- Email: [email protected]
Prediction error after pediatric cataract surgery with intraocular lens implantation: Contact versus immersion A-scan biometry
ARTICLE
Prediction error after pediatric cataract surgery with intraocular lens implantation: Contact versus immersion A-scan biometry Rupal H. Trivedi, MD, MSCR, M. Edward Wilson, MD
PURPOSE: To evaluate the accuracy of pediatric intraocular lens (IOL) calculations performed using contact and immersion A-scan biometry. SETTING: Storm Eye Institute, Charleston, South Carolina, USA. DESIGN: Evaluation of diagnostic test or technology. METHODS: Data from a prospective study of pediatric eyes that had in-the-bag implantation of an AcrySof SN60WF IOL and had refraction results available from 14 days to 3 months postoperatively were retrospectively analyzed. The contact and immersion A-scan biometry techniques were performed in each eye and compared. RESULTS: The mean age at surgery of the 22 patients (22 eyes) was 4.8 years G 4.1 (SD). The mean prediction error was C0.4 G 0.7 diopter (D) in the contact group and 0.4 G 0.8 D in the immersion group (P < .001) and the mean absolute prediction error, 0.7 G 0.4 D and 0.7 G 0.6 D, respectively (PZ.694). The absolute prediction error was less than 0.5 D in 5 eyes (23%) using the contact technique and 11 eyes (50%) using the immersion technique. The mean postoperative spherical equivalent was C2.9 G 2.5 D, which was significantly different from the mean predicted refraction for contact A-scan (3.3 G 2.8 D; PZ.010) but not immersion A-scan (2.5 G 2.5 D; PZ.065). CONCLUSIONS: There was a significant difference in prediction error between postoperative refractive results obtained with contact biometry and immersion A-scan biometry in children. Based on the results, the immersion A-scan technique is recommended for pediatric IOL power calculation. Financial Disclosure: Neither author has a financial or proprietary interest in any material or method mentioned. J Cataract Refract Surg 2011; 37:501–505 Q 2011 ASCRS and ESCRS Supplemental material available at www.jcrsjournal.org.
Intraocular lens (IOL) implantation in the eyes of children has become a common practice during pediatric cataract surgery.1 However, accurate determination of IOL power in these eyes remains a challenge. Refractive surprises after pediatric cataract surgery are common.2–11 Although a myopic shift in refraction that occurs with eye growth can result in a late refractive error, early refractive errors can be attributed to inaccuracy in IOL power calculation and are a preventable cause of postoperative refractive error. Any early refractive surprise makes early amblyopia management difficult and partly influences long-term refractive shift. Errors in measurement of the axial length (AL) of the globe are the most significant errors in IOL power Q 2011 ASCRS and ESCRS Published by Elsevier Inc.
calculation. They can account for an error of 2.5 diopters (D)/mm in IOL power, increasing to 3.75 D/mm, or even higher, in short eyes. The ultrasound AL of the eye is commonly measured using contact or immersion techniques. In the contact method, the probe touches the cornea. This can result in corneal compression and a shorter AL. Immersion A-scan eliminates corneal compression but requires more training and experience to perform properly. Immersion A-scan has been shown to give better results than contact biometry in adults.12–15 However, in pediatric cataract surgery, the contact biometry remains a common technique for measuring the AL of the globe, especially when the measurements are taken in the operating room with the patient under general anesthesia.16 0886-3350/$ - see front matter doi:10.1016/j.jcrs.2010.09.023
501
502
PREDICTION ERROR IN PEDIATRIC IOL SURGERY
In a retrospective review of pediatric eyes by Ben-Zion et al.,17 AL measurements were taken using contact A-scan in the first 138 eyes and using the immersion technique in the subsequent 65 eyes. The authors compared absolute prediction errors in IOL power calculation and found no difference between the 2 techniques. However, the study did not directly compare the AL measurements obtained using immersion biometry and contact biometry in the same eyes. To our knowledge, no study has compared the prediction error when the AL was measured by both the contact technique and the immersion technique in children. Our study compared the prediction error between the 2 techniques in children having cataract surgery; the AL was measured with both techniques in all eyes. PATIENTS AND METHODS This study complied with the U.S. Health Insurance Portability and Accountability Act. Institutional Review Board approval was obtained from the Medical University of South Carolina. In cases of bilateral cataract, only 1 eye was selected to prevent a correlation effect in statistical analysis. Details of the methods used were published in a prospective trial comparing AL measurements by the contact technique and the immersion technique in 50 eyes.16 In the current study, eyes that had primary in-the-bag implantation of an AcrySof SN60WF IOL (Alcon Laboratories, Inc.) and had refractive data available within 14 days to 3 months postoperatively were retrospectively analyzed. The same surgeon (M.E.W.) performed all the surgeries, and an experienced pediatric ophthalmologist performed all refractions manually with a retinoscope. Data collected included the patient’s age at the time of surgery, the AL measured using the contact technique and the immersion technique, the IOL power required to achieve emmetropia with both techniques (using the Holladay 1 formula and manufacturer’s A-constant of 118.7), the predicted refraction with both techniques with the implanted IOL power, the duration of follow-up, and the postoperative refraction. The refraction closest to 4 weeks after surgery was
Submitted: June 28, 2010. Final revision submitted: September 22, 2010. Accepted: September 28, 2010. From the Miles Center for Pediatric Ophthalmology, Storm Eye Institute, Department of Ophthalmology, Medical University of South Carolina, Charleston, South Carolina, USA. Supported by Grady Lyman Fund and in part by an unrestricted grant to the Medical University of South Carolina, Storm Eye Institute, from Research to Prevent Blindness, Inc., New York, New York, USA. Corresponding author: Rupal H. Trivedi, MD, MSCR, Medical University of South Carolina–Storm Eye Institute, 167 Ashley Avenue, Charleston, South Carolina 29425-5536, USA. E-mail: [email protected].
used in the calculations of prediction error. The spherical equivalent (SE) of the residual refractive error was recorded in diopters. For each surgery, the prediction error was calculated as the difference between the predicted postoperative refraction and the actual postoperative refraction. The prediction error between contact A-scan and immersion A-scan AL measurements was then compared. Statistical analysis was performed using the paired t test.
RESULTS Twenty-two eyes of 22 patients (12 boys; 55%) were analyzed. The mean age at the time of AL measurement and cataract surgery was 4.8 years G 4.1 (SD) (median 3.6 years; range 0.1 to 15.4 years). Ten patients (45%) were white, 9 (41%) were African-American, and 3 (14%) were in other ethnic categories. Table 1 shows the results obtained by both techniques. Keratometry (K) readings measured with an automated keratometer were used in both groups (mean K1, 43.0 G 1.9 D; mean K2, 44.8 G 2.1 D). The mean time of the postoperative refraction measurement was 36 G 18 days (median 31 days; range 15 to 88 days). The absolute prediction error was less than 0.5 D in 5 eyes (23%) using the contact technique and 11 eyes (50%) using the immersion technique (Table 2 and Figure 1). Figure 2 shows a histogram of the prediction error with measurements by contact and immersion biometry. Sixteen eyes (73%) had a positive prediction error using contact A-scan data, while a similar number of eyes had a negative prediction error using immersion data. The mean postoperative SE was C2.9 G2.5 D. This value was significantly
Table 1. Comparison of contact and immersion techniques. Technique Parameter Axial length (mm) Mean G SD Median Range IOL power for emmetropia (D) Mean G SD Median Range Prediction error (D) Mean G SD Median Range Absolute prediction error (D) Mean G SD Median Range
J CATARACT REFRACT SURG - VOL 37, MARCH 2011
Contact
Immersion
22.2 G 2.6 22.1 16.7 to 27.6
22.5 G 2.5 22.4 16.8 to 27.6
P Value !.001
!.001 25.8 G 8. 8 26.4 6.8 to 47.0
24.6 G 8.5 25.2 6.8 to 46.4
0.4 G 0.7 0.6 1.1 to 1.3
0.4 G 0.8 0.3 2.9 to 0.9
!.001
.694 0.7 G 0.4 0.6 0.06 to 1.30
0.7 G 0.6 0.5 0.03 to 2.90
PREDICTION ERROR IN PEDIATRIC IOL SURGERY
503
Emmetropia has progressed from being an exceptional result to an expected outcome in adult eyes having cataract surgery. Previous studies2–11 have documented the difficulties in achieving the refractive target after pediatric cataract surgery. Minimizing the refractive prediction error is a primary goal for all patients having cataract surgery, including children. An accurate postoperative refraction target would allow better planning of treatment for amblyopia and would potentially yield better visual acuity in the operated eye. In adults, the immersion technique has been shown to be more accurate than contact biometry.12–15,18 Inaccurate calculations of IOL power during adult cataract surgery are largely attributed to errors in AL measurement. In the current study, the AL measured by the contact technique was significantly shorter than the AL measured by the immersion technique (22.2 mm versus
22.5 mm; P!.001) in the same eye. This concurs with results in studies of adult eyes.12–15 The IOL power for emmetropia was 1.2 D stronger with the contact technique than with the immersion technique (25.8 D versus 24.6 D; P!.001). The prediction error was significantly different (C0.4 D with the contact technique and 0.4 D with the immersion technique). This finding concurs with results in another study,17 which found a prediction error of C0.23 D and 0.32 D with the contact technique and immersion technique, respectively. We found no significant difference in absolute prediction error between the contact and immersion A-scan biometry techniques (0.7 D and 0.7 D, respectively). Ben-Zion et al.17 also found no significant difference in absolute error (1.11 D and 1.03 D, respectively). Tromans et al.19 studied 52 pediatric eyes using contact A-scan biometry. They report a mean absolute prediction error of 1.40 D. Nihalani and VanderVeen20 report a mean absolute prediction error of 0.76 D using the Holladay 1 formula. L€ uchtenberg et al.21 assessed the predictability of postoperative refraction using the Holladay II formula in pediatric patients randomized to an optic-capture versus a no-optic-capture procedure. The Holladay II formula provided reliable IOL calculations in pediatric cataracts. The authors reported a tendency toward underachievement of the target refraction in younger patients. The predicted refraction correlated with the achieved postoperative refraction (total group, PZ.019), and was closer in the no-capture group (PZ.033) than in the optic-capture group (PZ.265). In our study, more than twice as many eyes achieved a postoperative refraction within G0.5 D of the estimated refraction with the immersion technique than with the contact technique (50.0% versus 22.7%). Based on this observation, despite the small sample size in our study, we believe that the use of the
Figure 1. Histogram of the absolute prediction error with measurements by contact and immersion biometry.
Figure 2. Histogram of prediction error with measurements by contact and immersion biometry.
Table 2. Number of eyes within various ranges of absolute prediction error by technique. Eyes (n) Absolute PE !0.50 D !1.00 D O1.00 D O2.00 D
Contact
Immersion
5 15 7 0
11 19 3 1
PE Z prediction error
different from the predicted refraction with the contact technique (mean 3.3 G 2. 8 D; PZ.010) but not from that with the immersion technique (2.5 G 2.5 D; PZ.065). DISCUSSION
J CATARACT REFRACT SURG - VOL 37, MARCH 2011
504
PREDICTION ERROR IN PEDIATRIC IOL SURGERY
immersion A-scan ultrasound technique results in a lower rate of large prediction errors. When the contact technique was used, the trend was toward a positive prediction error, whereas when the immersion technique was used, the trend was toward a negative prediction error. The mean postoperative SE was C2.9 D. The mean predicted refraction using the contact technique was 3.3 D. These results suggest that on average, a myopic error in predicted refraction of 0.4 D can be attributed to the use of the contact technique. If a lower AL value is entered (eg, contact AL measurement as opposed to immersion AL measurement) during IOL power calculation, it will result in the use of a stronger IOL power than required. This can lead to induced myopia in the postoperative refraction. We routinely use the immersion technique for AL measurement. However, had the IOL power values obtained with the contact technique been chosen and had the IOL power been aimed at emmetropia, the mean implanted IOL power would have been approximately 1.00 D greater (IOL power calculated using the Holladay 1 formula). Note that a difference of 1.00 D in IOL power results in an approximate difference of 0.75 D refraction at the spectacle plane. We entered the mean contact A-scan data and mean immersion A-scan data of all patients enrolled in our original prospective trial22 into the Holladay Consultant software; the mean K value was 44.0 D at 4 years of age (mean age in this cohort). The required IOL power for emmetropia would be 28.0 D with the contact technique and 27.0 D with the the immersion technique. If the child had IOL implantation using contact biometry data (28.0 D IOL), the predicted postoperative refraction using Holladay 1 formula would be 0.77 D. We may be able to minimize postoperative myopic prediction errors in refraction by routinely using immersion A-scan for children having cataract surgery. Our study has limitations. Although the AL was measured prospectively using both techniques, the postoperative refraction data were collected in a retrospective manner. This may have led to observation bias. The small sample size was another problem because it did not allow us to evaluate prediction errors at different AL ranges. For example, only 3 eyes had an AL less than 20.0 mm. Another limitation is that we used the manufacturer-provided A constant. Most physicians perform few pediatric cataract surgeries (!10 surgeries/year for 71.5% of respondents in a survey of American Society of Cataract and Refractive Surgery members1). Thus, most surgeons have insufficient data to calculate personalized IOL constant. Despite these limitations, an advantage of our study is that the AL was measured by both the contact technique and the immersion technique in all eyes, with
the order randomized to avoid measurement bias. The same surgeon performed all surgeries, and all eyes had in-the-bag implantation of the same IOL model. As pediatric IOL implantation becomes increasingly common, higher standards and more precise postoperative target refractions will be expected. Hence, it is crucial to find ways to decrease prediction errors in refraction. This study found a significant difference in prediction error obtained with contact versus immersion A-scan biometry in children. Immersion ultrasound can be easily performed in children under anesthesia just before the surgical preparation. A balanced salt solution is used as the immersion fluid during A-scan measurement, and the use of the fluid does not change the corneal clarity during the surgery. Making the change from contact to immersion A-scan ultrasound is well worth the small learning curve. Based on the results in our study, we recommend using immersion A-scan for pediatric IOL power calculation, which agrees with the published results for adult IOL power calculation. REFERENCES 1. Wilson ME Jr, Bartholomew LR, Trivedi RH. Pediatric cataract surgery and intraocular lens implantation; practice styles and preferences of the 2001 ASCRS and AAPOS memberships. J Cataract Refract Surg 2003; 29:1811–1820 2. Trivedi RH, Peterseim MM, Wilson ME Jr. New techniques and technologies for pediatric cataract surgery. Curr Opin Ophthalmol 2005; 16:289–293 3. Trivedi RH, Wilson ME Jr. IOL power calculation for children. In: Garg A, Lin JT, eds, Mastering Intraocular Lenses (IOLs); Principles, Techniques and Innovations. New Delhi, India, Jaypee Brothers, 2007; 84–91 4. Crouch ER, Crouch ER Jr, Pressman SH. Prospective analysis of pediatric pseudophakia: myopic shift and postoperative outcomes. J AAPOS 2002; 6:277–282 5. Plager DA, Lipsky SN, Snyder SK, Sprunger DT, Ellis FD, Sondhi N. Capsular management and refractive error in pediatric intraocular lenses. Ophthalmology 1997; 104:600–607; discussion by AW Biglan, 607 6. Weakley DR, Birch E, McClatchey SK, Felius J, Parks MM, Stager D. The association between myopic shift and visual acuity outcome in pediatric aphakia. J AAPOS 2003; 7:86–90 7. McClatchey SK, Hofmeister EM. Intraocular lens power calculation for children. In: Wilson ME Jr, Trivedi RH, Pandey SK, eds, Pediatric Cataract Surgery; Techniques, Complications, and Management. Philadelphia, PA, Lippincott, Williams & Wilkins, 2005; 30–38 8. Eibschitz-Tsimhoni M, Archer SM, Del Monte MA. Intraocular lens power calculation in children. Surv Ophthalmol 2007; 52:474–482 9. Fan DSP, Rao SK, Yu CBO, Wong CY, Lam DSC. Changes in refraction and ocular dimensions after cataract surgery and primary intraocular lens implantation in infants. J Cataract Refract Surg 2006; 32:1104–1108 10. Astle WF, Ingram AD, Isaza GM, Echeverri P. Paediatric pseudophakia: analysis of intraocular lens power and myopic shift. Clin Exp Ophthalmol 2007; 35:244–251 11. Barry J-S, Ewings P, Gibbon C, Quinn AG. Refractive outcomes after cataract surgery with primary lens implantation in infants. Br J Ophthalmol 2006; 90:1386–1389. Available at: http://www.
J CATARACT REFRACT SURG - VOL 37, MARCH 2011
505
PREDICTION ERROR IN PEDIATRIC IOL SURGERY
12.
13. 14.
15.
16.
17.
18.
19.
ncbi.nlm.nih.gov/pmc/articles/PMC1857464/pdf/1386.pdf. Accessed November 5, 2010 Hennessy MP, Chan DG. Contact versus immersion biometry of axial length before cataract surgery. J Cataract Refract Surg 2003; 29:2195–2198 Watson A, Armstrong R. Contact or immersion technique for axial length measurement? Aust N Z J Ophthalmol 1999; 27:49–51 Olsen T, Nielsen PJ. Immersion versus contact technique in the measurement of axial length by ultrasound. Acta Ophthalmol (Copenh) 1989; 67:101–102 Shammas HJ. A comparison of immersion and contact techniques for axial length measurement. Am Intra-Ocular Implant Soc J 1984; 10:444–447 Trivedi RH, Wilson ME. Axial length measurements by contact and immersion techniques in pediatric eyes with cataract. Ophthalmology 2010 Oct 29; [Epub ahead of print] Ben-Zion I, Neely DE, Plager DA, Ofner S, Sprunger DT, Roberts GJ. Accuracy of IOL calculations in children: a comparison of immersion versus contact A-scan biometry. J AAPOS 2008; 12:440–444 Schelenz J, Kammann J. Comparison of contact and immersion techniques for axial length measurement and implant power calculation. J Cataract Refract Surg 1989; 15:425–428 Tromans C, Haigh PM, Biswas S, Lloyd IC. Accuracy of intraocular lens power calculation in paediatric cataract surgery. Br J Ophthalmol 2001; 85:939–941. Available at: http://www.ncbi.nlm.nih.
gov/pmc/articles/PMC1724100/pdf/v085p00939.pdf. Accessed November 5, 2010 20. Nihalani BR, VanderVeen DK. Comparison of intraocular lens power calculation formulae in pediatric eyes. Ophthalmology 2010; 117:1493–1499 € chtenberg M, Kuhli-Hattenbach C, Fronius M, Zubcov AA, 21. Lu Kohnen T. Predictability of intraocular lens calculation using the Holladay II formula after in-the-bag or optic captured posterior chamber intraocular lens implantation in paediatric cataracts. Ophthalmologica 2008; 222:302–307 22. Trivedi RH, Wilson ME. Keratometry in pediatric eyes with cataract. Arch Ophthalmol 2008; 126:38–42. Available at: http://archopht.ama-assn.org/cgi/reprint/126/1/38.pdf. Accessed November 5, 2010
J CATARACT REFRACT SURG - VOL 37, MARCH 2011
First author: Rupal H. Trivedi, MD, MSCR Miles Center for Pediatric Ophthalmology, Storm Eye Institute, Department of Ophthalmology, Medical University of South Carolina, Charleston, South Carolina, USA
Prediction error after pediatric cataract surgery with intraocular lens implantation: Contact versus immersion A-scan biometry Rupal H. Trivedi, MD, MSCR, M. Edward Wilson, MD
PURPOSE: To evaluate the accuracy of pediatric intraocular lens (IOL) calculations performed using contact and immersion A-scan biometry. SETTING: Storm Eye Institute, Charleston, South Carolina, USA. DESIGN: Evaluation of diagnostic test or technology. METHODS: Data from a prospective study of pediatric eyes that had in-the-bag implantation of an AcrySof SN60WF IOL and had refraction results available from 14 days to 3 months postoperatively were retrospectively analyzed. The contact and immersion A-scan biometry techniques were performed in each eye and compared. RESULTS: The mean age at surgery of the 22 patients (22 eyes) was 4.8 years G 4.1 (SD). The mean prediction error was C0.4 G 0.7 diopter (D) in the contact group and 0.4 G 0.8 D in the immersion group (P < .001) and the mean absolute prediction error, 0.7 G 0.4 D and 0.7 G 0.6 D, respectively (PZ.694). The absolute prediction error was less than 0.5 D in 5 eyes (23%) using the contact technique and 11 eyes (50%) using the immersion technique. The mean postoperative spherical equivalent was C2.9 G 2.5 D, which was significantly different from the mean predicted refraction for contact A-scan (3.3 G 2.8 D; PZ.010) but not immersion A-scan (2.5 G 2.5 D; PZ.065). CONCLUSIONS: There was a significant difference in prediction error between postoperative refractive results obtained with contact biometry and immersion A-scan biometry in children. Based on the results, the immersion A-scan technique is recommended for pediatric IOL power calculation. Financial Disclosure: Neither author has a financial or proprietary interest in any material or method mentioned. J Cataract Refract Surg 2011; 37:501–505 Q 2011 ASCRS and ESCRS Supplemental material available at www.jcrsjournal.org.
Intraocular lens (IOL) implantation in the eyes of children has become a common practice during pediatric cataract surgery.1 However, accurate determination of IOL power in these eyes remains a challenge. Refractive surprises after pediatric cataract surgery are common.2–11 Although a myopic shift in refraction that occurs with eye growth can result in a late refractive error, early refractive errors can be attributed to inaccuracy in IOL power calculation and are a preventable cause of postoperative refractive error. Any early refractive surprise makes early amblyopia management difficult and partly influences long-term refractive shift. Errors in measurement of the axial length (AL) of the globe are the most significant errors in IOL power Q 2011 ASCRS and ESCRS Published by Elsevier Inc.
calculation. They can account for an error of 2.5 diopters (D)/mm in IOL power, increasing to 3.75 D/mm, or even higher, in short eyes. The ultrasound AL of the eye is commonly measured using contact or immersion techniques. In the contact method, the probe touches the cornea. This can result in corneal compression and a shorter AL. Immersion A-scan eliminates corneal compression but requires more training and experience to perform properly. Immersion A-scan has been shown to give better results than contact biometry in adults.12–15 However, in pediatric cataract surgery, the contact biometry remains a common technique for measuring the AL of the globe, especially when the measurements are taken in the operating room with the patient under general anesthesia.16 0886-3350/$ - see front matter doi:10.1016/j.jcrs.2010.09.023
501
502
PREDICTION ERROR IN PEDIATRIC IOL SURGERY
In a retrospective review of pediatric eyes by Ben-Zion et al.,17 AL measurements were taken using contact A-scan in the first 138 eyes and using the immersion technique in the subsequent 65 eyes. The authors compared absolute prediction errors in IOL power calculation and found no difference between the 2 techniques. However, the study did not directly compare the AL measurements obtained using immersion biometry and contact biometry in the same eyes. To our knowledge, no study has compared the prediction error when the AL was measured by both the contact technique and the immersion technique in children. Our study compared the prediction error between the 2 techniques in children having cataract surgery; the AL was measured with both techniques in all eyes. PATIENTS AND METHODS This study complied with the U.S. Health Insurance Portability and Accountability Act. Institutional Review Board approval was obtained from the Medical University of South Carolina. In cases of bilateral cataract, only 1 eye was selected to prevent a correlation effect in statistical analysis. Details of the methods used were published in a prospective trial comparing AL measurements by the contact technique and the immersion technique in 50 eyes.16 In the current study, eyes that had primary in-the-bag implantation of an AcrySof SN60WF IOL (Alcon Laboratories, Inc.) and had refractive data available within 14 days to 3 months postoperatively were retrospectively analyzed. The same surgeon (M.E.W.) performed all the surgeries, and an experienced pediatric ophthalmologist performed all refractions manually with a retinoscope. Data collected included the patient’s age at the time of surgery, the AL measured using the contact technique and the immersion technique, the IOL power required to achieve emmetropia with both techniques (using the Holladay 1 formula and manufacturer’s A-constant of 118.7), the predicted refraction with both techniques with the implanted IOL power, the duration of follow-up, and the postoperative refraction. The refraction closest to 4 weeks after surgery was
Submitted: June 28, 2010. Final revision submitted: September 22, 2010. Accepted: September 28, 2010. From the Miles Center for Pediatric Ophthalmology, Storm Eye Institute, Department of Ophthalmology, Medical University of South Carolina, Charleston, South Carolina, USA. Supported by Grady Lyman Fund and in part by an unrestricted grant to the Medical University of South Carolina, Storm Eye Institute, from Research to Prevent Blindness, Inc., New York, New York, USA. Corresponding author: Rupal H. Trivedi, MD, MSCR, Medical University of South Carolina–Storm Eye Institute, 167 Ashley Avenue, Charleston, South Carolina 29425-5536, USA. E-mail: [email protected].
used in the calculations of prediction error. The spherical equivalent (SE) of the residual refractive error was recorded in diopters. For each surgery, the prediction error was calculated as the difference between the predicted postoperative refraction and the actual postoperative refraction. The prediction error between contact A-scan and immersion A-scan AL measurements was then compared. Statistical analysis was performed using the paired t test.
RESULTS Twenty-two eyes of 22 patients (12 boys; 55%) were analyzed. The mean age at the time of AL measurement and cataract surgery was 4.8 years G 4.1 (SD) (median 3.6 years; range 0.1 to 15.4 years). Ten patients (45%) were white, 9 (41%) were African-American, and 3 (14%) were in other ethnic categories. Table 1 shows the results obtained by both techniques. Keratometry (K) readings measured with an automated keratometer were used in both groups (mean K1, 43.0 G 1.9 D; mean K2, 44.8 G 2.1 D). The mean time of the postoperative refraction measurement was 36 G 18 days (median 31 days; range 15 to 88 days). The absolute prediction error was less than 0.5 D in 5 eyes (23%) using the contact technique and 11 eyes (50%) using the immersion technique (Table 2 and Figure 1). Figure 2 shows a histogram of the prediction error with measurements by contact and immersion biometry. Sixteen eyes (73%) had a positive prediction error using contact A-scan data, while a similar number of eyes had a negative prediction error using immersion data. The mean postoperative SE was C2.9 G2.5 D. This value was significantly
Table 1. Comparison of contact and immersion techniques. Technique Parameter Axial length (mm) Mean G SD Median Range IOL power for emmetropia (D) Mean G SD Median Range Prediction error (D) Mean G SD Median Range Absolute prediction error (D) Mean G SD Median Range
J CATARACT REFRACT SURG - VOL 37, MARCH 2011
Contact
Immersion
22.2 G 2.6 22.1 16.7 to 27.6
22.5 G 2.5 22.4 16.8 to 27.6
P Value !.001
!.001 25.8 G 8. 8 26.4 6.8 to 47.0
24.6 G 8.5 25.2 6.8 to 46.4
0.4 G 0.7 0.6 1.1 to 1.3
0.4 G 0.8 0.3 2.9 to 0.9
!.001
.694 0.7 G 0.4 0.6 0.06 to 1.30
0.7 G 0.6 0.5 0.03 to 2.90
PREDICTION ERROR IN PEDIATRIC IOL SURGERY
503
Emmetropia has progressed from being an exceptional result to an expected outcome in adult eyes having cataract surgery. Previous studies2–11 have documented the difficulties in achieving the refractive target after pediatric cataract surgery. Minimizing the refractive prediction error is a primary goal for all patients having cataract surgery, including children. An accurate postoperative refraction target would allow better planning of treatment for amblyopia and would potentially yield better visual acuity in the operated eye. In adults, the immersion technique has been shown to be more accurate than contact biometry.12–15,18 Inaccurate calculations of IOL power during adult cataract surgery are largely attributed to errors in AL measurement. In the current study, the AL measured by the contact technique was significantly shorter than the AL measured by the immersion technique (22.2 mm versus
22.5 mm; P!.001) in the same eye. This concurs with results in studies of adult eyes.12–15 The IOL power for emmetropia was 1.2 D stronger with the contact technique than with the immersion technique (25.8 D versus 24.6 D; P!.001). The prediction error was significantly different (C0.4 D with the contact technique and 0.4 D with the immersion technique). This finding concurs with results in another study,17 which found a prediction error of C0.23 D and 0.32 D with the contact technique and immersion technique, respectively. We found no significant difference in absolute prediction error between the contact and immersion A-scan biometry techniques (0.7 D and 0.7 D, respectively). Ben-Zion et al.17 also found no significant difference in absolute error (1.11 D and 1.03 D, respectively). Tromans et al.19 studied 52 pediatric eyes using contact A-scan biometry. They report a mean absolute prediction error of 1.40 D. Nihalani and VanderVeen20 report a mean absolute prediction error of 0.76 D using the Holladay 1 formula. L€ uchtenberg et al.21 assessed the predictability of postoperative refraction using the Holladay II formula in pediatric patients randomized to an optic-capture versus a no-optic-capture procedure. The Holladay II formula provided reliable IOL calculations in pediatric cataracts. The authors reported a tendency toward underachievement of the target refraction in younger patients. The predicted refraction correlated with the achieved postoperative refraction (total group, PZ.019), and was closer in the no-capture group (PZ.033) than in the optic-capture group (PZ.265). In our study, more than twice as many eyes achieved a postoperative refraction within G0.5 D of the estimated refraction with the immersion technique than with the contact technique (50.0% versus 22.7%). Based on this observation, despite the small sample size in our study, we believe that the use of the
Figure 1. Histogram of the absolute prediction error with measurements by contact and immersion biometry.
Figure 2. Histogram of prediction error with measurements by contact and immersion biometry.
Table 2. Number of eyes within various ranges of absolute prediction error by technique. Eyes (n) Absolute PE !0.50 D !1.00 D O1.00 D O2.00 D
Contact
Immersion
5 15 7 0
11 19 3 1
PE Z prediction error
different from the predicted refraction with the contact technique (mean 3.3 G 2. 8 D; PZ.010) but not from that with the immersion technique (2.5 G 2.5 D; PZ.065). DISCUSSION
J CATARACT REFRACT SURG - VOL 37, MARCH 2011
504
PREDICTION ERROR IN PEDIATRIC IOL SURGERY
immersion A-scan ultrasound technique results in a lower rate of large prediction errors. When the contact technique was used, the trend was toward a positive prediction error, whereas when the immersion technique was used, the trend was toward a negative prediction error. The mean postoperative SE was C2.9 D. The mean predicted refraction using the contact technique was 3.3 D. These results suggest that on average, a myopic error in predicted refraction of 0.4 D can be attributed to the use of the contact technique. If a lower AL value is entered (eg, contact AL measurement as opposed to immersion AL measurement) during IOL power calculation, it will result in the use of a stronger IOL power than required. This can lead to induced myopia in the postoperative refraction. We routinely use the immersion technique for AL measurement. However, had the IOL power values obtained with the contact technique been chosen and had the IOL power been aimed at emmetropia, the mean implanted IOL power would have been approximately 1.00 D greater (IOL power calculated using the Holladay 1 formula). Note that a difference of 1.00 D in IOL power results in an approximate difference of 0.75 D refraction at the spectacle plane. We entered the mean contact A-scan data and mean immersion A-scan data of all patients enrolled in our original prospective trial22 into the Holladay Consultant software; the mean K value was 44.0 D at 4 years of age (mean age in this cohort). The required IOL power for emmetropia would be 28.0 D with the contact technique and 27.0 D with the the immersion technique. If the child had IOL implantation using contact biometry data (28.0 D IOL), the predicted postoperative refraction using Holladay 1 formula would be 0.77 D. We may be able to minimize postoperative myopic prediction errors in refraction by routinely using immersion A-scan for children having cataract surgery. Our study has limitations. Although the AL was measured prospectively using both techniques, the postoperative refraction data were collected in a retrospective manner. This may have led to observation bias. The small sample size was another problem because it did not allow us to evaluate prediction errors at different AL ranges. For example, only 3 eyes had an AL less than 20.0 mm. Another limitation is that we used the manufacturer-provided A constant. Most physicians perform few pediatric cataract surgeries (!10 surgeries/year for 71.5% of respondents in a survey of American Society of Cataract and Refractive Surgery members1). Thus, most surgeons have insufficient data to calculate personalized IOL constant. Despite these limitations, an advantage of our study is that the AL was measured by both the contact technique and the immersion technique in all eyes, with
the order randomized to avoid measurement bias. The same surgeon performed all surgeries, and all eyes had in-the-bag implantation of the same IOL model. As pediatric IOL implantation becomes increasingly common, higher standards and more precise postoperative target refractions will be expected. Hence, it is crucial to find ways to decrease prediction errors in refraction. This study found a significant difference in prediction error obtained with contact versus immersion A-scan biometry in children. Immersion ultrasound can be easily performed in children under anesthesia just before the surgical preparation. A balanced salt solution is used as the immersion fluid during A-scan measurement, and the use of the fluid does not change the corneal clarity during the surgery. Making the change from contact to immersion A-scan ultrasound is well worth the small learning curve. Based on the results in our study, we recommend using immersion A-scan for pediatric IOL power calculation, which agrees with the published results for adult IOL power calculation. REFERENCES 1. Wilson ME Jr, Bartholomew LR, Trivedi RH. Pediatric cataract surgery and intraocular lens implantation; practice styles and preferences of the 2001 ASCRS and AAPOS memberships. J Cataract Refract Surg 2003; 29:1811–1820 2. Trivedi RH, Peterseim MM, Wilson ME Jr. New techniques and technologies for pediatric cataract surgery. Curr Opin Ophthalmol 2005; 16:289–293 3. Trivedi RH, Wilson ME Jr. IOL power calculation for children. In: Garg A, Lin JT, eds, Mastering Intraocular Lenses (IOLs); Principles, Techniques and Innovations. New Delhi, India, Jaypee Brothers, 2007; 84–91 4. Crouch ER, Crouch ER Jr, Pressman SH. Prospective analysis of pediatric pseudophakia: myopic shift and postoperative outcomes. J AAPOS 2002; 6:277–282 5. Plager DA, Lipsky SN, Snyder SK, Sprunger DT, Ellis FD, Sondhi N. Capsular management and refractive error in pediatric intraocular lenses. Ophthalmology 1997; 104:600–607; discussion by AW Biglan, 607 6. Weakley DR, Birch E, McClatchey SK, Felius J, Parks MM, Stager D. The association between myopic shift and visual acuity outcome in pediatric aphakia. J AAPOS 2003; 7:86–90 7. McClatchey SK, Hofmeister EM. Intraocular lens power calculation for children. In: Wilson ME Jr, Trivedi RH, Pandey SK, eds, Pediatric Cataract Surgery; Techniques, Complications, and Management. Philadelphia, PA, Lippincott, Williams & Wilkins, 2005; 30–38 8. Eibschitz-Tsimhoni M, Archer SM, Del Monte MA. Intraocular lens power calculation in children. Surv Ophthalmol 2007; 52:474–482 9. Fan DSP, Rao SK, Yu CBO, Wong CY, Lam DSC. Changes in refraction and ocular dimensions after cataract surgery and primary intraocular lens implantation in infants. J Cataract Refract Surg 2006; 32:1104–1108 10. Astle WF, Ingram AD, Isaza GM, Echeverri P. Paediatric pseudophakia: analysis of intraocular lens power and myopic shift. Clin Exp Ophthalmol 2007; 35:244–251 11. Barry J-S, Ewings P, Gibbon C, Quinn AG. Refractive outcomes after cataract surgery with primary lens implantation in infants. Br J Ophthalmol 2006; 90:1386–1389. Available at: http://www.
J CATARACT REFRACT SURG - VOL 37, MARCH 2011
505
PREDICTION ERROR IN PEDIATRIC IOL SURGERY
12.
13. 14.
15.
16.
17.
18.
19.
ncbi.nlm.nih.gov/pmc/articles/PMC1857464/pdf/1386.pdf. Accessed November 5, 2010 Hennessy MP, Chan DG. Contact versus immersion biometry of axial length before cataract surgery. J Cataract Refract Surg 2003; 29:2195–2198 Watson A, Armstrong R. Contact or immersion technique for axial length measurement? Aust N Z J Ophthalmol 1999; 27:49–51 Olsen T, Nielsen PJ. Immersion versus contact technique in the measurement of axial length by ultrasound. Acta Ophthalmol (Copenh) 1989; 67:101–102 Shammas HJ. A comparison of immersion and contact techniques for axial length measurement. Am Intra-Ocular Implant Soc J 1984; 10:444–447 Trivedi RH, Wilson ME. Axial length measurements by contact and immersion techniques in pediatric eyes with cataract. Ophthalmology 2010 Oct 29; [Epub ahead of print] Ben-Zion I, Neely DE, Plager DA, Ofner S, Sprunger DT, Roberts GJ. Accuracy of IOL calculations in children: a comparison of immersion versus contact A-scan biometry. J AAPOS 2008; 12:440–444 Schelenz J, Kammann J. Comparison of contact and immersion techniques for axial length measurement and implant power calculation. J Cataract Refract Surg 1989; 15:425–428 Tromans C, Haigh PM, Biswas S, Lloyd IC. Accuracy of intraocular lens power calculation in paediatric cataract surgery. Br J Ophthalmol 2001; 85:939–941. Available at: http://www.ncbi.nlm.nih.
gov/pmc/articles/PMC1724100/pdf/v085p00939.pdf. Accessed November 5, 2010 20. Nihalani BR, VanderVeen DK. Comparison of intraocular lens power calculation formulae in pediatric eyes. Ophthalmology 2010; 117:1493–1499 € chtenberg M, Kuhli-Hattenbach C, Fronius M, Zubcov AA, 21. Lu Kohnen T. Predictability of intraocular lens calculation using the Holladay II formula after in-the-bag or optic captured posterior chamber intraocular lens implantation in paediatric cataracts. Ophthalmologica 2008; 222:302–307 22. Trivedi RH, Wilson ME. Keratometry in pediatric eyes with cataract. Arch Ophthalmol 2008; 126:38–42. Available at: http://archopht.ama-assn.org/cgi/reprint/126/1/38.pdf. Accessed November 5, 2010
J CATARACT REFRACT SURG - VOL 37, MARCH 2011
First author: Rupal H. Trivedi, MD, MSCR Miles Center for Pediatric Ophthalmology, Storm Eye Institute, Department of Ophthalmology, Medical University of South Carolina, Charleston, South Carolina, USA