The p-values shown were calculated by log-rank analysis and reflect a comparison of the behavior of the curves as a whole (considering all available data), rather than at a specific time point. During registration we request information about the recipient, the donor, the eye bank practices and the surgical procedure.
OVERVIEW AND CONTRIBUTOR INFORMATION
Current database
- Synopsis of the current database
- Survival of penetrating, lamellar and limbal grafts
In these and all subsequent survival curves, the number of grafts at risk in each stratum is shown in the table immediately below the curves, and graft survival at various intervals after transplantation is presented in the following table. Ten years after transplantation, all limbal grafts have failed, although some limbal grafts survive periods of less than ten years.
Grafts registered by state, territory and individual
- Grafts entered by Australian state and territory
- Contributors in each state
- The centre effect
- Outcome: influence of surgeon workload per year
- Outcome: according to era
- Outcome: whether followed by contributing surgeon or another practitioner.23
A comparison of graft survival since the inception of the Australian Corneal Transformation Registry is shown in Figure 1.7. Graft survival is better when the patient is followed elsewhere in both keratoconus and non-keratoconus cases.
DONORS AND EYE-BANKING
Cause of donor death
Closer inspection revealed that 48% of corneas in the trauma/accidental death group were used for patients with keratoconus, compared with 27% in the cardiac group, 31% in the cerebrovascular group, 29% in the malignancy group, and 29% in the causes of others in the death group and 26% in the respiratory group. Further, 53% of recipients in the trauma group were 40 years or younger at the time of transplantation compared with 26% in the cardiac group, 31% in the cerebrovascular group, 30% in the other death group, 29.% in the malignant diseases and 26% in the respiratory group.
Donor gender
Donor age
Corneal survival appears to be influenced by donor age when examining the total cohort. Donor age does not appear to affect graft survival in the subset of patients with keratoconus.
Death-to-enucleation and death-to-graft times
Death to enucleation times of 3 hours or less were compared with times of: more than 3 hours and up to 6 hours; over 6 hours and up to 9 hours; over 9 hours and up to 12 hours; and more than 12 hours. Of the 803 penetrating corneal grafts that were enucleated more than 12 hours after donor death, only 7 were >24 hours.
Corneal storage media
- Optisol storage time
The type of storage medium significantly affects corneal graft survival, with Optisol performing better than other media.
Donor procurement source
Multi-organ donors
- Multi-organ donors vs cadaveric donors
Tables and 2.9 compare sex, age, cause of death, death-to-enucleation times and death-to-graft times of multi-organ donors and cadaveric donors. Multi-organ donors tend to be relatively younger than cadaveric donors, and are more likely to die from cerebrovascular disease or trauma/accident/poisoning.
Primary non-functioning grafts
Summary of donor and eye-banking information
RECIPIENTS
Recipient age at graft
- Infant, child and adolescent recipients
- Congenital abnormalities
- Paediatric recipients
- Elderly recipients
Age at transplantation appears to have no significant effect on graft survival when keratoconus is excluded (Log Rank Statistic=8.09; df=4; p=0.0885). Recipients younger than 4 years appear to have a significantly poorer graft survival rate than older recipients, while those older than 14 years appear to have a significantly better graft survival rate.
Recipient gender
The significant difference in p value appears to be due to a poorer survival rate when corneas from male donors are used for female recipients. When indicators other than keratoconus were analyzed, poorer survival rates became apparent when both donor and recipient were male.
Pre-graft morbidities
- Vascularisation
- Inflammation
- Intraocular pressure (pre-graft or at graft)
A history of inflammation was considered to have occurred in any graft where there was a record of steroid administration during the 2-week period before transplantation, or where the grafted eye had undergone previous surgery. Inflamed in grafting and in the past Not in the past, inflamed in grafting Inflamed in the past, not in grafting Never. Recipients with elevated blood pressure at the time of transplant and in the past are compared with recipients in whom there was a history of elevated blood pressure but who had a normal blood pressure at the time of transplant, with those with elevated blood pressure at transplant but not in the past. , and with those in whom increased intraocular pressure was never recorded.
Increased intraocular pressure at any time before or during transplantation is a significant risk factor for corneal transplant failure (Log Rank Statistic=676.72; df=3; p<0.00001).
Main indication for graft
3 including corneal ectasia (63); unspecified corneal degeneration (14); rupture of Descemet's membrane (8); Salzmann's nodular dystrophy (7); unspecified change of the corneal membrane (6); macular degeneration (4); myopic degeneration (2);
Effect of specific indication for graft on graft survival
- Keratoconus, keratoconus with hydrops, Down syndrome or keratoglobus
- Bullous keratopathy
- Previous failed ipsilateral graft
- The effect of reason for failure of a previous graft
- Corneal dystrophy
The type of dystrophy does not appear to significantly affect graft survival (Log rank Statistic=6.10; df=5; p=0.296). Corneal transplants performed for an abscess or interstitial keratitis appear to have better graft survival than those for other nonherpetic infections. The consequences of corneal vascularization, inflammation, and elevated IOP before or at the time of transplantation all have a significant negative impact on graft survival.
Graft survival for bullous keratopathy, previous failed graft, burns, corneal ulcers, active HSV, infection, corneal perforations, endophthalmitis, and mycotic ulcers is relatively poor.
PROCEDURES AT TIME OF GRAFT AND INFLUENCE OF IOLs
Operative procedures at graft
- Host bed size of graft
- Accompanying procedures at graft
- Influence of vitrectomy
- Lens insertion
- Triple and staged procedures
Cataract extraction with an IOL inserted at the time of transplantation is categorized as a triple procedure. Triple procedures are defined as cataract removal and IOL insertion at the time of corneal transplantation. We observed no significant difference in graft survival between the cohort that underwent the triple procedure versus the cohort that underwent the staged procedure (Log Rank Statistic = 2.63; df = 1, p = 0.1049).
However, both cohorts show significantly better graft survival than the cohort in which a cataract was removed at the time of corneal transplantation, but who remained aphakic (Log Rank Statistic=106.26; df=1, p<0.00001).
Summary of effect of operative procedures at time of graft and influence of IOLs
POST-GRAFT EVENTS
- Reasons for graft failure
- Time to suture removal
- Post-graft complications
- Microbial keratitis/stitch abscess
- Uveitis
- Synechiae
- Post-graft vascularisation
- Post operative rise in intraocular pressure
- Rejection episodes since graft
- Post-graft operative procedures
- Refractive surgery
- Summary of post-graft events
The development of microbial keratitis or suture abscess is associated with poor corneal graft survival (Log Rank Statistic=145.73; df=1; p<0.00001). A postoperative increase in intraocular pressure is a significant risk factor for corneal graft failure (Log Rank Statistic=151.25; df=1; p<0.00001). Of the 2046 transplants in which an increase in intraocular pressure was recorded in the post-transplant period, this failed.
Postoperative surgery for glaucoma was performed on 406 (20%) of the grafts in which a postoperative increase in IOP was recorded.
VISUAL OUTCOME
Desired outcome
- Desired outcome and actual outcome achieved
Overall visual acuity
- Comparison where post-graft visual acuity is ≥6/18 or < 6/18
- Post-graft changes in visual acuity
In the latter group, there was a relative excess of transplants performed for pain relief or structural repair, bullous keratopathy, perforation or previous failed transplant, together with more aphakic, pseudophakic and elderly recipients, and more recipients who developed cystoid macular edema, retinal detachment or whose transplant has failed. Interestingly, 351 transplants reported as failure actually recorded better visual acuity at last follow-up than they did at the time of transplant. In most cases where the change in Snellen acuity after transplantation cannot be determined, the Registry is unaware of the pre-operative acuity.
Visual outcome related to presenting disease
- Keratoconus
- Fuchs' dystrophy
- Aphakic bullous keratopathy
- Pseudophakic bullous keratopathy
- Previous failed graft
- Herpetic infection
Seventy percent of all grafts performed for keratoconus achieved a posttransplant visual acuity of 6/12 or better, and 78% achieved a visual acuity of 6/18 or better. In the 88% of cases in which visual acuity was achieved both before and after transplantation, at least one line of improvement on the Snellen chart was achieved after transplantation. Twelve percent of grafts performed for aphakic bullous keratopathy achieved visual acuity of 6/12 or better, 19% achieved 6/18 or better, and 40% achieved 6/60 or better.
In 67% of cases reporting pre- and post-transplant visual acuity, at least one line of improvement on the Snellen chart was achieved post-transplant.
Factors affecting visual potential of the grafted eye
Astigmatism
Refractive surgery
- Correction following refractive surgery
- Has refractive surgery improved visual acuity?
Of the 11,737 transplants that did not undergo refractive surgery, visual acuity was provided both pre- and post-transplant for 7,501.
Triple procedures…
- Snellen acuity
- Visual improvement after graft: triple procedure
Post-graft correction
Summary of visual outcome after corneal transplantation
LAMELLAR AND LIMBAL GRAFTS
- Survival of lamellar and limbal grafts
- Desired outcome…
- Overall visual acuity
- Effect of surgeon workload
- Outcome: number of grafts performed
- Outcome: followup surgeon
- Lamellar grafts
- Limbal grafts
- Summary of factors relating to lamellar and limbal grafts
The main reasons for failure of lamellar and limbal grafts were failure of a previous corneal graft (27%), corneal scleral melting (14%) and infection (11%). Coster DJ, Williams KA, on behalf of all contributors to ACGR, The Australian Corneal Graft Registry (ACGR). Williams KA, Muehlberg SM, Lewis RF, Coster DJ, on behalf of all contributors to the Australian Corneal Graft Registry (ACGR).
Williams KA, Muehlberg SM, Lewis RF, Coster DJ, on behalf of all contributors to the Australian Cornea Registry.
COX PROPORTIONATE HAZARDS REGRESSION ANALYSIS
Methods
A multivariate model was used to investigate the combined effect of the variables on invasive graft survival, adjusted for all other variables in the model. In preliminary univariate analyses, each registered penetrating graft along with its archival follow-up records was treated as a separate and independent entity. Some patients had a history of more than one ipsilateral corneal graft, and some had a record of one or more grafts in the contralateral eye (Table 8.1).
Some variables indicated as significant in the univariate analysis were omitted due to non-convergence of the model or collinearity.
Final model
- Model 1…
- Model 2…
This model includes variables with a p-value of p<0.05, with variables eliminated stepwise, starting with the last significant variable. When comparing model 1 with model 2, there is no significant change in hazard ratios for most variables, except for some states (mainly state 2). This suggests that there is country-surgeon confounding, with some of the differences between countries explained by the performance of individual surgeons in that country.
Transformation of variables
Interpretation of the model
SUMMARY
- Grafts, contributors and era
- Corneal donors and eye banking
- Corneal graft recipients
- Operative procedures
- Causes of graft failure
- Visual outcome after corneal transplantation
- Lamellar and limbal grafts
- Risk factors for failure of penetrating grafts
In about 36% of cases where information on past history was available, the eye to be transplanted had been inflamed in the past or was inflamed at the time of transplantation. A penetrating corneal graft that is never cleaned and thinned in the immediate postoperative period is considered to be a nonfunctional primary graft. In 70% of cases, the only desired outcome for the corneal transplant was improved vision in the transplanted eye.
An intraocular lens was placed in the grafted eye in about 33% of the group.
METHODS AND DEFINITIONS
- Entry and follow-up
- Definition of risk factors
- Definition of graft failure, rejection and complications
- Statistical analyses
- Computer hardware and software
- Corneal graft registration form
- Corneal graft followup form
- References
Any existing graft that is replaced by another in the same eye, regardless of graft clarity and for whatever reason, is classified as a failed graft. For surviving transplants, trial time is calculated as the time between the date of transplant and the date the patient was last seen. For failed transplants, trial time is calculated as the time between the date of transplant and the date of failure.
If this is a repeat transplant in the grafted eye, please give the reason and date of failure of previous transplant.
ACGR PUBLICATIONS
Journal articles
Reports
