Abstract
Background
This study aimed to provide a comprehensive assessment of burden estimates and the secular trend of vision loss due to retinopathy of prematurity (ROP) among people younger than 20 years, at the global, regional and national levels.
Methods
Data were obtained from the Global Burden of Disease Study 2019 database. The average annual percentage change (AAPC) was calculated to quantify the temporal trends in the measures of vision loss.
Results
In 2019, the global age-standardised rates (ASRs) of prevalence per 100 000 population was 86.4 for vision loss, specifically, 35 for moderate vision loss, 19.9 for severe vision loss, 31.6 for blindness due to ROP among people younger than 20 years. Moreover, the ASR of years lived with disability per 100 000 was 10.6 for vision loss, specifically, 1.1 for moderate vision loss, 3.6 for severe vision loss, 5.9 for blindness, respectively. From 1990 to 2019, the ASR of prevalence of blindness and vision loss due to ROP significantly increased, while its burden slightly decreased. Males showed higher ASR of prevalence than females in 2019, whereas females have larger increasing trend than males from 1990 to 2019. The global highest ASR of disease burden was observed in South Asia and Southern sub-Saharan Africa, as well as low sociodemographic index (SDI) regions in 2019.
Conclusions
Globally, although the burden decreased, the prevalence of childhood and adulthood vision loss due to ROP continues to increase. Reasonable resource allocation and advanced intervention are recommended to prevent and control the vision loss due to ROP.
Keywords: Epidemiology, Ophthalmology
WHAT IS ALREADY KNOWN ON THIS TOPIC
Retinopathy of prematurity (ROP) is a retinal vascular disease affecting premature infants that can culminate in blindness.
In recent decades, the disease burden of ROP has been increasing in some countries, as screening promotion globally.
However, the prevalence and the disease burden of ROP among children and adolescents have not been estimated in detail.
WHAT THIS STUDY ADDS
This study presents the comprehensive estimations of the ROP burden of disease by age, sex and sociodemographic index (SDI) levels for 204 countries and territories between 1990 and 2019, using the most updated Global Burden of Disease data for 2019.
HOW THIS STUDY MIGHT AFFECT RESEARCH, PRACTICE OR POLICY
This study explores the spatiotemperal burden of ROP globally and provides a reference for formulating reasonable and effective ROP prevention and control policies among childhood and adolescents.
Introduction
Retinopathy of prematurity (ROP) is an ocular disease caused by retinal ophthalmic dysplasia in premature infants, leading to strabismus, amblyopia, cataract, glaucoma and even blindness, which seriously affects the quality of life of preterm infants. The proportion of infant blindness has reached 6%–18% according to figures released by the WHO. Generally speaking, ROP is the main cause of blindness in children of the world. Recently, a comprehensive study reported the burden of visual impairments due to 16 different diseases during the past three decades using the Global Burden of Diseases 2019 (GBD 2019). It is estimated that approximately 2 480 000 people were blind globally, and 423 000 people had moderate and severe visual impairments due to ROP among all age groups in 2019. To our acknowledge, as a serious threat to childhood blindness, there is no report on the burden of ROP among children and adolescents. According to population-based dataset, females and Hispanic ethnicity infants are more likely to suffer from ROP.1 However, other studies revealed that the prevalence rates of the ROP are more common in whites than in blacks or in boys more than girls,2 while ROP had no significant relationship with sex in Iran.3 Moreover, the distribution of ROP by age is also lacking.
Understanding the epidemiological characteristics of ROP is crucial for disease prevention. Therefore, we conducted a comprehensive estimation on the burden of ROP by age, sex and sociodemographic index (SDI) levels for 204 countries and territories between 1990 and 2019.
Methods
Data sources
Data for global children and adolescent (0–19 years) ROP from 1990 to 2019 were retrieved from IHME (Institute for Health Metrics and Evaluation) ‘GBD compare’ tool available online (https://vizhub.healthdata.org/gbd-compare/). The GBD 2019 codes can be accessed from https://ghdx.healthdata.org/gbd-2019/code. GBD study 2019 entails the data for 369 injuries and diseases, 286 causes of death and 87 risk factors in 204 countries and territories. Comprehensive methodology has been published previously.4
ROP definition
ROP was defined based on the International Classification of Diseases (ICD) codes (eg, P07.2–P07.39, P22–P22.9, P25–P28.9, P61.2 and P77–P77.9).
Vision loss was divided into three categories: moderate vision impairment (defined as visual acuity of ≥6/60 and <6/18), severe vision impairment (visual acuity of ≥3/60 and <6/60) and blindness (visual acuity of <3/60 or <10° visual field around central fixation).
Measures
The main outcomes of the current study include the total number (in thousand), crude rates (per 100 000 population) and age-standardised rates (per 100 000 population) of incidence, prevalence and years lived with disability (YLDs) due to ROP. Currently, YLDs refer to years of life lived with any health loss, which were calculated based on the prevalence of the eye disorders multiplied by its associated disability weight. In GBD datasets, disability weights use counts on a scale from 0 to 1, which represents the severity of health loss associated with a particular health state. The severity of health loss (blindness, severe and moderate vision loss) was according to surveys of the general public in five countries (Bangladesh, Indonesia, Peru, Tanzania and the USA) as well as an open internet survey. More information on YLD can be found from the IHME databases (https://ghdx.healthdata.org/), Generally, YLD rates were calculated by dividing the number of YLDs by the relevant population. Age-adjusted rates were calculated using the WHO 2000–2025 World Standard Population, which excluded the effects of total population size and age structure. The percentages for YLDs were calculated as the number of YLDs due to a specific disease divided by the total number of YLDs for all diseases.
Sociodemographic index
SDI is the geometric mean of lag-distributed income per capita, average educational attainment for people aged 15years and older, and the total fertility rate (in people aged <25 years) indicators. The SDI values ranged from 0 to 1, where 0 indicates that the development level related to health outcomes in the region is theoretically the lowest. When the SDI value is 1, it indicates that the region has the highest theoretical level of development related to health outcomes.
Statistical analysis
All analyses were accomplished using the R program (V.4.0.0). Moreover, all estimates were derived from the mean of 1000 draws, and 95% uncertainty intervals (UIs) were determined using the 2.5th and 97.5th percentiles of the ordered draws. The size and course of temporal trends on ROP from 1990 to 2019 were evaluated using the joinpoint regression analysis (Joinpoint regression software, V.4.9.0.0. Statistical Research and Applications Branch, National Cancer Institute, USA). The joinpoint regression analysis identified points called joinpoints, which denote a statistically significant change in a trend and determine the trends between joinpoints. The permutation test was applied using the Monte Carlo method. The grid search was used to fit the segmented regression function. Using the calendar year as the regression variable, the joinpoint regression analysis estimated the average annual percent change (AAPC) as a summary measure of the trend, with the corresponding 95% CI. We further analysed the strength and direction of the association between temporal trends of ROP burden and SDI using Spearman’s rank-order correlation. P value <0.05 was considered as statistical significance.
Ethics approval
This study followed the Guidelines for Accurate and Transparent Health Estimates Reporting (GATHER). Guangdong Provincial People’s Hospital Institutional Review Board has approved a waiver of informed consent.
Results
Global level
In 2019, there were 2 169 800 (95% UI 1714.5 to 2670.1, thousand) prevalent cases of blindness and vision loss due to ROP across the world, with an age-standardised point prevalence of 86.4 (95% UI 70.3 to 103.8) per 100 000 population, which showed an increasing trend with AAPC to be 0.08 (95% CI 0.06 to 0.1) among people younger than 20 years old (table 1). In 2019, ROP caused 951 000 (95% UI 542.4 to 1452.8, thousand) moderate vision loss (online supplemental table 1), 537 400 (95% UI 381.3 to 716.9, thousand) severe vision loss (online supplemental table 2) and 681 400 (95% UI 488.4 to 876.4, thousand) blindness (online supplemental table 3). The ASR of prevalence on ROP caused moderate vision loss, severe vision loss and blindness was 35 per 100 000 (95% UI 19.9 to 53.6), 19.9 per 100 000 (95% UI 14.1 to 26.5) and 31.6 per 100 000 (95% UI 24 to 39.4), respectively. Additionally, the trends of global prevalence of ROP caused moderate vision loss (AAPC 0.59; 95% CI 0.58 to 0.61) and severe vision loss (AAPC 0.34; 95% CI 0.32 to 0.35) were increasing, while the prevalence of ROP caused blindness showed decreasing trend (AAPC −0.68; 95% CI −0.84 to −0.52).
Table 1. Prevalence of retinopathy of prematurity by regions, 1990 and 2019.
| Regions | 1990 counts(thousand) | Age-standardised rate (per 100 000 population), 1990 | Crude rate (per 100 000 population), 1990 | 2019 counts(thousand) | Age-standardised rate (per 100 000 population), 2019 | Crude rate (per 100 000 population), 2019 | Average annual percent change | % change in number between 1990 and 2019 |
|---|---|---|---|---|---|---|---|---|
| Global | 1870.1 (1526.2 to 2216.4) | 80.4 (65.8 to 94.6) | 82.3 (67.1 to 97.5) | 2169.8 (1714.5 to 2670.1) | 86.4 (70.3 to 103.8) | 84.1 (66.5 to 103.5) | 0.08 (0.06 to 0.1) | 0.16 (0.07 to 0.26) |
| High SDI | 144.1 (106.3 to 185.4) | 66.3 (50.8 to 83.6) | 61.6 (45.4 to 79.2) | 152.7 (110.3 to 202.1) | 74.9 (56.2 to 96.7) | 69.1 (50 to 91.5) | 0.4 (0.37 to 0.44) | 0.06 (−0.01 to 0.13) |
| High-middle SDI | 258.4 (205.4 to 312.4) | 66.4 (53.7 to 78.8) | 63.8 (50.7 to 77.1) | 226 (177.5 to 283.2) | 75.3 (61.3 to 90.9) | 68.9 (54.1 to 86.3) | 0.26 (0.24 to 0.29) | −0.13 (−0.21 to −0.05) |
| Middle SDI | 576.2 (476.5 to 674.6) | 75.7 (63.8 to 88) | 75.1 (62.1 to 88) | 572.5 (462.2 to 702.1) | 82.2 (68.2 to 97.2) | 77.7 (62.8 to 95.3) | 0.12 (0.09 to 0.16) | −0.01 (−0.1 to 0.1) |
| Low-middle SDI | 571.9 (458.9 to 682.2) | 100.1 (80.7 to 118.6) | 100.2 (80.4 to 119.6) | 608.9 (474.1 to 755.1) | 93.8 (75 to 114.3) | 87.6 (68.2 to 108.6) | −0.46 (−0.49 to −0.43) | 0.06 (−0.04 to 0.19) |
| Low SDI | 318.5 (261.6 to 379.2) | 108.2 (89.2 to 127.9) | 107.8 (88.6 to 128.4) | 608.6 (483.1 to 742) | 106 (85.4 to 128.2) | 101.9 (80.9 to 124.3) | −0.19 (−0.23 to −0.16) | 0.91 (0.74 to 1.12) |
| Andean Latin America | 12.3 (9.4 to 15.6) | 67.3 (51.5 to 84.1) | 64.4 (49.1 to 81.7) | 13.4 (9.8 to 17.8) | 62.7 (47.2 to 81.8) | 56.4 (41.3 to 75.2) | −0.52 (−0.73 to −0.32) | 0.08 (−0.14 to 0.38) |
| Australasia | 4.3 (3.1 to 5.7) | 75.9 (57.4 to 98.3) | 68.1 (49.7 to 90.5) | 5 (3.6 to 6.7) | 76.6 (57.1 to 100.7) | 68.9 (49.6 to 93.3) | 0.05 (−0.03 to 0.12) | 0.16 (−0.01 to 0.35) |
| Caribbean | 11.5 (8.5 to 15.2) | 78.7 (59.6 to 102.4) | 76.1 (56.2 to 100.9) | 11.4 (8 to 15.7) | 76.6 (56 to 102.4) | 73.7 (51.6 to 101.3) | −0.12 (−0.2 to −0.03) | 0 (−0.13 to 0.15) |
| Central Asia | 16.5 (12.4 to 20.6) | 55 (42.7 to 67.7) | 52.2 (39.3 to 65.4) | 17.6 (13.2 to 22.3) | 56.5 (43.6 to 70.4) | 51.7 (38.9 to 65.7) | −0.03 (−0.1 to 0.04) | 0.07 (−0.07 to 0.23) |
| Central Europe | 17.9 (13 to 23.8) | 52 (39.3 to 67.2) | 46.5 (33.8 to 61.7) | 12.4 (8.6 to 16.9) | 59.6 (43.7 to 78.9) | 53 (36.6 to 72.2) | 0.45 (0.43 to 0.47) | −0.31 (−0.39 to −0.23) |
| Central Latin America | 49 (38.2 to 60.8) | 61.9 (48.4 to 75.9) | 59.5 (46.4 to 73.9) | 54.4 (41.2 to 70.6) | 66.4 (51 to 85.2) | 62.2 (47.1 to 80.8) | 0.16 (0.14 to 0.17) | 0.11 (−0.02 to 0.26) |
| Central sub-Saharan Africa | 19.3 (14 to 25.8) | 61.9 (44.6 to 82.4) | 61.4 (44.3 to 81.8) | 45 (31 to 61.8) | 65.4 (45.4 to 88.8) | 63.2 (43.6 to 86.8) | 0.12 (0.01 to 0.22) | 1.33 (0.97 to 1.77) |
| East Asia | 296.7 (250.3 to 346) | 64.7 (54.8 to 75.3) | 63.7 (53.8 to 74.3) | 201.5 (167.8 to 235.8) | 70.3 (59.7 to 81.2) | 64.8 (54 to 75.8) | 0.05 (−0.05 to 0.14) | −0.32 (−0.39 to −0.25) |
| Eastern Europe | 32.7 (24.5 to 42.4) | 52.4 (40 to 66.6) | 48.6 (36.3 to 63) | 26.9 (19.5 to 36) | 61.4 (45.7 to 80.1) | 56.8 (41.1 to 76) | 0.55 (0.52 to 0.57) | −0.18 (−0.26 to −0.08) |
| Eastern sub-Saharan Africa | 109.6 (86.6 to 133.1) | 99.3 (78.8 to 120.7) | 99 (78.2 to 120.3) | 206.4 (162 to 258.4) | 96.9 (76.9 to 119.1) | 92.3 (72.5 to 115.6) | −0.23 (−0.3 to −0.17) | 0.88 (0.69 to 1.12) |
| High-income Asia Pacific | 26.4 (20.5 to 33.2) | 55.2 (43.7 to 68.6) | 52.4 (40.7 to 65.8) | 18 (13.4 to 23.5) | 60.7 (46.3 to 78) | 55.7 (41.5 to 72.7) | 0.21 (0.18 to 0.24) | −0.32 (−0.39 to −0.23) |
| High-income North America | 61.1 (42.8 to 82.3) | 81.3 (59.2 to 107.4) | 75 (52.6 to 101) | 73.4 (50.5 to 100.5) | 88.5 (63.5 to 118.5) | 81.6 (56.1 to 111.8) | 0.32 (0.25 to 0.39) | 0.2 (0.11 to 0.28) |
| North Africa and Middle East | 173.7 (140.8 to 207.9) | 98.5 (80.7 to 117) | 96.3 (78.1 to 115.3) | 235.5 (182 to 299.3) | 105.9 (83.2 to 132.3) | 102.9 (79.6 to 130.8) | 0.23 (0.21 to 0.25) | 0.36 (0.19 to 0.52) |
| Oceania | 2.2 (1.6 to 2.9) | 64.9 (48 to 85.3) | 65.7 (48.6 to 86.5) | 3.8 (2.8 to 5.1) | 62.6 (45.8 to 83.1) | 62.6 (45.7 to 83.6) | −0.18 (−0.22 to −0.13) | 0.76 (0.39 to 1.25) |
| South Asia | 609.2 (480.7 to 733.2) | 112 (89 to 134) | 111.2 (87.7 to 133.8) | 672.5 (525.9 to 839.2) | 105.9 (84.2 to 129) | 96.8 (75.7 to 120.8) | −0.48 (−0.51 to −0.44) | 0.1 (−0.01 to 0.24) |
| Southeast Asia | 158.7 (131.6 to 186.4) | 73.4 (61.6 to 85.7) | 71.7 (59.5 to 84.3) | 146.8 (117.1 to 179.1) | 69.4 (56.6 to 82.9) | 65.1 (51.9 to 79.4) | −0.32 (−0.34 to −0.29) | −0.08 (−0.18 to 0.04) |
| Southern Latin America | 8.7 (6.6 to 11.3) | 49.7 (39.1 to 63.3) | 44.9 (34.2 to 58.5) | 10.2 (7.4 to 13.6) | 57.2 (43.1 to 73.7) | 51.1 (37.1 to 68) | 0.44 (0.42 to 0.47) | 0.17 (−0.04 to 0.43) |
| Southern sub-Saharan Africa | 33.2 (25.2 to 42.3) | 126.2 (94.6 to 161.1) | 126.6 (96.1 to 161.4) | 39.1 (28.6 to 51.2) | 131.8 (97.5 to 171.5) | 127.6 (93.6 to 167.3) | 0.03 (−0.06 to 0.12) | 0.18 (0.05 to 0.31) |
| Tropical Latin America | 58.6 (45.5 to 72.1) | 84.3 (65.1 to 104.5) | 84.2 (65.4 to 103.5) | 62.1 (46.9 to 80) | 97.6 (74.4 to 124) | 93 (70.2 to 119.7) | 0.36 (0.32 to 0.39) | 0.06 (−0.06 to 0.2) |
| Western Europe | 57.3 (43.5 to 73.5) | 64.6 (50.1 to 80.6) | 58.2 (44.2 to 74.7) | 54.4 (39.7 to 71.2) | 66.9 (50.3 to 85) | 58.9 (43 to 77.2) | 0.04 (0.02 to 0.07) | −0.05 (−0.14 to 0.05) |
| Western sub-Saharan Africa | 111.3 (89.7 to 130.7) | 103.6 (82.4 to 122.1) | 103.7 (83.5 to 121.8) | 260.1 (209.7 to 314) | 107.2 (86.1 to 128.7) | 104.7 (84.5 to 126.5) | 0.04 (0 to 0.08) | 1.34 (1.09 to 1.65) |
In addition, the global YLD numbers of blindness and vision loss due to ROP was 253 900 (95% UI 166.6 to 366.6, thousand) among people younger than 20 years old (table 2). The ASR of blindness and vision loss due to ROP was 10.6 per 100 000 (95% CI 7.1 to 15.1) in 2019, with downward trend (AAPC −0.19; 95% CI −0.24 to −0.13). The global trends of YLD on moderate vision loss (AAPC 0.59; 95% CI 0.58 to 0.61) (online supplemental table 4) and severe vision loss due to ROP (AAPC 0.34; 95% CI 0.32 to 0.35) were increasing (online supplemental table 5), whereas the blindness burden (AAPC −0.67; 95% CI −0.83 to −0.51) was decreasing from 1990 to 2019 (online supplemental table 6). Compared with 1990, the proportion of ROP-related blindness was reduced, whereas the proportion of ROP-related moderate and severe vision loss was increased.
Table 2. Years lived with disability of retinopathy of prematurity by regions, 1990 and 2019.
| Regions | 1990 counts(thousand) | Age-standardised rate (per 100 000 population), 1990 | Crude rate (per 100 000 population), 1990 | 2019 counts(thousand) | Age-standardised rate (per 100 000 population), 2019 | Crude rate (per 100 000 population), 2019 | Average annual percent change | % change in number between 1990 and 2019 |
|---|---|---|---|---|---|---|---|---|
| Global | 236.1 (156.7 to 340.5) | 10.2 (6.7 to 14.7) | 10.4 (6.9 to 15) | 253.9 (166.6 to 366.6) | 10.6 (7.1 to 15.1) | 9.8 (6.5 to 14.2) | −0.19 (−0.24 to −0.13) | 0.08 (−0.01 to 0.18) |
| High SDI | 15.9 (9.8 to 23.5) | 7.6 (4.8 to 11.2) | 6.8 (4.2 to 10) | 15.6 (9.4 to 23.6) | 8.2 (5.1 to 12.1) | 7 (4.3 to 10.7) | 0.14 (0.1 to 0.18) | −0.02 (−0.09 to 0.06) |
| High-middle SDI | 32.7 (21.4 to 47.5) | 8.6 (5.6 to 12.3) | 8.1 (5.3 to 11.7) | 26.1 (17 to 38) | 9.3 (6.1 to 13.3) | 8 (5.2 to 11.6) | −0.07 (−0.12 to −0.02) | −0.2 (−0.28 to −0.12) |
| Middle SDI | 75.4 (50.4 to 107.9) | 10 (6.7 to 14.2) | 9.8 (6.6 to 14.1) | 69.1 (45.9 to 99.5) | 10.4 (7 to 14.8) | 9.4 (6.2 to 13.5) | −0.15 (−0.18 to −0.12) | −0.08 (−0.17 to 0.03) |
| Low-middle SDI | 72.5 (47.5 to 104.8) | 12.7 (8.3 to 18.1) | 12.7 (8.3 to 18.4) | 71.6 (46.4 to 103) | 11.5 (7.6 to 16.5) | 10.3 (6.7 to 14.8) | −0.72 (-0.78 to -0.67) | −0.01 (-0.12 to 0.13) |
| Low SDI | 39.6 (25.4 to 57.5) | 13.4 (8.5 to 19.5) | 13.4 (8.6 to 19.5) | 71.4 (47.2 to 103.1) | 12.7 (8.3 to 18.4) | 12 (7.9 to 17.3) | −0.38 (−0.44 to −0.32) | 0.8 (0.61 to 1.04) |
| Andean Latin America | 1.6 (1 to 2.4) | 8.9 (5.7 to 13.2) | 8.4 (5.3 to 12.5) | 1.5 (0.9 to 2.2) | 7.4 (4.8 to 10.9) | 6.3 (4 to 9.4) | −1.01 (−1.15 to −0.86) | −0.06 (−0.28 to 0.24) |
| Australasia | 0.5 (0.3 to 0.7) | 9.1 (5.7 to 13.2) | 7.7 (4.7 to 11.6) | 0.5 (0.3 to 0.8) | 8.9 (5.5 to 13.4) | 7.4 (4.4 to 11.7) | −0.11 (−0.17 to −0.04) | 0.11 (−0.11 to 0.37) |
| Caribbean | 1.3 (0.8 to 1.9) | 9.1 (5.7 to 13.8) | 8.5 (5.1 to 12.8) | 1.1 (0.7 to 1.7) | 8.1 (5 to 12.2) | 7.2 (4.3 to 11.2) | −0.51 (−0.57 to −0.45) | −0.12 (−0.23 to 0.01) |
| Central Asia | 2.1 (1.3 to 3.1) | 7.2 (4.6 to 10.6) | 6.7 (4.2 to 9.8) | 2.1 (1.3 to 3.1) | 7.1 (4.6 to 10.3) | 6.2 (3.9 to 9.2) | −0.26 (−0.33 to −0.2) | 0 (−0.15 to 0.2) |
| Central Europe | 2 (1.2 to 3.1) | 6.2 (3.8 to 9.1) | 5.2 (3.1 to 7.9) | 1.2 (0.7 to 2) | 6.5 (3.9 to 9.9) | 5.3 (3 to 8.4) | 0.1 (0.07 to 0.13) | −0.38 (−0.45 to −0.29) |
| Central Latin America | 6.2 (4 to 9.1) | 8 (5.2 to 11.8) | 7.6 (4.9 to 11.1) | 6.2 (3.9 to 9.3) | 7.9 (4.9 to 11.6) | 7.1 (4.4 to 10.6) | −0.22 (−0.24 to −0.2) | 0 (−0.14 to 0.14) |
| Central sub-Saharan Africa | 2 (1.2 to 3.1) | 6.5 (3.8 to 10.2) | 6.4 (3.7 to 10) | 4.2 (2.4 to 6.7) | 6.3 (3.6 to 10) | 5.9 (3.3 to 9.5) | −0.3 (−0.37 to −0.24) | 1.07 (0.71 to 1.55) |
| East Asia | 40.2 (26.4 to 58.3) | 8.8 (5.7 to 12.6) | 8.6 (5.7 to 12.5) | 26.8 (17.8 to 39.4) | 9.6 (6.3 to 14) | 8.6 (5.7 to 12.7) | −0.03 (−0.1 to 0.04) | −0.33 (−0.42 to −0.24) |
| Eastern Europe | 3.7 (2.4 to 5.6) | 6.2 (4 to 9.2) | 5.5 (3.5 to 8.3) | 2.8 (1.7 to 4.3) | 6.9 (4.2 to 10.2) | 6 (3.5 to 9.1) | 0.3 (0.23 to 0.38) | −0.24 (−0.35 to −0.11) |
| Eastern sub-Saharan Africa | 13.2 (8.1 to 19.6) | 12 (7.3 to 17.8) | 11.9 (7.4 to 17.7) | 23.7 (14.8 to 34.7) | 11.4 (7.2 to 16.7) | 10.6 (6.6 to 15.5) | −0.4 (−0.47 to −0.32) | 0.79 (0.55 to 1.1) |
| High-income Asia Pacific | 3.3 (2.1 to 4.9) | 7.1 (4.5 to 10.3) | 6.6 (4.1 to 9.7) | 2 (1.3 to 3.1) | 7.2 (4.6 to 10.7) | 6.3 (3.9 to 9.5) | −0.14 (−0.17 to −0.1) | −0.38 (−0.47 to −0.28) |
| High-income North America | 6.1 (3.6 to 9.5) | 8.7 (5.2 to 13) | 7.5 (4.4 to 11.7) | 7 (4.1 to 11) | 9.1 (5.5 to 13.8) | 7.8 (4.6 to 12.2) | 0.16 (0.08 to 0.23) | 0.15 (0.08 to 0.23) |
| North Africa and Middle East | 22.6 (14.8 to 32.4) | 12.9 (8.5 to 18.5) | 12.5 (8.2 to 18) | 26.9 (17.4 to 39) | 12.5 (8.2 to 18) | 11.8 (7.6 to 17) | −0.22 (−0.24 to −0.19) | 0.19 (0.04 to 0.37) |
| Oceania | 0.3 (0.2 to 0.4) | 8 (5 to 12) | 8.2 (5 to 12.2) | 0.5 (0.3 to 0.7) | 7.5 (4.6 to 11.5) | 7.5 (4.5 to 11.7) | −0.27 (−0.32 to −0.23) | 0.7 (0.33 to 1.17) |
| South Asia | 77.4 (50.4 to 111.3) | 14.2 (9.3 to 20.3) | 14.1 (9.2 to 20.3) | 80.4 (52 to 117.1) | 13.3 (8.6 to 19.1) | 11.6 (7.5 to 16.9) | −0.69 (−0.8 to −0.57) | 0.04 (−0.09 to 0.2) |
| Southeast Asia | 20.8 (13.5 to 30.2) | 9.7 (6.3 to 14) | 9.4 (6.1 to 13.7) | 18.1 (11.8 to 26.3) | 8.8 (5.8 to 12.5) | 8 (5.2 to 11.6) | −0.53 (−0.6 to −0.47) | −0.13 (−0.24 to 0) |
| Southern Latin America | 1.1 (0.7 to 1.7) | 6.7 (4.2 to 9.8) | 5.8 (3.6 to 8.8) | 1.2 (0.7 to 1.7) | 6.8 (4.4 to 10) | 5.8 (3.6 to 8.7) | 0 (−0.03 to 0.03) | 0.03 (−0.18 to 0.31) |
| Southern sub-Saharan Africa | 3.6 (2.2 to 5.5) | 13.6 (8.3 to 20.5) | 13.8 (8.5 to 20.8) | 3.8 (2.4 to 5.7) | 12.9 (8 to 19.3) | 12.4 (7.7 to 18.6) | −0.36 (−0.45 to −0.28) | 0.05 (−0.06 to 0.19) |
| Tropical Latin America | 7.3 (4.7 to 10.8) | 10.5 (6.7 to 15.3) | 10.6 (6.7 to 15.5) | 7.1 (4.4 to 10.7) | 11.5 (7.2 to 17.2) | 10.7 (6.6 to 16) | 0.05 (0.02 to 0.09) | −0.03 (−0.18 to 0.14) |
| Western Europe | 6.6 (4.2 to 9.9) | 7.9 (5.1 to 11.5) | 6.8 (4.3 to 10) | 5.7 (3.5 to 8.6) | 7.6 (4.8 to 11.2) | 6.2 (3.8 to 9.4) | −0.29 (−0.32 to −0.27) | −0.14 (−0.22 to −0.05) |
| Western sub-Saharan Africa | 14 (8.6 to 20.9) | 13.1 (7.8 to 19.9) | 13.1 (8 to 19.5) | 30.9 (19.8 to 45) | 12.9 (8.2 to 18.9) | 12.4 (8 to 18.1) | −0.17 (−0.23 to −0.11) | 1.2 (0.93 to 1.58) |
Regional level
The global number of prevalent cases of blindness and vision loss due to ROP increased from 1 870 100 (95% UI 1526.2 to 2216.4, thousand) in 1990 to 2 169 800 (95% UI 1714.5 to 2670.1, thousand) in 2019. South Asia, East Asia, and North Africa and Middle East experienced the largest number of prevalent cases in 1990, while the highest in 2019 were found in South Asia, Western sub-Saharan Africa, and North Africa and Middle East (table 1). In 2019, the ASR of prevalence on blindness and vision loss due to ROP (per 100 000 population) was highest in Southern sub-Saharan Africa (131.8 (95% UI 97.5 to 171.5)), Western sub-Saharan Africa (107.2 (95% UI 86.1 to 128.7)) and North Africa and Middle East (105.9 (95% UI 83.2 to 132.3)). Central Asia (56.5 (95% UI 43.6 to 70.4)), Southern Latin America (57.2 (95% UI 43.1 to 73.7)) and Central Europe (59.6 (95% UI 43.7 to 78.9)) had the lowest ASR for people younger than 20 years old (table 1). From 1990 to 2019, the highest ASR of children and adolescents with moderate vision loss and severe vision loss due to ROP was observed in Southern Latin America and Southern Latin America, respectively. All GBD regions showed downward trends of the ASR of prevalence with blindness due to ROP, except high-income North America with the largest increase (AAPC 0.08; 95% UI 0.02 to 0.14).
The global number of YLDs on blindness and vision loss due to ROP increased from 236 100 (95% UI 156.7 to 340.5, thousand) in 1990 to 253 900 (95% UI 166.6 to 366.6, thousand) in 2019. South Asia, East Asia, and North Africa and Middle East had the highest numbers of YLDs due to anaemia in 1990, while in 2019, the highest numbers were found in South Asia, Western sub-Saharan Africa, and North Africa and Middle East (table 2). South Asia (13.3 (95% UI 8.6 to 19.1)), Southern sub-Saharan Africa (12.9 (95% UI 8 to 19.3)) and Western sub-Saharan Africa (12.9 (95% UI 8.2 to 18.9)) had the highest ASR of YLD from blindness and vision loss due to ROP. Specifically, ROP caused worldwide ASR of YLD to be 5.9 per 100 000 (95% UI 3.7 to 8.9) for blindness in 2019, 1.1 per 100 000 (95% UI 0.5 to 1.9) and 3.6 per 100 000 (95% UI 2.1 to 5.7) of all causes for moderate and severe vision loss, respectively (table 2). Except high-income North America, all GBD regions showed a decrease in the age-standardised point YLD of blindness due to ROP from 1990 to 2019, with the largest decreases being in Andean Latin America (AAPC −2.34; 95% CI −2.54 to −2.15), South Asia (AAPC −1.44; 95% CI −1.74 to −1.13), and North Africa and Middle East (AAPC −1.24; 95% CI −1.31 to −1.17).
National level
In 2019, the national ASR of prevalence of blindness and vision loss due to ROP ranged from 147.9 to 35.5 cases per 100 000 population. Afghanistan (147.9 (95% UI 106.8 to 197.1)), Yemen (146.8 (95% UI 100.7 to 205.2)) and Niger (144.6 (95% UI 105.1 to 191.6)) had the highest age-standardised point prevalence rates of blindness and vision loss due to ROP in 2019. In contrast, Barbados (35.5 (95% UI 16.6 to 56.5)), Taiwan (Province of China) (38.6 (95% UI 22.3 to 57)) and Sweden (39 (95% UI 24.2 to 54.1)) had the lowest rates among people younger than 20 years (online supplemental table 7). South Africa had the largest ASR on prevalence of moderate vision loss, Afghanistan had the largest ASR on prevalence of severe vision loss, and Mali had the largest ASR on prevalence of blindness. The changes in the ASR of prevalence, from 1990 to 2019, differed substantially between countries (figure 1). Greece, Georgia and Bahrain showed the largest increases over the period 1990–2019, while Cambodia, Mozambique and Ghana showed the largest decreases.
Figure 1. Global map of retinopathy of prematurity–related vision impairment prevalence changes from 1990 to 2019 among people younger than 20 years old.
The national ASR of YLD rates of blindness and vision loss due to ROP varied in 2019 from 2.7 to 18.5 cases per 100 000 population. The highest rates were observed in Afghanistan (18.5 (95% UI 11.4 to 27.8)), Mali (17.7 (95% UI 8.9 to 29)) and Ethiopia (17.3 (95% UI 11 to 25.4)). The ASR of YLD with moderate vision loss, severe vision loss and blindness due to ROP was in South Africa (2.4 per 100 000; 95% UI 1.2 to 4.2), Afghanistan (7.1 per 100 000; 95% UI 3.8 to 12.4) and Equatorial Guinea (12.1 per 100 000; 95% UI 6.2 to 19.9) in 2019, respectively (online supplemental table 8). Georgia, Greece and Ukraine showed the largest increases in the ASR of YLD from 1990 to 2019, while Cabo Verde, Cambodia and Ghana whose YLD rate decreased largely during this period (figure 2).
Figure 2. Geographic distribution of age-standardised years lived with disability per 100 000 population in nations with retinopathy of prematurity–related vision impairment from 1990 to 2019 among people younger than 20 years old.
Sex patterns
Although both male (AAPC=0.06; 95% CI 0.04 to 0.08) and female (AAPC=0.09; 95% CI 0.07 to 0.11) ASR of prevalence on blindness and vision loss were increasing from 1990 to 2019, the increasing trends of ASR of prevalence for female was slightly higher than that of male. In 2019, the ASR of prevalence of male and female were similar, with 86.4 per 100 000 (95% UI 69.9 to 105) and 86.5 per 100 000 (95% UI 70.6 to 104.2), respectively (figure 3).
Figure 3. Global age-standardised prevalence of retinopathy of prematurity–related vision impairment by sex from 1990 to 2019.
The ASR of YLD of blindness and vision loss due to ROP decreased significantly in both male and female younger than 20 years old, but the decline in male was slightly higher than that in female (AAPC −0.2; 95% CI −0.26 to −0.14 vs −0.18; 95% CI −0.22 to −0.13).
Global trends by age group
The prevalent cases in adolescents (10–14 years old) were slightly higher than any other age groups, reaching 552 000 in 2019 (figure 4). The prevalence rate of old adolescents (15–19 years old) was the highest (86.8 per 100 000 population). The number of ROP-related YLDs in older children (61 566 in 1990 to 64 131 in 2019), younger adolescents (57 206 in 1990 to 65 437 in 2019) and older adolescents (54 951 in 1990 to 64 514 in 2019) increased by 4.1%, 14.4% and 14.7% respectively, while children less than 5 years old (62 400 in 1990 to 59 768 in 2019) decreased by 4.2% (figure 5).
Figure 4. Global retinopathy of prematurity–related vision impairment prevalent cases by age from 1990 to 2019.
Figure 5. Global retinopathy of prematurity-related vision impairment burden by age from 1990 to 2019. YLDs, years lived with disability.
Global trends by SDI
The prevalent cases of blindness and vision loss due to ROP were increased in low SDI (91%) areas, while they were decreased in high-middle SDI (13%) areas (table 1). In 2019, the prevalent cases were about four times in high SDI (152 700; 95% UI 110.3 to 202.1, thousand) areas compared with low (608 600; 95% UI 483.1 to 742, thousand) or low-middle SDI (608 900; 95% UI 474.1 to 755.1, thousand) areas. The AAPC of ASR of prevalence in low SDI and low-middle SDI areas were decreasing, while it is increasing in high SDI, high-middle SDI and middle SDI areas (table 2). The YLD count also increased only in low SDI areas (80%), while it decreased in high-middle SDI (20%) areas (table 2). In 2019, the ASR of YLD on low SDI areas (12.7 per 100 000 person-year) was more than 1.5 times higher than that in high SDI areas (8.2 per 100 000 person-year). The joinpoint regression indicated that the ASR of YLD in all regions have been decreasing, except for high SDI (AAPC 0.14; 95% CI 0.1 to 0.18).
Discussion
In the present study, we found that the global trends of prevalence on blindness and vision loss due to ROP among children and adolescents were increasing, whereas the disease burden was decreased in the previous years. The ROP-related age-standardised prevalence and YLD rates declined in all GBD regions but even increased in high-income North America. Some subdevelopment countries such as Afghanistan have the higher burden of blindness and vision loss due to ROP. Female showed higher increasing trends on prevalence than male, while they have lower downward trends of YLDs than males. Old adolescents have higher disease burden than other age groups.
A recent study using GBD datasets indicated that approximately 15.2 million preterm babies (9.4%) complicated with ROP-related moderate/severe vision loss or blindness.5 However, these findings were originated for all age groups combined, and there is no comprehensive report on the burden of blindness due to ROP among children and adolescents. Given ROP is the main cause of visual impairments for children and adolescents in many regions worldwide, thus, it is very important to investigate the disease burden among those subjects. Generally, we found that the age-standardised prevalence rates have steadily increased during the past 30 years. These may be related to the rapid advancement of neonatal care that improves preterm infants’ survival.6 7 Notably, we found that the disease burden declined due to advancement of ROP screening programme and early interventions, particularly in recent years.5 Nevertheless, we should recognise that the advanced screening techniques and treatments on ROP are still unevenly distributed.8,11 For instance, the major prevalent and YLD rates of blindness and vision loss were in developing regions and countries, while their higher increasing trends was observed in developed regions. These findings highlight that early diagnosis and timely treatments are urgently needed in the world.12 Specifically, the trends of global prevalence of ROP caused moderate vision loss and severe vision loss were increasing, while the prevalence of ROP caused blindness showed decreasing trend. Therefore, although the blindness caused by ROP has been alleviated to some extent, moderate to severe visual impairment is still a concern, which reflects the significance of early treatment for ROP from another perspective.
Low and middle SDI locations experienced a considerable burden of blindness and vision loss due to ROP, and main burdens were reported in Africa and Asia. The highest prevalence and YLD for ROP-related moderate and severe blindness have been observed in South Africa and Afghanistan in 2019. Additionally, the disease burden also showed an upward trend in low SDI regions. However, it is noteworthy that the prevalence rate of ROP-related blindness and vision loss has still increased in high SDI regions and the same condition has also occurred in high-income North America.
In the present study, the ASR of prevalence on blindness and vision loss due to ROP in both male and female were increasing from 1990 to 2019, the ASR of prevalence in female was slightly higher than that in male. In a previous cross-sectional analysis of hospitalised children in the USA from 2009 to 2018, it was found that the incidence of ROP was higher in female than in male.1 However, other studies have found that the prevalence of ROP is not different between male and female.13,15 Ludwig et al used the National Patient Sample database provided by the Healthcare Research and Quality Bureau to find that women account for approximately 52% of cases of ROP.16 Although the prevalence of ROP between genders is still controversial, there has been no report on the gender differences in visual impairment caused by ROP.
The strength of the current study includes comprehensively estimating the global, regional and national burden of blindness and vision loss due to ROP by age and sex level among children and adolescents. However, this study also has several limitations. First, the quality of the available data is still limited. We found that some countries have no reports on the date regarding ROP, although GBD study groups have estimated the data using Bayesian meta-regression tool. Second, the burden of vision loss due to ROP may be underestimated due to using only the diagnostic criteria by the ICD code for ROP.
Conclusion
This study demonstrates the global, regional and national prevalence and burden changes of ROP-related blindness and vision loss during the past 30 years, stratified by age and gender. The global trends of prevalence on blindness and vision loss due to ROP among children and adolescents were increasing, whereas the disease burden was decreased in the previous years. Therefore, early screening programmes and advanced treatments are effective in preventing ROP-related blindness and vision loss and are still needed in regions with high prevalence and disease burden.
Supplementary material
Acknowledgements
The estimates used in this article are based on the GBD data and methodologies. We appreciate the visionary global health leadership of the Institute for Health Metrics and Evaluation (IHME) in Seattle, Washington, and the contribution of all anonymous collaborators, without whom this report would not be possible.
Footnotes
Funding: This study was supported by the General Project of Natural Science Foundation of Shandong Province (ZR2021MH411), GDPH Supporting Fund for Talent Program (KY0120220263) and Funding by Science and Technology Projects in Guangzhou (202102080291).
Provenance and peer review: Not commissioned; externally peer reviewed.
Patient consent for publication: Not applicable.
Ethics approval: This study was approved by the Ethics Committee on Biomedical Research of Guangdong Provincial People's Hospital; ID: KY-Q-2022-495-01. Participants gave informed consent to participate in the study before taking part.
Map disclaimer: The inclusion of any map (including the depiction of any boundaries therein), or of any geographic or locational reference, does not imply the expression of any opinion whatsoever on the part of BMJ concerning the legal status of any country, territory, jurisdiction or area or of its authorities. Any such expression remains solely that of the relevant source and is not endorsed by BMJ. Maps are provided without any warranty of any kind, either express or implied.
Patient and public involvement: Patients and/or the public were not involved in the design, or conduct, or reporting or dissemination plans of this research.
Data availability free text: Publicly available datasets were analysed in this study. These data can be found here: http://ghdx.healthdata.org/gbd-results-tool.
References
- 1.Thangamathesvaran L, Wang J, Repka MX, et al. Trends in retinopathy of prematurity care in the United States 2009-2018: a nationwide analysis using national inpatient sample. Ophthalmol Retina. 2023;7:360–6. doi: 10.1016/j.oret.2022.10.010. [DOI] [PubMed] [Google Scholar]
- 2.Sabri K, Manktelow B, Anwar S, et al. Ethnic variations in the incidence and outcome of severe retinopathy of prematurity. Can J Ophthalmol. 2007;42:727–30. doi: 10.3129/i07-136. [DOI] [PubMed] [Google Scholar]
- 3.Venook AP, Shuman MA, Corash L. Prophylactic heparin in APL. Blood. 1987;70:886–7. [PubMed] [Google Scholar]
- 4.Vos T, Lim SS, Abbafati C. Global burden of 369 diseases and injuries in 204 countries and territories, 1990-2019: a systematic analysis for the global burden of disease study 2019. Lancet. 2020;396:1204–22. doi: 10.1016/S0140-6736(20)30925-9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Zhang R-H, Liu Y-M, Dong L, et al. Prevalence, years lived with disability, and time trends for 16 causes of blindness and vision impairment: findings highlight retinopathy of prematurity. Front Pediatr. 2022;10:735335. doi: 10.3389/fped.2022.735335. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Stoll BJ, Hansen NI, Bell EF, et al. Trends in care practices, morbidity, and mortality of extremely preterm neonates, 1993-2012. JAMA. 2015;314:1039–51. doi: 10.1001/jama.2015.10244. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Fierson WM, et al. Screening examination of premature infants for retinopathy of prematurity. Pediatrics. 2018;142:e20183061. doi: 10.1542/peds.2018-3061. [DOI] [PubMed] [Google Scholar]
- 8.Athikarisamy SE, Lam GC. Screening and surveillance for retinopathy of prematurity: a Wilson and Jungner framework approach. J Glob Health. 2023;13:03028. doi: 10.7189/jogh.13.03028. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Desurmont M-G, Bremond-Gignac D, Torchin H, et al. Retinopathy of prematurity detection: a retrospective quality improvement project before-after implementation of retinal digital imaging for screening. Eur J Pediatr. 2023;182:3093–9. doi: 10.1007/s00431-023-04951-z. [DOI] [PubMed] [Google Scholar]
- 10.deCampos-Stairiker MA, Coyner AS, Gupta A, et al. Epidemiologic evaluation of retinopathy of prematurity severity in a large telemedicine program in India using artificial intelligence. Ophthalmology. 2023;130:837–43. doi: 10.1016/j.ophtha.2023.03.026. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Wilkinson AR, Adams GGW, Fleck BW, et al. UK screening and treatment of retinopathy of prematurity updated 2022 guidelines. Early Hum Dev. 2023;177–178:105715. doi: 10.1016/j.earlhumdev.2023.105715. [DOI] [PubMed] [Google Scholar]
- 12.Bujoreanu Bezman L, Tiutiuca C, Totolici G, et al. Latest trends in retinopathy of prematurity: research on risk factors, diagnostic methods and therapies. Int J Gen Med. 2023;16:937–49. doi: 10.2147/IJGM.S401122. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Lad EM, Hernandez-Boussard T, Morton JM, et al. Incidence of retinopathy of prematurity in the United States: 1997 through 2005. Am J Ophthalmol. 2009;148:451–8. doi: 10.1016/j.ajo.2009.04.018. [DOI] [PubMed] [Google Scholar]
- 14.Ahmed MAT, Duncan M, Kent A, et al. Incidence of retinopathy of prematurity requiring treatment in infants born greater than 30 weeks' gestation and with a birthweight greater than 1250 G from 1998 to 2002: a regional study. J Paediatr Child Health. 2006;42:337–40. doi: 10.1111/j.1440-1754.2006.00868.x. [DOI] [PubMed] [Google Scholar]
- 15.Skrehot HC, Bhatnagar A, Weng CY. Epidemiology of retinopathy of prematurity in the US from 2003 to 2019-reply. JAMA Ophthalmol. 2023;141:1090. doi: 10.1001/jamaophthalmol.2023.4616. [DOI] [PubMed] [Google Scholar]
- 16.Ludwig CA, Chen TA, Hernandez-Boussard T, et al. The epidemiology of retinopathy of prematurity in the United States. Ophthalmic Surg Lasers Imaging Retina. 2017;48:553–62. doi: 10.3928/23258160-20170630-06. [DOI] [PubMed] [Google Scholar]