE 2.
The frequency of red-green CVD in all students by sex based on the two failing criteria Sex Boys
Girls
8 (1.6) (0.8Y3.1) 9 (1.8) (0.9Y3.4) 490 (98.4) 489 (98.1) 498 (100.0)
1 (0.2) (0.0Y1.1) 3 (0.6) (0.2Y1.7) 501 (99.8) 499 (99.4) 502 (100.0)
CVD status Abnormal Normal Total
Total
n (%) (95% CI) More More More More
than than than than
3 errors 2 errors 3 errors 2 errors
Optometry and Vision Science, Vol. 91, No. 11, November 2014
Copyright © American Academy of Optometry. Unauthorized reproduction of this article is prohibited.
9 (0.9) (0.4Y1.7) 12 (1.2) (0.7Y2.0) 991 (99.1) 988 (98.8) 1000 (100.0)
Color Vision Deficiency in IranVMomeni-Moghaddam et al.
from multiple sites/schools, avoiding several potential sources of bias.22 These include sampling bias arising from studying a limited number of schools where relatives/siblings could confound the results and selection bias from recruiting only self-referred volunteers. We used a controlled and carefully selected illumination source, unlike the variability of daylight or other light sources used in some other studies.11,21 Use of the 38-plate Ishihara test has been deemed the best screening test for the identification of red-green inherited CVD repeatedly in the literature.22 Best refractive correction, when needed, was in place for testing. It should be emphasized that the Ishihara color plates do not test for blue-yellow defects, and as a result, their prevalence is not known from this work. We also used a rather strict testing protocol and pass/fail criterion, which would actually cause an increase in false positives theoretically (i.e., labeling a ‘‘fail’’ of the test as only three or four plates missed would result in an artificially higher prevalence, that is, a higher sensitivity and a lower specificity; labeling a ‘‘fail’’ of the test as more than five plates, for example, would result in a lower prevalence, that is, a lower sensitivity but a higher specificity). Limiting time per plate and presenting the plates at the prescribed test distance, which are not always done clinically, adhere to the test instructions and could theoretically decrease the specificity of the test. Additionally, the color testing results were confirmed by repeat testing with another examiner, ensuring their repeatability, which could also result in a decrease of the observed prevalence. This additional confirmation of the reliability of the results has not typically been reported in the literature with regard to large-scale population-based color vision studies. Finally, all subjects underwent complete ophthalmic examination, ensuring that acquired color vision defects did not confound our findings; this could also reduce the measured prevalence. Binocular testing was used in this study, as this is the common clinical way to test for congenital CVD, but it is possible that congenital CVD could be missed in some cases with interocular physiological variations between the eyes (e.g., macular pigment or the crystalline lens). This could have contributed to a lower prevalence. The estimated prevalence of inherited CVD in Zahedan, Iran, is less than 1% and significantly lower in boys. The low prevalence is among the lowest found in the literature when considering all populations worldwide and is unexpected.
CONCLUSIONS Color vision deficiency, as detected by the Ishihara test, makes up about 1% of the elementary schoolYaged population in southeastern Iran.
ACKNOWLEDGMENTS This research was supported by the health promotion research center of Zahedan University of Medical Sciences. The authors would like to thank the participants who made this study possible. None of the authors have any proprietary interests or conflicts of interest related to this submission. Received February 9, 2014; accepted August 21, 2014.
1375
REFERENCES 1. Eskridge JB, Amos JF, Bartlett JD. Clinical Procedures in Optometry Philadelphia, PA: J.B. Lippincott; 1991. 2. Swanson WH, Cohen JM. Color vision. Ophthalmol Clin North Am 2003;16:179Y203. 3. Duckman RH, ed. Visual Development, Diagnosis, and Treatment of the Pediatric Patient. Philadelphia, PA: Lippincott Williams & Wilkins; 2006. 4. Deeb SS. The molecular basis of variation in human color vision. Clin Genet 2005;67:369Y77. 5. Grosvenor T. Primary Care Optometry 5th ed. Boston, MA: Butterworth-Heinemann; 2007:285Y6. 6. Schwartz SH. Visual Perception: A Clinical Orientation, 4th ed. New York, NY: McGraw Hill; 2010. 7. Citirik M, Acaroglu G, Batman C, Zilelioglu O. Congenital color blindness in young Turkish men. Ophthalmic Epidemiol 2005; 12:133Y7. 8. Pramanik T, Sherpa MT, Shrestha R. Color vision deficiency among medical students: an unnoticed problem. Nepal Med Coll J 2010; 12:81Y3. 9. Oriowo OM, Alotaibi AZ. Colour vision screening among Saudi Arabian children. S Afr Optom 2008;67:56Y61. 10. Gupta Y, Sukul RR, Gupta M, Phougat A, Jain R, Varshney A. School eye survey in rural population in UP, India. Nepal J Ophthalmol 2011;3:78Y9. 11. Farokhfar A. [Prevalence of color blindness in the primary school students in Sari Township 1999]. J Mazandaran Univ Med Sci 2001;11:57Y62. 12. Niroula DR, Saha CG. The incidence of color blindness among some school children of Pokhara, Western Nepal. Nepal Med Coll J 2010;12:48Y50. 13. Songden SD, Ike EE. Determination of colour vision using Ishihara and Dvorine plates. J Med Tropics 2010;12:14Y7. 14. Mehra KS. Incidence of colour blindness in Indians. Br J Ophthalmol 1963;47:485Y7. 15. Dargahi H, Einollahi N, Dashti N. Color blindness defect and medical laboratory technologists: unnoticed problems and the care for screening. Acta Med Iran 2010;48:172Y7. 16. Zarrabi Z, Sadighian M. Incidence of colour blindness (colour defect) among Iranian primary school children. Acta Med Iran 1974;17:70Y2. 17. Yasmin A, Jahan N, Akhter R. Assessment of colour blindness and erythrocyte G6PD enzyme status among the school children of Dhaka City. J Bangladesh Soc Physiol 2009;4:64Y70. 18. Chia A, Gazzard G, Tong L, Zhang X, Sim EL, Fong A, Mei Saw S. Red-green colour blindness in Singaporean children. Clin Experiment Ophthalmol 2008;36:464Y7. 19. Cabrero FJ, Ortiz MA, Mesa MS, Fuster V, Moral P. Red-green colour blindness in the Tormes-Alberche Valley (Avila-Central Spain). Anthropol Anz 1997;55:295Y301. 20. Qian YS, Chu RY, He JC, Sun XH, Zhou XT, Zhao NQ, Hu DN, Hoffman MR, Dai JH, Qu XM, Pao KE. Incidence of myopia in high school students with and without red-green color vision deficiency. Invest Ophthalmol Vis Sci 2009;50:1598Y605. 21. Modarres M, Mirsamadi M, Peyman GA. Prevalence of congenital color deficiencies in secondary-school students in Tehran. Int Ophthalmol 1996;20:221Y2. 22. Birch J. Worldwide prevalence of red-green color deficiency. J Opt Soc Am (A) 2012;29:313Y20.
Optometry and Vision Science, Vol. 91, No. 11, November 2014
Copyright © American Academy of Optometry. Unauthorized reproduction of this article is prohibited.
1376 Color Vision Deficiency in IranVMomeni-Moghaddam et al. 23. Bittles AH. The role and significance of consanguinity as a demographic variable. Popul Dev Rev 1994;20:561Y84. 24. Birch J. Identification of red-green colour deficiency: sensitivity of the Ishihara and American Optical Company (Hard, Rand and Rittler) pseudo-isochromatic plates to identify slight anomalous trichromatism. Ophthalmic Physiol Opt 2010;30:667Y71. 25. Cole BL. Assessment of inherited colour vision defects in clinical practice. Clin Exp Optom 2007;90:157Y75. 26. Iranian peoples. Wikipedia, The Free Encyclopedia; 2014. Available at: http://en.wikipedia.org/wiki/Iranian_peoples. Accessed August 14, 2014. 27. Statistical Centre of Iran (SCI). Emigration Report; 1990. Available at: http://www.sci.org.ir/SitePages/report_90/Emigration_report.aspx. Accessed February 5, 2014.
28. Saadat M, Ansari-Lari M, Farhud DD. Consanguineous marriage in Iran. Ann Hum Biol 2004;31:263Y9. 29. Shah A, Hussain R, Fareed M, Afzal M. Prevalence of red-green color vision defects among Muslim males and females of Manipur, India. Iran J Public Health 2013;42:16Y24.
Hamed Momeni-Moghaddam Optometry Department Parastar Str., Ahmad Abad Ave Mashhad, Iran e-mail: [email protected]
Optometry and Vision Science, Vol. 91, No. 11, November 2014
Copyright © American Academy of Optometry. Unauthorized reproduction of this article is prohibited.
ORIGINAL ARTICLE
Color Vision Deficiency in Zahedan, Iran: Lower than Expected Hamed Momeni-Moghaddam*, Jason S. Ng†, Hassan Robabi*, and Farshid Yaghubi‡
ABSTRACT Purpose. To estimate the prevalence of congenital red-green color vision defects in the elementary school students of Zahedan in 2012. Methods. In this cross-sectional study, 1000 students with a mean (TSD) age of 9.0 (T1.4) years were selected randomly from a large primary school population. Color vision was evaluated using the Ishihara pseudoisochromatic color plates (38-plate edition). A daylight fluorescent tube was used as an illuminant C equivalent (i.e., 860 lux, color rendering index greater than 92, and color temperature = 6500 K). Having more than three misreadings on the test was considered a failing criterion. Data were analyzed in SPSS version 17 software using W2 tests. Results. Nine students (0.9%) made more than three errors on the Ishihara test. Based on this criterion, the prevalence of red-green color vision deficiency in girls and boys was 0.2 and 1.6% (p = 0.02), respectively. Conclusions. The prevalence of red-green color vision deficiency was found to be significantly lower in Zahedan than comparable reports in the literature. (Optom Vis Sci 2014;91:1372Y1376) Key Words: color vision deficiency, color blindness, prevalence, Iran
C
olor vision can be defined as a characteristic of visual sense that enables a person to discriminate between light of different wavelengths.1 This is attributed to the presence of three types of cone photoreceptors in the retina in patients with normal color vision.2 When normal trichromatic vision does not exist, it is called anomalous color vision, color vision deficiency (CVD), or most commonly the somewhat erroneous term color blindness.3 Color vision deficiency can be classified by etiology as acquired or congenital. Clinically, these conditions are further divided into partial CVD (red-green and blue-yellow) and total color blindness (achromatopsia). The most common form overall is congenital red-green CVD, which is an X-linked recessive (XLR) trait and, by definition, it is more common in male than in female subjects.4 This type of CVD has two subtypes, protan and deuteran, each
*MSc † OD, PhD ‡ BSc Department of Optometry, School of Paramedical Sciences, Mashhad University of Medical Sciences, Mashhad, Iran (HM-M); Southern California College of Optometry, Marshall B. Ketchum University, Fullerton, California (JSN); and School of Nursing (HR) and School of Rehabilitation (FY), Zahedan University of Medical Sciences, Zahedan, Iran.
of which is further subdivided into dichromacy (protanopia and deuteranopia) and anomalous trichromacy (protanomaly and deuteranomaly).2 The dichromatic state is characterized by missing one type of cone photopigment (usually M- and, to a lesser extent, L-cones), and in anomalous trichromacy, there is a shift in cone peak sensitivity toward the lower or higher wavelengths owing to a change in one of these pigments.5 Generally, M-cone (green) photopigment anomalies are the cause of the redgreen CVD in 75% of color-deficient men.6 X-linked recessive anomalies of color vision are prevalent throughout the world and more so in men because they have only one X chromosome and thus would more commonly inherit an XLR trait. Women have two X chromosomes and would need to inherit defective genes on both X chromosomes. Standard references cite that up to 8% of the European male population and 0.4% of the female population have anomalous trichromatic color vision or dichromacy.6 These percentages are somewhat different in other areas and among other ethnicities.7 Some previous studies reported CVD prevalences of 0.9 to 9.6% using the Ishihara isochromatic color plates.8Y22 Although isolated familial marriages may pose little risk to increasing the general prevalence of recessive genetic disorders, repeated consanguineous marriages within a group are more problematic. After repeated generations of familial marriage, the actual
Optometry and Vision Science, Vol. 91, No. 11, November 2014
Copyright © American Academy of Optometry. Unauthorized reproduction of this article is prohibited.
Color Vision Deficiency in IranVMomeni-Moghaddam et al.
genetic relationship between two people is closer than the most immediate relationship would suggest. Recessive genetic disorders can affect any family; however, marriages between close relatives significantly increase the probability that children will be affected by a recessive genetic disorder, including XLR disorders such as red-green color deficiency. Alleles that are rare in large populations can randomly increase to high frequency in small groups within a few generations because of the founder effect and accelerated genetic drift in a breeding pool of restricted size. Genetic drift explains how disadvantageous alleles can become common in small populations.23 The higher frequency of familial marriage in Zahedan compared with other regions of Iran persuaded us to design this study to estimate the prevalence of CVD among elementary schoolYaged students in southeastern Iran. A higher prevalence than normally reported was expected owing to higher rates of familial marriage. The selection of elementary schoolYaged students as the lowest age level of students in Iran will assist in prevention of some problems that children may encounter in the following years such as failure in academic achievement or job selection. It would also help a child and his or her educators to know the status of a child’s color vision. Also, although a range of diseases can produce acquired CVDs, in the studied population (elementary school students), acquired CVDs are less common than hereditary CVDs. In addition, management of a child with CVD through patient education and possibly using tinted lenses could improve his or her quality of life.
METHODS In this cross-sectional study, the target population was primary school students (grades 1 to 5). The period of elementary school in Iran is 5 years. In total, 1000 students were selected as the sample out of a population of 63,187 students using a cluster random sampling strategy. The city of Zahedan has two educational areas with 100 schools in each area. The schools in each area were considered as a cluster. A total of 50 clusters were randomly selected based on a list of schools obtained from the Zahedan education department. The students to be screened were randomly selected based on a random number table, so that 20 students were in each cluster. The nature, purpose, and methods of this research work were clearly explained to the school administrators and parents of the children in advance in their native language, and informed consent was obtained from all of them. Also, because of the potential existence of a social stigma for CVD, they were assured that the children’s information would be kept confidential in accordance with the tenets of the Declaration of Helsinki. All participants were transferred to a clinic for an extensive eye examination. Those who met inclusion criteria were entered into the study. It was determined that the enumerated students who did not attend the examination process be contacted again, and if they were not able to participate in the study because of personal problems, or they were excluded from the study, other students would be randomly selected based on the random number table to replace them. Although we planned for the possibility of needing to exclude subjects that could not be found or recruited the first time, all subjects originally recruited were tested. That is, no student declined to participate in the study and our random replacement procedure was never used.
1373
Inclusion criteria were absence of anterior and posterior segment pathologies diagnosed with slit lamp biomicroscopy and direct ophthalmoscopy, no history of eye trauma, and no use of ocular or systemic drugs at the time of evaluation. For each student, refractive error was determined with retinoscopy and subjective refraction. The mean and SD of the bestcorrected decimal visual acuity in the right and left eyes of the students with normal and abnormal color vision status was 0.96 T 0.12 (95% confidence interval [CI], 0.95 to 0.98), 0.95 T 0.13 (95% CI, 0.95 to 0.96) and 0.95 T 0.08 (95% CI, 0.95 to 0.96), 0.97 T 0.60 (95% CI, 0.94 to 1.00), respectively. The mean visual acuity was not different between the students with normal and abnormal color vision status (p = 0.9 for the right eye and p = 0.6 for the left eye). Anterior and posterior segment evaluation was done using slit lamp biomicroscopy and direct ophthalmoscopy to rule out any organic causes of color deficiency. Evaluation of color vision was performed using the Ishihara isochromatic color plates (38 plates) under binocular viewing conditions, because inherited defects are bilateral conditions and resultantly an inherited color vision defect is generally not missed owing to binocular testing. A daylight fluorescent tube with a color rendering index of greater than 92 and a color temperature of 6500 K was used for this purpose because this source has illumination characteristics close enough (80 fc or 860 lux) to the Macbeth Illuminant C to make it a reasonable substitute.5,23 Illumination was measured with a lux meter (Hagner Photometer, model EC1, Sweden) at the level of test surface, a distance about 75 cm from the subject. With the best correction in a trial frame and while the head was stable, the color vision plates were held about 75 cm from the subject by another examiner, parallel to the subject’s face and perpendicular to their line of sight. The sequence of pages was demonstration (page 1), transformation (pages 2 to 9), vanishing (pages 10 to 17), hidden digits (pages 18 to 21), and diagnostic plates (pages 22 to 25). Each plate was presented for 3 to 5 seconds. If there were three or less misreadings on plates 2 to 21 of the 38-plate test version, we considered the child to have passed the color vision test because a failing criterion of more than three errors has a sensitivity of 97.7% in the detection of CVD.24 Also, Cole25 has cited in a review study that three or four errors will indicate a probable CVD and more than five errors will show a certain CVD. After data collection, the data were analyzed by W2 tests in the software program Statistical Package for the Social Sciences (SPSS version 17, SPSS Inc, Chicago, IL). In all tests, the significance level was considered to be 0.05.
RESULTS Of the 1000 students under study from grade 1 to grade 5, 502 (50.2%) were girls. The age range was 7 to 11 years. The mean (TSD) age was 9.0 (T1.4) years, with no significant difference in mean age between the two sexes (p = 0.60). The number of misreadings on plates 2 to 21 of the 38-plate test version in the two sexes is displayed in Table 1. Based on a failing criterion, more than two and three misreadings, the sex distribution of color vision status is shown in Table 2.
Optometry and Vision Science, Vol. 91, No. 11, November 2014
Copyright © American Academy of Optometry. Unauthorized reproduction of this article is prohibited.
1374 Color Vision Deficiency in IranVMomeni-Moghaddam et al. TABLE 1.
Frequency distribution of the number of misreadings separately in the two sexes No. misreadings Sex
One
Two
Three
More than three
Boys Girls Total
13 7 20
11 4 15
9 3 12
8 1 9
As can be seen in Table 2, based on a failing criterion of more than three errors, 0.9% (95% CI, 0.4 to 1.7) of students likely had CVD, and the prevalence in girls and boys was 0.2% (95% CI, 0.0 to 1.1) and 1.6% (95% CI, 0.8 to 3.1), respectively. With attention to a failing criterion of more than two errors, the prevalence of CVD in all students and separately in girls and boys was 1.2% (95% CI, 0.7 to 2.0), 0.6% (95% CI, 0.2 to 1.7), and 1.8% (95% CI, 0.9 to 3.4), respectively. The difference between the two sexes was found to be statistically significant (p = 0.02, Fisher exact test). The mean (TSD) age in students with and without CVD was 8.8 (T1.3) and 9.0 (T1.4) years, respectively, which was not found to be significantly different (p = 0.70, independent samples t test).
DISCUSSION Based on our findings, the prevalence of CVD in Zahedan province appears to be two times lower in girls (0.2 vs. 0.4%) and five times lower in boys (1.6 vs. 8.0%), compared with their European counterparts based on a failing criterion of more than three misreadings. There have been very few large-scale studies addressing the prevalence of inherited CVD in Iran and none was conducted in Zahedan to our knowledge. A recent study by Birch22 examined the prevalence of red-green CVD worldwide. In that study, Iran is grouped into the ‘‘European’’ group, one of the five anthropologically defined groups as the author describes. The Iranian people (also known as Iranic people) are a diverse IndoEuropean ethno-linguistic group that comprises the speakers of Iranian languages.26 This grouping is somewhat reasonable given that Modarres et al.21 found prevalences of red-green CVD of 8.18 and 0.43% in male and female subjects, respectively, in the secondary school population of Tehran, Iran. They stated that their study was the ‘‘first study to determine prevalence of congenital color blindness in Iran,’’ but in fact,
several other studies were conducted previously. Zarrabi and Sadighian16 determined the prevalence of CVD in primary school children in Tehran in the 1970s and found prevalences of 2.04 and 0.53% in male and female subjects, respectively. Similarly, Dargahi et al.15 and Farokhfar11 found male prevalence of CVD around 2%. All other previous studies, except for Modarres et al.,21 found similar prevalences to this present study. The discrepancy in prevalences between Zarrabi and Sadighian16 and Modarres et al.,21 at least for Tehran, may be due to the fact that migration into and possibly the diversity of the gene pool in Tehran have increased according to the Statistical Center of Iran.27 Such increases in migration and settlement of people with an increased diversity in backgrounds push the population more toward Hardy-Weinberg equilibrium as Birch suggests.22 A large, randomly mating population in Hardy-Weinberg equilibrium would show stable congenital CVD prevalences of about 8 and 0.4% for male and female subjects, respectively. In Zahedan, the population is somewhat smaller than that in Tehran and relatively more isolated (i.e., less migration). The familial marriage rates are higher than those in Tehran28 and the hypothesis was that a higher prevalence rate of CVD would be found,29 but unexpectedly the prevalence in this study was significantly lower than that reported by Modarres et al.21 and significantly lower in other ethnic groups such as Asians and Africans.22 The reason for this finding is unclear. Although familial marriage rates are relatively high in Zahedan, it is not a small isolated ethnic group. Although we used random sampling in this study and the population we sampled from was rather large, the data appear to indicate that we have identified a particular extreme in a local population. The female prevalence of CVD in this study is not necessarily significantly different from European populations when the CI is considered. However, the male prevalence of CVD in this study is significantly lower compared with European populations and even in the aforementioned studies. This is not easily explainable. With a point estimate of 1.6% for male prevalence, one would expect based on theory alone that the female point estimate of prevalence would be 0.03% (i.e., 1.6% squared), but the estimate found in this study was a tenth of that. It is possible that the sample size was insufficient to reliably estimate female prevalence given that it is already expected to be relatively low. Another aspect to consider is whether the lower prevalence found in this study could be attributed to our methodology. We recruited at random from a large population of schoolchildren
TABLE 2.
The frequency of red-green CVD in all students by sex based on the two failing criteria Sex Boys
Girls
8 (1.6) (0.8Y3.1) 9 (1.8) (0.9Y3.4) 490 (98.4) 489 (98.1) 498 (100.0)
1 (0.2) (0.0Y1.1) 3 (0.6) (0.2Y1.7) 501 (99.8) 499 (99.4) 502 (100.0)
CVD status Abnormal Normal Total
Total
n (%) (95% CI) More More More More
than than than than
3 errors 2 errors 3 errors 2 errors
Optometry and Vision Science, Vol. 91, No. 11, November 2014
Copyright © American Academy of Optometry. Unauthorized reproduction of this article is prohibited.
9 (0.9) (0.4Y1.7) 12 (1.2) (0.7Y2.0) 991 (99.1) 988 (98.8) 1000 (100.0)
Color Vision Deficiency in IranVMomeni-Moghaddam et al.
from multiple sites/schools, avoiding several potential sources of bias.22 These include sampling bias arising from studying a limited number of schools where relatives/siblings could confound the results and selection bias from recruiting only self-referred volunteers. We used a controlled and carefully selected illumination source, unlike the variability of daylight or other light sources used in some other studies.11,21 Use of the 38-plate Ishihara test has been deemed the best screening test for the identification of red-green inherited CVD repeatedly in the literature.22 Best refractive correction, when needed, was in place for testing. It should be emphasized that the Ishihara color plates do not test for blue-yellow defects, and as a result, their prevalence is not known from this work. We also used a rather strict testing protocol and pass/fail criterion, which would actually cause an increase in false positives theoretically (i.e., labeling a ‘‘fail’’ of the test as only three or four plates missed would result in an artificially higher prevalence, that is, a higher sensitivity and a lower specificity; labeling a ‘‘fail’’ of the test as more than five plates, for example, would result in a lower prevalence, that is, a lower sensitivity but a higher specificity). Limiting time per plate and presenting the plates at the prescribed test distance, which are not always done clinically, adhere to the test instructions and could theoretically decrease the specificity of the test. Additionally, the color testing results were confirmed by repeat testing with another examiner, ensuring their repeatability, which could also result in a decrease of the observed prevalence. This additional confirmation of the reliability of the results has not typically been reported in the literature with regard to large-scale population-based color vision studies. Finally, all subjects underwent complete ophthalmic examination, ensuring that acquired color vision defects did not confound our findings; this could also reduce the measured prevalence. Binocular testing was used in this study, as this is the common clinical way to test for congenital CVD, but it is possible that congenital CVD could be missed in some cases with interocular physiological variations between the eyes (e.g., macular pigment or the crystalline lens). This could have contributed to a lower prevalence. The estimated prevalence of inherited CVD in Zahedan, Iran, is less than 1% and significantly lower in boys. The low prevalence is among the lowest found in the literature when considering all populations worldwide and is unexpected.
CONCLUSIONS Color vision deficiency, as detected by the Ishihara test, makes up about 1% of the elementary schoolYaged population in southeastern Iran.
ACKNOWLEDGMENTS This research was supported by the health promotion research center of Zahedan University of Medical Sciences. The authors would like to thank the participants who made this study possible. None of the authors have any proprietary interests or conflicts of interest related to this submission. Received February 9, 2014; accepted August 21, 2014.
1375
REFERENCES 1. Eskridge JB, Amos JF, Bartlett JD. Clinical Procedures in Optometry Philadelphia, PA: J.B. Lippincott; 1991. 2. Swanson WH, Cohen JM. Color vision. Ophthalmol Clin North Am 2003;16:179Y203. 3. Duckman RH, ed. Visual Development, Diagnosis, and Treatment of the Pediatric Patient. Philadelphia, PA: Lippincott Williams & Wilkins; 2006. 4. Deeb SS. The molecular basis of variation in human color vision. Clin Genet 2005;67:369Y77. 5. Grosvenor T. Primary Care Optometry 5th ed. Boston, MA: Butterworth-Heinemann; 2007:285Y6. 6. Schwartz SH. Visual Perception: A Clinical Orientation, 4th ed. New York, NY: McGraw Hill; 2010. 7. Citirik M, Acaroglu G, Batman C, Zilelioglu O. Congenital color blindness in young Turkish men. Ophthalmic Epidemiol 2005; 12:133Y7. 8. Pramanik T, Sherpa MT, Shrestha R. Color vision deficiency among medical students: an unnoticed problem. Nepal Med Coll J 2010; 12:81Y3. 9. Oriowo OM, Alotaibi AZ. Colour vision screening among Saudi Arabian children. S Afr Optom 2008;67:56Y61. 10. Gupta Y, Sukul RR, Gupta M, Phougat A, Jain R, Varshney A. School eye survey in rural population in UP, India. Nepal J Ophthalmol 2011;3:78Y9. 11. Farokhfar A. [Prevalence of color blindness in the primary school students in Sari Township 1999]. J Mazandaran Univ Med Sci 2001;11:57Y62. 12. Niroula DR, Saha CG. The incidence of color blindness among some school children of Pokhara, Western Nepal. Nepal Med Coll J 2010;12:48Y50. 13. Songden SD, Ike EE. Determination of colour vision using Ishihara and Dvorine plates. J Med Tropics 2010;12:14Y7. 14. Mehra KS. Incidence of colour blindness in Indians. Br J Ophthalmol 1963;47:485Y7. 15. Dargahi H, Einollahi N, Dashti N. Color blindness defect and medical laboratory technologists: unnoticed problems and the care for screening. Acta Med Iran 2010;48:172Y7. 16. Zarrabi Z, Sadighian M. Incidence of colour blindness (colour defect) among Iranian primary school children. Acta Med Iran 1974;17:70Y2. 17. Yasmin A, Jahan N, Akhter R. Assessment of colour blindness and erythrocyte G6PD enzyme status among the school children of Dhaka City. J Bangladesh Soc Physiol 2009;4:64Y70. 18. Chia A, Gazzard G, Tong L, Zhang X, Sim EL, Fong A, Mei Saw S. Red-green colour blindness in Singaporean children. Clin Experiment Ophthalmol 2008;36:464Y7. 19. Cabrero FJ, Ortiz MA, Mesa MS, Fuster V, Moral P. Red-green colour blindness in the Tormes-Alberche Valley (Avila-Central Spain). Anthropol Anz 1997;55:295Y301. 20. Qian YS, Chu RY, He JC, Sun XH, Zhou XT, Zhao NQ, Hu DN, Hoffman MR, Dai JH, Qu XM, Pao KE. Incidence of myopia in high school students with and without red-green color vision deficiency. Invest Ophthalmol Vis Sci 2009;50:1598Y605. 21. Modarres M, Mirsamadi M, Peyman GA. Prevalence of congenital color deficiencies in secondary-school students in Tehran. Int Ophthalmol 1996;20:221Y2. 22. Birch J. Worldwide prevalence of red-green color deficiency. J Opt Soc Am (A) 2012;29:313Y20.
Optometry and Vision Science, Vol. 91, No. 11, November 2014
Copyright © American Academy of Optometry. Unauthorized reproduction of this article is prohibited.
1376 Color Vision Deficiency in IranVMomeni-Moghaddam et al. 23. Bittles AH. The role and significance of consanguinity as a demographic variable. Popul Dev Rev 1994;20:561Y84. 24. Birch J. Identification of red-green colour deficiency: sensitivity of the Ishihara and American Optical Company (Hard, Rand and Rittler) pseudo-isochromatic plates to identify slight anomalous trichromatism. Ophthalmic Physiol Opt 2010;30:667Y71. 25. Cole BL. Assessment of inherited colour vision defects in clinical practice. Clin Exp Optom 2007;90:157Y75. 26. Iranian peoples. Wikipedia, The Free Encyclopedia; 2014. Available at: http://en.wikipedia.org/wiki/Iranian_peoples. Accessed August 14, 2014. 27. Statistical Centre of Iran (SCI). Emigration Report; 1990. Available at: http://www.sci.org.ir/SitePages/report_90/Emigration_report.aspx. Accessed February 5, 2014.
28. Saadat M, Ansari-Lari M, Farhud DD. Consanguineous marriage in Iran. Ann Hum Biol 2004;31:263Y9. 29. Shah A, Hussain R, Fareed M, Afzal M. Prevalence of red-green color vision defects among Muslim males and females of Manipur, India. Iran J Public Health 2013;42:16Y24.
Hamed Momeni-Moghaddam Optometry Department Parastar Str., Ahmad Abad Ave Mashhad, Iran e-mail: [email protected]
Optometry and Vision Science, Vol. 91, No. 11, November 2014
Copyright © American Academy of Optometry. Unauthorized reproduction of this article is prohibited.
