Glaucoma Risk In Children Hinges On Steroid Therapy
Overview
The use of glucocorticoids after pediatric cataract surgery is essential for preventing inflammation, but it poses risks such as secondary glaucoma and suppression of the hypothalamic-pituitary-adrenal axis. This study aimed to evaluate glaucoma outcomes in children receiving either high-dose or low-dose glucocorticoid therapy post-surgery.
The cohort study included Danish children under 10 who had cataract surgery, and were treated with either high or low doses of glucocorticoids. Data was retrospectively collected from 2010-2016 and prospectively until 2021. High-dose treatment involved 0.5–1.0 mg of depot dexamethasone or methylprednisolone, followed by a regimen of 6-8 dexamethasone drops tapered weekly. Low-dose treatment consisted of 6 drops for 3 days, then 3 drops for 18 days. The incidence of sustained ocular hypertension or glaucoma (lasting >3 months) was compared between the groups.
A total of 267 children (388 eyes) were included. The high-dose group, comprising 95 children (133 eyes), had a median follow-up of 89 months, while the low-dose group, with 173 children (255 eyes), had a median follow-up of 40.5 months. Children with axial lengths of 18 mm or greater in the low-dose group exhibited a lower risk of developing glaucoma, as shown by survival curves.
Low-dose glucocorticoid treatment reduced the risk of glaucoma in children with axial lengths of 18 mm or more. This protective effect was not observed in children with shorter eyes, suggesting that high-dose glucocorticoids should be used cautiously in pediatric cataract surgery.
Introduction
Effective control of postoperative inflammation is critical in pediatric cataract surgery, as inadequate management can result in complications such as synechiae, pupillary membranes, and pupil-block glaucoma. Glucocorticoid eye drops are commonly used to manage inflammation, but their use carries the risk of increased intraocular pressure (IOP), potentially leading to sight-threatening secondary glaucoma. Research indicates that 25% to 75% of children may be “steroid responders,” meaning their IOP rises above 21 mmHg or increases by 5–10 mmHg from baseline following glucocorticoid treatment.
Additionally, prolonged use of glucocorticoids can cause suppression of the hypothalamic-pituitary-adrenal (HPA) axis, which can lead to a life-threatening Addisonian crisis. Given these risks, achieving a balance between controlling inflammation and avoiding glucocorticoid-induced ocular hypertension (OHT) is essential. Currently, there is no established consensus on the optimal postoperative treatment regimen following pediatric cataract surgery. However, evidence suggests that reducing the dose and frequency of glucocorticoids may lower the incidence of OHT and glaucoma in children.
In Denmark, since 2010, pediatric cataract surgeries have been centralized at two specialized centers: Rigshospitalet (RH) and Aarhus University Hospital (AUH). To further understand the relationship between glucocorticoid dosing and the development of secondary glaucoma, we conducted a multi-site cohort study examining postoperative outcomes in pediatric cataract patients.
Method
This study analyzed all children who underwent surgery for congenital, infantile, or developmental cataract at two specialized pediatric cataract centers in Denmark—Rigshospitalet (RH) and Aarhus University Hospital (AUH)—between January 1, 2010, and December 31, 2021. The inclusion criteria focused on children who had cataract surgery before the age of 10, while those who had secondary cataracts due to trauma, uveitis, steroid treatment, infections, cancer therapy, or were diagnosed with ocular hypertension (OHT) or glaucoma prior to surgery were excluded. Cases were identified using hospital administrative records.
Data collection involved standardized forms capturing information on the cause of cataract, ocular and systemic conditions, age at diagnosis and surgery, surgical approach, and glucocorticoid use. The review of medical records was retrospective for surgeries performed between 2010 and 2016, and prospective for surgeries from 2017 to 2021. Ethical approval was granted by relevant authorities, and no additional patient examinations were required per Danish research ethics guidelines.
Cataract surgeries were performed by specialized surgeons using either anterior or posterior approaches. In the anterior approach, lens aspiration and vitrectomy were performed, with posterior rhexis omitted in older children (≥7 years) who could later undergo Nd:YAG laser treatment if necessary. The posterior approach involved pars plana incision and vitrectomy. At AUH, intraocular lens (IOL) implantation was initially restricted to children over 2 years, but from 2017, it included those over 6 months. At RH, IOL was implanted in children older than 6 months throughout the study. Postoperative follow-up visits were scheduled frequently in the first year and annually thereafter until the age of 10.
Glucocorticoid treatment post-surgery was classified into high-dose and low-dose regimens. The high-dose treatment involved subconjunctival glucocorticoid (methylprednisolone or dexamethasone) and frequent topical dexamethasone drops, tapering over several weeks. In contrast, the low-dose treatment mainly consisted of topical dexamethasone drops with lower total glucocorticoid exposure. The low-dose regimen was standard at AUH for the entire study period and became the norm at RH after 2017.
The study’s primary outcome was the incidence of glaucoma or OHT after three months post-surgery, while the secondary outcome was the occurrence of visual axis opacity (VAO). Glaucoma was diagnosed based on criteria from the World Glaucoma Association, which included factors like intraocular pressure exceeding 21 mmHg, optic nerve head changes, and other ocular abnormalities. OHT was defined as intraocular pressure above 21 mmHg.
Statistical Analysis
To assess the incidence of glaucoma between the low-dose and high-dose groups, Average Treatment Effect (ATE) survival curves (Díaz et al., 2019) were applied, adjusting for axial length (AL) and intraocular lens (IOL) implantation as confounders. The analysis included 95% confidence intervals calculated for paired data (eyes). Age at surgery, which correlated highly with AL and IOL implantation, was evaluated separately. However, it was found to have no significant impact on the treatment outcome, so the analysis was focused solely on AL and IOL implantation. The Aalen’s additive hazards model (Martinussen & Scheike, 2006) was employed to evaluate the treatment effect in a multivariate regression context, using the same confounders as in the ATE analysis. Both ATE survival curves and multivariable regression analysis yielded similar conclusions regarding the treatment effect. Robust standard errors had minimal influence on the results and were thus not used in further analyses.
The axial length cut-off value was determined by visually inspecting the scatter plot in highlighted in the study, selecting the point that best differentiated between eyes with and without glaucoma or ocular hypertension (OHT) at the end of follow-up. Kaplan–Meier survival curves were utilized to estimate the treatment effect on the incidence of treatment for visual acuity outcomes (VAO). Due to the partially retrospective nature of the study, any missing data from incomplete registration were excluded from the analysis. The normality of data was verified using Q–Q plots, and variance was assessed using the Fligner-Killeen test. For non-normal distributions but equal variances, the Mann-Whitney U-test was used for group comparisons. The statistical analysis was conducted using R software, version 1.4.1717 (R Core Team, 2020).
Result
This study included 267 children (120 females), with a total of 388 eyes undergoing pediatric cataract surgery. The median follow-up period for the cohort was 51.9 months (IQR: 29.3–85.9). A genetic cause of cataract was identified in 43 children (16%). Additionally, 24 children (11%) had known syndromes or conditions, such as Down syndrome (5 cases), Nance Horan syndrome (3), and Lowe syndrome (1), among others.
Ocular malformations were observed in 43 eyes, with specific conditions like posterior embryotoxon (5 eyes), iris coloboma (5 eyes), and microphthalmia (16 eyes) being noted. Surgery was performed on children younger than 6 months in 42% of cases at one hospital and 30% at another. The median follow-up was similar at both centers. The study also explored axial length (AL) at the time of surgery, noting that children with AL <18 mm had a higher incidence of ocular malformations compared to those with AL ≥18 mm.
At the end of the follow-up, 12.5% of eyes with microphthalmia developed glaucoma or ocular hypertension (OHT). A significant number of children received intraocular lenses (IOLs), particularly in the high-dose treatment group. Glaucoma or OHT was diagnosed in 21 children (34 eyes), with the majority being open-angle glaucoma. Some of the children had additional risk factors, such as ocular malformations or syndromic associations, while others had no known abnormalities apart from the cataract.
Within two years after cataract surgery, the risk of developing glaucoma or OHT was 10% in the high-dose group and 1% in the low-dose group. However, no significant difference was observed between the two groups after six years. A total of 57 children (69 eyes) required treatment for visual axis opacification (VAO), either through Nd:YAG laser or vitrectomy, with the latter being more commonly performed on younger children.
The study concluded that children with AL ≥18 mm at the time of surgery had a lower risk of glaucoma or OHT when treated with the low-dose regimen. In contrast, children with AL <18 mm did not show the same benefit. Additionally, those younger than 6 months at the time of surgery had an increased risk of glaucoma or OHT, but age did not significantly interact with treatment in the adjusted analysis.
Conclusion
In our initial analysis of the incidence of glaucoma or ocular hypertension (OHT), there was no observed difference between low-dose and high-dose glucocorticoid treatments over six years. However, after stratifying by axial length (AL) below or above 18 mm and adjusting for intraocular lens (IOL) implantation, a significant difference emerged. Specifically, after five years, eyes with an AL ≥18 mm receiving low-dose treatment had a reduced risk of secondary glaucoma or OHT compared to those on high-dose treatment. In contrast, no dose-dependent effect was observed in eyes with AL <18 mm. The lack of a significant difference in glaucoma risk at the six-year mark between the treatment groups was attributed to eyes with AL <18 mm.
The increased incidence of glaucoma or OHT in children with AL <18 mm may be linked to younger age at surgery, more frequent ocular anomalies, and immature eye structures, potentially affecting the trabecular meshwork, which regulates aqueous outflow. Alternatively, smaller eyes or those with malformations may be more susceptible to microtubular changes induced by dexamethasone, independent of the glucocorticoid dose. However, detecting any clear differences in glaucoma risk between the treatment regimens may have been hindered by the relatively small sample size. Children who underwent surgery before six months of age had a higher glaucoma or OHT risk regardless of the treatment regime, likely due to their stronger inflammatory response, which may necessitate higher glucocorticoid doses.
Other studies have similarly identified a correlation between younger age at surgery and glaucoma risk. For instance, the IOLunder2 study found that increasing age at surgery and larger ALs lowered glaucoma risk, aligning with our findings. Furthermore, a substantial proportion of children in the IOLunder2 study received intensive postoperative glucocorticoid treatment, akin to the high-dose regimen in our study.
While previous research suggested a higher glaucoma risk in children with aphakia, this was not corroborated by the 10-year results of the Infant Aphakia Study or the five-year results from the IOLunder2 study. Notably, children in Denmark receiving IOL implantation after six months of age showed a correlation between age at surgery and IOL implantation.
We did not observe an increased long-term risk of treatment for visual axis opacification (VAO) in the low-dose group, indicating that the low glucocorticoid dose was sufficient to control inflammation. However, there was a higher risk of VAO treatment in the low-dose group within the first five years, which could lead to a greater likelihood of requiring vitrectomies under anesthesia, as younger children often cannot cooperate with Nd:YAG laser membranectomy.
This study’s limitations include its non-interventional design, which may have introduced selection bias and unknown confounders, potentially affecting the validity of the results. The wide range of ages at surgery, varying from a few weeks to school age, also contributed to the variability in glaucoma risk across the cohort. Additionally, the surgical techniques employed may have differed over the study period, as multiple surgeons performed the procedures. Particularly, the posterior technique used in the low-dose group before 2017 at AUH differed significantly. Moreover, the high-dose regimen was only implemented at RH until 2017, based on surgeon preferences, but was distinctly higher than the low-dose regimen, allowing for clear categorization. Finally, we were unable to verify the adherence to prescribed eye drop regimens.
Despite these limitations, the study’s strength lies in its population-based approach, encompassing every child who underwent cataract surgery in Denmark over a 12-year period. Further research, ideally in the form of a randomized controlled trial with a larger cohort of children with AL <18 mm, is needed to confirm these findings.
In conclusion, this study reports a reduced risk of glaucoma five years after cataract surgery in children with AL ≥18 mm who received low-dose glucocorticoid treatment compared to high-dose treatment. This effect was not observed in children with shorter axial lengths. Additionally, there was no significant difference between the two treatment groups in long-term risks for VAO or pupillary membranes.
Oncology Related Tools
Other
Latest Research
