Abstract
Purpose.:
To determine the prevalence and identify the types of glaucoma in the Akinyele district of Oyo State in southwestern Nigeria.
Methods.:
Residents of Akinyele district of Oyo State in southwestern Nigeria aged 40 years and older were randomly selected in a stratified manner. All participants underwent comprehensive ophthalmic examination, including visual acuity assessment, anterior segment biomicroscopy, IOP measurement, gonioscopy, optic nerve head and disc evaluation, and central visual field assessment. Glaucoma was diagnosed using the International Society of Geographical and Epidemiological Ophthalmology (ISGEO) classification scheme.
Results.:
A sample of 811 subjects (90% response rate) was examined. The crude prevalence of all forms of glaucoma was 7.3% (95% confidence interval [CI] 5.5%–9.1%) with an age and sex standardized rate of 6.9% (95% CI 6.88%–6.92%). Primary open angle glaucoma was found in 6.2% (95% CI 4.5%–7.8%) and primary angle closure glaucoma in 0.2% (95% CI 0.0%–0.6%). Secondary glaucoma accounted for 0.9% of the cases, with couching and neovascular process being the main causes (0.2% each). Prevalence of glaucoma increased significantly with increasing age (P for trend < 0.05).
Conclusions.:
The high prevalence of glaucoma (7.3%) in the Akinyele district in southwestern Nigeria is comparable with those in predominantly black populations in the Akwapim-South district of Ghana and Barbados. Primary open angle glaucoma remains the most prevalent form of glaucoma.
This study, a cross-sectional population-based epidemiologic survey, had a target population of 43,420 subjects aged 40 years and older, resident in the Akinyele district of Oyo State in southwestern Nigeria.
A sample of 902 subjects was randomly selected in a stratified manner.
The study was approved by the institutional review board of the University of Ibadan/University College Hospital Ibadan. The study was fully explained to all the subjects and they all provided written informed consent. The research adhered to the tenets of the Declaration of Helsinki.
The sample size was calculated assuming a 95% CI (
z = 1.96), a glaucoma prevalence (p) of 4.5% in the target population with the precision level (d) set at 2%, and a design effect of 1.75 to account for cluster sampling, assuming a nonresponse rate of 20%.
13
The sampling frame for the survey was drawn from the 2006 census data for the local government area and used to generate a list of the settlements in the local government area. Four settlements were randomly selected from each of the 12 wards using the random number package on the EPI 2000 statistical software (CDC, Atlanta, GA). Individuals aged 40 years or older who had been resident in the selected study settlement for at least 1 year were considered eligible for inclusion in the survey.
The number of subjects selected in each settlement was based on probability proportional to size (PPS).
A general clinical history, including history of eye trauma, surgery, couching and chronic drug use, loss of vision, eye ache, and family history of glaucoma blindness, was taken and recorded on a standard questionnaire.
All subjects had a comprehensive ophthalmic examination that included visual acuity testing using the standard Snellen chart or the tumbling E chart depending on the level of literacy of the subject. Visual acuity assessment was done at a distance of 20 feet, in the open, and in daylight in a shaded environment. Acuity was recorded as the lowest line of the chart where at least half of the letters in the row were correctly identified. Each eye was tested separately. Intraocular pressure (IOP) measurement was with the Goldman applanation tonometer (Haag Streit, Bern, Switzerland) with the average of two consecutive readings taken as the IOP for each eye.
Anterior segment examination was done with the slit lamp (BM Model; Haag Streit), specifically directed at detecting signs of pigment dispersion, pseudo exfoliation, and other secondary causes of glaucoma.
Gonioscopy was also performed on all subjects using the one mirror Goldmann goniolens (Haag Streit). The grading of the anterior chamber angle was done using the Shaffer grading system.
14
The optic disc was also examined using a 78D Volks lens (Volks Optical, Inc., Mentor, OH) at ×16 magnification after adequate pupillary dilation. The margins of the cup were defined by stereoscopic view as the point of maximum inflection of the vessels crossing the neuroretinal rim. A measuring eyepiece graticle (Haag Streit) was used in measuring the vertical optic diameter and cup diameter. Also noted were the presence of notching on the disc rim and any violation of the ISNT rule. In the normal disc, the neuroretinal rim (NRR) has a characteristic configuration, with the rim width being thickest in the inferior region followed by the superior sector, the nasal sector, and the temporal sector (ISNT rule).
15 The vertical cup-to-disc ratio (VCDR) was used as an index of structural glaucomatous damage. If the optic disc was not visible after pharmacological dilation, “fundus not seen” was recorded.
Administration of the questionnaire and initial ocular examination were all performed in the field. Examination sites used included such places as the town hall, school classrooms, and the compounds of the traditional leaders.
Subjects with suspected glaucomatous disc damage as evidenced by VCDR of 0.7 or greater in either eye, VCDR asymmetry of 0.2 or more not explained by difference in disc diameter, were referred to the base hospital for central visual field (CVF) assessment. To ensure good response, subjects from the different wards were pooled and transported to the base hospital within 2 weeks of the initial examination.
Figure 1 shows the flow chart for the methodology.
Subjects with distant visual acuity of less than 6/9 during the initial examination had streak retinoscopy (Keeler; Halma Co., Windsor, UK) and subjective refraction before perimetry. Refraction data were based on subjective refraction value. The visual field test was done by a trained perimetrist.
White-on-white semiautomated perimetry (Humphrey 740; Carl Zeiss Meditec, Inc., Dublin, CA) was performed using near refraction. The Swedish Interactive Threshold Algorithms (SITA) Fast 24-2 mode was used, and in the case of an abnormal result, followed by a SITA standard. However, if the result was found to be unreliable, a repeat field was done on another day. The reliability indices were set at less than 30%.
A visual field regarded as indicative of glaucoma is one with a glaucoma hemifield result outside the normal limit and a cluster of three contiguous points at the 5% level on the pattern deviation plot.
A random sample of 1 in every 30 subjects with VCDR of 0.5 or less and IOP of 21 mm Hg or lower also had central visual field assessment.
The survey team consisted of 3 ophthalmologists, 2 field assistants, and 12 community health extension workers (CHEW). A 2-week presurvey training was organized, during which the field assistants were trained on questionnaire administration and the CHEWs on visual acuity assessment. Their competence was assessed through field observation during the pilot survey in a nonselected cluster.
Intraocular pressure measurements and gonioscopy were performed by two ophthalmologists (OO and BO), whereas VCDR assessment was performed only by one (OA).
For the purpose of validation, 46 randomly selected subjects were reexamined by a more senior ophthalmologist (AA) with respect to IOP measurement, VCDR assessment, and gonioscopy. The agreement between the readings and grading was assessed using the k statistics. There was a high level of interobserver agreement (weighted k score ≥ 0.85 for all parameters).
The definitional criteria of ISGEO
12 were used in categorizing the subjects, and are briefly described in the following sections.
A criterion 3 diagnosis was given if it was not possible to examine the optic disc, and eyes had visual acuity less than 20/400, presence of relative afferent pupillary defect with IOP of 26 mm Hg or higher, and/or evidence of glaucoma surgery or medical records confirming glaucomatous visual morbidity.
A subject was categorized as having open angle glaucoma (OAG) on fulfilling one of the ISGEO criteria and the angles were adjudged open on gonioscopy. Angle closure glaucoma (ACG) was diagnosed in subjects with anatomically occludable angle evidenced by less than 90° of the posterior (pigmented) trabecular meshwork being visible on gonioscopy without indentation and fulfilling any of the criteria for glaucoma diagnosis under the ISGEO protocol. An OAG or ACG that was thought to have resulted from another ocular or systemic condition was labeled secondary glaucoma.
All data were entered into a database. Data analysis was done using the Statistical Package for Social Sciences version 17 software (SPSS, Inc., Chicago, IL).
Frequency tables were generated to observe the pattern of variable distribution among the subjects. Comparison of categorical variables between groups was by χ2 test, whereas continuous variables were compared using the t-test. Confidence intervals were calculated for rates. Level of significance was set at P less than 0.05.
A total of 811 subjects were screened of a total of 902 who were enumerated for the study. This represents 1.8% of the total population of subjects eligible for the study in the district and an overall participation rate of 90%. Participation was similar across all age groups, ranging from 85.2% to 95.0%. There was a slightly higher female participation (92.3%) than male (87.4%).
There were 389 (48%) male and 422 (52%) female subjects. The mean age of the participants was 58.0 ± 12.76 years. The demographic characteristics of the enumerated and examined subjects are summarized in
Table 1.
Table 1 Demographic Characteristics of the Enumerated and Examined Subjects
Table 1 Demographic Characteristics of the Enumerated and Examined Subjects
| Demographic Variable | Enumerated | Respondents |
| Age, y |
| 40–49 | 304 | 259 (85.2%) |
| 50–59 | 247 | 226 (91.5%) |
| 60–69 | 223 | 206 (92.4%) |
| 70–79 | 101 | 96 (95.0%) |
| >80 | 27 | 24 (89.0%) |
| Mean age ± SD, y | 56.5 ± 10.3 | 58 ± 12.8 |
| Median age, y | 56 | 57 |
| Sex, n (%) |
| Male | 445 (49.0) | 389 (48.0) |
| Female | 457 (51.0) | 422 (52.0) |
The mean IOPs in the right and left eyes were 15.70 ± 5.00 mm Hg (791 eyes) and 16.20 mm Hg ± 5.24 mm Hg (790 eyes), respectively. However, for subjects diagnosed with glaucoma, the mean IOPs were 19.00 ± 8.20 mm Hg and 20.00 ± 7.40 mm Hg for the right and left eyes, respectively. The difference was statistically significant (P < 0.05).
Mean IOPs for men and women was 16.0 ± 5.22 mm Hg and 15.9 ± 5.11 mm Hg, respectively (P > 0.05).
Figure 2 compares the IOP pattern in the right eyes of respondents with and without glaucoma. The pattern in subjects with glaucoma is shifted to the right compared with the study population without glaucoma.
The mean VCDR was 0.43 ± 0.6 for the right eyes (751 eyes) and 0.44 ± 0.17 for the left eyes (748 eyes). However, among subjects categorized as having glaucoma, it was 0.83 ± 0.10 and 0.87 ± 0.10 for the right and left eyes, respectively. It was not possible to assess the optic nerve head in 60 right eyes and 63 left eyes because of media opacity.
Gonioscopic examination was performed on 787 right eyes, of which 779 (99.0%) had open angle and 8 (1.0%) had occludable angle. Of the 783 left eyes that had gonioscopic grading of the angles, 772 (98.6%) had open angle and 11 (1.4%) had occludable angle.
Fifty-one subjects comprising 25 that met criterion 1 of the ISGEO classification and 26 subjects who were randomly selected had CVF test.
Of the 25 subjects that were categorized as having glaucoma using the ISGEO criterion 1, 2 (8.0%) had normal CVF results for both eyes, 7 (28.0%) could not complete a satisfactory visual field test, while 16 (64.0%) had reproducible glaucomatous field changes in at least one eye.
Among the randomly selected group, 1 subject had reproducible glaucomatous field changes in one eye, 2 subjects could not complete satisfactory field testing, and the remaining 23 subjects had normal field results.
At the final evaluation, 59 subjects were found to have glaucoma: 2 (3.4%) with primary angle closure glaucoma (PACG), 7 (11.9%) with secondary glaucoma, and 50 (84.7%) with POAG. The identifiable causes in subjects with secondary glaucoma were lens related in four cases, of which two were cases of couching, one was complicated cataract with background trauma, and one was postcataract surgery; whereas the others included two cases of neovascular process and one subject with pseudoexfoliation (PEX).
The crude prevalence of glaucoma among the study population was 7.3% (95% CI 7.28%–7.32%). Six (0.7%) respondents had been previously diagnosed with glaucoma. Mean duration since diagnosis was 5.50 ± 4.13 years.
The prevalence of glaucoma in the study population using the three definitional criteria of ISGEO is given in
Table 2.
Table 2 Prevalence of Glaucoma by Diagnostic Criteria
Table 2 Prevalence of Glaucoma by Diagnostic Criteria
| ISGEO Criteria | No. of Respondents | Prevalence |
| Criterion 1 | 16 | 2.0% (1.0–2.9) |
| Criterion 2 | 39 | 4.8% (3.3–6.3) |
| Criterion 3 | 4 | 0.5% (0.0–1.0) |
| Total | 59 | 7.3% (5.5–9.1) |
Using the 2006 Nigeria national census figure,
16 the age and sex standardized prevalence of glaucoma among the population aged 40 years and older was 6.9% (95% CI 6.88%–6.92%).
Table 3 shows the prevalence of the various forms of glaucoma.
Table 3 Prevalence of the Various Forms of Glaucoma
Table 3 Prevalence of the Various Forms of Glaucoma
| Glaucoma Type | No. of Subjects | Crude Prevalence |
| % | (95% CI) |
| All glaucoma | 59 | 7.3 | (5.5–9.1) |
| Primary glaucoma | 52 | 6.4 | (4.7–8.1) |
| POAG | 50 | 6.2 | (4.5–7.8) |
| PACG | 2 | 0.2 | (0.0–0.6) |
| Secondary | 7 | 0.9 | (0.2–1.5) |
The prevalence of glaucoma was 4.6% for those aged 40 to 49 years and 16.7% for those aged 80 years and older.
Table 4 shows the age and sex prevalence of the various forms of glaucoma among the 811 subjects.
Table 4 Age- and Sex-Specific Prevalence of the Various Forms of Glaucoma Among the Study Population
Table 4 Age- and Sex-Specific Prevalence of the Various Forms of Glaucoma Among the Study Population
| | N | All Glaucoma | POAG | PACG | Secondary Glaucoma |
| n | % (95% CI) | n | % (95% CI) | n | % (95% CI) | n | % (95% CI) |
| Male | | | | | | | | | |
| 40–49 | 108 | 7 | 6.48 (1.84–11.12) | 7 | 6.48 (1.84–11.12) | — | 0 (—) | — | 0 (—) |
| 50–59 | 99 | 7 | 7.07 (2.0–12.1) | 7 | 7.07 (2.0–12.1) | — | 0 (—) | — | 0 (—) |
| 60–69 | 105 | 8 | 7.6 (2.5–12.7) | 8 | 7.6 (2.5–12.7) | — | 0 (—) | — | 0 (—) |
| 70–79 | 59 | 10 | 16.9 (7.4–26.5) | 6 | 10.2 (2.5–17.9) | — | 0 (—) | 4 | 6.8 (0.36–13.2) |
| ≥80 | 18 | 3 | 16.7 (0.6–34.0) | 3 | 16.7 (0.6–34.0) | — | 0 (—) | — | 0 (—) |
| All | 389 | 35 | 9.0 (6.1–11.8) | 31 | 8.0 (5.3–10.7) | — | 0 (—) | 4 | 1.0 (0.03–2.0) |
| Female | | | | | | | | | |
| 40–49 | 151 | 5 | 3.3 (0.5–6.2) | 5 | 3.3 (0.5–6.2) | — | 0 (—) | — | 0 (—) |
| 50–59 | 127 | 6 | 4.7 (1.0–8.4) | 5 | 3.9 (0.6–7.3) | 1 | 0.8 (0.7–2.3) | — | 0 (—) |
| 60–69 | 101 | 8 | 7.9 (2.7–13.2) | 5 | 5.0 (0.7–9.2) | 1 | 1.0 (0.09–2.9) | 2 | 2.0 (0.07–4.7) |
| 70–79 | 37 | 4 | 10.8 (0.8–20.8) | 3 | 8.1 (0.7–16.9) | — | 0 (—) | 1 | 2.7 (0.03–7.9) |
| ≥80 | 6 | 1 | 16.7 (−13.2 to 46.5) | 1 | 16.7 (−13.2 to 46.5) | — | 0 (—) | — | 0 (—) |
| All | 422 | 24 | 5.7 (3.5–7.9) | 19 | 4.5 (2.5–6.5) | 2 | 0.5 (0.0–1.1) | 3 | 0.7 (0.0–1.5) |
| Male and female | | | | | | | | | |
| 40–49 | 259 | 12 | 4.6 (2.1–7.1) | 12 | 4.6 (2.1–7.1) | — | 0 (—) | — | 0 (—) |
| 50–59 | 226 | 13 | 5.8 (2.7–8.8) | 12 | 5.3 (2.4–8.2) | 1 | 0.4 (−0.4 to 1.3) | — | 0 (—) |
| 60–69 | 206 | 16 | 7.8 (4.1–11.4) | 13 | 6.3 (3.0–9.6) | 1 | 0.5 (−0.46 to 1.4) | 2 | 1.0 (0–2.3) |
| 70–79 | 96 | 14 | 14.6 (7.5–21.6) | 9 | 9.4 (3.5–15.2) | — | 0 (—) | 5 | 5.2 (0.7–9.6) |
| ≥80 | 24 | 4 | 16.7 (1.8–31.6) | 4 | 16.7 (1.8–31.6) | — | 0 (—) | — | 0 (—) |
| All | 811 | 59 | 7.3 (5.5–9.1) | 50 | 6.2 (4.5–7.8) | 2 | 0.2 (0–0.5) | 7 | 0.9 (0.2–1.5) |
| Age-adjusted rate | | | 6.9 (6.88–6.92) | | 6.1 (6.08–6.12) | | | | |
Six (10.2%) of the glaucoma patients had been previously diagnosed and three were still on glaucoma medication at the time of the survey.
Among the respondents diagnosed with POAG, three (5.8%) were bilaterally blind, representing 0.37% of the examined population, whereas five (9.6%) were blind in one eye, representing 0.6% of the examined population.
To the best of the knowledge of these authors, this is the first attempt to estimate the prevalence of glaucoma using the ISGEO criteria among a West African population.
The multistage stratified sampling technique coupled with the high response rate ensured that the studied population was a fair representation of the total population of Akinyele district.
Active community participation was encouraged during the survey by engaging the community leaders and the resident traditional birth attendants during the enumeration. Also, free transportation was provided for participants who needed to be moved to the base hospital. Furthermore, the survey was conducted between the months of October and January, a period when there was not much farming work, as 46.6% of the enumerated subjects were peasant farmers. These factors could have accounted for the high participation rate.
Using the ISGEO criteria for glaucoma diagnosis, the observed crude prevalence of all types of glaucoma in this study population was 7.3% (age-adjusted prevalence of 6.9%). The observed prevalence of POAG was 6.2%. The result of this cross-sectional survey further substantiated the evidence that the prevalence of glaucoma is much higher in people of African descent (black race) than in other racial groups, and that the predominant form of glaucoma is the primary open angle type.
2–4,17 The study population included all Nigerians from the Yoruba ethnic group.
More men aged 70 years and older were enumerated and participated in the survey compared with women in the same age group (male to female ratio 1.9:1.0). This contradicts the findings in most population-based surveys where there tend to be more female participants across all age groups. This incidental finding could be due to the sociocultural practice whereby elderly women relocate to cities to assist in the care of their grandchildren. This practice is very common Nigerians from the southwest. Because glaucoma prevalence tends to increase with increasing age, this could have resulted in a little underestimation of the true burden of glaucoma among this semirural population.
An advantage of the ISGEO criteria
12 is that it takes into consideration eyes in which visual field testing or disc evaluation is impossible, which was not captured in most surveys on glaucoma prevalence before the adoption of the ISGEO classification. Consequently, direct comparison of the prevalence rates of glaucoma among the present study population and the rates reported in some of these other studies among people of African descent must be done with caution. For such comparison to be meaningful, there is the need to take into consideration differences in the methodology and diagnostic criteria used.
Ntim-Amponsah et al.
8 reported an 8.4% prevalence rate for POAG in the Akwapim–South district of Ghana, a country located along the coastal area of the West African subregion just like southwestern Nigeria. The survey included those aged 30 to 39 years, but adjusting for those aged 40 years and older would give a prevalence rate much higher than was found in the Akinyele district (9.2%). The results from these two studies appear to be the closest among the various population-based surveys on glaucoma, considering that the participants were both from black African communities located in the same tropical region and as such may have some genetic linkage. However, the use of a nonsystematic sampling method (volunteer sample), which would most likely result in unhealthy volunteer effect and inclusion of first-degree relations of glaucoma patients might have resulted in the much higher prevalence rate.
The prevalence rate of 3.1% reported for POAG in Kongwa district of Tanzania, East Africa,
5 is much lower than was found in the Akinyele district even though both studies were conducted among Africans of the same age group (40 years and older). In the Kongwa study, glaucoma diagnosis was based strictly on disc appearance and visual field assessment. Therefore, subjects who would have fallen under criterion 3 of the ISGEO classification were excluded. Also 98.3% of the study population had visual field assessment compared with only 6.3% (comprising 3.1% who met the ISGEO Criteria 1 and 3.2% who were randomly selected) in the present survey. Also in the Kongwa study,
5 a VCDR of 0.5 and above was used as the cutoff. The present study would obviously have captured more subjects if visual field testing had been done for all subjects, and the VCDR cutoff was lowered to 0.5, especially as one of the randomly selected subjects had reproducible field compatible with glaucoma diagnosis. Obviously this might have further widened the gap in the difference in rates between the two studies. Closest in comparison is the criterion 1 diagnosis of the present study and the definition 2 criteria of the Kongwa study,
5 which used a VCDR of 0.7 or more and a definite and reliable CVF in defining the glaucoma subject. This gave a prevalence of 1.7% (95% CI 1.3%–2.2%) compared with the 2.0% (95% CI 1.0%–2.9%) in the present survey (
Table 3).
The prevalence of 7.0% reported among the black population in the Barbados Eye study (BES)
4 is very close to what was found in Akinyele district. However, the BES used very strict criteria based on CVF and disc appearance, similar to that in the Kongwa study,
5 and as such captured mainly those corresponding to criterion 1 diagnostic criteria of the ISGEO protocol.
12
It is worthy to note that the exact effect that differences in methodology could have on calculated prevalence might be difficult to predict with good precision unless a meta-analysis of previous and future glaucoma surveys in which different methods are used is done.
Despite the differences in prevalence rates reported between studies, the results from this survey has added to the weight of evidence that POAG is several times more prevalent among black Africans compared with the Caucasian population.
It has been postulated that the high prevalence of POAG seen in Barbados
4 and St. Lucia
18 can be linked to their African ancestral origin. This was explained by the genetic drift theory, whereby migrants carry more of certain risk alleles for POAG than their ancestral African population.
18 The people of central Tanzania studied in the Kongwa survey were not included in the migration to the New World to the same extent as their West African counterparts.
5
The potential weakness in the use of VCDR as a defining factor for glaucoma is that the VCDR depends on disc size, which varies among individuals.
2,12 Larger optic disc area and therefore cup has been found in African American than in European American individuals.
2,12,19 However, reports have shown that the 97.5th and 99.5th percentile figures for VCDR distribution among Asian and the Dutch individuals are similar.
20 There is limited information on the distribution of VCDR among people of African descent.
5 In this study, the use of +78-diopter lens with slit lamp biomicroscope enhanced the accuracy of the evaluation of the structure of the optic nerve head. The same examiner performed the tests, thus intraobserver variability was reduced.
The mean VCDR in this population for the right and left eye was similar to that reported in the Kongwa study
5 and in the Rotterdam study.
21 The lower value for VCDR reported by Ekwerekwu and Umeh
9 in the southeastern part of Nigeria could have been due to the use of direct ophthalmoscope in assessing the optic nerve head.
The mean IOP in the right eye and left eye in this study population was 15.7 mm Hg and 16.2 mm Hg, respectively. This is similar to values reported in previous studies,
5,8,22 but lower than the mean IOP reported in the Barbados study.
4
The difference in IOP value between studies might be due to such factors as the type of instrument used for measurement and the time of measurement. Most of the respondents in the present study were examined between 9 AM and 3 PM; it is therefore possible that the peak diurnal variation was missed in quite a number of them. Central corneal thickness (CCT) was also not measured in this study and, as such, the effect of CCT on the distribution of IOP could not be evaluated. In this study population, the mean IOP value was, as expected, higher in the glaucomatous subjects than in the general population.
The inclusion of raised IOP in the presence of visual acuity less than 20/400 as diagnostic criteria under the ISGEO protocol becomes even more important in developing countries with high prevalence of unoperated cataract and cornea scarring from trachoma and vitamin A deficiency. Subjects who fell into this group were excluded in previous studies
4,5 and this in a way could have resulted in underestimation of the true burden of glaucoma as previously observed.
All primary ACG cases diagnosed in this study were in women, similar to the finding among Brazilian individuals.
23 Also among a Chinese population, Foster and Johnson
24 observed a higher prevalence of ACG in women. In the present study, this observation could have been a chance finding, as the authors have observed male subjects with ACG in daily practice and considering the limited sample size.
The low prevalence of secondary glaucoma recorded among respondents in Akinyele district was similar to what was reported among the Tanzanian population (5/135). One case of PEX with bilateral glaucoma was found. This may be an indication that PEX may not be as rare as was assumed among Nigerian individuals but rather that its detection may require a higher index of suspicion. The study among Zulu individuals reported a prevalence of 2.1% for secondary glaucoma, with PEX glaucoma ranking highest among the causes of the secondary glaucoma.
10
Pseudo aphakic glaucoma resulting from couching was responsible for most of the cases of secondary glaucoma in this study. Gilbert et al.
25 reported couching as accounting for 24.3% of procedures for cataract in southwestern Nigeria. Such practice is more rampant in the rural communities, further strengthening the need for a structured eye care program in the Akinyele district.
The number of subjects with previously diagnosed glaucoma was very low and compares with what has been reported in some previous studies.
8,26,27 However, in studies among Caucasian individuals,
21,28 rates between 49% and 75% were reported. This could be a reflection of the better standard of the health care delivery system, especially in the areas of access and awareness creation, as glaucoma remains asymptomatic in its early stage among all populations. Because only a few (3) of the previously diagnosed subjects were still on glaucoma medication, the effect of this on the overall mean IOP of the subjects with glaucoma is not likely to be significant even though this effect was not evaluated for in the survey.
Even though a much larger sample size would have been preferred for this survey, this was not possible because of limited resources, a common problem militating against population-based studies in the developing world. Despite this limitation, this survey has further substantiated evidence on the much higher prevalence of POAG among people of African descent and even more so among those from the West African subregion, considering that most of the black population in Barbados are of West African ancestral origin.
This study reports the prevalence of glaucoma in an indigenous African population domiciled in West Africa. The prevalence of POAG was higher than that reported for Hispanic and European populations and for Africans in East Africa. The results obtained are closer to reported values among other peoples of West African origin. Primary open angle glaucoma constituted 84.7% of all the diagnosed glaucoma.
There is still a need for a larger population study in this subregion using the ISGEO criteria to provide even more robust data on glaucoma prevalence and allow for better data comparison.