Journal of Refractive Surgery

Original Article 

Evaluation of Femtosecond Laser Versus Manual Clear Corneal Incisions in Cataract Surgery Using Spectral-Domain Optical Coherence Tomography

Xiaogang Wang, MD, PhD; Zhe Zhang, MD; Xiaoyan Li, MD; Lisha Xie, MD; Haining Zhang, MD; Douglas D. Koch, MD; Li Wang, MD, PhD; Suhua Zhang, MD

Abstract

PURPOSE:

To investigate the wound healing changes of clear corneal incisions performed with a femtosecond laser or a steel bevel knife in cataract surgery using spectral-domain optical coherence tomography.

METHODS:

Prospectively, consecutive patients undergoing either femtosecond laser–assisted cataract surgery (femtosecond laser group) using the LenSx laser (Alcon Laboratories, Inc., Fort Worth, TX) or traditional cataract surgery using steel blades (control group) were enrolled. The incidence of posterior wound gape, Descemet's membrane detachment, posterior wound retraction, and the inner and outer corneal incision thicknesses were assessed.

RESULTS:

Compared to the control group, the femtosecond laser group had significantly lower incidences of posterior wound gape at 1 day (P = .012) and Descemet's membrane detachment at 1 week (P = .030), 1 month (P = .048), and 3 months (P = .048), and a significantly higher incidence of posterior wound retraction at 1 month (P < .001) and 3 months (P < .001) postoperatively. The inner and outer corneal incision thicknesses decreased over time in both groups.

CONCLUSIONS:

The clear corneal incisions created by femtosecond laser demonstrated a lower incidence of posterior wound gape and Descemet's membrane detachment. However, the higher posterior wound retraction incidence, which indicated that there was wound remodeling, was apparent.

[J Refract Surg. 2018;34(1):17–22.]

Abstract

PURPOSE:

To investigate the wound healing changes of clear corneal incisions performed with a femtosecond laser or a steel bevel knife in cataract surgery using spectral-domain optical coherence tomography.

METHODS:

Prospectively, consecutive patients undergoing either femtosecond laser–assisted cataract surgery (femtosecond laser group) using the LenSx laser (Alcon Laboratories, Inc., Fort Worth, TX) or traditional cataract surgery using steel blades (control group) were enrolled. The incidence of posterior wound gape, Descemet's membrane detachment, posterior wound retraction, and the inner and outer corneal incision thicknesses were assessed.

RESULTS:

Compared to the control group, the femtosecond laser group had significantly lower incidences of posterior wound gape at 1 day (P = .012) and Descemet's membrane detachment at 1 week (P = .030), 1 month (P = .048), and 3 months (P = .048), and a significantly higher incidence of posterior wound retraction at 1 month (P < .001) and 3 months (P < .001) postoperatively. The inner and outer corneal incision thicknesses decreased over time in both groups.

CONCLUSIONS:

The clear corneal incisions created by femtosecond laser demonstrated a lower incidence of posterior wound gape and Descemet's membrane detachment. However, the higher posterior wound retraction incidence, which indicated that there was wound remodeling, was apparent.

[J Refract Surg. 2018;34(1):17–22.]

The architecture of clear corneal incisions (CCIs) plays an important role in the outcome of phacoemulsification. Improper CCI construction can lead to a variety of complications, including Descemet's membrane detachments, wound leaks, and excessive surgically induced astigmatism. A previous study demonstrated that it was hard to predictably make manual three-plane, two-plane, or single-plane tunnels with steel or diamond keratomes.1

Femtosecond laser–assisted cataract surgery technology can provide reproducible corneal incisions, capsulotomies, and nuclear fragmentation.2,3 The obvious advantage of a CCI created by femtosecond laser is the ability to make wounds of the desired configuration and dimensions.

Optical coherence tomography (OCT), which is a non-contact, non-invasive technology, has been used to evaluate CCI architecture made with steel blades. CCI construction by laser (incision length, incision location, and angle of incision) and surgically induced changes in CCI features (Descemet's membrane detachment and posterior wound gape) in the early postoperative period (up to 1 month) have been reported.4–8 In a previous study evaluating the healing changes of manual CCIs, Descemet's membrane detachment and posterior wound gape were noted in the early postoperative period, whereas posterior wound retraction emerged later and was present in 90% of the studied eyes after 3 years.9 In this study, we investigated the incidence of these three CCI features and related CCI quantitative parameters in patients who were undergoing femtosecond laser–assisted cataract surgery.

Patients and Methods

Patients

This study was performed at the Shanxi Eye Hospital (Taiyuan, Shanxi, China). The research protocol was approved by the institutional review board of Shanxi Eye Hospital and performed according to the tenets of the Declaration of Helsinki. Written informed consent was obtained from each patient after explaining the nature of this study.

Prospectively, consecutive patients undergoing either femtosecond laser–assisted cataract surgery (femtosecond laser group) using the LenSx laser (Alcon Laboratories, Inc.) or traditional cataract surgery using steel blades (control group) were enrolled between April and July 2014. Inclusion criteria were described in detail in our previous study10: senile cataracts, no systemic disease, dilated pupil size of 8 mm or larger, nuclear cataract grade 3 (Lens Opacities Classification System III),11 corneal endothelial cell count greater than 1,200/mm2,1 no pathological alteration of the anterior segment (eg, keratoconus, zonular dialysis, pseudoexfoliation syndrome, or corneal opacity), no retinal diseases impairing visual function, no previous anterior or posterior segment surgery, and no intraoperative or postoperative complications.

Cataract Surgery

All patients in both groups were prescribed levofloxacin 0.5% (Santen Pharmaceutical Co., Ltd., Osaka, Japan) and pranoprofen 0.1% (Niflan 0.1%; Senju Pharmaceutical Co., Ltd., Senju Seiyaku, Japan) eye drops for use in the operative eye four times daily beginning 48 hours before surgery. One experienced surgeon (SZ) performed all of the surgeries.

For the femtosecond laser group, capsulotomy, lens fragmentation, one two-plane primary cataract incision, and one single-plane paracentesis incision were created using the femtosecond laser as described in our previous study.10 The femtosecond laser–related parameters for those steps are summarized in Table 1.

Parameters for Primary Incision, Paracentesis Incision, Capsulotomy, and Lens Fragmentation Using the Femtosecond Laser

Table 1:

Parameters for Primary Incision, Paracentesis Incision, Capsulotomy, and Lens Fragmentation Using the Femtosecond Laser

After the femtosecond laser treatments were completed, the patients were transferred to a sterile operating room for the next steps in the procedure. The primary clear corneal incision and paracentesis incision were opened with blunt dissection. An Infiniti Vision System (Alcon Laboratories, Inc., Fort Worth, TX) was used for phacoemulsification and a foldable intraocular lens (Tecnis ZCB00; Abbott Medical Optics, Abbott Park, IL) was inserted into the capsular bag.

For the control group, a single-plane clear corneal incision at a meridian approximately 120° was created with a 2.2-mm single-bevel steel knife (Alcon Laboratories, Inc.) and a manual continuous curvilinear capsulorhexis (approximately 5.5 mm in diameter) was made. The Infiniti Vision System was used for phacoemulsification. Finally, a foldable intraocular lens (Tecnis ZCB00) was inserted into the capsular bag.

Data Acquisition

All patients received postoperative OCT examinations at 1 day, 1 week, 1 month, and 3 months after the surgical procedure. OCT scans of the cornea were captured with the spectral-domain OCT system RTVue-XR Avanti (Optovue, Inc., Fremont, CA) with a supplemental Cornea-Anterior Module technique. This OCT system was operated at an axial scan speed of 70 kHz using an 840-nm wavelength superluminescent diode with a bandwidth of 45 nm. It had a scan length of 8 mm for the corneal single-line scan model and the depth resolution was up to 5 μm. The OCT scans were repeated three times during the same visit. All of the images in this study were captured by an experienced technician (LSX), and the image quality score was more than 27, which was good for data analysis (software version 2015.1.0.90). Every location of interest was scanned and captured three times, and the image with the highest image quality was chosen for further analysis.

CCI Architecture Parameters

For CCI architecture features of each group, the presence of Descemet's membrane detachment, posterior wound gape, and posterior wound retraction (which was defined as an abrupt step-off or recession of the central edge of the posterior wound surface) were assessed and double checked by the same two experienced ophthalmologists (ZZ, XL) (Figure 1).

Optical coherence tomography images of (A) Descemet's membrane detachment, (A and B) posterior wound gape, and (C) posterior wound retraction.

Figure 1.

Optical coherence tomography images of (A) Descemet's membrane detachment, (A and B) posterior wound gape, and (C) posterior wound retraction.

The CCI quantitative parameters included inner and outer corneal incision thickness (Figure 2). All measurements were collected three times, and the average of the data was calculated for the final analysis.

A cross-sectional clear corneal incision image showing the outer corneal incision thickness (the perpendicular corneal thickness of the external wound opening is marked with a yellow line) and the inner corneal incision thickness (the perpendicular corneal thickness of the internal wound opening is marked with a green line).

Figure 2.

A cross-sectional clear corneal incision image showing the outer corneal incision thickness (the perpendicular corneal thickness of the external wound opening is marked with a yellow line) and the inner corneal incision thickness (the perpendicular corneal thickness of the internal wound opening is marked with a green line).

Statistical Analyses

Statistical analyses were performed with commercial software (version 13.0; SPSS, Inc., Chicago, IL). Chi-square tests were used to compare the incidence of Descemet's membrane detachment, posterior wound gape, and posterior wound retraction between groups. An independent sample t test was performed to compare the quantitative measurement results between groups. A repeated measures analysis of variance was used to investigate the variation tendency of incidence of Descemet's membrane detachment, posterior wound gape, and posterior wound retraction, as well as the CCI quantitative data for each group. All tests had a significance level of 5%.

Results

Fifty-eight eyes of 40 patients were included in the femtosecond laser group and 34 eyes of 26 patients were included in the control group. There were no significant differences in age between the two groups (P = .173). The effective phacoemulsification time and phacoemulsification energy of the femtosecond laser group were both lower than the control group (P < .001 and P = .041, respectively). The characteristics of patients in each group are summarized in Table 2.

Patient Characteristics

Table 2:

Patient Characteristics

Incidences of posterior wound gape, Descemet's membrane detachment, and posterior wound retraction are shown in Figure 3. For both groups, the incidences of Descemet's membrane detachment and posterior wound gape decreased over time (all P < .05). In addition, for the femtosecond laser group, the incidence of posterior wound retraction increased over time (all P < .05).

The incidence of (A) posterior wound gape, (B) Descemet's membrane detachment, and (C) posterior wound retraction at each follow-up time point in the femtosecond laser–assisted (femto) and manual (control) cataract surgery groups.

Figure 3.

The incidence of (A) posterior wound gape, (B) Descemet's membrane detachment, and (C) posterior wound retraction at each follow-up time point in the femtosecond laser–assisted (femto) and manual (control) cataract surgery groups.

Compared to the control group, the incidence of posterior wound gape at 1 day postoperatively was significantly lower in the femtosecond laser group (P = .012) but no significant differences were found at later time periods. The femtosecond laser group had a lower incidence of Descemet's membrane detachment compared to the control group at 1 week, 1 month, and 3 months postoperatively (all P < .05). Moreover, there was no evidence of Descemet's membrane detachment in eyes that had a follow-up period longer than 1 month in the femtosecond laser group. There was no posterior wound retraction noted in the control eyes at any of the follow-up time points. In the femtosecond laser group, posterior wound retraction was first seen at 1 month postoperatively with an incidence of 38% (22 of 58) that increased to 50% (29 of 58) at 3 months (P = .031).

Table 3 shows the mean inner and outer corneal incision thickness values at each time point for both groups. The mean inner corneal incision thickness of the femtosecond laser group at 1 month was significantly thinner compared to that of the control group (P = .004). The outer corneal incision thickness was statistically thinner in the femtosecond laser group than the corresponding values for the control group at each time point (all P < .05; Table 3). The inner and outer corneal incision thicknesses all decreased over time in both groups (all P < .05; Figure 4).

Inner and Outer Corneal Incision Thickness (CIT)

Table 3:

Inner and Outer Corneal Incision Thickness (CIT)

The (A) inner and (B) outer corneal incision thicknesses at each follow-up time point in the femtosecond laser–assisted (femto) and manual (control) cataract surgery groups.

Figure 4.

The (A) inner and (B) outer corneal incision thicknesses at each follow-up time point in the femtosecond laser–assisted (femto) and manual (control) cataract surgery groups.

Discussion

In our study, the major differences we found between the two groups were: (1) the femtosecond laser group had a lower incidence of posterior wound gape, Descemet's membrane detachment, and thinner inner and outer corneal incision thickness; (2) the femtosecond laser group had a higher incidence of posterior wound retraction; and (3) no eyes in the control group had posterior wound retraction at any of the follow-up time points.

The incidence of posterior wound gape in the control group for this study was slightly higher than some previous studies.6,8,9 This trend could be attributed to the inconsistent wound construction (single-plane, two-plane, or three-plane), different material properties of the diamond and steel blades for CCI, and variations in surgical technique. However, these results could also be due to different CCI sizes and imaging analysis technology. A significantly lower posterior wound gape incidence was found at 1 day postoperatively in the femtosecond laser group. Moreover, a relatively lower prevalence of posterior wound gape was demonstrated at 1 week, 1 month, and 3 months postoperatively in the femtosecond laser group. We hypothesize that this difference may depend on the tunnel incision geometry differences between the groups. Compared to the single oblique plane manual CCI, the femtosecond laser group received a two-plane CCI with a partial lamellar cut positioned parallel to the collagen lamellae, which may have improved the shearing force effects of the stromal collagen lamellae across the whole depth of the cornea.12–14

The prevalence of Descemet's membrane detachment in the control group in this study was within the large range of previous reports on Descemet's membrane detachment.9,15–17 Similar to Grewal and Basti,17 no eyes presented Descemet's membrane detachment in the femtosecond laser group in our study, which was significantly lower compared to the control group (9%) at 1 month. Descemet's membrane detachment, which potentially hindered the local endothelium pump mechanism, played an important role in wound healing. The relatively low incidence of Descemet's membrane detachment in the femtosecond laser group suggested there was a potential advantage of the femtosecond laser for creating a CCI. The creation procedure of CCI using the LenSx femtosecond laser is from the inner cornea (anterior chamber) to the outer cornea (epithelium). Therefore, some bubbles should be observed in the anterior chamber during the CCI creation, and the Descemet's membrane detachment should be an unusual complication for the femtosecond laser.18 However, the blunt dissection, phacoemulsification manipulation through the incision, and incision hydration may contribute to the emergence of Descemet's membrane detachment in the femtosecond laser group.

Posterior wound retraction was defined as an abrupt step-off or recession of the central edge of the posterior wound surface. Different from previous studies,9,17,19 a high prevalence of posterior wound retraction in eyes with manual CCI was not demonstrated in this study. This outcome may be due to the relatively high incidence of posterior wound gape in this study. The posterior wound retraction was undefinable in posterior wound gape cases, in which posterior wound margins were still separated. Similar to the study by Grewal and Basti,17 38% of eyes in the femtosecond laser group showed posterior wound retraction at 1 month postoperatively. Moreover, the prevalence of posterior wound retraction increased to 50% at 3 months postoperatively in the femtosecond laser group, which was similar to the findings in the study by Wang et al.9 The higher incidence of posterior wound retraction in the femtosecond laser group from 1 month postoperatively may indicate potential remodeling of the CCI due to molecule dissociation, endothelial cell necrosis, and biomechanical and thermal changes from the femtosecond laser.20 Moreover, a previous study using transmission electron microscopy confirmed the presence of necrotic keratocytes (ie, cell death accompanied with releasing lysosomal enzymes and other components from membrane-bound intracellular compartments) after femtosecond laser corneal flap formation, which was different from the predominant keratocyte apoptosis (ie, a gentler form of cell death in which most intracellular components remain confined in membrane-bound apoptotic bodies) after microkeratome corneal flap formation and may play an important role in posterior wound retraction in this study.21 Furthermore, the posterior wound retraction may have caused changes in the posterior and anterior corneal curvature, total corneal power, and corneal astigmatism, among other factors, which should be investigated further in the future.

The relatively thin inner and outer corneal incision thickness at each time point in the femtosecond laser group may demonstrate that the femtosecond laser–assisted cataract surgery caused less peripheral corneal incision edema than the manual procedure. This result emphasized that less mechanical and thermal stress was performed at the incision site after lens fragmentation using the femtosecond laser.

Our study had several limitations. The single line scan of the OCT measurements can be influenced by reproducibility. Therefore, we repeated the measurements three times and performed three measurements during each visit to decrease this bias. In addition, the relatively small sample size, non-plane–matched incisions, and short follow-up period are other limitations of this study. Possible factors that influenced the incision architecture, such as stretching and tearing of the wound from surgical instruments, and direct wound damage by ultrasound power, should also be considered in future studies. Additionally, longitudinal studies as a function of different incision designs would be useful.

We found a significant difference in the architecture of CCIs created by a femtosecond laser and a manual procedure. The femtosecond laser CCIs demonstrated a lower prevalence of posterior wound gape and Descemet's membrane detachment, as well as thinner inner corneal incision thickness and outer corneal incision thickness. However, the higher posterior wound retraction incidence indicated that there was wound remodeling. Further studies are desirable.

References

  1. Calladine D, Ward M, Packard R. Adherent ocular bandage for clear corneal incisions used in cataract surgery. J Cataract Refract Surg. 2010;36:1839–1848. doi:10.1016/j.jcrs.2010.06.058 [CrossRef]
  2. He L, Sheehy K, Culbertson W. Femtosecond laser-assisted cataract surgery. Curr Opin Ophthalmol. 2011;22:43–52.
  3. Masket S, Sarayba M, Ignacio T, Fram N. Femtosecond laser-assisted cataract incisions: architectural stability and reproducibility. J Cataract Refract Surg. 2010;36:1048–1049. doi:10.1016/j.jcrs.2010.03.027 [CrossRef]
  4. Rao B, Zhang J, Taban M, McDonnell P, Chen Z. Imaging and investigating the effects of incision angle of clear corneal cataract surgery with optical coherence tomography. Opt Express. 2003;11:3254–3261. doi:10.1364/OE.11.003254 [CrossRef]
  5. Fine IH, Hoffman RS, Packer M. Profile of clear corneal cataract incisions demonstrated by ocular coherence tomography. J Cataract Refract Surg. 2007;33:94–97. doi:10.1016/j.jcrs.2006.09.016 [CrossRef]
  6. Torres LF, Saez-Espinola F, Colina JM, et al. In vivo architectural analysis of 3.2 mm clear corneal incisions for phacoemulsification using optical coherence tomography. J Cataract Refract Surg. 2006;32:1820–1826. doi:10.1016/j.jcrs.2006.06.020 [CrossRef]
  7. Schallhorn JM, Tang M, Li Y, Song JC, Huang D. Optical coherence tomography of clear corneal incisions for cataract surgery. J Cataract Refract Surg. 2008;34:1561–1565. doi:10.1016/j.jcrs.2008.05.026 [CrossRef]
  8. Fukuda S, Kawana K, Yasuno Y, Oshika T. Wound architecture of clear corneal incision with or without stromal hydration observed with 3-dimensional optical coherence tomography. Am J Ophthalmol. 2011;151:413–419.e1. doi:10.1016/j.ajo.2010.09.010 [CrossRef]
  9. Wang L, Dixit L, Weikert MP, Jenkins RB, Koch DD. Healing changes in clear corneal cataract incisions evaluated using Fourier-domain optical coherence tomography. J Cataract Refract Surg. 2012;38:660–665. doi:10.1016/j.jcrs.2011.10.030 [CrossRef]
  10. Wang L, Zhang Z, Koch DD, Jia Y, Cao W, Zhang S. Anterior chamber interleukin 1beta, interleukin 6 and prostaglandin E2 in patients undergoing femtosecond laser-assisted cataract surgery. Br J Ophthalmol. 2016;100:579–582. doi:10.1136/bjophthalmol-2015-307586 [CrossRef]
  11. Chylack LT Jr, Wolfe JK, Singer DM, et al. The Lens Opacities Classification System III. The Longitudinal Study of Cataract Study Group. Arch Ophthalmol. 1993;111:831–836. doi:10.1001/archopht.1993.01090060119035 [CrossRef]
  12. Smolek MK, McCarey BE. Interlamellar adhesive strength in human eyebank corneas. Invest Ophthalmol Vis Sci. 1990;31:1087–1095.
  13. Shin TJ, Vito RP, Johnson LW, McCarey BE. The distribution of strain in the human cornea. J Biomech. 1997;30:497–503. doi:10.1016/S0021-9290(97)84433-8 [CrossRef]
  14. Serrao S, Lombardo G, Schiano-Lomoriello D, Ducoli P, Rosati M, Lombardo M. Effect of femtosecond laser-created clear corneal incision on corneal topography. J Cataract Refract Surg. 2014;40:531–537. doi:10.1016/j.jcrs.2013.11.027 [CrossRef]
  15. Xia Y, Liu X, Luo L, et al. Early changes in clear cornea incision after phacoemulsification: an anterior segment optical coherence tomography study. Acta Ophthalmol. 2009;87:764–768. doi:10.1111/j.1755-3768.2008.01333.x [CrossRef]
  16. Calladine D, Tanner V. Optical coherence tomography of the effects of stromal hydration on clear corneal incision architecture. J Cataract Refract Surg. 2009;35:1367–1371. doi:10.1016/j.jcrs.2009.03.036 [CrossRef]
  17. Grewal DS, Basti S. Comparison of morphologic features of clear corneal incisions created with a femtosecond laser or a keratome. J Cataract Refract Surg. 2014;40:521–530. doi:10.1016/j.jcrs.2013.11.028 [CrossRef]
  18. Nosé RM, Rivera-Monge MD, Forseto AS, Nosé W. Descemet membrane detachment in femtosecond laser-assisted cataract surgery. Cornea. 2016;35:562–564. doi:10.1097/ICO.0000000000000773 [CrossRef]
  19. Calladine D, Packard R. Clear corneal incision architecture in the immediate postoperative period evaluated using optical coherence tomography. J Cataract Refract Surg. 2007;33:1429–1435. doi:10.1016/j.jcrs.2007.04.011 [CrossRef]
  20. Schultz T, Tischoff I, Ezeanosike E, Dick HB. Histological sections of corneal incisions in OCT-guided femtosecond laser cataract surgery. J Refract Surg. 2013;29:863–864. doi:10.3928/1081597X-20131029-01 [CrossRef]
  21. Santhiago MR, Wilson SE. Cellular effects after laser in situ keratomileusis flap formation with femtosecond lasers: a review. Cornea. 2012;31:198–205. doi:10.1097/ICO.0b013e3182068c42 [CrossRef]

Parameters for Primary Incision, Paracentesis Incision, Capsulotomy, and Lens Fragmentation Using the Femtosecond Laser

StepsLaser Energy (μJ)StructureWidth (mm)Delta Values
Primary incision6Two-plane2.2N/A
Paracentesis incision6Single-plane1.2N/A
Capsulotomy65.2 mmN/A300 μm above and below the plane of the capsule
Lens fragmentation106 quadrantsN/A500 μm above and 800 μm below the plane of the capsule

Patient Characteristics

CharacteristicFemto GroupControl Group
No. of patients/eyes40/5826/34
Male/female20/2013/13
Age (y)57 ± 1558 ± 14
Effective phaco time (seconds)17 ± 932 ± 13
Phaco energy (%)16 ± 1320 ± 5
Suction time (seconds)178 ± 43N/A
Laser time (seconds)32 ± 6N/A

Inner and Outer Corneal Incision Thickness (CIT)

CITFemto Group (58 Eyes)Control Group (34 Eyes)Pa,b
Inner (μm)
  1 day1,035 ± 721,081 ± 115.063
  1 week918 ± 102957 ± 142.219
  1 month763 ± 74814 ± 85.004
  3 months741 ± 59768 ± 76.089
Outer (μm)
  1 day815 ± 60908 ± 85< .001
  1 week793 ± 57838 ± 104.044
  1 month749 ± 54802 ± 74< .001
  3 months739 ± 46790 ± 72.002
Authors

From Shanxi Eye Hospital, Shanxi, People's Republic of China (XW, ZZ, XL, LX, HZ, SZ); and the Department of Ophthalmology, Cullen Eye Institute, Baylor College of Medicine, Houston, Texas (DDK, LW).

Dr. Koch reports personal fees from Alcon, Carl Zeiss Meditec AG, and Johnson & Johnson, outside the submitted work. Dr. Li Wang reports personal fees from Hoya Surgical Optics and Carl Zeiss Meditec AG, outside the submitted work. The remaining authors have no financial or proprietary interest in the materials presented herein.

Supported by the National Natural Science Foundation of China under Grant No. 81501544 and the Department of Science and Technology (research grant 20140313014-9), Shanxi Province, China, and an unrestricted grant from Research to Prevent Blindness, Inc., New York, New York (DDK, LW).

AUTHOR CONTRIBUTIONS

Study concept and design (DDK, LW, SZ); data collection (ZZ, XL, LX, HZ); analysis and interpretation of data (XW, ZZ, XL, LX, HZ, LW, SZ); writing the manuscript (XW); critical revision of the manuscript (XW, ZZ, XL, LX, HZ, DDK, LW, SZ); statistical expertise (XW, DDK); administrative, technical, or material support (ZZ, XL, LX, HZ, LW, SZ); supervision (DDK, SZ)

Correspondence: Suhua Zhang, MD, Shanxi Eye Hospital, Shanxi, People's Republic of China. E-mail: cfykys@163.com

Received: July 27, 2017
Accepted: October 25, 2017

10.3928/1081597X-20171109-01

Sign up to receive

Journal E-contents