PURPOSE: To identify the speed of corneal neovascularization (CNV) after penetrating keratoplasty (PK) and to evaluate the influence of surgery-related factors on postkeratoplasty CNV in keratoconus patients. DESIGN: Prospective, longitudinal, observational, case series study. METHODS: All consecutive primary PKs performed for keratoconus by four experienced surgeons between January 1, 2000 and December 31, 2002 at our department were included (n=66 patients). Standardized corneal photographs taken preoperatively and at 6 weeks, 3, 6, 12, 18, and 24 months postoperatively were evaluated. Limbus suture distance (LSD), limbus graft distance (LGD), inner suture angle (ISA), and maximal extension of CNVs in digitized pictures with 100-fold magnification were measured at each of the 16 suture turning points at every timepoint. RESULTS: Forty-four (67%) out of 66 corneal grafts developed some degree of CNV after PK, most commonly from around the 12 o’clock position. The mean speed of CNVgrowth was 114 μm/month with the fastest growth occurring during the first six weeks after PK. There was an inverse correlation between CNV and LSD, LGD, and ISA (all P< .001). Ninety percent of all CNVs developed with LSD < 406 μm and with LGD < 1000 μm. Superior limbal localization between 11 and 1 o’clock is an independent risk factor for postoperative neovascularization, too (P< .001). CONCLUSIONS: Small LSD, small LGD, and narrow stitching with small ISA were identified as potentially modifiable surgical risk factors for CNVafter PK. The speed of CNV outgrowth was most pronounced in the first weeks after PK making early postoperative controls for CNV growth and initiation of antiangiogenic treatment important. PURPOSE: To identify the speed of corneal neovascularization (CNV) after penetrating keratoplasty (PK) and to evaluate the influence of surgery-related factors on postkeratoplasty CNV in keratoconus patients. DESIGN: Prospective, longitudinal, observational, case series study. METHODS: All A consecutive primary PKs performed for keratoconus by four experienced surgeons between January 1, 2000 and December 31, 2002 at our department were included (n = 66 patients). Standardized corneal photographs were preoperatively and at 6 weeks, 3, 6, 12, 18, and 24 months postoperatively were evaluated. Limbus suture distance (LSD), limbus graft distance (LGD), inner suture angle (ISA), and maximal extension of CNVs in digitized pictures with 100-fold magnification were measured at each of 16 suture turning points at every timepoint. RESULTS: Forty-four (67%) out of 66 corneal grafts developed some degree of CNV after PK, most commonly from around the 12 o’clock position. The mean speed of CNVgr owth was 114 μm / month with the fastest growth occurring during the first six weeks after PK. There was an inverse correlation between CNV and LSD, LGD, and ISA (all P <.001). Ninety percent of all CNVs developed with LSD < 406 μm and with LGD <1000 μm Superior limbal localization between 11 and 1 o’clock is an independent risk factor for postoperative neovascularization, too (P <.001). CONCLUSIONS: Small LSD, small LGD, and narrow stitching with small ISA were identified as potentially modifiable surgical risk factors for CNVafter PK. The speed of CNV outgrowth was the most pronounced in the first weeks after PK making early postoperative controls for CNV growth and initiation of antiangiogenic treatment important.