Impact of venting incision on graft detachment in Descemet stripping automated endothelial keratoplasty.

PURPOSE: To evaluate the impact of venting incision on graft detachment in Descemet stripping automated endothelial keratoplasty (DSAEK).
MATERIALS AND METHODS: Retrospective comparative study on DSAEK procedure performed with (n = 13) and without (n = 39) venting incision. Patients were treated with DSAEK for various causes of endothelial decompensation between February 2014 and April 2016. The rate of postoperative graft detachments documented on anterior segment optical coherence tomography was compared in both groups.
RESULTS: Paracentral venting incision decreases the rate of graft displacement. Non-venting group (nvg) had 28.2% graft displacement compared to venting group (vg) which had 7.6% graft displacement.
CONCLUSION: Venting incision is a simple and safe manoeuvre to facilitate graft adherence in DSAEK, especially in caseswith presence of high-risk factors for graft detachment.

Introduction

Descemet stripping automated endothelial keratoplasty (DSAEK) offers several advantages over penetrating keratoplasty, including faster visual recovery and increased refractive predictability.[12] However, early graft detachment is a considerable challenge, accounting for an average of 14.5% of early complications after DSAEK.[3456] Various attempts to surmount this complication have been made, including sweeping across the host corneal surface to expel interface fluid, roughening of the host stroma, and use of air–fluid exchange systems.[78910] Venting incisions, in the form of four equally spaced mid-peripheral incisions with the use of 15° blade in the recipient cornea to the graft interface, were first described by Price and Price.[11] They and other authors showed that the incisions facilitated graft adhesion by draining interface fluid.[1213] The present study aims to evaluate the impact of venting incision on graft detachment in DSAEK by comparing patients who have undergone DSAEK for various reasons with and without venting incisions.

Materials and Methods

It was a retrospective comparative consecutive case series of DSAEK performed in 52 eyes for various causes of endothelial decompensation from February 2014 to April 2016 by 3 surgeons (1 cornea surgeon for venting group [vg] who did all the DSAEK procedures with venting incision and 2 cornea surgeons for the non-venting group [nvg]). Patients in need of DSAEK were randomly distributed among the three surgeons and were divided into two groups those with (n = 13) and without (n = 39) venting incision. Indications were pseudophakic bullous keratopathy (n = 33; vg = 61%, nvg = 64%), aphakic bullous keratopathy (n = 4; vg = 7.69%, nvg = 7.69%), Fuchs' dystrophy (n = 4; vg = 7.69%, nvg = 7.69%), congenital hereditary endothelial dystrophy (n = 2; vg = 0%, nvg = 5%), failed graft (n = 6; vg = 7.7%, nvg = 12.8%), pseudophakic bullous keratopathy + anterior chamber intraocular lenses (ACIOL) (n = 1; vg = 0%, nvg = 2.5%), pseudophakic bullous keratopathy Status post (S/P) trabeculectomy (n = 1; vg = 0%, nvg = 2.5%), and S/P Descemet membrane detachment (S/P C3F8) (n = 1; vg = 7.69%, nvg = 0%). Thus, indications for DSAEK were not strikingly different in two groups though the sample size was different. All patients underwent full ophthalmic evaluation including visual acuity, refractive error, slit-lamp examination, applanation tonometry, indirect ophthalmoscopy, ultrasound, pre- and postoperative anterior segment optical coherence tomography (ASOCT), Pre- and postoperative astigmatism, and complications were also noted. The rate of postoperative graft detachments was compared and analyzed in both the groups (substantiated with ASOCT) [Figures 1 and 2].

Figure 1

Anterior segment optical coherence tomography images of nonventing group. (i, ii, iii): graft detachment on the first postoperative day (a) and their correspondingimages showing graft attachment after rebubbling (b)

Figure 2

Anterior segment optical coherence tomography image of graft detachment in venting group. (a) First postoperative day (b) postrebubbling

Common surgical technique

An 8–8.5 mm epithelial debridement was done followed by Descemet's membrane stripping with inverse Sinskey hook. Surgical peripheral iridectomy was done in all the cases. The size of endothelial graft (endolenticule) was same for both the groups. Endolenticule preparation was done with 350 μ blade with Moria ALK system, so the range of endolenticule thickness was comparable in both the groups (All the tissues were preserved in MK media). Endolenticule insertion was done. Air fill was complete in both the groups. All the patients were advised a strict supine position right from shifting them from OT complex till postoperative day 1.

Surgical variation for venting group

Midperipheral, perpendicular, transcorneal, radial 3–4 venting incisions at 0°, 90°, 180°, and 270° were made with 20G MVR (Ovation International, India) [Figures 3 and 4] after surgical peripheral iridectomy. After complete air fill, interface fluid expression was facilitated by corneal massage and sweeping the corneal surface with a sponge [Video 1].

Figure 3

Anterior segment optical coherence tomography image showing venting incision in Descemet stripping automated endothelial keratoplasty

Figure 4

Clinical photograph showing venting incision in Descemet stripping automated endothelial keratoplasty in diffuse illumination (a) and parallelepiped illumination (b).

Scleral fixated IOL part was done by vitreoretinal surgeon wherever needed.

Statistical analysis

For analysis, the participants were divided into two groups: vg and nvg. Age, best-corrected visual acuity (BCVA), pre- and post-operative astigmatism, and rate of rebubbling were analyzed by comparing the means between groups. Graft adherence in combined cases was also analyzed in both the groups. Non-normally distributed data were logarithmically transformed. Differences in graft detachment and rate of rebubbling were evaluated with Chi-square test for trend. A Chi-square test was also done to test the association between vg and nvg on graft adherence in combined cases.

Data were presented as mean ± standard deviation (SD) with a 95% confidence interval (95% CI) and relative risk (RR), respectively.

Statistical analyses were performed with Stata®/MP 14 Software (Developer: StataCorp), testing the null hypothesis of identical groups. Significance level was set at a P = 0.05.

Yates's continuity correction was also carried out to prevent overestimation of statistical significance of small data. However, it did not have much impact.

Results

Out of 52 patients, venting incisions were made in 13 patients. The remaining 29 procedures were performed without venting incisions. The age group of 11–81 years was included. Vg included 6 males and 7 females and nvg included 24 males and 15 females. Follow-up period was 1–32 months, mean follow-up period being 12.3 months.

Nvg had 28.2% graft displacement [Figure 1: ia, iia, iiia] compared to vg which had 7.6% graft displacement (substantiated by ASOCT) [Figure 2a] though the association was not found to be statistically significant on Pearson's Chi-squared test (P = 0.128) due to very small sample size in vg who underwent rebubbling. Interface gap was measured in all cases of graft displacement. In nvg, rebubbling [Figure 1: ib, iib, iiib] was done in 0–3 days in 4 patients, 4–7 days in 6 patients, and >1 week in 1 patient. In vg, rebubbling was done in only 1 patient on the 2nd postoperative day [Figure 2b].

There were five cases (4 aphakic bullous keratopathy and 1 pseudophakic bullous keratopathy s/p complicated pseudophakia) of combined intraocular lens (IOL)explantation + scleral-fixated IOL + DSAEK. Out of 3 cases in nvg, 2 showed graft displacement, whereas in vg (n = 2), graft remained adherent in both the cases. Thus, the rate of graft displacement in high-risk group was less in vg though the association was not found to be not statistically significant (P = 0.128).

Endothelial rejection was seen in two patients with venting incision and one patient of nvg. Graft failure was eventually seen in four cases of nvg compared to one patient of vg. Four patients of nvg showed secondary glaucoma whereas the same was not found in any of the patients of vg.

None of the patients in either group showed epithelial ingrowth or infectious keratitis in our study.

A Chi-square test was done to find the association of astigmatism on two groups. The resulting P value was found to be 0.437 showing that the effect of astigmatism was comparable in both the groups. There was no statistically significant difference in pre- or postoperative astigmatism between vg and nvg. Furthermore, BCVA turned out to be statistically uniform between the two groups.

Discussion

The eye is a closed chamber with intraocular pressure equalized throughout during the air compression phase of DSAEK. Increased air compression pressure will have no effect on the rate of interface fluid dispersion once a steady pressure is achieved throughout. Initial inflation to a higher pressure may drive more fluid from the interface as a function of shape change due to corneal elasticity. However, sustained high-pressure tamponade will make little or no difference to the volume of retained interface fluid. Venting incisions create pressure differential across the donor lenticule (anterior chamber pressure on the endothelial side and atmospheric pressure on the stromal side).[13]

We have placed venting incision before endolenticule insertion itself, so once endolenticule is inserted, and during AC air fill, interface fluid gets drained off automatically which could be observed as fluid droplet at the venting incision immediately after AC air fill.

A study done by Hovlykke et al.[14] showed that venting incisions done in 266 DSAEK patients did not significantly alter BCVA, astigmatism, Contrast Sensitivity Function (CSF) aberrations, or reduce the risk of graft detachment in triple or nontriple procedures. Our study goes well with this study in terms of postoperative BCVA and astigmatism. The only difference is that they used 23G MVR whereas we used 20G MVR (Ovation International, India) for making vents.

Venting incision during case series of 111 DSAEK Asian patients achieved a low postoperative dislocation rate of 2.2% in a study done by Mehta and Tan[15](Asian eyes are smaller with shallow AC and thick brown irides as compared to Caucasian eyes).

Knecht et al.[16] found that venting incisions resulted in the removal of interface fluid in only two of six cases assessed using intra-operative ASOCT.

Although difference in adherence rate was statistically not significant, two of in vitro studies[1213] concluded to have beneficial effect of venting incision. So to assess whether this benefit transcends in vivo cases, the study has been carried out. Medline and literature search showed only one study to assess the impact of venting incision on detachment rate. No study in Asian population has been carried out to either support or refute in vitro studies till date. Because of multiple dynamic factors playing a role in vivo and because of limited sample size, we could not get statistically significant difference.

In a case series of 150 patients undergoing DSEK with vent incisions done by Hannush et al.,[17] the occurrence of deep infectious keratitis was 2%. However, they hypothesize that these patients would have developed their infectious keratitis whether the vent incisions were present or not.

Suh et al.[18] described five cases of epithelial ingrowth after DSAEK. All five cases involved corneal or limbal incisions that may have provided the entryway for epithelial cells. However, our study did not show epithelial ingrowth in any of the cases of either group.

Limitations of our study are that there was a small sample size. Sample size was variable in both the groups. It was not a single surgeon study though all the surgeons were in their same learning curves.

Conclusion

Venting incision is a simple and safe manoeuvre to facilitate graft adherence in DSAEK particularly in cases with presence of high-risk factors for graft detachment.

Financial support and sponsorship

Nil.

Conflicts of interest

There are no conflicts of interest.