In this case series study, five patients were female (seven eyes) and one was male (two eyes). The mean (± SD) age was 11.16 ± 7.9 years (range: 4–25 years). Slit-lamp microscope showing partial cataract morphology of each patient is presented in Fig. 1. The preoperative spherical power ranged from − 7.5 D to -42 D. The mean follow-up time was 36 (36–36) months (range: 24–48 months). The characteristics of patients’ preoperative data, follow-up time, and endpoint data for BCVA, MTFoutoff, OSI, and AXL are presented in Table 1. In case 2 and case 3, high myopia progression was shown according to the previous medical record. A family history of high myopia was also collected. For case 1, the patient’s father suffered from congenital cataract, high myopia of both eyes, and retinal detachment of the left eye. The father of case 6 also suffered from high myopia. All surgeries were uneventful and no postoperative complications occurred during the entire follow-up period.
Patients’ refractive and uncorrected visual acuity (UCVA) outcomes are presented in Table 2. IOL was selected to achieve a goal refraction of ~ + 2 D in cases 1 and 2, whose ages were 4 and 5 years old, respectively. In the remaining cases, the goal refraction was plano to -2 D. Initial spherical correction (treatment achieved) averaged 18.25 ± 11.15 D (range: 7.75-44.0 D). Five eyes (56%) were corrected to within ± 1.0 D of spherical goal refraction. Of the remaining four eyes, two were undercorrected (residual myopia of -1.25 and − 2.25 D) and two were overcorrected (0 and + 2 D). Hyperopic and myopic shift (regression) during the follow-up interval is reported in Table 2 as a regression rate, expressed as (final postop refraction - initial postop refraction)/ follow-up time. Eight eyes (0.89%) exhibited myopia shift (~-0.5 D/year except for case 3), whereas case 5 (age 25) showed hyperopic change.
Table 2
Refractive and uncorrected visual acuity outcomes by case number and operative eye of patients, organized by age at the onset
Case No.
|
Pre-op Sphere diopter (D)
|
Goal
Refraction(D)
|
Initial postop
Refraction(D)
|
Spherical correction
achieved(D)
|
Final postop
refraction(D)
|
Follow-up
(years)
|
Regression
(D/year)
|
Preop cylinder(D)
|
Initial postop cylinder (D)
|
Final postop cylinder (D)
|
Preop UDVA
|
Postop UDVA
|
1 Right
|
-9
|
1.64
|
-1.25
|
7.75
|
-4
|
4
|
-0.68
|
1.5
|
1.25
|
0.5
|
0.1
|
0.3
|
2 Left
|
-10
|
1.91
|
1
|
11
|
-1.3
|
4
|
-0.57
|
1
|
1
|
0.75
|
NA
|
0.5
|
3 Right
|
-42
|
-2.15
|
2
|
44
|
-2
|
3
|
-1.3
|
2
|
1.5
|
1.5
|
0.03
|
0.3
|
3 Left
|
-26.5
|
-2.18
|
0
|
26.5
|
-3
|
3
|
-1.0
|
3.5
|
3.5
|
3.5
|
0.04
|
0.3
|
4 Right
|
-10
|
0.38
|
1.5
|
11.5
|
-0.5
|
3
|
-0.67
|
1.75
|
3
|
1.25
|
0.1
|
0.5
|
4 Left
|
-12
|
0.54
|
0.75
|
12.75
|
-0.5
|
3
|
-0.41
|
1.75
|
2
|
1
|
0.1
|
0.6
|
5 Left
|
-15.5
|
-0.38
|
-2.25
|
13.25
|
-1.5
|
3
|
0.25
|
4.5
|
3.5
|
3.5
|
0.04
|
0.6
|
6 Right
|
-19
|
-0.65
|
-0.5
|
18.5
|
-1.5
|
2
|
-0.5
|
1
|
1.5
|
1.5
|
0.1
|
0.7
|
6 Left
|
-21
|
-1.97
|
-2
|
19
|
-3
|
2
|
-0.5
|
1
|
1
|
1
|
0.1
|
0.4
|
Pre-op preoperative, D diopter, UCVA uncorrected visual acuity (presented as Snellen decimal visual acuity) |
All nine eyes had some degree of astigmatism before surgery. Post-operatively, cylinder remained unchanged in two eyes, diminished in six eyes, and increased by 0.5 D in one eye. All patients’ UCVA improved ≥ 2 lines postoperatively. Due to residual myopia and astigmatism, BCVA was used for comparison and analysis.
LogMAR BCVA, MTFcutoff, and OSI at pre-operative, 3 months, 12 months, 24 months, and endpoint were compared using one-way repeated measures of anova, as well as AXL at preoperative, 12 months, 24 months, and endpoint. A line graph was drawn to show the mean ± SD of each parameter at each follow-up time (Fig. 2). P-value was displayed if the change was significant (p<0.05).
Specifically, for LogMAR BCVA comparison, the Greenhouse & Geisser method (e = 0.403) was used as the variance and covariance matrices were not equal according to Mauchly's spherical hypothesis test (χ2 = 30.386, p = 0.001). LogMAR BCVA was statistically significant among different time points (F (1.611, 11.276) = 12.948, p = 0.02, partial η2 = 0.649). LogMAR BCVA was significantly improved at 24-month (95%CI: 0.015–0.878, p = 0.042) and endpoint (95%CI: 0.008–0.909, p = 0.046) follow-ups, but the improvement was not significant at 3-month (95%CI: -0.019-0.797, p = 0.063) and 12-month (95%CI: -0.016-0.787, p = 0.062) follow-ups. No significant difference was found among post-operative follow-ups (p>0.05).
For MTFcutoff comparison, Mauchly's spherical hypothesis test (χ2=-8.782, p = 0.059) was used, and a statistical significant difference was found among different time points (F(4, 28) = 16.113, p<0.001, partial η2 = 0.697). MTFcutoff was significantly improved at 12-month (95%CI: -39.692–2.591, p = 0.025), 24-month (95%CI: -39.976–7.983, p = 0.005) and endpoint (95%CI: -43.221–8.782, p = 0.005) follow-ups, but the improvement was not significant at 3-month (95%CI: -14.695-5.663, p = 1.000) follow-up. No significant difference was found among post-operative follow-ups (p>0.05).
For OSI comparison, the Greenhouse & Geisser method (e = 0.345) was used as the variance and covariance matrices were not equal according to Mauchly's spherical hypothesis test (χ2 = 39.069, p<0.001). OSI was statistically significant among different time points (F (1.380, 9.662 = 9.790, p = 0.008, partial η2 = 0.583). OSI was significantly improved at 12-month (95%CI: 0.217–8.033, p = 0.038), 24-month (95%CI: 1.442–7.983, p = 0.007) and endpoint (95%CI: 1.315–8.035, p = 0.008) follow-ups, but the improvement was not significant at 3-month (95%CI: -3.040-5.190, p = 1.000) follow-up. No significant difference was found among post-operative follow-ups (p>0.05).
For AXL comparison, the Greenhouse & Geisser method (e = 0.384) was used as the variance and covariance matrices were not equal according to Mauchly's spherical hypothesis test (χ2 = 27.664, p<0.001). AXL was statistically significant among different time points (F (1.151, 9.205 = 8.893, p = 0.013, partial η2 = 0.526), but postoperative AXL at all follow-up times did not change significantly compared to pre-operative status (p>0.05). However, AXL at 24 months was significantly longer than AXL at 12 months (95%CI: -0.607–0.153, p = 0.002).