Risk of Retinal Detachment After Pediatric Cataract Surgery

Abstract

Purpose.: To determine the long-term risk of retinal detachment following pediatric cataract surgery and to identify risk factors for retinal detachment.

Methods.: We included all children (aged 0 to 17 years) who during the time period of 1977 to 2005 underwent pediatric cataract surgery in Denmark, excluding cataract cases caused by trauma, or acquired systemic or acquired ocular pathology, and cases with ocular anomalies associated with the development of retinal detachment. Cases of cataract were ascertained from the mandatory Danish National Patient Register, and information on retinal detachment was based on medical chart review.

Results.: Among 1043 eyes of 656 children undergoing surgery for pediatric cataract, 25 eyes (23 children) developed retinal detachment at a median time of 9.1 years after surgery. The overall 20-year risk of retinal detachment was 7% (95% confidence interval [CI]: 3%–11%) among cataract patients. In otherwise normal children having isolated cataract, the risk was 3% (95% CI: 0%–7%). A significantly higher risk of developing retinal detachment was found in children with mental retardation (23% [95% CI: 9%–35%]) or in cataract cases with other ocular or systemic anomalies (16% [95% CI: 6%–24%]).

Conclusions.: The estimated overall risk of retinal detachment 20 years after pediatric cataract surgery was 7%, but only 3% for isolated cataract. Particularly high risks of retinal detachment after cataract surgery were associated with mental retardation and having other ocular or systemic diseases.

Introduction

Among adults, it is well-known that the risk of retinal detachment after cataract surgery is higher compared with eyes not undergoing cataract surgery. Thus, studies have reported a 7.5 to 8.7 times higher risk 6 or more years after cataract surgery. ^1,2 The risk of retinal detachment in adults remains higher even 20 years after cataract surgery. ³  

Among adults without previous cataract surgery, the overall risk of retinal detachment is estimated to be 0.03% per person year. ¹ In comparison, the occurrence of retinal detachment among children is very rare. Based on data from the Danish National Patient Register, retinal detachment occurs in approximately 0.001% of all children aged 0 to 17 years (la Cour, unpublished observations, 2008). 

In children, it is recognized that retinal detachment is a late complication to cataract surgery. The condition has been reported to occur on average 23 to 33 years after the primary surgery, ^4–6 with around one-third occurring within the first 10 years. ^4,7 Previous studies on the risk retinal detachment after pediatric cataract surgery have short-term follow-up times (3.5 to 6.8 years) and are based merely on aphakic eyes. ^8–10 Accordingly, there is a lack of large, long-term studies on the risk of retinal detachment after surgery for pediatric cataract including both pseudophakic and aphakic eyes. It is also unknown if different surgical procedures for pediatric cataract have an influence on the risk of later detachment. Before 1990 in Denmark, the most common cataract surgery was lensectomy without posterior capsulorhexis and anterior vitrectomy. After 1990, many surgeons changed surgical technique and included posterior capsulorhexis and anterior vitrectomy as standard procedures in small children undergoing surgery for pediatric cataract. 

Based on the long existing Danish national registries, we therefore established a population-based cohort of children who had undergone surgery for pediatric cataract during the period of 1977 to 2005. The aim was to evaluate the long-term risk of retinal detachment following pediatric cataract surgery and to look for risk factors for developing retinal detachment in these patients. 

Methods

Study Cohort

Since 1977, all discharge diagnoses and surgical procedures of patients admitted to public hospitals have been registered in the Danish National Patient Register (NPR). Since 1995 also outpatients have been registered. Based on the information from the NPR, we established a cohort of all children, who were diagnosed with pediatric cataract during the period of 1977 to 2001 before the age of 18 years. We validated the cataract diagnosis by subsequently contacting all relevant hospitals to retrieve and review the children's medical records. Additional case information was obtained from consultant private ophthalmologists attended by the children. We obtained the complete medical record, including all follow-up visits and consultation data. In Denmark, surgery for retinal detachment is performed only at public hospitals in both children and adults. Information about possible death and emigration was obtained by linkage with the Danish Civil Registration System. ¹¹ For this study, only cataract cases (diagnosed between 1977 and 2001) who underwent cataract surgery during the period of 1977 to 2005 were included in the study cohort. Cataract cases caused by trauma, or considered due to acquired systemic (e.g., diabetes) or acquired ocular pathology (e.g., uveitis), were excluded. ^12,13 We furthermore excluded cataract cases with ocular conditions considered to be highly associated with the development of detachment in itself or after cataract surgery (persistent fetal vasculature [PFV]; n_eyes = 43) and retinopathy of prematurity (ROP; n_eyes = 2). 

Information on retinal detachment following pediatric cataract surgery was obtained from the medical records. 

Permission to receive data from the national registries was obtained from the Danish Data Protection Agency (2000-41-0285). The study was approved by the Scientific-Ethical Committees for Copenhagen and Frederiksberg ([KF] 01-253/00), and it adhered to the tenets of the Declaration of Helsinki. 

Follow-up

All analyses were based on eyes as the unit. Time at risk was defined for each eye as the time from cataract surgery until time of diagnosis of retinal detachment, date of last visit to an ophthalmologist, emigration or death, date of relevant trauma (n_eyes = 1), or posterior vitrectomy (n_eyes = 3), whichever came first. Accordingly, eyes with retinal detachment prior to cataract surgery were not included in the analyses (n_eyes = 3). The median follow-up time since cataract surgery was 6.8 years (25th quartile, 2.6 years; 75th quartile, 12.8 years). 

Statistical Analyses

Rate ratios of retinal detachment following surgery of pediatric cataract according to various characteristics were estimated by a Cox proportional hazards regression model with time since surgery as underlying time scale using the procedure PROC PHREG in advanced analytics software (SAS; SAS Institute, Inc., Cary, NC, USA). Age at surgery was analyzed by including age in the regression model with the following categories (in months): 0 to 5, 6 to 11, 12 to 23, 24 to 47, and ≥48 months. A trend was fitted by allowing a linear effect of age at surgery (in months). The outcome for two eyes from the same person can not necessarily be considered independent. The traditional standard errors were therefore substituted by robust standard errors that allows for such a clustered data structure, ¹⁴ resulting in slightly wider confidence intervals on estimated rate ratios. The cumulative risk of retinal detachment by years since cataract surgery was estimated by Kaplan-Meier estimates using statistical software (Stata 12; StataCorp LP, College Station, TX, USA). 

Results

During the time period of 1977 to 2005, a total of 1043 eyes of 656 children had a validated diagnosis of pediatric cataract, underwent cataract surgery, and were without other known risk factors for developing retinal detachment (see Methods section). 

In total, 25 eyes (23 children) with retinal detachment were observed in the study. Median time between cataract surgery and development of retinal detachment was 9.1 years (25th quartile, 5.2; 75th quartile, 16.9 years). Two children developed bilateral retinal detachments. We found only a slight insignificant (P = 0.73) male predominance: 56% (14/25). 

In the Table, the rate ratios of developing retinal detachment according to possible risk factors are shown. Retinal detachment was significantly higher in eyes of children with mental retardation (rate ratio [RR] = 9.59 [95% CI: 3.89, 23.61], P < 0.001) and in eyes with other ocular or systemic diseases (clinical manifestation; RR = 8.52 [95% CI: 3.28, 22.15], P < 0.001), but also in children having pars plana lensectomy (RR = 4.66 [95% CI: 1.05, 2.63], P = 0.04). However, after adjusting for mental retardation and clinical manifestation, only mental retardation remained significant (RR = 5.26 [95% CI: 1.24, 22.36], P = 0.03), whereas clinical manifestation (RR = 3.39, [95% CI: 0.81, 14.15], P = 0.09) and pars plana surgery (RR = 2.34 [95% CI: 0.38, 14.43], P = 0.36) became insignificant. All other adjusted rate ratios were below 1.7 and insignificant. 

Table

 

RRs of Developing Retinal Detachment After Pediatric Cataract Surgery According to Various Potential Risk Factors

Variable	Eyes With Retinal Detachment*	Operated Eyes*	Follow-up, y	Unadjusted RR (95% CI)	P	Adjusted† RR (95% CI)	P
Type of surgery
Lensectomy	18	707	6638	1 (reference)	0.21	1 (reference)	0.39
Lensectomy, anterior vitrectomy	7	311	1894	2.00 (0.67, 5.99)		1.62 (0.54, 4.83)
Unspecified‡	0	28	220
Posterior capsule
Intact	14	558	5138	1 (reference)	0.55	1 (reference)	0.78
Not intact	11	460	3394	1.30 (0.55, 3.08)		1.14 (0.45, 2.93)
Unspecified‡	0	28	220
Surgical technique
Not pars plana	23	991	8500	1 (reference)	0.04	1 (reference)	0.36
Pars plana	2	55	252	4.66 (1.05, 20.72)		2.34 (0.38, 14.43)
Surgery for secondary cataract
None	15	555	4598	1 (reference)	0.40	1 (reference)	0.90
At least once	10	491	4154	0.68 (0.27, 1.69)		0.94 (0.36, 2.45)
Etiology
Nonhereditary	20	678	5798	1 (reference)	0.13	1 (reference)	0.77
Hereditary	5	368	2954	0.47 (0.18, 1.25)		1.20 (0.36, 3.98)
Laterality
Unilateral	2	192	1412	1 (reference)	0.34	1 (reference)	0.84
Bilateral	23	854	7340	2.04 (0.48, 8.71)		1.16 (0.27, 5.09)
Primary intraocular lens implantation
No	25	742	7394
Yes	0	304	1358
Morphology
Other	12	574	4697	1 (reference)	0.58	1 (reference)	0.51
Nuclear	13	468	4005	1.24 (0.57, 2.69)		1.30 (0.60, 2.80)
Secondary intraocular lens implantation
No	24	995	8139	1 (reference)	0.61	1 (reference)	0.79
Yes	1	51	613	0.61 (0.08, 4.42)		0.75 (0.09, 6.41)
Secondary glaucoma
No	18	957	7969	1 (reference)	0.64	1 (reference)	0.92
Yes	7	89	783	1.35 (0.39, 4.65)		1.06 (0.33, 3.42)
Surgery for secondary glaucoma
No	21	990	8328	1 (reference)	0.06	1 (reference)	0.43
Yes	4	56	424	3.13 (0.94, 10.36)		1.62 (0.49, 5.43)
Age at cataract surgery, mo
0–5	8	209	2029	1.77 (0.63, 4.94)	0.55	0.90 (0.32, 2.52)	0.99
6–11	5	103	971	2.25 (0.60, 8.42)		1.14 (0.27, 4.91)
12–23	2	74	652	1.55 (0.32, 7.50)		0.91 (0.18, 4.53)
24–47	2	140	1398	0.67 (0.14, 3.31)		0.65 (0.11, 3.96)
≥48	8	520	3702	1 (reference)		1 (reference)
Trend				1.00 (0.99, 1.01)	0.80	1.01 (1.00, 1.01)	0.23
Surgeon experience
≥50 surgeries	4	192	1302	1.24 (0.41, 3.73)	0.71	0.78 (0.22, 2.84)	0.71
<50 surgeries	20	812	6920	1 (reference)		1 (reference)
Missing	1	42	530
Clinical manifestation
Isolated cataract	6	807	6456	1 (reference)	<0.0001	1 (reference)	0.09
Cataract plus§	19	239	2296	8.52 (3.28, 22.15)		3.39 (0.81, 14.15)
Mental retardation
Yes	16	148	1351	9.59 (3.89, 23.61)	<0.0001	3.94 (1.02, 15.18)	0.04
No	8	851	7034	1 (reference)		1 (reference)
Missing	1	47

*  For time-dependent variables, the exposed is counted at any time during follow-up.

†  Adjusted for mental retardation and clinical manifestation. Mental retardation was chosen by forward inclusion. However, as clinical manifestation had an essential impact on the RR for mental retardation, we also chose to adjust for clinical manifestation.

‡  Excluded from analyses.

§  Among the 25 eyes with retinal detachment, nine eyes had another ocular anomaly apart from cataract: four had microcornea, four aniridia, and one had tapetoretinal dystrophy. Two eyes of two children had rubella and four eyes with retinal detachment were in two children with Down syndrome. Mental retardation was seen in 16 cases (eyes). Six eyes (isolated cataract) were not associated with other ocular or systemic anomalies. Among the cataract cases who did not develop retinal detachment, microcornea and aniridia were the most common other ocular anomaly comprising around one third of all “cataract plus” cases, that is the same percentage as among the retinal detachment cases.

Due to the higher correlation between mental retardation and clinical manifestation, we investigated their interaction further. No statistically significant interaction was observed (P = 0.43). However, in children without mental retardation, eyes of children having other ocular or systemic diseases had a significantly higher rate of retinal detachment (RR = 4.39 [1.05–18.33], P = 0.04); and in children whose cataract was associated with other ocular or systemic disease, having mental retardation was associated with a significantly higher rate of retinal detachment (RR = 8.12 [1.23–53.48], P = 0.03). RR in eyes in children having both mental retardation and other ocular or systemic diseases compared with children having neither was 14.12 (4.80–1.4), P < 0.001. Due to these additional results, we emphasize in the following both the association with mental retardation and clinical manifestation. 

Age at surgery did not appear to be associated with risk of retinal detachment. Among children (304 eyes) who received a primary intraocular lens implantation, no cases developed retinal detachment. 

We also compared surgeons within the present study who had performed more than 50 pediatric cataract surgeries with those having performed less. There was no difference between the two groups on the risk of retinal detachment (RR = 1.24 [95% CI: 1.05, 20.72] P = 0.71). 

There was no difference in the risk estimates for retinal detachment after cataract surgery performed before 1990 and after 1990 (RR = 1.16 [95% CI: 0.47, 2.89] P = 0.75), when the majority of eyes had primary surgery with posterior capsulorhexis and anterior vitrectomy performed. 

Figure (A) shows the cumulative risk of retinal detachment by years since cataract surgery. The overall risk of retinal detachment 20 years after pediatric cataract surgery remained 7% (95% CI: 3%, 11%). Including cases with PFV and ROP, the risk was 7% (95% CI: 4%, 11%). 

Figure

 

(A–C) Cumulative risk of retinal detachment by years since cataract surgery overall and by clinical manifestation and mental retardation estimated using the Kaplan-Meier estimator. The rate ratios are estimated in univariate Cox models.

Figure (B) shows the cumulative risk of retinal detachment by years since cataract surgery by clinical manifestation. In patients with other ocular or systemic diseases apart from pediatric cataract, the 20-year risk was 16% (95% CI: 6%, 24%) compared with 3% (95% CI: 0%, 7%) in isolated cataract. Figure (C) shows the cumulative risk of retinal detachment by years since cataract surgery by mental retardation. The 20-year risk was 23% (95% CI: 9%, 35%) in patients with mental retardation compared with 3% (95% CI: 0%, 6%) in patients without mental retardation. 

In 52% (13/25) of the retinal detachment eyes, surgery was not performed because it was considered pointless due to fibrosis. Eight of these eyes had no light perception, two had light perception, one had Snellen visual acuity (VA) of 0.03, one had 0.1, and in one eye, no VA was available. 

Among the eyes who underwent retinal detachment surgery (12 eyes), three eyes went blind with no light perception, one had hand movements, five eyes had VA < 0.1, in one eye VA was 0.5, and in another VA < 0.8. In one eye, no VA was available. 

Discussion

Retinal detachment after pediatric cataract surgery is recognized as a postoperative complication that for the most part occurs more than 10 years after the primary surgery. ^4–7 Previous studies reporting an incidence of retinal detachment after pediatric cataract surgery are based on merely aphakic cases and short follow-up times (mean 3.5–6.8 years), with the incidence reported to be 1% to 3.2%. ^8–10 In accordance, the risk of retinal detachment 5 years after surgery in our study was 0.8%. 

Two studies found an increased risk of complications, including retinal detachment, after pediatric cataract surgery when the children had other anomalies (ocular ⁸ and/or systemic ^8,9 ). This is in accordance with our results where mental retardation and/or having other ocular/systemic diseases were associated with an increased risk of retinal detachment. Having other ocular diseases may influence the anatomical structure of the eye in general, which in association with cataract surgery may lead to an increased risk of retinal detachment. Even though we excluded cases with reported relevant trauma to the operated eyes we cannot exclude that eye rubbing or unknown eye trauma of children who suffer from other systemic diseases or who are mentally retarded may have occurred. Irrespective of the cause, these findings stress the need for regular eye examinations of children having undergone pediatric cataract surgery. 

We observed only a slightly insignificant male predominance (56% of cases) among those with retinal detachment. A male predominance has previously been reported in some studies of retinal detachment after congenital cataract surgery. ^4–7,10,15 Among adults, the risk of retinal detachment is higher among males, whether occurring after cataract surgery ² or occurring without previous cataract surgery. ¹⁶  

In pediatric cataract surgery, postoperative complications such as glaucoma and secondary cataract have been shown to be more common in children who have had cataract surgery performed at an early age. ^9,17–19 In one study, the occurrence of retinal detachment was found to be higher among children undergoing cataract surgery that were aged 1 year or older ¹⁷ ; our study did not support that age of surgery affects the risk of retinal detachment. 

We did not find that the risk of retinal detachment varied by surgical technique. In adult cataract surgery, it is shown that a posterior capsule complication during cataract surgery increases the risk of retinal detachment significantly. ²⁰ It is therefore reasonable to speculate that the performance of a posterior capsulotomy and anterior vitrectomy in children would be associated with a higher risk of retinal detachment. Before 1990 in Denmark, the most common primary pediatric cataract procedure was lensectomy without posterior capsulorhexis and anterior vitrectomy. After 1990, many surgeons changed surgical technique and included posterior capsulorhexis and anterior vitrectomy as standard procedures in small children in order to reduce the risk of visual axis opacities. The posterior capsule in young children does not stay clear if not primarily removed, and secondary removal of the posterior capsule will lead to an open capsule. In this study, the comparison between different surgical techniques and between primarily leaving an intact posterior capsule or not showed no difference in the risk of retinal detachment. Furthermore, there was no difference in the risk of retinal detachment for cataract surgery before or after 1990. In a similar study based only on aphakic eyes, there was also no association between posterior capsulotomy/anterior vitrectomy and retinal detachment. ¹⁰  

We did not find any retinal detachment cases among the children who had an intraocular lens implanted. Primary intraocular lens implantation is a more modern surgery usually performed in older children with otherwise normal eyes and accordingly, these eyes have less follow-up, are older, and do not have other risk factors for developing retinal detachment. Therefore, the seemingly protective effect of primary lens implantation is not likely. 

One limitation of this study is the low number of retinal detachment cases. In addition, it would have been interesting to compare the axial lengths of retinal detachment eyes with eyes without retinal detachment since axial length in adults is shown to be associated with a higher risk of retinal detachment. ²¹ This parameter was unfortunately not available postoperatively in many cases. We did not compare postoperative refraction due to the inclusion of both aphakic and pseudophakic eyes and the difficulty in comparing these according to—for example—levels of myopia, which in adults is known to increase the risk of retinal detachment. Finally, it should be noted that due to the design, the 20-year risk estimates are based on older and newly diagnosed cases in varying degrees. Thus, the evaluation of the risk 10 to 20 years after operation is mostly based on children diagnosed in the early calendar years of the study, whereas evaluation of the risk in the first years after operation is based on cases diagnosed in both early and more recent calendar years. 

Our study had several strengths. The population-based design made it possible to identify all children who underwent surgery for pediatric cataract in Denmark from 1977 to 2005. Due to a long tradition of registration in Denmark, we were able to follow the cases for an expanded period, for the first time providing a long-term risk estimation of the development of retinal detachment after pediatric cataract surgery based on an entire nation. 

To conclude, retinal detachment is estimated to occur in 3% of children within the first 20 years after surgery for isolated pediatric cataract. In mentally retarded children, this figure rose to nearly one out of four, representing the single most important risk factor for retinal detachment in this study. However, children with other ocular and/or systemic disease were also found to have a substantially increased risk of retinal detachment after cataract surgery. 

Acknowledgments

The authors alone are responsible for the content and writing of the paper. 

Disclosure: B. Haargaard, None; E.W. Andersen, None; A. Oudin, None; G. Poulsen, None; J. Wohlfahrt, None; M. la Cour, None; M. Melbye, None