Development and validation to predict visual acuity and keratometry two years after corneal crosslinking with progressive keratoconus by machine learning.

Purpose: To explore and validate the utility of machine learning (ML) methods using a limited sample size to predict changes in visual acuity and keratometry 2 years following corneal crosslinking (CXL) for progressive keratoconus. Methods: The study included all consecutive patients with progressive keratoconus who underwent CXL from July 2014 to December 2020, with a 2 year follow-up period before July 2022 to develop the model. Variables collected included patient demographics, visual acuity, spherical equivalence, and Pentacam parameters. Available case data were divided into training and testing data sets. Three ML models were evaluated based on their performance in predicting case corrected distance visual acuity (CDVA) and maximum keratometry (Kmax) changes compared to actual values, as indicated by average root mean squared error (RMSE) and R-squared (R2) values. Patients followed from July 2022 to December 2022 were included in the validation set. Results: A total of 277 eyes from 195 patients were included in training and testing sets and 43 eyes from 35 patients were included in the validation set. The baseline CDVA (26.7%) and the ratio of steep keratometry to flat keratometry (K2/K1; 13.8%) were closely associated with case CDVA changes. The baseline ratio of Kmax to mean keratometry (Kmax/Kmean; 20.9%) was closely associated with case Kmax changes. Using these metrics, the best-performing ML model was XGBoost, which produced predicted values closest to the actual values for both CDVA and Kmax changes in testing set (R2 = 0.9993 and 0.9888) and validation set (R2 = 0.8956 and 0.8382). Conclusion: Application of a ML approach using XGBoost, and incorporation of identifiable parameters, considerably improved variation prediction accuracy of both CDVA and Kmax 2 years after CXL for treatment of progressive keratoconus.

Independent-effect comparison of five crosslinking procedures for Progressive Keratoconus based on Keratometry and the ABCD Grading System using Generalized Estimating Equations (GEE).

PURPOSE: Corneal collagen crosslinking (CXL) is an effective treatment for progressive keratoconus. Multiple CXL modalities are clinically available. The present study compared the 1 year outcomes of five types of CXL procedures for progressive keratoconus in a Chinese population using generalized estimating equations (GEE). METHODS: This retrospective study included 239 eyes in 171 patients with keratoconus who underwent CXL and were followed up for 1 year. Five CXL procedures were assessed, including Accelerated Transepithelial CXL, Iontophoresis CXL for 10 min, CXL plus phototherapeutic keratectomy (CXL-plus-PTK), High-Fluence Accelerated CXL, and Accelerated CXL. Patients treated with the Accelerated CXL procedure represented the reference group. Primary outcomes were visual acuity change, spherical equivalence, endothelial cell density, mean keratometry (Kmean), maximum keratometry (Kmax), minimum corneal thickness (MCT), and the ABCD Grading System, consisting of A (staging index for ARC; ARC = anterior radius of curvature), B (staging index for PRC, PRC = posterior radius of curvature), and C (staging index for MCT) values 1 year postoperatively compared to baseline. Secondary outcomes were corrected GEE comparisons from each procedure versus the Accelerated CXL group. RESULTS: The Accelerated Transepithelial CXL group had lower performance than the Accelerated CXL group according to Kmean and Kmax. The CXL-plus-PTK group performed significantly better than the reference group as reflected by Kmax (ß = -0.935, P = 0.03). However, the CXL-plus-PTK group did not perform as well for B and C, and the Iontophoresis CXL group performed better for C. CONCLUSIONS: The CXL-plus-PTK procedure was more effective than the Accelerated CXL procedure based on Kmax, and the Iontophoresis CXL procedure performed better on the C value based on the ABCD Grading System.

A novel analysis of Scheimpflug total corneal refractive power following corneal cross-linking in mild to moderate keratoconus.

AIM: To detect an earlier improvement in mild to moderate keratoconus following corneal cross-linking (CXL) with total corneal refractive power (TCRP) using ray tracing method. METHODS: A total of 40 eyes of 30 consecutive patients who underwent CXL for progressive keratoconus were retrospectively enrolled. The following keratometric parameters provided by Pentacam HR, including maximum keratometry (Kmax), steepest keratometry (Ksteep), 3 mm zonal TCRP centered over corneal apex (TCRPapex,zone 3 mm), zonal mean keratometry and TCRP centered over corneal cone (Kmcone,zone and TCRPcone,zone 1, 2, 3 mm) were evaluated preoperatively and 1, 3, 6, and 12mo postoperatively. Groups 1 and 2 were defined based on Kmax at postoperative 1mo as improved (the initial improvement group) or worsen (the initial deterioration group) compared to the preoperative level. RESULTS: In the overall group, only keratometric parameters based on ray tracing method displayed significant improvement early at 3mo postoperatively, in which TCRPcone,zone 1 mm and 2 mm exhibited the largest flattening (0.57 D and 0.53 D, respectively). In Group 1, only Kmax, Kmcone,zone 2 mm and TCRPcone,zone 2 mm showed significant improvement initially at 1mo postoperatively, in which Kmax exhibited the largest improvement (1.05 D), followed by TCRPcone,zone 2 mm (0.82 D). In Group 2, only keratometric parameters based on ray tracing method and Kmcone,zone 3 mm showed slight but not significant improvement early at 3mo, in which TCRPcone,zone 3 mm displayed the most improvement (0.19 D), followed by TCRPcone,zone 2 mm (0.15 D). CONCLUSION: The findings indicate that a 2 mm zonal TCRP centered over Kmax could earlier detect keratometric improvement by CXL compared to other commonly used parameters in mild to moderate keratoconic eyes.

Clinical observations of iridociliary cysts and their changes after implantable collamer lens implantation in myopic patients.

AIM: To observe the characteristics of iridociliary cysts in myopic patients and evaluate the influences on the position and safety of implantable collamer lens (ICL) after surgery. METHODS: Totally 270 eyes of 135 patients who underwent ICL surgery for the corrections of myopia were included in this study. Preoperative and postoperative morphology of iridociliary cysts were observed in ultrasonic biomicroscopy (UBM) image. RESULTS: A total of 138 iridociliary cysts were found in 88 eyes of 50 patients among 270 eyes of 135 patients before surgery (37%). Twenty-five patients had cysts in one eye (50%) and 25 had cysts in both eyes (50%). The prevalence of iridociliary cysts was negatively correlated with age, but no gender difference (P>0.05). The incidence of iridociliary cysts was much less in eyes with myopia greater than -9.00 D (P<0.05). The diameter of the largest cyst was 1.96 mm and the smallest cyst was 0.24 mm, with a majority within the range of 0.5 to 1.0 mm. Most of the cysts were located in the inferior temporal quadrant. One year after ICL implantation, 51 iridociliary cysts (37%) remained unchanged, 47 cysts (34%) decreased in size, and 40 cysts (29%) disappeared. Most of cysts that changed after surgery were smaller than 1.0 mm (P<0.05) and located in the nasal and temporal sides around the haptics of implantable lens. All the ICL were in their original position. CONCLUSION: Iridociliary cysts are commonly seen in myopic eyes. The cysts have no impact on the safety of ICL surgery. Some cysts may decrease in size or disappear after ICL implantation.

Clinical and microstructural changes with different iontophoresis-assisted corneal cross-linking methods for keratoconus.

AIM: To compare the clinical and microstructural changes induced by different transepithelial iontophoresis-assisted corneal cross-linking (I-CXL) methods for keratoconus. METHODS: A total of 42 eyes of 42 patients with progressive keratoconus were divided into two groups. Group A received I-CXL for 5min, while group B received I-CXL for 10min. Visual acuity, optical coherence tomography (OCT), specular microscopy and confocal microscopy were evaluated preoperatively and at 1, 3, 6, and 12mo postoperatively. RESULTS: Twelve months after the operation, uncorrected visual acuity (UCVA) and corrected distance visual acuity (CDVA) were improved in both groups, with a better outcome in the I-CXL 10min group (P=0.025, 0.021, respectively). Kmax values decreased by 0.94±3.00 D in the I-CXL 10min group (P=0.033) but increased by 1.87±3.29 D in the I-CXL 5min group (P=0.012). OCT scans showed that the demarcation line was most visible and substantially deeper in the I-CXL 10min group. Confocal microscopy showed greater anterior stromal keratocyte decreases in the I-CXL 10min group than in the I-CXL 5min group at 3 and 6mo postoperatively (P<0.001); however, anterior stromal keratocytes and subbasal nerve density were not significantly different between the two groups at 12mo postoperatively. CONCLUSION: I-CXL for 10min more effectively halts the progression of keratoconus than I-CXL for 5min after 12mo of follow-up. However, long-term studies are needed to evaluate the efficacy and safety of I-CXL.

The Effect of the Crystalline Lens on Central Vault After Implantable Collamer Lens Implantation.

PURPOSE: To identify associations between crystalline lens-related factors and central vault after Implantable Collamer Lens (ICL) (Staar Surgical, Monrovia, CA) implantation. METHODS: This retrospective clinical study included 320 eyes from 186 patients who underwent ICL implantation surgery. At 1 year after surgery, the central vault was measured using anterior segment optical coherence tomography. Preoperative anterior chamber depth, lens thickness, lens position (lens position = anterior chamber depth + 1/2 lens thickness), and vault were analyzed to investigate the effects of lens-related factors on postoperative vault. RESULTS: The mean vault was 513 ± 215 µm at 1 year after surgery. Vault was positively correlated with preoperative anterior chamber depth (r = 0.495, P < .001) and lens position (r = 0.371, P < .001), but negatively correlated with lens thickness (r = -0.262, P < .001). Eyes with vaults of less than 250 µm had shallower anterior chambers, thicker lenses, and smaller lens position than eyes in the other two vault groups (which had vaults ≥ 250 µm) (P < .001). Eyes with both anterior chamber depth less than 3.1 mm and lens position less than 5.1 mm had greatly reduced vaults (P < .001). CONCLUSIONS: The crystalline lens could have an important influence on postoperative vault. Eyes with a shallower anterior chamber and a forward lens position will have lower vaults. [J Refract Surg. 2017;33(8):519-523.].

Prevention and management of collagen copolymer phakic intraocular lens exchange: causes and surgical techniques.

PURPOSE: To analyze causes leading to explantation and exchange of Implantable Collamer Lens phakic intraocular lenses (pIOLs) and examine the surgical techniques of pIOL exchange. SETTING: Wuhan Aier Eye Hospital, Wuhan, China. DESIGN: Retrospective analysis of prospectively collected data. METHODS: Different techniques were used to explant the pIOLs based on the amount of vaulting. Preoperative and postoperative data were collected to evaluate possible causes leading to pIOL exchange and the safety of pIOL exchange. RESULTS: Of 616 myopic eyes with previous pIOL implantation, 16 eyes of 15 patients having pIOL exchange were reviewed. Eight surgeries (50%) were performed because of low vaulting (≤100 µm) and another 8 (50%) because of too high vaulting (≥1000 µm). The causes leading to low vaulting included increased crystalline lens thickness (≥4.0 mm), low anterior chamber depth (ACD) (<3.1 mm), and a too-small pIOL (<12.0 mm). The high vaulting was primarily the result of oversized white-to-white (WTW) measurements with Scheimpflug pachymetry (Pentacam) or the use of a digital caliper by an unskilled examiner and of ciliary body cysts. Six months after pIOL exchange, the vaulting ranged from 162 to 715 µm. No anterior subcapsular cataracts or other complications occurred during the observation period. CONCLUSIONS: The main causes of abnormal vaulting and subsequent pIOL exchange included a thick crystalline lens, low ACD, too-small pIOL, WTW measurement error, and ciliary body cysts. Phakic IOL explantations were individually designed based on the vaulting to achieve proper safety and efficacy. FINANCIAL DISCLOSURE: No author has a financial or proprietary interest in any material or method mentioned.

[Endothelial keratoplasty].

Endothelial keratoplasty (EK) is a new surgical method in treatment of corneal endothelial keratopathy, which retains healthy portions of a patient's cornea while replacing diseased endothelium with healthy donor tissue. In the past decade EK has been used increasingly and has shown better results over standard penetrating keratoplasty. The emergence and improvement of EK demonstrate the convert of corneal transplantation to refined and refractive surgery. EK, which has favorable refractive outcomes and potential low rejection rate, opens a new era for the treatment of endothelial dysfunction. However, there are still some aspects requiring to be studied, such as the methods of implantation to decrease graft dislocation rate and endothelial cell loss, use of femtosecond laser in EK to improve refractive errors, and the long-term result of immune rejection rate.

[Clinical features of 233 cases of keratoconus].

OBJECTIVE: To analyze clinical features of keratoconus. METHODS: The case history of keratoconus patients who underwent penetrating keratoplasty, epikeratophakia or deep lamellar keratoplasty in our hospital during January 1993 to March 2002 was reviewed, and inquiry form was sent to those who without complete data. Patients were followed during January 2000 to March 2002. Two hundred and thirty three cases of keratoconus were enrolled in this study with 216 males and 17 females. Patient's family history, systemic complications, refractive status, progress of disease, eye axial length and anterior chamber depth were recorded. RESULTS: The clinical natural history of these patients followed the model of "myopia-astigmatism-poor spectacle correction-acute corneal hydrops or stromal scar". The presenting age of myopia ranged from 6 to 29 years, with a mean age of 13.8 years, 10.1% (22/219) of the patients were younger than 10 years, 67.1% (147/219) were between 11 and 15 years. The interval time between emerging myopia and poor spectacle correction (best corrected visual acuity less than 20/200) was ranged from 1 to 20 years, of which 85.2% (161/189) were shorter than 6 years. Twenty percent (47/233) of the patients had acute corneal hydrops in one eye at the age of 12 to 31 years, and 4.3% of them (2/47) resulted in corneal perforation. Thirty-four patients had corneal stromal scars 1 to 13 years after keratoconus developed, and 94.1% of them (32/34) were within 8 years. Topography examination demonstrated a tendency of the cone base position progress: first, paracentral-enlargement following dominant astigmatism axis, then, enlargement along smaller astigmatism axis, and finally, the cone base moving centrally. The eye axial length averaged at (25.9 +/- 1.4) mm, the depth of anterior chamber (AC) were (4.3 +/- 0.3) mm, 89.7% (279/311) eyes had an axial length longer than 24 mm. All cases had an asymmetry clinical progress course. Two cases had Down syndrome, 2 with intelligence defect. No family history was noted in this study. CONCLUSIONS: Keratoconus in Han nationality of Shandong Province has a relatively earlier age onset. These eyes tend to have a deeper AC with higher incidence of longer axial length.

[Evaluation of antifungal chemotherapeutic effects on fungal keratitis by confocal microscopy].

OBJECTIVE: To evaluate the validity of confocal microscopy in estimating curative effect and in directing the treatment for fungal keratitis in the process of antifungal chemotherapy. METHODS: Fifty-eight patients, who were confirmed fungal infection by confocal microscopy, were selected from 328 patients with fungal keratitis. All patients received routine topical and/or oral antifungal medication, and were examined by confocal microscopy once a week and one week after discontinuation of the treatment. The density of hyphae in the corneal lesion, the configuration of inflammatory cells and keratocyte were recorded. Antifungal chemotherapy was adjusted according to examination results and medicines were changed accordingly. If no hyphae were detected by confocal microscopy, antifungal medication was maintained for one week and then discontinued. All patients were followed up for two months to ensure no relapse of fungal infection. RESULTS: Fifty three patients were cured. The area of corneal lesions began to reduce 7 days after the beginning of antifungal chemotherapy. Confocal microscopy examination revealed that the hypha positive sites and the density of hypha were reduced gradually; inflammatory cells also decreased, the configuration of corneal lesion was transformed from asymmetry to symmetry; and normal keratocytes could be detected gradually. After 14 days of treatment, ulcers healed up in 37 cases and no hyphae and inflammatory cells were found in 23 cases. After 28 days of treatment, all corneal ulcers healed up; hyphae and inflammatory cells were completely disappeared in 31 patients, but a few hyphae still could be found in 22 patients. Antifungal chemotherapy was tapered gradually if no hyphae and inflammatory cells were detected by confocal microscopy. There was no relapse of fungus infection during 2-month follow-up. Infection deteriorated in the other five patients within 7 days, which showed increased density of hypha and inflammatory cells under confocal microscopy examination. All of them were treated with a penetrating keratoplasty to save the eyeball. CONCLUSIONS: Confocal microscopy is an ideal method for the evaluation of curative effects of fungal keratitis in the process of antifungal chemotherapy. This is also a valuable objective tool in directing antifungal medication.

[A retrospective medical history analysis of the patients with severe fungal keratitis].

OBJECTIVE: To investigate the cause of delayed diagnosis and treatment in severe fungal keratitis. METHODS: Retrospective analysis of the treatment history was performed in patients with severe fungal keratitis which attended to our hospital during 2000 to 2001. The main parameters evaluated were nature of access with the health care services, examinations undertaken, initiated treatment, and the time interval between presentation to each medical contact and to our center. Finally, the management undertaken at our center was evaluated. RESULTS: 69 patients (eyes) were enrolled into this investigation, of which 86% (59/69) was local residents in Shandong province, 93% (64/69) was farmer. Penetrating keratoplasty (PKP) were performed for 63 cases with a transplantation of donor cornea > 8.0 mm. However, 6 eyes were enucleated because of the severe endophthalmitis. Most of patients sought medical care following a same procedure: village and towns medical service, county hospital, city medical center. The time interval between the first medical contact and presentation to our center was 22.9 days on average. 78% (54/69) patients attended to county hospitals, only 24% (13/54) patients got correct diagnosis of fungal keratitis. Smear examination had been undertaken only in 22% (12/54) of cases, the positive rate was 50%. In 30 patients who initially contact to city medical centers, 63% (19/30) was diagnosed as fungal keratitis, 73% (22/30) was given smear examination with positive result of 73% (16/22). The positive rate of corneal scraping for fungus was 93% (64/69). 46% (35/69) patients were prescribed antifungal agents of fluoconazole. CONCLUSION: Several factors, including ignoring traumatic history and basic examination techniques, lacking of effective ophthalmic antifungal agents, cornea donor, without qualified ophthalmologists, lead to inappropriate management in fungal keratitis.

[Cataract extraction and intraocular lens implantation after high-risk penetrating keratoplasty].

OBJECTIVE: To study the indication and incision of phacoemulsification and intraocular lens (IOL) implantation after high-risk penetrating keratoplasty (PKP) and to evaluate the clinical results. METHODS: Twenty-six eyes of 25 patients who underwent high-risk PKP were treated with phacoemulsification and IOL implantation. The high-risk reasons included chemical burn of cornea, severe infective corneal perforation, total keratoplasty, diameter of the corneal graft larger than 8.5 mm, secondary keratoplasty and PKP combined with cataract extraction and anterior segment vitrectomy. Systemic and local steroid administration was given and surgical indication and incision were chosen according to the condition of the eyes. The mean interval time between PKP and IOL implantation was 11 months. Visual acuity, refractive status, central corneal endothelial cell density before and after IOL implantation (3 months) and allograft rejection were recorded. The mean follow-up after IOL placement was 17 months. RESULTS: Uncorrected visual acuity was 50/100 or better in 7 (27%) eyes and corrected visual acuity was 50/100 or better in 12 (46%) eyes. Fourteen (54%) eyes had refractive errors within 2 diopters. The mean refractive cylinders before and after IOL implantation were 5.70 D and 4.70 D, respectively (paired t test, t = 1.471, P = 0.154). The endothelial cell density before and after IOL implantation was 1883 cells/mm(2) and 1760 cells/mm(2), respectively (t = 1.934, P = 0.065) and the rate of endothelial cell loss was 6.5%. Only one graft rejection (4%) occurred 7 months after IOL implantation (anterior chamber IOL). CONCLUSION: It is safe and effective to perform phacoemulsification and IOL implantation after high-risk PKP when paying attention to the perioperative treatment and selecting the surgical indication and the position of the incision.