Purpose
To identify the possibility and characteristics of late increase in the cumulative risk of retinal detachment (RD) after cataract extraction and intraocular lens (IOL) implantation.
Design
Prospective cohort study of submitted claims records.
Methods
Records of 9388 consecutive insurance beneficiaries who underwent cataract extraction and IOL implantation between August 1, 1999 and December 1, 2001 were collected from the Bureau of National Health Insurance. Data included each patient’s demographic characteristics, medical history, refractive status, axial length (AL), type of cataract extraction, and intraoperative complications. Posterior capsulotomy, diagnostic procedures, and treatment for retinal complications and other ocular diseases were identified by codes from physicians’ billing records yearly to the end of 2007.
Results
The cumulative 8-year RD rate was 2.31% at the end of the follow-up. Factors with significant effect on the risk of RD included being male and younger, having an increased AL, and having a history of RD in the fellow eye. The cumulative risk of RD after cataract extraction and IOL implantation increased with time. There was a significantly late increased risk of RD after 4 years in the group of patients with high myopia (AL, > 26 mm; P = .026), but not in the group of moderate myopia (AL 23 to 26 mm, P = .068) or nonmyopia (AL, < 23 mm; P = .066). The late increase appeared in male group, but not in the female group, even in those with high myopia (male, P = .042; female, P = .068).
Conclusions
Our study showed that patients with high myopia may be at increased risk for late pseudophakic RD after 4 years, especially in males.
The myopic eye generally is considered vulnerable to a range of ocular pathologic features, such as glaucoma, cataract, retinal detachment (RD), and macular degeneration, especially when the level is more than 6 diopters. Unfortunately, the prevalence of myopia in young people has increased substantially over recent decades and now is approximately 10% to 25% in the West and 60% to 80% in the East. Inasmuch as the World Health Organization reports myopia as the leading cause of visual impairment, myopia-related complications remain an important concern for ophthalmologists. Myopia is a proven significant risk factor for RD after cataract extraction. RD after cataract extraction is a serious and potentially sight-threatening complication that virtually always necessitates further surgical intervention. According to a long-term population-based study, the cumulative risk of RD after extracapsular cataract extraction (ECCE) and phacoemulsification was increased up to 20 years after surgery. Being male, being younger, having myopia, and having increased axial length (AL) were associated significantly with RD. Our previous studies also supported the significance of the above factors. Moreover, the extended follow-up study showed the hint of a late wave of increased risk. Therefore, we propose to identify the possibility and characteristics of late increase in the cumulative risk of RD after cataract extraction and intraocular lens (IOL) implantation.
Methods
We performed a prospective cohort study of insurance claims and medical charts of insurance beneficiaries who underwent ECCE (including phacoemulsification) and IOL implantation at the Sixth Branch of the Bureau of National Health Insurance of Taiwan between August 1, 1999 and December 1, 2001. Each patient was identified with a coded number in the Bureau of National Health Insurance databank (to which we did not have access) and no invasive procedures were involved. Data from these charts, including demographic characteristics; medical history; refractive status; AL; type of cataract extraction; and complications of capsular rupture, vitreous loss, or both were reviewed and recorded. Codes listed in physicians’ billing records regarding posterior capsulotomies performed with a neodymium:yttrium–aluminum–garnet (Nd:YAG) laser, diagnostic procedures, and treatments for retinal complications and other ocular diseases were identified and recorded from the date of operation to the end of 2007. If two cataract surgeries were identified for one patient during the study, the first procedure was chosen as the index event.
We excluded all cases of cataract surgery combined with other ocular procedures: corneal grafts, glaucoma treatments, or other posterior segment procedures. Exclusion criteria included age less than 18 years, history of ocular trauma, history of retinal breaks including lattice degeneration in the operative eye, and development of endophthalmitis. The detailed information of the excluded cases was not recorded and not analyzed. Subjects were followed up until one of the following occurred: an episode of RD was detected, any other intraocular procedure unrelated to RD repair was performed, the subject died, or the study period ended. Those who received other intraocular procedure unrelated to RD repair or who died during follow-up were assumed to change the risk for RD and were counted as being lost to follow-up.
Statistical Analysis
Average numeric values were calculated as mean ± standard deviation. The person-years of follow-up for each subject were calculated from the date of enrollment to one of the following: 1) date of diagnosis of newly developed RD, 2) date of any other intraocular procedure unrelated to RD repair, 3) date of death, or 4) the end of the study. Incidence rates were calculated by dividing the number of RD incidents by the number of person-years of follow-up. A Cox proportional hazards model was used to estimate the relative risk of RD by risk factor: gender, age, concurrent systemic diseases, medical history, AL, type of surgery, intraoperative complications, and subsequent neodymium: yttrium–aluminum–garnet (Nd:YAG) posterior capsulotomy. Significance levels were determined with use of two-tailed tests, and 95% confidence intervals for relative risks were also calculated. All statistical analyses were conducted with SPSS software version 12.0 (SPSS Inc, Chicago, Illinois, USA).
Results
A total of 9388 cases were collected, including 5235 females and 4153 males. One thousand and eighty-four cases were counted as lost to follow-up before the end of the study, including those who experienced other intraocular procedure unrelated to RD repair (n = 44) or those who died during follow-up (n = 1,040). The age of patients ranged from 18 to 99 years, with a mean of 65.96 ± 9.33 years. Mean follow-up time was 79.21 ± 15.56 months (range, 5.1 to 103.3 months). The mean age was 65.84 ± 9.55 years for males and 66.06 ± 0.15 years for females ( P = .56). Five hundred and three patients (5.37%) were 50 years of age or younger, 1935 (20.61%) were 50 to 60 years of age, and 6950 (74.03%) were older than 60 years. Eyeball AL ranged 15.28 to 34.68 mm, with a mean length of 22.85 ± 3.60 mm. An AL of less than 23 mm was found in 4445 eyes (47.34%), whereas in 4394 eyes (46.80%), the AL was 23 to 26 mm. Five hundred and forty-nine eyes (5.85%) had an AL of 26 mm or more. Refractive data were missing or unreliable (lenticular myopia) in some patients with high cataract density; we therefore decided to forgo analysis of refractive status as a risk factor. Cataract operations were performed by extracapsular lens extraction in 3550 eyes (37.81%) and by phacoemulsification in 5838 eyes (62.19%).
The cumulative 8-year RD rate in our study group was 2.31% in the total group. The mean interval between cataract extraction and diagnosis of RD was 40.6 ± 13.34 months (range, 1.0 to 82.93 months). Factors significantly affecting RD risk after cataract extraction included male gender ( P = .01), age younger than 50 years ( P = .002), AL, RD history in the fellow eye ( P = .005), and phacoemulsification ( P = .013; Table ). The greater the AL, the higher the risk of pseudophakic RD was (pairwise comparison for AL 23 to 26 mm vs AL of more than 26 mm, P = .0003). Systemic diseases such as diabetes and hypertension did not show significant correlation with an increased risk of pseudophakic RD.
| No. of Patients in Cataract | 8-year RD | Crude | Adjusted | |||
|---|---|---|---|---|---|---|
| Number (%) | Rate (%) | RR (95% CI) | P value | RR (95% CI) | P value | |
| Gender | ||||||
| Female | 5235 (55.8) | 1.52 | 1 | 1 | ||
| Male | 4153 (44.1) | 3.28 | 1.75 (1.14 to 2.66) | .01 | 1.72 (1.12 to 2.64) | .013 |
| Age (years) | ||||||
| ≦50 age | 503 (5.4) | 6.65 | 2.69 (1.44 to 5.00) | .002 | 2.35 (1.26 to 4.38) | .007 |
| 50 < age ≦ 60 | 1935 (20.6) | 2.57 | 1.15 (0.68 to 1.95) | .614 | 1.16 (0.69 to 1.95) | .567 |
| >60 age | 6950 (74) | 2.01 | 1 | 1 | ||
| DM (−) | 8132 (86.6) | 2.27 | 1 | 1 | ||
| DM (+) | 1256 (13.4) | 2.61 | 1.06 (0.87 to 1.94) | .864 | 1.39 (0.84 to 2.29) | .198 |
| Hypertension (+) | 1331 (14.2) | 2.78 | 1 | 1 | ||
| Hypertension (−) | 8057 (85.8) | 2.24 | 0.80 (0.45 to 1.41) | .434 | 1.22 (0.70 to 2.14) | .491 |
| AL (mm) | ||||||
| <23 | 4445 (47.3) | 1.57 | 1 | 1 | ||
| 23 to 25 | 4394 (46.8) | 2.44 | 1.78 (1.10 to 2.88) | .02 | 1.59 (0.97 to 2.60) | .066 |
| ≧26 | 549 (5.8) | 6.14 | 4.61 (2.50 to 8.50) | <.001 | 4.19 (2.25 to 7.80) | <.001 |
| Type | ||||||
| ECCE | 3550 (37.8) | 1.66 | 1 | |||
| PHAC | 5838 (62.2) | 2.91 | 1.78 (1.13 to 2.79) | .013 | 0.55 (0.37 to 0.81) | .003 |
| Intraoperative complication (−) | 9293 (99) | 2.32 | 1 | 1 | ||
| Intraoperative complication (+) | 95 (1) | 1.32 | 1.30 (0.18 to 9.31) | .797 | 0.84 (0.12 to 6.09) | .864 |
| Capsulotomy (+) | 688 (7.2) | 4.72 | 1.55 (0.78 to 3.09) | .214 | 1.45 (0.80 to 2.65) | .224 |
| Capsulotomy (−) | 8700 (92.8) | 2.16 | 1 | 1 | ||
| RD history (+) | 6 (0.1) | 33.33 | 17.34 (2.41 to 124.64) | .005 | 10.31 (1.40 to 75.83) | .022 |
| RD history (−) | 9382 (99.9) | 2.28 | 1 | 1 | ||
Intraoperative complications (posterior capsule rupture, vitreous loss, or both) occurred in 95 eyes (1.01%), but did not significantly affect the risk of developing RD after cataract surgery ( P = .797). Because the cost of cataract extraction was covered by case-payment policies, surgeons received no extra fees for performing vitrectomies or other related procedures, but did face a greater risk of litigation. Complications may have been underreported for this reason. Posterior capsulotomies performed with an Nd:YAG laser were carried out on 668 eyes after cataract extraction (mean, 57.67 ± 12.93 months; range, 2.77 to 97.8 months). In the full study group, patients who received an Nd:YAG laser posterior capsulotomy were no different in their risk of developing RD ( P = .214; Table ).
Cataract extractions in our group were performed more often by phacoemulsification than by traditional ECCE, especially in the second year (56.5% in the first year and 71.8% in the second year). Surgeons were transitioning from ECCE to phacoemulsification during the study period. Phacoemulsification appeared to be significantly more risky than ECCE in the study group as a whole ( P = .013). However, the difference disappeared in cases collected in the second year (first year, P = .016; second year, P = .964, Student t test), although the other risk factors (gender, age, AL) remained the same.
The cumulative risk of RD increased with time ( Figure 1 ). Moreover, the survival curves revealed that patients in different AL ranges ran differently on the change in cumulative risk of RD after cataract extraction ( Figure 2 ). There was a significantly late increased risk of RD after 4 years in the group of patients with high myopia, but not in the group of patients with moderate myopia or nonmyopia (AL > 26 mm, P = .026; AL 23 to 26 mm, P = .068; AL less than 23 mm, P = .066; Wilcoxon signed-rank test). This change appeared in the male group, but not in the female group ( Figure 3 ; male, P = .042; female, P = .068; Wilcoxon signed-rank test). In the male group, both moderate and high myopic patients showed a significantly late increased risk of RD 4 years after cataract extraction and IOL implantation (AL > 26 mm, P = .024; AL 23 to 26 mm, P = .027; Wilcoxon signed-rank test). Because young age is also an important risk factor for pseudophakic RD, we analyzed the age factor as well. Those who aged 50 years or younger showed a tendency toward late increase, but one that was not significant ( Figure 4 ; age < 50 years, P = .068; age 50 to 60 years, P = .180; age > 60 years, P = .109; Wilcoxon signed-rank test).
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
