Methods

In this consecutive interventional case series, we compiled the data of consecutive patients undergoing DALK for moderate (48 diopters [D] < mean keratometry < 55 D) to advanced (mean keratometry ≥55 D or immeasurable keratometry) keratoconus between January 14, 2004 and February 9, 2013. Indications for surgery included contact lens intolerance or poor corrected visual acuity. Ethics Committee approval was obtained from the Ophthalmic Research Center, affiliated to Shahid Beheshti University of Medical Sciences, Tehran, Iran, to use the patients’ data. Informed consent for tissue donation was obtained from patients’ relatives, and the study adhered to the tenets of the Declaration of Helsinki.

Keratoconus was diagnosed clinically, based on slit-lamp findings (stromal thinning, Fleischer ring, and Vogt striae) and keratometry, and confirmed by elevation topography (Orbscan II system; Bausch & Lomb, Rochester, New York, USA). Patients with a history of vernal keratoconjunctivitis; other ocular diseases such as cataract, glaucoma, and retinal disease; any previous ocular surgery; or graft opacity were excluded from this study. Additionally, no deep stromal scar or defect in the DM indicating previous hydrops was observed. Complete preoperative ocular examinations were performed, including uncorrected visual acuity (UCVA) and best-spectacle corrected visual acuity (BSCVA) using the Snellen acuity chart, slit-lamp examination, tonometry, dilated funduscopy, and manifest refraction (when possible). The Orbscan II system was used to measure central and peripheral corneal thickness. Anterior chamber depth and vitreous length measurements were performed using A-scan biometry (A/B scan; Sonomed Inc, Lake Success, New York, USA).

All eyes were operated on by a single experienced anterior segment surgeon (M.A.J.) using general anesthesia and the big-bubble technique, as described in detail elsewhere. The size of the recipient trephine was 2.50 mm less than the vertical corneal diameter. After trephination to approximately 80% of the peripheral corneal thickness with a Hessburg–Barron suction trephine (Katena, Denville, New Jersey, USA), a 27 gauge needle was inserted into the stroma to the center of the cornea. Air was gently injected into the mid stroma until a big bubble was formed extending to the border of the trephination. After big-bubble formation, debulking of the anterior two-thirds of the corneal stroma was performed with a crescent blade (Alcon Laboratories, Fort Worth, Texas, USA). Thereafter, a peripheral paracentesis was performed to reduce intraocular pressure, and the bubble was punctured with a 15-degree slit knife (Alcon Laboratories). Viscoelastic material (Coatel; Bausch & Lomb, Waterford, Ireland) was injected to protect the DM from manipulation, and the rest of the corneal stroma was completely excised. The viscoelastic material was then completely washed out before proceeding to graft suturing. In cases in which the big bubble could not be created after several attempts, manual stromal dissection down to the DM was progressively performed using a crescent knife. As much corneal stroma as possible was removed to create a smooth recipient bed with uniform thickness. In both groups, a donor cornea, oversized by 0.25 mm and without the DM and endothelium, was punched from the endothelial side with the Barron punch (Katena) and then sutured to the recipient bed using a combined suturing technique consisting of a 16-bite single running and 8-bite interrupted 10-0 nylon sutures. The tension of running and separate sutures was adjusted intraoperatively with a handheld keratoscope. At the conclusion of the operation, cefazolin 100 mg and betamethasone 4 mg were injected subconjunctivally.

Follow-up examinations were scheduled 1, 3, 7, and 30 days and 3, 6, and 12 months postoperatively and at least 3 months after complete suture removal and every 6 months thereafter. Postoperatively, the patients were medicated with topical chloramphenicol every 6 hours for 30 days and topical betamethasone 0.1% every 6 hours gradually tapered over 5 months. Premature suture removal was performed for any suture-related problems such as loosening or suture-tract vascularization. For control of postoperative astigmatism >4 D, the tension of running sutures was adjusted and interrupted sutures were selectively removed starting 2 months after the operation, based on the keratometry readings. Selective interrupted suture removal was continued over the follow-up period until an acceptable keratometric astigmatism was achieved. Afterward, the remaining sutures were left in place until they became ineffective. All sutures had been removed by the time of the final examination. The postoperative visual and refractive outcomes evaluated in this study consisted of UCVA and BSCVA (expressed in logarithm of the minimal angle of resolution), mean keratometry, keratometric astigmatism, and manifest refraction. The topography was performed at least once after complete suture removal in all cases, and more often if deemed necessary by the surgeon.

Results

A total of 194 consecutive eyes (102 right eyes) from 181 patients (128 male patients) with keratoconus underwent DALK during the study period. The mean patient age was 27.9 ± 8.2 years (range: 14–52 years). Thirty-four eyes (17.5%) had moderate keratoconus, and severe keratoconus was observed in 160 cases (82.5%). A bare DM was successfully achieved in 162 eyes (83.5%), whereas in 32 eyes (16.5%), layer-by-layer manual stromal dissection was performed to reach the predescemetic level.

Table 1 provides data that are relevant to preoperative visual acuity, keratometric measurements, and spherical equivalent refraction. Table 2 presents eye characteristics including central and peripheral corneal thickness, anterior chamber depth, and vitreous length before surgery. The mean recipient trephination size was 7.96 ± 0.11 mm (range: 7.50–8.25 mm). The mean postoperative follow-up was 35.9 ± 18.2 months (range: 12–89 months). The mean duration of steroid therapy was 116 ± 31 days (range: 84–145 days). Time interval from surgery to initial suture removal was 10.4 ± 8.9 months (range: 1.5–18 months), from surgery to running suture removal was 16.7 ± 9.0 months (range: 4.5–26 months), and from surgery to complete suture removal was 17.7 ± 8.9 months (range: 9–26 months). Suture-related complications in the 194 eyes included stitch-associated abscesses (n = 36, 18.6%), suture-tract vascularization (n = 24, 12.4%), and premature loosening (n = 51, 26.3%). Stromal graft rejection occurred in 2 eyes (1.0%) and was successfully treated with frequent topical betamethasone 0.1% gradually tapered over 3–6 weeks. No significant change was observed in refraction after improvement of the rejection episodes (refraction was −5.0 −6.0 × 140 before stromal rejection and −4.75 −5.5 × 150 after 71 days in 1 eye; in the other eye, corresponding values were −5.0 −5.5 × 180 before rejection and −5.25 −5.5 × 5 after 94 days.) Steroid-induced high IOP was observed in 27 eyes (13.9%). All of the cases were controlled medically. No eyes developed lens opacity during the follow-up period.

Corneal topography was obtained once in 121 eyes (62.4%), twice in 63 eyes (32.5%), and 3 times in 10 eyes (5.1%) after complete suture removal. We did not encounter recipient corneal thinning indicating inadequate trephination, or topographic changes suggesting the recurrence of keratoconus in grafts.

Compared with preoperative values, visual acuity, mean keratometry, keratometric astigmatism, and spherical equivalent refraction significantly improved postoperatively ( Table 1 ). In univariate analysis, postoperative mean keratometry was significantly associated with preoperative mean keratometry (r = 0.23, P = .008), anterior chamber depth (r = 0.23, P = .02), time interval from surgery to initial suture removal (r = −0.17, P = .03), time interval from surgery to running suture removal (r = −0.29, P = .001), suture loosening (r = 0.15, P = .04), and suture-tract vascularization (r = 0.15, P = .04). Postoperative spherical equivalent refraction demonstrated a significant association with preoperative vitreous length (r = −0.21, P = .03) and suture-tract vascularization (r = −0.12, P = .02).

ANCOVA revealed that the preoperative mean keratometry ( P = .007), time interval from surgery to running suture removal ( P = .01), and suture-tract vascularization ( P = .04) significantly influenced the postoperative mean keratometry. It was revealed that for each diopter increase in the preoperative mean keratometry, postoperative mean keratometry increased by 0.10 D (95% confidence interval [CI]: 0.03–0.17 D). Each month decrease in the time interval from surgery to running suture removal increased postoperative mean keratometry by 0.06 D (95% CI: 0.015–0.106 D). The presence of suture-tract vascularization corresponded with an increase in the postoperative mean keratometry by 1.23 D (95% CI: 0.005–2.46 D). The same analysis showed that only vitreous length predicted postoperative spherical equivalent ( P = .03); each millimeter increase in vitreous length increased postoperative myopia by −0.39 D (95% CI: −0.03 to −0.75 D).

Postoperative keratometric astigmatism failed to exhibit any significant correlation with preoperative, surgical, and postoperative variables. Additionally, postoperative refractive outcomes were not correlated with patient age and sex, central and peripheral corneal thickness, recipient trephination size, surgical technique (big-bubble vs manual dissection DALK), duration of steroid administration, and elevated IOP.