Cheng 2009.

Methods	Study design: randomised multicenter clinical trial Number randomised (total and per group): total 80 eyes of 80 patients; 40 eyes of 40 patients in each group Number analysed (total and per group): total 68 eyes of 68 patients; 29 eyes in the FLEK group and 39 eyes in the PKP group Exclusions and loss to follow‐up: In the FLEK group, excluded patients include two keratitis and two corneal ulcers before the surgery, three primary graft failures, two died, one immunologic graft failure and one corneal fold in the endothelial graft after the surgery; In the PKP group, one was excluded due to lost to follow‐up for health‐related issues Study follow‐up: 12 months
Participants	Country: the Netherlands Age (mean ± SD, range): 69.0 ± 8.8 in the FLEK group (n = 36); 71.4 ± 11.3 in the PKP group (n = 40); four patients in the FLEK group were excluded due to keratitis and corneal ulcers before receiving the intervention Gender: 21/36 (58.3%) in the FLEK group and 27/40 (67.5%) in the PK group were women Inclusion criteria: Fuchs’ endothelial dystrophy, pseudophakic bullous keratopathy, or posterior polymorphous dystrophy, age ≥ 18 years, and best spectacle‐corrected visual acuity (BSCVA) < 20/50 Exclusion criteria: mental retardation, previous PK, and human leukocyte antigen typed keratoplasty
Interventions	Treatment or intervention 1: femtosecond laser‐assisted endothelial keratoplasty (FLEK) Control or intervention 2: penetrating keratoplasty (PKP) General procedures: “In the FLEK group, the donor endothelial graft was prepared with the 30‐kHz femtosecond laser (AMO‐Intralase, Irvine, CA) as previously described. In the recipient, a 5.0‐mm corneoscleral incision and two limbal paracenteses were made. The Descemet membrane was scored with a Price‐Sinskey hook (Moria, Anthony, France), and a circle of 7.5‐mm Descemet membrane and endothelium was stripped from the posterior stroma. A 15‐degree blade was used to make four transcorneal incisions in the midperipheral recipient cornea to drain fluid between the recipient cornea and endothelial graft. An 8.0‐mm donor corneal disc was trephined from the corneoscleral button, and the endothelial graft was removed from the anterior cornea. The endothelial surface of the endothelial graft was coated with a small layer of viscoelastic material (Healon, AMO, Uppsala, Sweden), gently folded into a taco‐configuration, and inserted using a Goosey forceps (Moria, Anthony, France). The corneoscleral incision was closed with four 10‐0 nylon sutures. An air bubble was injected to unfold the endothelial graft and press the endothelial graft against the recipient cornea. After 20 min, the bubble was partly removed, and two drops of tropicamide minims 0.5% (Chauvin Benelux, Brussel, Belgium) were instilled to avoid a pupillary block. However, after two cases of pupillary block had occurred, a peripheral iridectomy was routinely performed. Patients were instructed to lie in supine position during 24 hr, to maximize the pressure of the remaining air bubble against the endothelial graft. Postoperatively, all patients received topical dexamethasone 0.1% drops (Ratiopharm, Zaandam, The Netherlands) 6 times daily and chloramphenicol 0.5% drops (Ratiopharm, Zaandam, The Netherlands) three times daily in a tapering dose." "In the PK group, the recipient cornea was trephined using a 7.75‐ or 8.0‐mm Hessburg‐Barron vacuum trephine, and the donor cornea was trephined with an 8.0‐ or 8.25‐mm disposable trephine. In all cases, a combined suturing technique of a running 11‐0 nylon suture with eight interrupted 10‐0 nylon sutures was used. Selective suture removal was based on topographic astigmatism pattern. Postoperatively, all patients received topical dexamethasone 0.1% drops (Ratiopharm, Zaandam, The Netherlands) six times daily and chloramphenicol 0.5% drops (Ratiopharm, Zaandam, The Netherlands) three times daily in a tapering dose.”
Outcomes	Primary outcome(s): per cent of eyes with postoperative refractive astigmatism ≤ 3.0 diopters (D) Secondary outcome(s): topographical astigmatism (EyeMap EH‐290, Alcon), uncorrected visual acuity (UCVA), best spectacle‐corrected visual acuity (BSCVA), manifest refraction, endothelial cell density (Noncon Robo, SP 8000, Konan, Hyogo, Japan), and intraoperative and postoperative complications Measurements taken, and intervals at which outcomes assessed: "The UCVA and BSCVA were determined using the Early Treatment of Diabetic Retinopathy Study letter charts and were converted to logarithm of the minimum angle of resolution measurements. Vision levels of counting fingers, hand movements, light perception, and no light perception were substituted by logarithm of the minimum angle of resolution values of 1.7, 2.0, 2.5, and 3.0, respectively. Preoperatively, and at 3, 6, and 12 months follow‐up, all visual parameters were recorded, and the endothelial cell density was measured at 3, 6, and 12 months follow‐up." Unit of analysis (individual or eye): eye Other issues with outcome assessment (e.g., quality control for outcomes if any): None
Notes	Study dates: patients were recruited between April 2005 and April 2007 Funding source(s): a grant of the Netherlands Organization for Health Research and Development (ZonMw) in the program Health Care Efficiency Research Declaration of interest: the authors declared no conflicts of interest Publication language: English
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	“Consecutive cases from each clinical population were randomly assigned to treatment by FLEK or by PK. The randomization code was generated using a permuted block size of 2.”
Allocation concealment (selection bias)	High risk	Allocation cannot be concealed with a block size of 2
Blinding of participants (performance bias)	Unclear risk	Not specified
Blinding of outcome assessment (detection bias) All outcomes	Unclear risk	Not specified
Incomplete outcome data (attrition bias) All outcomes	High risk	“In the FLEK group, four patients did not receive the allocated treatment due to significant preoperative events and were eventually excluded from the study analysis. All patients in the PK group received the allocated treatment.” The proportions lost to follow‐up were not low: 29/40 in FLEK group and 39/40 in PKP group and the study investigators did not perform intention‐to‐treat analysis
Selective reporting (reporting bias)	Unclear risk	Protocol not available
Other bias	Low risk	No other source of bias identified