If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
Keratoconus is defined as central and/or paracentral corneal thinning and steepening with consequent irregular astigmatism. To date, no prospective study has reported its prevalence in Central Europe. This study aimed to evaluate the prevalence of keratoconus assessed with Scheimpflug imaging in a profile of general medicine practice in Poland.
Methods
This cross-sectional study collected data by inviting all patients aged 10–80 years subscribed to a single general practitioner in the Elbląg area, Poland. Corneal measurements were performed using a rotating Scheimpflug corneal tomography system. Three masked examiners established the diagnosis of keratoconus in the right or left eye. Fleiss’ kappa coefficient (κ) was used to qualitatively analyze the interexaminer agreement.
Results
Of 1,705 patients invited to the study, 771 agreed to participate (45.2%). The mean age of the 728 patients (60.6% female and 39.4% male) with gradable images in at least one eye was 53.6 ± 16.5 years. All of the examiners classified the same four cases as keratoconus (0.549%, 549 per 100,000, 95% confidence interval 150 to 1,400 per 100,000), of which two were bilateral. Fleiss’ kappa showed almost perfect agreement among the three examiners regarding the diagnostic decision (κ = 0.784, 95% confidence interval 0.750 to 0.817, p < 0.001). The keratoconus cases were all women (p = 0.106), aged 48, 69, 70, and 71 years old.
Conclusion
The prevalence of keratoconus in this area was higher than that previously demonstrated in many European countries. Underestimation of the prevalence has been postulated but never demonstrated.
The overall prevalence of keratoconus varies greatly depending on geographical, genetic, environmental, and cultural factors, as well as the diagnostic criteria and procedures used. In Russia, for example, the prevalence of keratoconus was reported to be 0.3 per 100,000, while in Lebanon, it was 3,300 per 100,000 [
]. Particularly high prevalence has been reported in studies from the Middle East, while in European studies the prevalence is evidently lower. For example, a Dutch population study reported a prevalence of 265 cases per 100,000 [
]. A significant shortfall of several currently published studies is that they did not evaluate the prevalence in a general population (but for example in army recruits [
]), and moreover, they did not evaluate the ethnicity of the patients.
A meta-analysis by Hashemi et al. which included 7,158,241 participants found the overall prevalence of 138 per 100,000 population (95% confidence interval [CI]: 114–162 per 100,000) [
]. In contrast to corneal topography, corneal tomography is able to determine the elevation of the front and back corneal surfaces and reconstructs the pachymetric mapping, which significantly improves the sensitivity and specificity for detecting corneal ectasia [
]. In a recent study by Torres Netto et al., the prevalence of keratoconus assessed with Scheimpflug tomography in Riyadh, Saudi Arabia, was estimated as 4.79% (95% CI: 2.96 to 6.62) [
To date, no prospective study has reported on the prevalence of keratoconus in central Europe. This study aimed to evaluate the prevalence of keratoconus in a profile of patients in a general medical practice. The study group consisted of people of Polish ethnic origin because it is well known that keratoconus varies greatly depending on geographical, genetic, environmental factors. To the author's knowledge, it was the first study to survey the rate of keratoconus in North Europe using tomography.
2. Methods
2.1 Study design
Cross–sectional study involving patients subscribed to a general practitioner (GP).
2.2 Patient acquisition
Healthcare in Poland is guaranteed by the constitution, and all medical activities are overseen by the Polish Ministry of Health. The National Health Fund (Polish: Narodowy Fundusz Zdrowia) manages the public healthcare, and requires every patient to choose a GP who will manage the healthcare and refer to specialists when needed. Most primary care practices in Poland are solo practices or small group practices with 2–4 full-time GPs, who share office space but not patients; one GP usually serves up to 2,500–2,800 patients. Although distance does not seem to be an essential decision-making factor in the choice of doctor, patients are likely to consult a GP in their residential area or easily accessible by public transport.
The participants for this study were recruited by contacting patients subscribed to a single GP in the Elbląg branch of the Hygeia Clinic, Poland, aged 10–80 years and of Polish ethnicity. Every eligible patient subscribed to the practice received a phone call and was invited to make an appointment for the examination, which was offered free of charge. Subscribed patients who visited the practice for a GP consultation were also invited to take part in the examination. The study adhered to the tenets of the Declaration of Helsinki. Written consent was obtained for all patients and/or their guardians, as required by Polish law. The subjects were asked whether they had undergone laser vision correction prior to the examination. The examination protocol was approved by the local bioethical committee (Komisja Bioetyczna przy Izbie Lekarskiej w Gdańsku, approval no. KB-13/20).
2.3 Examination
The study was performed from August to October 2020, during the SARS-CoV-2 pandemic; however, in this summer period the number of cases in Polabnd was low. The patients and examiners wore face masks and direct interpersonal contact was as short as possible. On presentation, all participants signed informed consent forms after receiving an explanation of the study purpose. All patients were required to answer whether they had undergone prior corneal refractive surgery. Images were obtained with the Scansys Anterior Segment Analyzer TA 517 (MediWorks Precision Instruments, Shanghai, China); the device has a Conformité Européenne (CE) and Food and Drug Administration (FDA) approval. The Scansys device applies an infrared camera and Scheimpflug camera to obtain high-resolution corneal tomography images. For each scan the device collects 107,520 data points and acquires 28 corneal tomography images during a one-second scan. After an automatic analysis, Scansys provides a series of topographic maps of corneal curvature, thickness and anterior/posterior surface elevation.
The images were taken in a dim room in the opening hours of the practice (between 9 AM and 6 PM). Three images were obtained by experienced examiners (K.P., and J.Ś.) for each eye in random order, but only the one best-quality scan was analyzed. A scan was considered as having sufficient quality if the captured area was adequate, no eye movement occurred during capturing of the scan, and the device could create stereometric models of the cornea; the image quality was assessed automatically by the software (“OK” sign indicated proper acquisition quality). No additional measurements were taken if the quality of the obtained images was insufficient. Three experienced independent examiners (KP, PK, and IT) blindly classified the topographic images of the cases.
2.4 Keratoconus diagnosis
In the recent software update (version 1.0.041.039), Scansys introduced a parameter named keratoconus probability (KCP). It employs an artificial intelligence algorithm to estimate the risk of keratoconus. Because we could not obtain information on how the device calculates KCP, and this parameter has not been validated in larger studies, KCP was recorded for the images, but we decided to manually evaluate the risk of keratoconus. The diagnosis of keratoconus was made based on the RETICS (Red Temática de Investigación Cooperativa en Salud) classification [
Comparison of three-dimensional optical coherence tomography and combining a rotating Scheimpflug camera with a Placido topography system for forme fruste keratoconus diagnosis.
Toprak I, Vega A, Alió Del Barrio JL, Espla E, Cavas F, Alió JL. Diagnostic Value of Corneal Epithelial and Stromal Thickness Distribution Profiles in Forme Fruste Keratoconus and Subclinical Keratoconus. Cornea. Published online August 4, 2020. doi:10.1097/ICO.0000000000002435.
A manifest keratoconus diagnosis was based on the presence of tomographical map patterns for keratoconus (round, oval, superior steep, inferior steep, irregular, inferior-steep asymmetric bow tie, superior-steep asymmetric bow tie, and symmetric or asymmetric bow tie with skewed radial axis [SRAX] > 21 degrees), accompanied by focal steepening (front elevation > 7 µm and/or back elevation > 17 µm) and corresponding corneal thinning, 3-mm inferior–superior (I-S) keratometric asymmetry > 1.4 D or central Kmean > 47.2 D. Keratoconus suspect was defined based on subtle tomographical abnormalities (inferior or superior steep asymmetric bow tie and symmetric or asymmetric bow tie with SRAX < 21 degrees), front elevation at the thinnest point ≤ 7 μm and/or back elevation at the thinnest point ≤ 17 μm, thinnest pachymetry (TP) > 470 μm and/or 3 mm I-S keratometric asymmetry < 1.4 D. Normality was defined as none of these findings.
] the prevalence of keratoconus in Norway was 192.1/100,000. Based on the results of the aforementioned investigation and the age range of the subjects, the minimum required sample size was deemed as 548 individuals, assuming the margin of error at 5% (common choice in biomedical sciences) and with the confidence level of 99%. This high confidence level was chosen to ensure high precision of the estimates, given that the study was not a multicentre study, and the selection of its participants was not strictly random. The sample size calculation has taken into account the potentially greater senstivity of corneal tomography compared to older imaging techniques, and the number of patients was overestimated by at least 25% as a certain number of drop-outs, screening failures, and defective quality corneal scans were expected.
Descriptive statistics were used for basic demographic and prevalence data. The main outcome of the study was the prevalence of keratoconus within the population; only the eyes determined as keratoconus, not suspects, were used for calculating prevalence data. The overall prevalence of keratoconus manifest was reported as a percentage and cases per 100,000 with a 95% CI. The chi-squared test was used to compare gender between the keratoconic and non-keratoconic participants, whereas age was compared between these two groups using the Mann-Whitney U test because parametric test assumptions were not met. Fleiss’ kappa coefficient (κ) was used for quantitative analysis of the inter-rater agreement among the three examiners [
Fleiss JL, Levin B, Paik MC. Statistical Methods for Rates and Proportions. Wiley Series in Probability and Statistics. Published online 2003. doi:10.1002/0471445428.
]. A p-value less than 0.05 at 95% CI was accepted as statistically significant.
3. Results
3.1 Study population
The patient acquisition flow chart is presented in Fig. 1. The mean age of the patients subscribed to the practice is 47.41 ± 17.98 years. Following the phone invitation, 771 patients (475 women and 296 men) living in Elbląg city and the surrounding area presented for the examination. Eight have stated to have undergone laser vision correction prior to the examination, and thus were excluded (Fig. 1). Both eyes were screened during the examination; despite performing three examinations it was not possible to obtain gradable images in (i) 80 eyes due to inadequate image area (ii) 26 eyes due to eye movement (iii) 15 eyes due to a non-existent eye or device error. Thus, 1,405 out of 1,526 images were deemed gradable. In 728 patients it was possible to obtain gradable images in at least one eye. The mean age of participants undergoing the final analysis was 53.6 ± 16.5 years; 441 (60.6%) were women and 287 (39.4%) were men, all participants were white.
The simulated flat keratometry (K1) value was 42.92 ± 1.54 D, and the simulated steep keratometry (K2) was 43.85 ± 1.62 D; the mean keratometry value was 43.33 ± 2.17 D. The mean superior keratometric power at 6 mm was 43.68 ± 1.97 D, while the mean inferior keratometric power at 6 mm was 43.45 ± 2.08 D; this resulted in an inferior–superior asymmetry of 0.23 ± 0.82 D. The average central corneal thickness was 559.12 ± 38.25 µm. The mean keratoconus probability index was 1.53 ± 4.99%.
3.3 Inter-rater agreement, prevalence and incidence of keratoconus
All of the examiners classified the same four cases as keratoconus (0.549%, 549 per 100,000, 95% CI: 150 to 1,400 per 100,000). Fleiss’ kappa showed substantial agreement among the three examiners regarding the diagnostic decision (κ = 0.784, 95% CI: 0.750 to 0.817, p < 0.001). The keratoconus cases were all women (p = 0.106) and aged 48, 69, 70, and 71 years old; two cases were bilateral. Three keratoconic patients were classified as RETICS Grade II, and one as RETICS Grade I (Fig. 2); none of the patients were aware that they have keratoconus. Table 1 depicts the diagnostic distribution based on the opinions of the three examiners. The prevalence of keratoconus suspects was established as 0.27–0.69% (Fig. 3).
Fig. 2Superior keratoconus in a 71-year-old female patient.
According to the Central Statistical Office of Poland, the life expectancy in Poland in 2018 was approximately 77.6 years. Keratoconus is rarely observed in the first 14 years, so the time span of contribution is 63.6 years [
]. Thus, the relative incidence was calculated as 8.63 (95% CI 2.36 to 29.48) per 100,000 residents per year.
4. Discussion
To the best of our knowledge, this is the first cross-sectional study evaluating the prevalence of keratoconus in Europe, and certainly in Central Europe. Compared to other European countries, Poland is very homogenous in terms of nationality and ethnicity. Based on data from the Central Statistical Office of Poland, 97.5% of persons living in the Elbląg region recognize themselves as having Polish nationality and ethnicity [
GUS. Główny Urząd Statystyczny / Spisy Powszechne / Narodowe Spisy Powszechne / Narodowy Spis Powszechny 2002 - wyniki / Wyniki Narodowego Spisu Powszechnego Ludności i Mieszkań 2002 w zakresie deklarowanej narodowości oraz języka używanego w domu. Accessed September 2, 2021. https://stat.gov.pl/spisy-powszechne/narodowe-spisy-powszechne/narodowy-spis-powszechny-2002/wyniki-narodowego-spisu-powszechnego-2002-narodowosci-oraz-jezyka/.
]. Our study indicated that the prevalence of keratoconus in this area was higher than previously demonstrated in many European countries (549 cases per 100,000; Table 2). One of the highest rates of 1,190 per 100,000 was reported by Santiago et al. in France [
] Nevertheless, their study could not be considered representative for the French population; it analyzed the prevalence among 18–25 year old French Armed Forces staff, but not the general population. The investigation by Ljubic et al. estimated the population prevalence of keratoconus based on a database analysis of patients attending to a contact-lens center [
]. Moreover, the diagnosis of keratoconus was made based on keratometric values and base curve of prescribed rigid contact lenses, but not clinical symptoms or corneal imaging techniques. Some of the studies have also calculated the prevalence based on the number of symptomatic in-patients with keratoconus and the population of the area that the clinic serves [
]. Such an approach might be strongly estimative and would presumably omit patients with early-stages of keratoconus. Finally some other investigations analyzed the national disease coding systems to detect keratoconus [
Kristianslund O, Hagem AM, Thorsrud A, Drolsum L. Prevalence and incidence of keratoconus in Norway: a nationwide register study. Acta Ophthalmol. Published online November 16, 2020. doi:10.1111/aos.14668.
]. Medical coding systems are known to reduce redundancy of healthcare and improve patient health outcomes; on the other hand, unintentional misspecification, unbundling, and upcoding are not uncommon and lead to coding inaccuracy [
]. These database studies did not provide criteria on how keratoconus was diagnosed, but could similarly omit patients with early-stages of keratoconus.
Table 2Prevalence and incidence of keratoconus in currently published European studies.
Kristianslund O, Hagem AM, Thorsrud A, Drolsum L. Prevalence and incidence of keratoconus in Norway: a nationwide register study. Acta Ophthalmol. Published online November 16, 2020. doi:10.1111/aos.14668.
Most of the European studies reporting the prevalence of keratoconus did not try or were not able to evaluate the ethnicity or demographical details of the examined patients. Only the study conducted in the United Kingdom by Pearson et al. provided such information [
]. We aimed to evaluate the prevalence solely in Polish individuals, who are considered as West Slavs; other nationalities or ethnicities were excluded. Because the borders of Poland have changed in the last 70 years, patients aged over 80 years were also excluded. Keratoconus is considered as a disease of young adults in their 20 s, with the clinical onset is at puberty and develops until the 3rd to 4th decade of life [
]. In these terms, the age of the patients subscribed to the GP practice and invited to participate in the study was high (47.41 ± 17.98 years). Aging is a problem of several European populations; in Poland, low fertility rate is a result of changes in lifestyle and procreation behavior: women are having their first child later and have fewer children [
]. The attractiveness of the biggest cities with good job opportunities contribute to high migration from this area and could explain why the age of the population invited for the study is slightly higher than the average age of the Polish population.
Another issue is the age of individuals who agreed to participate in the study, which was slightly higher than that of the entire GP database population. This could be explained by the fact that some patients involved in their full-time work could not have time to take part in this study, while older persons may have more free time and inclination to volunteer for such a study. Theoretically, because of the chronicity of keratoconus, similar prevalence rates could be expected in older and younger age groups. Nevertheless, most of the published studies have noted a low percentage of keratoconus in patients over 50 years of age [
]. The marked absence or loss of older KC patients has been observed by specialist corneal surgeons, but is not fully explained as has been observed by specialist corneal surgeons [
]. There are several potential hypotheses that could explain the increased mortality observed in keratoconus, including the higher prevalence of comorbidities such as mitral valve prolapse, asthma, obesity, and obstructive sleep apnea in this population. Another possible factor is selection bias in the published studies [
]. The prevalence rate demonstrated shown in this study is high considering the rates in previously published studies and particularly the relatively high age of the examined population
Gender differences in keratoconus prevalence have been reported previously. The low gender equality index for Poland could help explain the higher percentage of women presenting for the examination [
European Institute for Gender Equality. Gender Equality Index 2020: Poland. Accessed March 7, 2021. https://eige.europa.eu/sites/default/files/documents/20203357_mhag20010enn_002.pdf#:∼:text=Progress%20on%20gender%20equality%20in%20Poland%20since%202010,higher%20increase%20%28%2B%200.6%20points%29%20since%202017.%20Poland%E2%80%99s.
Kristianslund O, Hagem AM, Thorsrud A, Drolsum L. Prevalence and incidence of keratoconus in Norway: a nationwide register study. Acta Ophthalmol. Published online November 16, 2020. doi:10.1111/aos.14668.
]; in our study keratoconus was found only in female patients (p=0.106). It has also been shown that keratoconus patients are often unaware of their condition; in this study, none of the patients were aware that they have keratoconus. In a study by Hashemi et al. 42% of keratoconus patients in Mashhad, Iran, were unaware of their condition and it was bilateral in each case [
]. In a study by Millodot et al. nine out of the 23 subjects diagnosed with keratoconus in Jerusalem, Israel did not know about their condition (40%) [
]. Although there is evidence supporting a genetic contribution to the disease, a lack of awareness among family members of patients with keratoconus has also been shown [
The main limitation of this study is its restricted small sample size. This was partially associated with a relatively low participation rate (771/1,705; 45.2%). The participation rate for epidemiologic studies has been declining during the past 30–40 years [
]. Women were shown to more likely participate in a clinical study based on the influence of friends, family, researchers as well as general altruistic considerations (odds ratio 3.14, 3.45, 3.46, 8.45 respectively) [
]; this could explain the higher participation rate of women in our study. The SARS-CoV-2 pandemic in Poland could further influence this because taking part in this study, as it could not be considered as an unnecessary health consultation. Perceiving a risk to self is the principal reason why people make a decision to participate in epidemiological studies [
]. Feeling of fear, caused by the actual or perceived threat of danger or harm would explain why participation rates on cancer are expected to have higher participations rates than ophthalmic studies [
]. Because extreme efforts to increase participation rates may introduce more bias into the study, we did not generate excessive interest in corneal tomography for this study [
]. Finally, it is much harder to perform corneal topography or tomography on patients with keratoconus than on those without the disease. It is possible that some of the subjects who had “inadequate image area” may actually have had keratoconus. Keratoconus patients may also have a higher grade of eyelid laxity compared to controls, and the relationship between floppy eyelid syndrome and keratoconus has been documented [
The prevalence of keratoconus in this area was higher than that previously demonstrated in many European countries. The limitations of the previously published European studies on keratoconus are mainly associated with the study design: they might not be representative for the whole population, and usually did not include information on the ethnicity of the patients. Despite a limited group size, this appears to be the first cross-sectional study performed in Europe. Underestimation of the true prevalence of keratoconus has been postulated but never demonstrated.
Financial disclosure
No author has a financial or proprietary interest in any material or method mentioned.
Declaration of Competing Interest
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
Acknowledgements
The authors would like to thank OPTOtech Medical, Niepołomice, Poland for providing the device for the study, Dr Jacek Świeczka for obtaining excellent corneal tomography images, Prof. Jos Rozema from the Visual Optics Lab Antwerp for a critical discussion of the following work.
Financing
The study was partially financed by Lens – Centrum Diagnostyki i Mikrochirurgii Oka, Olsztyn, Poland.
References
Krachmer J.H.
Feder R.S.
Belin M.W.
Keratoconus and related noninflammatory corneal thinning disorders.
Comparison of three-dimensional optical coherence tomography and combining a rotating Scheimpflug camera with a Placido topography system for forme fruste keratoconus diagnosis.
Toprak I, Vega A, Alió Del Barrio JL, Espla E, Cavas F, Alió JL. Diagnostic Value of Corneal Epithelial and Stromal Thickness Distribution Profiles in Forme Fruste Keratoconus and Subclinical Keratoconus. Cornea. Published online August 4, 2020. doi:10.1097/ICO.0000000000002435.
Fleiss JL, Levin B, Paik MC. Statistical Methods for Rates and Proportions. Wiley Series in Probability and Statistics. Published online 2003. doi:10.1002/0471445428.
GUS. Główny Urząd Statystyczny / Spisy Powszechne / Narodowe Spisy Powszechne / Narodowy Spis Powszechny 2002 - wyniki / Wyniki Narodowego Spisu Powszechnego Ludności i Mieszkań 2002 w zakresie deklarowanej narodowości oraz języka używanego w domu. Accessed September 2, 2021. https://stat.gov.pl/spisy-powszechne/narodowe-spisy-powszechne/narodowy-spis-powszechny-2002/wyniki-narodowego-spisu-powszechnego-2002-narodowosci-oraz-jezyka/.
Kristianslund O, Hagem AM, Thorsrud A, Drolsum L. Prevalence and incidence of keratoconus in Norway: a nationwide register study. Acta Ophthalmol. Published online November 16, 2020. doi:10.1111/aos.14668.
European Institute for Gender Equality. Gender Equality Index 2020: Poland. Accessed March 7, 2021. https://eige.europa.eu/sites/default/files/documents/20203357_mhag20010enn_002.pdf#:∼:text=Progress%20on%20gender%20equality%20in%20Poland%20since%202010,higher%20increase%20%28%2B%200.6%20points%29%20since%202017.%20Poland%E2%80%99s.
00050-4&id=gr1.jpg){kind=link}
00050-4&id=gr2.jpg){kind=link}
00050-4&id=gr3.jpg){kind=link}