Abstract
PURPOSE:
To describe the epidemiology of ocular and periocular tumors in patients presenting to a multi-tier ophthalmology hospital network in India using the electronic medical records (EMRs) system.
METHODS:
A retrospective, observational, referral, hospital-based study of 1,142,098 patients. The data were collected using the in-house developed EyeSmart EMR system.
RESULTS:
During a 6-year study period, 9633 (0.8%) new patients were diagnosed with eye tumors. Of the 9633 patients, 5209 (54%) were male and 4424 (46%) were female. Of all tumors, 6372 (65%) were benign, 282 (3%) were pre-malignant, and 3089 (32%) were malignant in nature, respectively. Overall, the three most common tumors included retinoblastoma (n = 1167, 12%), ocular surface squamous neoplasia (n = 957, 10%), and conjunctival nevus (n = 903, 9%). The three most common benign tumors included conjunctival nevus (n = 903, 9%), eyelid nevus (n = 358, 4%), and orbital dermoid cyst (n = 344, 4%). The three most common malignant tumors included retinoblastoma (n = 1167, 12%), ocular surface squamous neoplasia (n = 957, 10%), and sebaceous gland carcinoma (n = 202, 2%). The most common tumor in 0–10 years' age group was retinoblastoma (n = 1163, 42%), 11–30 years was conjunctival nevus (n = 408, 16%), and > 30 years was ocular surface squamous neoplasia (n = 801, 17%), respectively.
CONCLUSIONS:
The present study results indicate the incidence and distribution of ocular and periocular tumors in a large cohort in India. Retinoblastoma is the most common tumor encountered in a referral-based comprehensive ophthalmic oncology practice in India. The use of EMRs enables to capture the structured information and big data analysis of the same.
Keywords: Electronic medical record, eye, oncology, tumors
Introduction
Cancer is a leading cause of death worldwide. Based on GLOBOCAN worldwide estimates produced by the International Agency for Research on Cancer, the burden of cancer is rising each year with an estimated 14.1 million new cancer cases in 2012[1] versus 18.1 million new cases in 2018.[2] Lung and breast cancers are the most frequently diagnosed cases of cancer, whereas eye cancer is rare. The incidence rate of eye cancer in the United States was estimated at 1 in 100,000 population in 1977.[3] The American Cancer Society's estimates for the incidence of eye cancer in the United States for 2018 are 3540 cases, and most cancers are ocular melanoma.[4] Among all cancers, the incidence of eye cancer in India is estimated to be <0.5%.
Ophthalmic oncology is a highly specialized branch of ophthalmology dealing with eye tumors, both benign and malignant. While the incidence of eye tumors is low compared to other tumors, there is limited literature regarding the incidence, demographics, and types of eye tumors. In our extensive literature search, there were no studies elaborating the epidemiology of ocular and periocular tumors. India carries a huge burden of cancer and is believed to be growing over the years. This growing burden of cancer in India could be related to epidemiological transition, better cancer detection strategies, and better cancer data capture.[5] In this study, we describe the epidemiology of ocular and periocular tumors in patients presenting to a multi-tier ophthalmology hospital network in India using the electronic medical records (EMRs) system.
Methods
This was a retrospective, observational hospital-based study of 8869 patients diagnosed with ocular or periocular tumors presenting between September 2013 and August 2018 – 4 tertiary centers located at Hyderabad, Bhubaneswar, Vishakhapatnam, and Vijayawada of the L V Prasad Eye Institute (LVPEI) network spread across four states in India. Written informed consent was obtained from all participants regarding the use of their clinical data for research purpose. If the subject was a minor, a written informed consent was obtained from parents or guardian. No identifiable information of the patient was used for analytical purposes. The study adhered to the Tenants of Declaration of Helsinki. This study was reviewed and approved by an Institutional Review Board (ethics committee) of LVPEI before the study began. Each patient underwent a comprehensive ophthalmic examination, and data were entered into a browser-based EMR system (EyeSmart EMR).
Data retrieval and processing
The data of 10,862 patients diagnosed with ocular or periocular tumors were retrieved from the EMR database across the ophthalmology network. Of these, 1229 patients who did not have a confirmed ocular or periocular tumor were excluded from the study. All patients with a definitive diagnosis of benign or malignant tumor either by clinical examination or confirmed histopathology were included in the study. The columns of 9633 patients, which included the data on patient demographics, ocular diagnosis, tumor status, and anatomical category were exported for the analysis. An interactive dashboard of the data was also developed.
Statistical analysis
Descriptive statistics using mean (± standard deviation) and median (± range) were used to elucidate the demographic data. All for age, gender, diagnosis, and anatomical category were drawn by using Microsoft Excel 2013 (Microsoft Corporation, Redmond, USA).
Results
Incidence of eye tumors
Of the 1,142,098 new patients presenting to the four tertiary centers of the LVPEI between September 2013 and August 2018 in the multi-tier hospital network, 9633 (0.8%) patients were diagnosed with eye tumors. An interactive dashboard of the data can be accessed at the following link – www.lvpei.org/aeye/oncology.html.
Age
The mean age of the patients was 32 years, whereas the median age was 30 years (range: <1–91 years). Of the 9,633 patients, 3057 (32%) were children <16 years and 6576 (68%) were adults. The majority of the patients (2347, 24%) presented between 0 and 10 years of age. The age-decade wise distribution of patients with eye tumors is presented in Table 1.
Table 1.
Age distribution of patients with eye tumors
| Age (years) | n=9633, n (%) |
|---|---|
| 0-10 | 2347 (24) |
| 11-20 | 1235 (13) |
| 21-30 | 1242 (13) |
| 31-40 | 1156 (12) |
| 41-50 | 1241 (13) |
| 51-60 | 1135 (12) |
| 61-70 | 822 (9) |
| 71-80 | 361 (4) |
| 81-90 | 91 (1) |
| 91-100 | 3 (<1) |
Gender
Of the 9633 patients, 5209 (54%) were male and 4424 (46%) were female.
Anatomical category
The 9743 ocular and periocular tumor diagnoses were segregated into the nine anatomical categories. The most common anatomical part involved was the conjunctiva with 3296 (34%) cases, and the least was lacrimal sac with 11 (<1%) cases. The anatomical distribution of the tumors is presented in Table 2.
Table 2.
Anatomical distribution of eye tumors
| Anatomical category | n=9743, n (%) |
|---|---|
| Conjunctiva | 3296 (34) |
| Eyelid | 2451 (25) |
| Orbit | 1764 (18) |
| Retina | 1383 (14) |
| Choroid | 572 (6) |
| Iris | 150 (2) |
| Caruncle | 81 (1) |
| Ciliary body | 35 (<1) |
| Lacrimal sac | 11 (<1) |
Tumor status
There were 9743 diagnosis instances made in 9633 patients over a 6-year period. In them, 6372 (65%) were benign, 282 (3%) were premalignant, and 3089 (32%) were malignant in nature, respectively. The most common anatomical part involved among benign tumors was the eyelid with 2081 (21%) cases, among premalignant tumors was the conjunctiva with 271 (3%) and among the malignant tumors was the retina with 1173 (12%) cases. The tumor status distribution according to the anatomical part is presented in Table 3.
Table 3.
Tumor status and anatomical category
| Anatomical category | Benign (n=6372; 65%), n (%) | Premalignant (n=282; 3%), n (%) | Malignant (n=3089; 32%), n (%) | Total (n=9743), n (%) |
|---|---|---|---|---|
| Conjunctiva | 2008 (21) | 271 (3) | 1017 (10) | 3296 (34) |
| Eyelid | 2081 (21) | 11 (<1) | 359 (4) | 2451 (25) |
| Orbit | 1449 (15) | 0 (0) | 315 (3) | 1764 (18) |
| Retina | 210 (2) | 0 (0) | 1173 (12) | 1383 (14) |
| Choroid | 370 (4) | 0 (0) | 202 (2) | 572 (6) |
| Iris | 148 (2) | 0 (0) | 2 (<1) | 150 (2) |
| Caruncle | 79 (1) | 0 (0) | 2 (<1) | 81 (1) |
| Ciliary body | 21 (<1) | 0 (0) | 14 (<1) | 35 (<1) |
| Lacrimal sac | 6 (<1) | 0 (0) | 5 (<1) | 11 (<1) |
Tumor distribution
Including both benign and malignant eye tumors, the three most common tumors were retinoblastoma (n = 1167, 12%), ocular surface squamous neoplasia (n = 957, 10%), and conjunctival nevus (n = 903, 9%). Among the benign tumors, the three most common tumors included conjunctival nevus (n = 903, 9%), eyelid nevus (n = 358, 4%), and orbital dermoid cyst (n = 344, 4%). Among the malignant tumors, the three most common tumors included retinoblastoma (n = 1167, 12%), ocular surface squamous neoplasia (n = 957, 10%), and sebaceous gland carcinoma (n = 202, 2%). The most common tumor in the eyelid was eyelid cyst (n = 661, 7%); caruncle was caruncle inclusion cyst (n = 24, <1%); conjunctiva was conjunctival nevus (n = 903, 10%); iris was iris pigment epithelial cyst (n = 101, 1%); ciliary body was ciliary body cyst (n = 21, <1%); retina was retinoblastoma (n = 1167, 12%); choroid was choroidal melanoma (n = 145, 2%); orbit was dermoid cyst (n = 344, 4%); and lacrimal sac was papilloma (n = 4, <1%) [Table 4].
Table 4.
Details of most common eye tumors
| Anatomical location | Most common benign tumor | n (%) | Median age (IQR) | Male: female | Most common malignant tumor | n (%) | Median age (IQR) | Male: female |
|---|---|---|---|---|---|---|---|---|
| Eyelid | Eyelid cyst | 661 (7) | 42 (25-58) | 1.2:1 | Sebaceous gland carcinoma | 202 (2) | 61 (52-69) | 0.9:1 |
| Caruncle | Inclusion cyst | 24 (<1) | 29 (11-45) | 2:1 | Sebaceous gland carcinoma* | 1 (<1) | 42 (NA) | 0:1 |
| Conjunctiva | Conjunctival nevus | 903 (10) | 17 (9-30) | 1.2:1 | Ocular surface squamous neoplasia | 957 (2) | 49 (36-63) | 2.2:1 |
| Iris | Iris pigment epithelial cyst | 101 (1) | 25 (10-52) | 1.9:1 | Iris melanoma** | 1 (<1) | 50 (NA) | 0:1 |
| Ciliary body | Ciliary body cyst | 21 (<1) | 45 (39-55) | 0.9:1 | Ciliary body melanoma | 6 (<1) | 45 (42-56) | 1:1 |
| Retina | Vasoproliferative tumor | 55 (<1) | 35 (28-46) | 1.9:1 | Retinoblastoma | 1167 (12) | 4 (2-8) | 1.3:1 |
| Choroid | Choroidal hemangioma | 114 (1) | 39 (28-50) | 1.2:1 | Choroidal melanoma | 145 (2) | 50 (39-58) | 1.4:1 |
| Orbit | Dermoid cyst | 344 (4) | 15 (4-25) | 1.2:1 | Lymphoma | 118 (1) | 55 (46-63) | 1.8:1 |
| Lacrimal sac | Papilloma | 4 (<1) | 53 (33-68) | 3:1 | Lymphoma | 3 (<1) | 61 (53-64) | 1:2 |
*There was one case each of sebaceous gland carcinoma and melanoma, **There was one case each of iris melanoma and iris metastasis. NA: Not applicable, IQR=Inter-quartile range
Decade wise distribution
The most common tumor in 0–10 years' age group was retinoblastoma (n = 1163, 42%), 11–30 years was conjunctival nevus (n = 408, 16%), and >30 years was ocular surface squamous neoplasia (n = 801, 17%), respectively. The tumor status distribution according to the decade wise is presented in Table 5.
Table 5.
Tumor distribution based on the decade of presentation
| Age (years) | Most common tumor | n (%) | Most common benign tumor | n (%) | Most common malignant tumor | n (%) |
|---|---|---|---|---|---|---|
| 0-10 | Retinoblastoma | 1163 (42) | Conjunctival nevus | 244 (18) | Retinoblastoma | 1163 (95) |
| 11-20 | Conjunctival nevus | 261 (23) | Conjunctival nevus | 261 (23) | OSSN | 39 (41) |
| 21-30 | Conjunctival nevus | 147 (12) | Conjunctival nevus | 147 (12) | OSSN | 107 (64) |
| 31-40 | OSSN | 163 (14) | Conjunctival nevus | 81 (10) | OSSN | 163 (60) |
| 41-50 | OSSN | 193 (15) | Xanthelasma | 117 (14) | OSSN | 193 (15) |
| 51-60 | OSSN | 164 (14) | Xanthelasma | 128 (18) | OSSN | 164 (42) |
| 61-70 | OSSN | 157 (19) | Eyelid nevus | 47 (10) | OSSN | 157 (46) |
| 71-80 | OSSN | 91 (25) | Eyelid pilosebaceous cyst | 23 (13) | OSSN | 91 (52) |
| 81-90 | OSSN | 32 (34) | Conjunctival papilloma | 3 (11) | OSSN | 32 (49) |
| 91-100 | OSSN* | 1 (33) | NA | 0 (0) | OSSN* | 1 (33) |
*There was one case each of OSSN, eyelid squamous cell carcinoma, and choroidal melanoma. OSSN: Ocular surface squamous neoplasia, NA: Not applicable
Discussion
This study gives an overview of ocular and periocular tumor diagnoses in a referral center setup in India. Adaptation to EyeSmart EMR allowed big data analysis to describe the epidemiology of eye tumors. Overall, including both benign and malignant eye tumors, the most common tumor was retinoblastoma (n = 1167, 12%). The burden of retinoblastoma is huge with the largest number of cases reported in India. It is estimated that Asia–Pacific region carries the global burden of retinoblastoma with India having the highest incidence of the disease. The predicted annual incidence of retinoblastoma in 2013 was the highest for India with an incidence of 1486 cases (range, 1335–1638 cases).[6] Our study also revealed that retinoblastoma is the most common tumor encountered in a comprehensive ophthalmic oncology practice. With the highest burden of retinoblastoma in India, it is prudent to train more ophthalmologists in the country to detect and treat retinoblastoma, thus ensuring accessibility of care for these children.
In our study, the most common benign tumor was conjunctival nevus (n = 903, 10%). In a review of 5002 conjunctival tumors by Shields et al., 52% were benign, 18% were premalignant, and 30% were malignant.[7] The most common tumor was conjunctival nevus constituting 23% of all conjunctival tumors.[7] In our study, including 3296 conjunctival lesions, 2008 (61%) were benign, 271 (8%) were premalignant, and 1017 (31%) were malignant. Conjunctival nevus was the most common conjunctival tumor constituting 27% of all conjunctival tumors. The higher proportion of benign conjunctival lesions in our population compared to the West (61% vs. 52%) could be related to referral bias. With fewer trained ocular oncologists in the country, even benign lesions get referred to the tertiary care center for a definitive diagnosis. Awareness and improved training about common benign ocular conditions among general ophthalmologists may help to decrease the load of benign tumors in the tertiary care center.
The three most common malignant tumors in our study included retinoblastoma (n = 1167, 12%), ocular surface squamous neoplasia (n = 957, 10%), and sebaceous gland carcinoma (n = 202, 2%). In the West, ocular melanoma is the most common ocular cancer.[4] Of all ocular melanomas, uveal melanoma is common. The world incidence of uveal melanoma is estimated at 6679–7095 new cases each year with Caucasian non-Hispanic population constituting >65% cases.[8] However, uveal melanoma is rare in non-White population. In our study, there were only 152 new cases of uveal melanoma including melanoma of the iris (n = 1), ciliary body (n = 6), and choroid (n = 145) during the 6-year period. Uveal melanoma was the fifth most common malignant tumor, and retinoblastoma was the most common malignant tumor in our population accounting for 95% of malignant tumors in the first decade of life. Worldwide, ocular surface squamous neoplasia is the third most common cancer after retinoblastoma and melanoma. In our study, ocular surface squamous neoplasia was the most common conjunctival malignant tumor and the second most common ocular malignant tumor. It was the most common malignant tumor beyond 10 years of age. All patients in the first and second decades of life presenting with OSSN had associated xeroderma pigmentosum. The association between xeroderma pigmentosum and young age of detection of OSSN is well established.[9] Sebaceous gland carcinoma is rare in the West accounting for only 1%–3% cases of all eyelid malignancies. However, in India, sebaceous gland carcinoma is the most common eyelid malignancy accounting for 53% cases of all eyelid malignancies.[10] In this study, sebaceous gland carcinoma was the most common eyelid malignancy and third most common ocular malignant tumor.
The study has a few limitations. Since this is a hospital-based study, there could be a referral bias with underrepresentation of benign cases. However, the strength of the study is a large cohort of patients, accurate diagnosis by a trained ocular oncologist, and a patient representation from a diverse geography of India. The use of EMR systems enables the mining of large datasets to inform policy-makers and to formulate health-care strategies.[11] The ability to analyze digital clinical information from EMR systems also assists to understand evolving trends with regard to the burden of ocular diseases to design the effective prevention and treatment measures for the patients.[12,13,14,15]
Conclusions
The use of uniform EMR system in our outpatient service of LVPEI network laid foundation for big data analysis in a rapid, accurate, and efficient manner. An interactive dashboard of the data was developed and can be accessed at the following link – www.lvpei.org/aeye/oncology.html. This study elaborates the epidemiology of eye tumors in a referral based comprehensive ophthalmic oncology setup in India. In this study, benign tumors comprised majority (65%) of the cases. Benign tumors can be managed and followed up in secondary care centers, whereas malignant tumors may need care at a tertiary center. Improved training in ophthalmic oncology among the general ophthalmologists and oncologists can help in stratification of cases to secondary versus tertiary care centers allowing easy accessibility to health care. Thus, big data analytics using EMR data plays an important role in strategic planning of health-care framework for ophthalmic oncology in India.
Financial support and sponsorship
This study was financially supported by The Operation Eyesight Universal Institute for Eye Cancer (SK) and Hyderabad Eye Research Foundation (SK), Hyderabad, India.
Conflicts of interest
There are no conflicts of interest.
Acknowledgment
The authors wish to acknowledge the support of Mr Ranganath Vadapalli, Mr Mohammad Pasha, and the entire Department of eyeSmart EMR & AEye.
References
- 1.Torre LA, Bray F, Siegel RL, Ferlay J, Lortet-Tieulent J, Jemal A. Global cancer statistics, 2012. CA Cancer J Clin. 2015;65:87–108. doi: 10.3322/caac.21262. [DOI] [PubMed] [Google Scholar]
- 2.Latest Global Cancer Data by International Agency for Research on Cancer. 2018. [Last accessed on 2018 Dec 27]. Available from: https://www.who.int/cancer/PRGlobocanFinal.pdf .
- 3.Kleinstein RN, Lehman HF. Incidence and prevalence of eye cancer. Am J Optom Physiol Opt. 1977;54:49–51. doi: 10.1097/00006324-197701000-00008. [DOI] [PubMed] [Google Scholar]
- 4.American Cancer Society's Estimates for Eye Cancer in the United States for 2018. 2018. [Last accessed on2018 Dec 27]. Available from: https://www.cancer.org/cancer/eye-cancer/about/key-statistics.html .
- 5.Smith RD, Mallath MK. History of the growing burden of cancer in India: From antiquity to the 21st century. J Glob Oncol. 2019;5:1–5. doi: 10.1200/JGO.19.00048. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Usmanov RH, Kivelä T. Predicted trends in the incidence of retinoblastoma in the asia-pacific region. Asia Pac J Ophthalmol (Phila) 2014;3:151–7. doi: 10.1097/APO.0000000000000060. [DOI] [PubMed] [Google Scholar]
- 7.Shields CL, Alset AE, Boal NS, Casey MG, Knapp AN, Sugarman JA, et al. Conjunctival tumors in 5002 cases. Comparative analysis of benign versus malignant counterparts. The 2016 James D. Allen Lecture. Am J Ophthalmol. 2017;173:106–33. doi: 10.1016/j.ajo.2016.09.034. [DOI] [PubMed] [Google Scholar]
- 8.Kivelä T. The epidemiological challenge of the most frequent eye cancer: Retinoblastoma, an issue of birth and death. Br J Ophthalmol. 2009;93:1129–31. doi: 10.1136/bjo.2008.150292. [DOI] [PubMed] [Google Scholar]
- 9.Kaliki S, Jajapuram SD, Maniar A, Mishra DK. Ocular and periocular tumors in xeroderma pigmentosum: A study of 120 Asian Indian patients. Am J Ophthalmol. 2019;198:146–53. doi: 10.1016/j.ajo.2018.10.011. [DOI] [PubMed] [Google Scholar]
- 10.Kaliki S, Bothra N, Bejjanki KM, Nayak A, Ramappa G, Mohamed A, et al. Malignant eyelid tumors in India: A study of 536 Asian Indian patients. Ocul Oncol Pathol. 2019;5:210–9. doi: 10.1159/000491549. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Das AV, Kammari P, Vadapalli R, Basu S. Big data and the eyeSmart electronic medical record system – An 8-year experience from a three-tier eye care network in India. Indian J Ophthalmol. 2020;68:427–32. doi: 10.4103/ijo.IJO_710_19. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Das AV, Rath S, Naik MN, Ali MJ. The incidence of lacrimal drainage disorders across a tertiary eye care network: Customization of an indigenously developed electronic medical record system-eyeSmart. Ophthalmic Plast Reconstr Surg. 2019;35:354–6. doi: 10.1097/IOP.0000000000001257. [DOI] [PubMed] [Google Scholar]
- 13.Donthineni PR, Kammari P, Shanbhag SS, Singh V, Das AV, Basu S. Incidence, demographics, types and risk factors of dry eye disease in India: Electronic medical records driven big data analytics report I. Ocul Surf. 2019;17:250–6. doi: 10.1016/j.jtos.2019.02.007. [DOI] [PubMed] [Google Scholar]
- 14.Das AV, Kammari P, Guizie EB, Varada R, Singh G. Prevalence of ocular disorders in Liberia: A retrospective study using the eyeSmart electronic medical record system. J Glob Health Rep. 2019;3:e2019033. [Google Scholar]
- 15.Das AV, Podila S, Prashanthi GS, Basu S. Clinical profile of pterygium in patients seeking eye care in India: Electronic medical records-driven big data analytics report III. Int Ophthalmol. 2020;40:1553–63. doi: 10.1007/s10792-020-01326-3. [DOI] [PubMed] [Google Scholar]
