Repeat corneal transplantation in Southern China: Indications, surgical technique, outcomes, and risk factors for repeat keratoplasty failure

Abstract

Purpose:

To report the indications, surgical techniques, and outcomes of repeat keratoplasty and evaluate the risk factors for graft failure in the Chinese population.

Methods:

The medical records of 216 patients (243 cases) who underwent at least two keratoplasties at a leading eye hospital in southern China between 2011 and 2020 were retrospectively reviewed. Indications and surgical procedures for repeat corneal transplantation were analyzed. Kaplan–Meier survival analysis was used to determine the graft survival rate after repeat keratoplasty. A multivariable survival model was used to assess the risk factors.

Results:

Repeated keratoplasties increased continuously from 2011 to 2020 (P = 0.002). The most common primary indication was infectious keratitis (38.7%), and the most common reason for repeat keratoplasty was graft rejection (30.04%). Regraft techniques included penetrating keratoplasty (PK) in 165 cases (67.9%), deep lamellar keratoplasty (DALK) in 52 cases (21.40%), and endothelial keratoplasty (EK) in 26 cases (10.7%). Median survival was 5.3, 6.8, and 6.4 years for PK, DALK, and EK, respectively. The 5-year survival rate was 53.5%, 66.6%, and 69.8% for PK, DALK, and EK, respectively. The median LogMAR visual acuity was 1.4 for PK, 0.75 for DALK, and 1.2 for EK at the end of the follow-up. Multivariate analysis revealed that graft rejection is a risk factor for repeat keratoplasty failure (P = 0.002).

Conclusions:

DALK and EK may provide better outcomes than PK in treating graft failure. Preventing and treating postoperative graft rejection may be key to improving regraft survival. These findings will aid in the management of failed corneal grafts.

Corneal blindness can cause reversible blindness and can be managed by healthy donor cornea transplantation. With the use of keratoplasty and increasing life expectancy, regrafting is now a common indication for keratoplasty in developed countries[1,2,3] and has been on the rise in developing countries[4,5] in recent decades. Repeat keratoplasty is technically more challenging and has lower graft survival rates than initial keratoplasty procedures, with graft survival rates ranging from 28% to 81% after repeat corneal transplantation.[6,7,8,9] Few studies on repeat corneal transplantation outcomes and risk factors have been conducted.[9,10] Studies on repeat keratoplasty in China are limited, especially regarding the outcomes and risk factors for graft failure in repeat corneal transplants.

We aimed to quantify the number of repeat keratoplasties performed at one of the largest corneal transplant centers in China to analyze the indications and surgical techniques and to investigate the outcomes and factors influencing the risk of repeat graft failure.

Methods

Study subjects

Cases of repeat keratoplasty performed at a leading eye hospital in southern China from January 1, 2011, to December 31, 2020, were included. Eyes with two or more repeat keratoplasties were categorized as having multiple repeat keratoplasties. The demographic data, primary indications, direct reasons, and surgical procedures for repeat keratoplasty were reviewed retrospectively. The length of time between the primary graft and the regraft and the size of the regraft were recorded. Ophthalmologic procedures, including glaucoma and cataract surgery, ocular comorbidities, including glaucoma and corneal neovascularization, and systemic conditions, including hypertension and diabetes, were also noted. Postoperative graft clarity was also recorded at every visit.

The primary outcomes were graft survival and best-corrected visual acuity (BCVA) after repeat keratoplasty and the risk factors associated with regraft failure. Rates of repeat keratoplasty and graft complications were also evaluated. This study adhered to the tenets of the Declaration of Helsinki and was approved by our institutional ethics committee.

Classification and definition

The primary indications for keratoplasty in this study were categorized as follows: infectious keratitis, non-infectious keratitis, bullous keratopathy, corneal malnutrition, keratoconus, corneal leukoma, corneal trauma, corneal tumors, and other causes. Bacterial, viral, fungal, amoebic, and unknown pathogens cause infectious keratitis. Corneal dystrophies include Fuchs’ endothelial dystrophy and other dystrophies involving the corneal stroma. Corneal leukoplakia characterizes congenital corneal opacities and scarring. Corneal trauma can include thermal, mechanical, or chemical injury.

The direct reasons for repeat keratoplasty were as follows: primary condition recurrence, graft dissolution, opacification, rejection, endothelial decompensation, graft infection (excluding primary condition recurrence), trauma, and unspecified indications. Graft rejection was diagnosed as graft edema occurring more than 2 weeks postoperatively and the presence of at least one of the following signs: endothelial rejection lines, infiltrating corneal debris, or anterior segment inflammation. Endothelial decompensation was defined as an irreversible decrease in endothelial cell density and loss of graft clarity.

Three keratoplasty procedures were performed: penetrating keratoplasty (PK), deep lamellar keratoplasty (DALK), and endothelial keratoplasty (EK).

Graft failure was defined as the irreversible central opacity of a previously clear corneal graft postoperatively. Graft survival was defined as the time to graft failure or end of follow-up.

Statistical analysis

Statistical analyses were performed using SPSS version 24.0 (IBM Corp., Armonk, NY, USA). Categorical data are presented as frequencies (n) and percentages (%). Visual acuity was converted into LogMAR for statistical analysis. Baseline characteristics were compared for the three surgical techniques by using Fisher’s exact test for count data, and the Mann-Whitney test was performed for continuous data. The Pearson correlation coefficient was used to analyze trends in the annual number of repeat keratoplasties. Graft survival rates were analyzed using Kaplan–Meier survival curves, and the differences between groups were compared using the log-rank test. A Wilcoxon signed rank test was performed to compare pre- and postoperative VA and for different types of regrafts. We applied the Cox proportional hazards model to determine the hazard ratio (HR) and corresponding 95% confidence interval (95% CI) to identify potential risk factors associated with the subsequent failure of repeat keratoplasty. A two-sided P value of <0.05 was considered statistically significant.

Results

Demographics

In total, 243 cases of repeat corneal transplantation involving 216 patients were included, of whom 142 were male (65.74%) and 74 were female (34.26%). The median age at the time of surgery was 56 (interquartile range [IQR]: 43–66) years. The median interval between the first and repeat keratoplasty was 2 (IQR: 0.5–5) years. The median size of regrafts was 8 (IQR: 7.75–8 mm) mm. The above-baseline characteristics were comparable between the three surgical techniques [Supplemental Table 1]. The median follow-up duration was 17 (IQR: 7–3) months. Among the 243 repeat corneal transplant cases, 202 (83.13%) underwent the procedure once, 33 (13.58%) twice, seven (2.88%) three times, and one (0.41%) four times.

Supplemental Table 1.

Baseline characteristics comparing repeat PK, DALK, and EK

Characteristic		PK		DALK		EK		P
Median age (y)		58		50		53		0.08
Sex (male:female)		112:165		32:20		14:12		0.330
Median (IQR) duration between first and repeat graft (y)		2 (1.2–5)		1.5 (0.7–5)		2.2 (1–4)		0.120
Size of regraft (mm)		8 (7.7–8)		7.7 (7.7–8)		8 (7.7–8)		0.687

PK: Penetrating keratoplasty, DALK: Deep lamellar keratoplasty, endothelial keratoplasties EK: Endothelial keratoplasty, IQR: Interquartile range

Primary indications and direct reasons for repeat corneal transplantation

As presented in Table 1, the most common primary indications were infectious keratitis (n = 94, 36.68%), including viral keratitis (n = 67, 27.57%), fungal keratitis (n = 13, 5.35%), bacterial keratitis (n = 11, 4.53%), and acanthamoeba keratitis (n = 3, 1.23%). Corneal leukoma was the second most common primary indication (n = 48, 19.75%).

Table 1.

Primary indications and direct reasons for repeat keratoplasty

Primary indications		n (%)		Direct reasons		n (%)
Infectious keratitis		94 (36.68)		Graft rejection		73 (30.04)
Corneal leukoma		48 (19.75)		Endothelial decompensation		54 (22.22)
Bullous keratopathy		33 (13.58)		Recurrence		41 (16.87)
Corneal malnutrition		24 (9.88)		Graft infection		38 (15.63)
Corneal trauma		17 (7)		Graft opacification		20 (8.23)
Non-infectious keratitis		11 (4.53)		Graft dissolution		14 (5.76)
Keratoconus		10 (4.12)		Others		3 (1.23)
Corneal tumors		3 (1.23)
Others		3 (1.23)

The direct reasons for repeat corneal transplantation are listed in Table 1, with graft rejection being the most common reason for regrafting (n = 73, 30.04%), followed by primary corneal endothelial decompensation (n = 54, 22.22%).

Primary indications and direct reasons for multiple regrafts are listed in Supplemental Table 2. The most common primary indications were infectious keratitis and bullous keratopathy, while the most common reason for regrafting was graft rejection.

Supplemental Table 2.

Primary indications and direct reasons for multiple regrafts

Primary indications		n (%)		Direct reasons		n (%)
Infectious keratitis		11 (26.83)		Graft rejection		16 (39.02)
Bullous keratopathy		11 (26.83)		Endothelial decompensation		9 (21.95)
Corneal leukoma		7 (17.07)		Recurrence		8 (19.51)
Corneal malnutrition		5 (12.20)		Graft infection		3 (7.32)
Corneal trauma		5 (12.20)		Graft opacification		2 (4.88)
Keratoconus		1 (2.44)		Others		2 (4.88)
Corneal tumors		1 (2.44)		Graft dissolution		1 (2.44)

Procedures of corneal transplantation

Between January 2011 and December 2020, 2577 corneal transplants were performed, with 243 (9.42%) being repeat corneal transplants. The annual count of repeat corneal transplantations significantly increased during the study period (r = 0.853, P = 0.002; Fig. 1). PK remained the primary surgical approach for repeat corneal transplants, while the proportion of EK showed an increasing trend (r = 0.520, P = 0.123; Fig. 1), and DALK showed a decreasing trend (r = −0.415, P = 0.233).

Figure 1.

(a) Annual number of repeat PK, DALK, and EK from 2011 to 2020. (b) The proportion of EK in repeat keratoplasty demonstrated an increasing trend

PK was performed in 165 cases (67.90%), DALK in 52 (21.40%), and EK in 26 (10.70%). In cases of PK failure, the predominant choice for subsequent repeat corneal transplantation was PK, followed by EK. For DALK failure, the preferred procedure for subsequent repeat corneal transplantation was DALK, followed by PK. For EK failure, PK was the primary choice for subsequent repeat corneal transplantation, followed by EK [Supplemental Table 3].

Supplemental Table 3.

Surgical procedures for primary and repeat keratoplasty

Primary keratoplasty		Repeat keratoplasty						Total
		PK		DALK		EK
PK		129 (85.43%)		4 (2.65%)		18 (11.92%)		151 (100%)
DALK		28 (28.57%)		48 (62.33%)		1 (1.30%)		77 (100%)
EK		8 (53.33%)		0 (0.00%)		7 (46.67%)		15 (100%)

PK: Penetrating keratoplasty, DALK: Deep lamellar keratoplasty, endothelial keratoplasties EK: Endothelial keratoplasty

Complications of repeat corneal transplantation

During the follow-up period, graft rejection was the most common complication in 62 cases (25.51%). The interval between transplantation and rejection ranged from 0.66 to 92 months, with a mean interval of 16.85 ± 18.48 months. Among the rejected cases, graft clarity after anti-rejection treatment was achieved in 28 (45.16%) patients. However, graft failure occurred in 34 cases despite aggressive anti-rejection treatment. We analyzed the relationship between glaucoma, corneal neovascularization, concurrent surgery, systemic disease, and rejection. Cases with corneal neovascularization exhibited higher rates of rejection than those without corneal neovascularization (P = 0.028).

Graft failure occurred in 49 cases after repeat PK. The causes of failure included 26 (53.06%) cases of rejection, 12 (24.49%) cases of graft opacification, eight (16.33%) cases of infectious keratitis, two (4.08%) cases of graft dissolution, and one (2.04%) case of endothelial decompensation. Graft failure occurred in 16 cases after repeat DALK. The causes of failure included five (31.35%) cases of graft opacification, five (31.35%) cases of rejection, three (18.75%) cases of infectious keratitis, two (12.50%) cases of graft dissolution, and one (6.25%) case of endothelial decompensation. Graft failure occurred in six cases after repeat EK. The causes of failure included three (50%) cases of rejection, two (33.33%) cases of graft opacification, and one (16.67%) cases of endothelial decompensation.

The second most common complication was elevated postoperative intraocular pressure (IOP), which occurred in 37 cases (15.22%). IOP could be controlled with medication in most cases, except in three patients who underwent vitrectomy for intraocular inflammation.

Outcomes and risk factor

Overall median survival was 6 years, with a survival rate of 84.5% at 1 year and 59.1% at 5 years. The first regrafts survived for a median of 6 years, and multiple regrafts survived for 3.3 years. The Kaplan–Meier curve showed that the cumulative survival proportion was similar for the first and multiple regrafts during the first year after transplantation. Thereafter, it declined faster for multiple regrafts, from 62.7% for the first regrafts to 35.3% for multiple regrafts at the end of 5 years [Fig. 2]. However, the difference was insignificant according to the log-rank test (P = 0.191).

Figure 2.

Kaplan–Meier cumulative proportion of graft survival in the first and multiple regrafts

The outcomes of three surgical techniques are listed in Supplemental Table 4. Median survival was 5.3 years for PK, 6.8 years for DALK, and 6.4 years for EK. The Kaplan–Meier curve showed that the cumulative survival proportion was 87.9% for PK during the first year after transplantation, slightly higher than that for DALK (75.2%) and EK (82.7%). Subsequently, graft survival declined faster for PK and was 53.5% for PK, 66.6% for DALK, and 69.8% for EK at the end of 5 years. At 7 years, graft survival remained stable for DALK and EK but dropped to 28.5% for PK [Fig. 3]. However, the log-rank test did not show a significant difference (P = 0.862), probably because of different trends over time.

Supplemental Table 4.

Outcomes of repeat PK, DALK, and EK

Surgical techniques		1-year survival (%)		5-year survival (%)		Survival time (y)		LogMAR BCVA (IQR)
PK		87.9%		53.5%		5.3		1.4 (0.7–2.8)
DALK		75.2%		66.6%		6.8		0.75 (0.5–1.7)
EK		82.7%		69.8		6.4		1.2 (0.7–2.2)

PK: Penetrating keratoplasty, DALK: Deep lamellar keratoplasty, endothelial keratoplasties EK: Endothelial keratoplasty, IQR: interquartile range

Figure 3.

Kaplan–Meier cumulative proportion of graft survival of repeat PK, DALK, and EK

All cases had preoperative visual acuity records with a median BCVA of LogMAR 2.9. However, only 158 cases had postoperative BCVA records. For the cases that had both pre- and postoperative BCVA records, median BCVA improved from LogMAR 2.9 to 1.3 (P < 0.001) at the end of the follow-up. The median LogMAR BCVA was 1.4 for PK, 0.75 for DALK, and 1.2 for EK at the end of the follow-up. BCVA in DALK was significantly better than that in PK.

Cox regression analysis was used to analyze the risk factors associated with repeat keratoplasty failure [Table 2]. In the multivariate regression analysis, graft rejection was a risk factor for repeat keratoplasty failure (HR = 2.623; 95% CI: 1.427–4.826; P = 0.002).

Table 2.

Risk factors for regraft failure

		Univariate				Multivariate
		HR		P		HR (95%CI)		P
Gender		1.093		0.738
Age		1.004		0.557
Primary indication		0.827		0.480
Regraft number		1.559		0.161
Duration between primary and repeat graft		0.942		0.824
Graft size		0.580		0.938
Systemic disease		0.980		0.953
Previous glaucoma		0.296		0.091
Corneal neovascularization		0.970		0.910
Concurrent surgical procedure		0.636		0.332
Postoperative high IOP		0.934		0.833
Graft rejection		2.160		0.003		2.624 (1.427–4.826)		0.002

HR: Hazard ratio, IOP: Intraocular pressure

Discussion

Repeat keratoplasty has become increasingly common in developed and developing countries, including China, over the last decade. This study is the first to evaluate graft survival and risk factors for graft failure after long-term repeat keratoplasty in a Chinese population.

We reviewed the annual data of keratoplasties performed at a leading eye hospital in southern China over the past 10 years, with 2577 grafts, including 243 regrafts. The overall percentage of repeat corneal transplants was 9.4%, which was slightly higher than the proportion of regrafts performed in other large centers in China.[4,11] When compared to New Zealand’s 18.3%,[10] the United Kingdom’s 21.1%[12] and Canada’s 17.1%,[13] the proportion of repeat corneal transplants in China was relatively low. However, this study shows that the annual regrafting rate gradually increased from 2011 to 2020. As donor shortages ease and surgical techniques improve, repeat corneal transplants are expected to increase over the next decade.

In this study, infectious keratitis was the most common primary indication for keratoplasty, followed by corneal leukoma, accounting for 38.68% and 19.75% of cases, respectively. This trend is reasonable because the most common indications for keratoplasty in China have been infectious keratitis and corneal leukoma over the last three decades.[11,14] Keratoconus in New Zealand[10] and keratoconus and corneal dystrophy in France[15] have been reported to be the most common primary indications for repeat corneal transplantation, accounting for only 4.12% and 9.88% of the indications in this study, respectively, indicating that there are significant differences in the primary indications for repeat corneal transplantation between developing and developed countries. The most common reasons for repeat keratoplasty were graft rejection (30%) and endothelial decompensation (22.2%). This pattern has been similarly reported in both developing and developed countries.[10,16]

The majority of repeat keratoplasty procedures were PK (165, 67.9%), and the smaller were DALK (52, 21.4%) and EK (26, 10.7%), which is consistent with the sequence of primary corneal transplants. The number of EK surgeries performed increased between 2012 and 2020. Among regraft procedures following primary PKs, regraft PKs account for 85.43% of cases, follwed by regraft EKs at 11.92%. This trend agreed with the repeat corneal transplant surgical techniques reported by Dipika for New Zealand.[10] EK has been used as a method of repeat keratoplasty in developed countries since 2015.[17] The growing trend of EK surgery was also reported by an ophthalmology center in eastern China.[16] Our results show that EK has become an alternative to PK after failed PK surgery in China.

The mean survival period in this study was 5.1 years, with survival rates of 84.5% at 1 year and 59.1% at 5 years. To our knowledge, our study is the first on the survival rate after long-term repeat corneal transplantation in Chinese patients. Chen et al.[18] reported a 1-year survival rate of 85.2% and a 5-year survival rate of 79.2% for short-term repeat corneal transplants (repeat transplant intervals of 1 year or less) in eastern China. This implies that long-term graft survival may be higher for short-term repeat corneal transplants than long-term ones. This potential disparity could be because long-term graft failure tends to be related to chronic causes, including graft rejection or endothelial dystrophy, which, as in our study, may reduce repeated graft success rates.

Few studies have analyzed graft survival after repeat keratoplasty in PK and reported that the long-term survival of repeat PK varies from 50% to 66% at 5 years.[7,8,10] Very limited data have evaluated the outcomes of repeated ALK and PK. The UK National Transplant Registry study showed that repeat PK was associated with worse long-term graft survival than repeat DALK.[12] More studies have compared the outcomes of secondary PK and EK and reported conflicting results. The Australian Corneal Graft Registry reported that repeat PK had superior graft survival compared with repeat DALK.[19] In contrast, Ang et al.[20] showed that repeat PK had poorer survival rates than EK in Singapore. A recent report from the UK Transplant Registry found no difference in graft survival between secondary PK and EK.[12] We also compared the graft survival of the three corneal transplant techniques. Similar to previous studies, short-term survival for DALK and EK was lower than that for PK.[15,21] This may be related to the technical difficulties associated with DALK and EK, resulting in high graft failure rates in a short period. However, long-term survival was better for DALK and EK than for PK, and the regraft survival time was longer for DALK and EK than for PK; however, the differences were not statistically significant. These results demonstrate that DALK and EK can be used as alternatives to PK surgery for failed corneal grafts, potentially providing better long-term outcomes while avoiding the disadvantages of PK.

Several studies have demonstrated that survival after keratoplasty decreases with repeat keratoplasty procedures.[10,22,23] In this study, we found a decrease in survival time with repeat keratoplasties. However, the cumulative survival proportion was similar for the first and multiple regrafts during the first year after transplantation. Thereafter, it declined faster for multiple regrafts at 35.3% for multiple regrafts and 62.7% for the first regrafts at 5 years. The number of regrafts had a hazard ratio of 1.559 in the univariate analysis, but this did not reach statistical significance (P = 0.161).

We analyzed the risk factors for repeat keratoplasty and found that graft rejection was associated with an increased risk of graft failure. Graft rejection was the most common postoperative complication, occurring in 62 cases (25.51%), with 34 cases ultimately failing despite aggressive anti-rejection therapy. The response rate in our study was more optimistic than the reports by Bersudsky et al.,[9] who found that almost all repeat grafts did not respond to conservative treatment and ultimately failed. Most international studies have shown that graft rejection is a risk factor for graft failure.[9,10,24,25] Ethnicity is thought to be associated with transplant rejection; for example, non-European and black people are more likely to demonstrate graft rejection.[3,24] Our findings revealed that graft rejection is the most important risk factor for repeat corneal graft failure in Asian populations. We also found that patients with corneal vascularization were at a higher risk of rejection, which was consistent with the results of other studies.[26,27] The second most common postoperative complication was elevated IOP (18.9%). In most cases, IOP can be controlled with medication, and elevated IOP is not a risk factor for graft failure in the regression analysis.

This study had some limitations. First, it was retrospective and had limitations such as unequal follow-up times and loss of follow-up. Patients in remote areas or those who recover well from corneal transplantation are more likely to be lost. Therefore, there may have been undetected or unreported rejections or failures in this study. Second, because of the high dropout rate, the median follow-up period was 17 months, which may have led to overestimating the 5-year graft survival rate of repeat grafts. Third, the number of eyes undergoing DALK and EK was lower than those undergoing PK. Therefore, statistical errors may have affected the results of the comparisons between subgroups and should be interpreted with caution.

Conclusion

This study represents the first comprehensive examination of indications, surgical techniques, outcomes, and risk factors for repeat corneal transplantation in a Chinese population. Infectious keratitis emerged as the primary leading indication, while graft rejection was the most common reason for repeat keratoplasty. PK is the major surgical procedure used for repeated corneal transplants. However, there is potential for DALK and EK to offer better outcomes and be used more widely to treat graft failure. Specific risk factors for repeat keratoplasty failure in a Chinese population were identified. Preventing and treating postoperative corneal graft rejection may be key to improving the survival of repeat corneal transplants.

Financial support and sponsorship

This study was funded by the National Key R and D Program of China (grant no. 2019YFC0840708).

Conflicts of interest

There are no conflicts of interest.