Journal of Refractive Surgery

The articles prior to January 2013 are part of the back file collection and are not available with a current paid subscription. To access the article, you may purchase it or purchase the complete back file collection here

Results of Transverse Keratotomies for Astigmatism After Penetrating Keratoplasty: A Retrospective Study of 48 Consecutive Cases

Jean-Jacques Saragoussi, MD; Alain Abenhaim, MD; Naji Waked, MD; Harry R Koster, MD; Yves J M Pouliquen, MD

Abstract

ABSTRACT

Background: High astigmatism is still a common complication of penetrating keratoplasty which often limits the final corrected visual acuity. Surgical correction of high astigmatism persistent after suture removal remains rather controversial. In our present study, we used the technique of transverse keratotomies for the surgical correction of high astigmatism following penetrating keratoplasty because of its simplicity and corrective potential.

Methods: We present a retrospective clinical study based on the results of 48 consecutive transverse keratotomy procedures performed on the graft. Three eyes were operated on twice, for a total of 51 procedures. The astigmatism had to be stable for at least 6 months after suture removal, with poor corrected visual acuity with spectacles or contact lenses. The operative protocol (number of incisions, optical zone size) was decided based on the degree of astigmatism. The mean postoperative observation period was 5.11 ± 3.93 months (range, 1 to 24 months).

Results: The mean preoperative cylinder was 8.96 ± 2.22 diopters and 4.91 ± 1.79 D postoperatively. The mean cylindrical change was 4.51 ± 2.77 D. The spherical equivalent remained unchanged in most cases. In a majority of cases, the corrected visual acuity for distance and near vision was improved. No cases of persistent graft edema caused by immunologic rejection or endothelial failure were observed. Moreover, in no case was there worsening of the best corrected visual acuity.

Conclusions: The technique of transverse incisions in the graft to correct or reduce high postkeratoplasty astigmatism is simple, efficient, and relatively safe. The main problem was poor predictability. (Refract Corneal Surg 1992;8:33-38.)

Abstract

ABSTRACT

Background: High astigmatism is still a common complication of penetrating keratoplasty which often limits the final corrected visual acuity. Surgical correction of high astigmatism persistent after suture removal remains rather controversial. In our present study, we used the technique of transverse keratotomies for the surgical correction of high astigmatism following penetrating keratoplasty because of its simplicity and corrective potential.

Methods: We present a retrospective clinical study based on the results of 48 consecutive transverse keratotomy procedures performed on the graft. Three eyes were operated on twice, for a total of 51 procedures. The astigmatism had to be stable for at least 6 months after suture removal, with poor corrected visual acuity with spectacles or contact lenses. The operative protocol (number of incisions, optical zone size) was decided based on the degree of astigmatism. The mean postoperative observation period was 5.11 ± 3.93 months (range, 1 to 24 months).

Results: The mean preoperative cylinder was 8.96 ± 2.22 diopters and 4.91 ± 1.79 D postoperatively. The mean cylindrical change was 4.51 ± 2.77 D. The spherical equivalent remained unchanged in most cases. In a majority of cases, the corrected visual acuity for distance and near vision was improved. No cases of persistent graft edema caused by immunologic rejection or endothelial failure were observed. Moreover, in no case was there worsening of the best corrected visual acuity.

Conclusions: The technique of transverse incisions in the graft to correct or reduce high postkeratoplasty astigmatism is simple, efficient, and relatively safe. The main problem was poor predictability. (Refract Corneal Surg 1992;8:33-38.)

High astigmatism is still a common complication of penetrating keratoplasty which often limits the final corrected visual acuity.

Several methods have been applied and combined in attempts to reduce its frequency. Trephination with pneumatic fixation systems1,2 reduces the risk of incision defects and facilitates good donorrecipient apposition. Adjusting the tension of a single continuous suture under photokeratoscopic control3 or removal of tight interrupted sutures4-5 significantly reduces postoperative astigmatism.

Surgical correction of high astigmatism persistent after suture removal remains rather controversial. Despite improvements in instrumentation, the results of corneal wedge resections,6 relaxing incisions,6-8 or Ruiz technique,9 remain poorly predictable.10-12

Transverse incisions (T-cuts) have been used by Ruiz in association with radial incisions (trapezoidal keratotomies) for the correction of high myopic astigmatisms. Thornton13 proposed using T-cuts alone or in combination with radial keratotomies for the correction of congenital astigmatism. The Ruiz procedure has further been modified by reducing the number of transverse incisions (five pairs initially). A study of transverse incisions on cadaver eyes14 has demonstrated little increase in effect for more than two pairs of transverse incisions tangentially placed at optical zone of 5 and 7 mm. In our present study, we used the technique of transverse keratotomies for the surgical correction of high astigmatism following penetrating keratoplasty because of its simplicity and corrective potential.

Table

Table 1Distribution of Initial Corneal Pathology Requiring Corneal Transplantation (n = 48)Table 2Protocol of Transverse Incisions in Grafts of ≥* 8-Millimeter DiameterFigure 1: Biomicroscopy view of transverse incision on grafts.

Table 1

Distribution of Initial Corneal Pathology Requiring Corneal Transplantation (n = 48)

Table 2

Protocol of Transverse Incisions in Grafts of ≥* 8-Millimeter Diameter

Figure 1: Biomicroscopy view of transverse incision on grafts.

MATERIALS AND METHODS

We present a retrospective clinical study based on the results of 48 consecutive transverse keratotomy procedures performed by two surgeons (J.J. S. and A.A.) between February 1987 and November 1990.

Our series included 48 eyes with high postkeratoplasty astigmatism of 47 patients with a mean age of 43.75 years (range, 27 to 70 years). Three eyes were operated on twice, for a total of 51 procedures. Table 1 shows the distribution of these eyes as a function of the initial corneal disease. Keratoconus represented more than half of the cases (27/48).

Indications were restricted to corneal grafts presenting no sign of immunologic rejection and having 4.00 or more diopters of astigmatism by keratometry (Javal). The astigmatism had to be stable for at least 6 months after suture removal, with poor corrected visual acuity with spectacles or contact lenses.

The surgical aim was not to fully correct the astigmatism, but to adequately reduce it so as to permit a better visual acuity with complementary optical correction.

The surgical procedure was the same for all patients. After topical (Oxybuprocaine 0.40%) anesthesia and subconjunctival (Hdocaine 1%) anesthesia, the steeper meridians were located using a Mendez graduated ring. Incisions were performed with the vertical border of a double-edged diamond knife set at 80% of the thinnest central corneal thickness (ultrasonic pachymeter, DGH 500, velocity 1640 m/s). One to four incisions, 3 mm in length, were tangentially placed at 5-, 6-, or 7-millimeter optical zones (Fig 1) with standard optical axis centration. Afterward, incisions were gently irrigated with BSS, and the cornea treated with a few drops of chloramphenicol. An ocular patch was placed for 24 hours, followed by combination topical Neomyein/Dexamethasone drops three times a day for 3 weeks.

The operative protocol (number of incisions, optical zone size) was decided based on the degree of astigmatism (Table 2). Surgical decision was not based on intraoperative keratometric readings. When one incision is used, it must be placed on the steepest semimeridian as noted by photokeratoscopy.

The mean postoperative observation period was 5.11 ± 3.93 months (range, 1 to 24 months).

Postoperative clinical data were analyzed, with particular attention directed to the variations of measured astigmatism (Javal keratometer) and corrected visual acuities.

RESULTS

Effectiveness

The Astigmatic Value. The mean preoperative astigmatism was 8.96 ± 2.22 D (range, 4.00 to 15.00 D, n = 48). Postoperatively, astigmatism was 4.91 ± 1.79 D (range, 2.00 to 10.00 D).

Figure 2: Distribution of astigmatic power (in diopters) before and after surgery (n = 48).Figure 3: Variations of the astigmatic power, measured by Javal keratometer and expressed in diopters (n = 48).Table 3Corrective Effect (D) of Different Protocols in This Series (n = 47)*

Figure 2: Distribution of astigmatic power (in diopters) before and after surgery (n = 48).

Figure 3: Variations of the astigmatic power, measured by Javal keratometer and expressed in diopters (n = 48).

Table 3

Corrective Effect (D) of Different Protocols in This Series (n = 47)*

Figure 2 depicts the distribution of astigmatic values before and after surgery.

The mean spherical equivalent was not modified by the transverse incisions. Overall, for the 51 procedures, it was -3.27 ± 3.72 D (range, -12.00 to +5.00 D) preoperatively and -3.23 ± 4.28 D (range, - 12.00 to + 7.00 D) postoperatively. Nevertheless, individual analysis of each case showed frequent but moderate myopic or hyperopic shift in an unpredictable manner.

Astigmatic Change. The mean change in measured astigmatism (with Javal keratometer) was 4.51 ± 2.77 D (range, 2 to 13, n = 48) (Fig 3).

Of particular interest is the effect of the protocol of cylindric change experienced (Table 3). The mean cylindric change was 2.93 D with only one incision tangentially placed at 6-millimeter optical zone, 3.57 D when a couple of incisions were performed with an optical zone of 6 mm, 4.00 D when a couple of incisions were performed with an optical zone of 5 mm, and 5.78 D with two pairs of incisions placed at optical zones of 5 and 7 mm. With the same protocol, a marked variation of the results exists.

Figure 4: Distribution of distance spectacle corrected decimal visual acuities before and after surgery (n = 48).

Figure 4: Distribution of distance spectacle corrected decimal visual acuities before and after surgery (n = 48).

Distance Corrected Visual Acuity. The mean best spectacle corrected decimal distance visual acuity was 0.27 (range, 0.05 to 0.80) before surgery and 0.42 (range, 0.05 to 0.90) after surgery. Figure 4 shows the distribution of the best spectacle corrected visual acuities before and after surgery. Although 17% of eyes had corrected visual acuity more than or equal to 20/40 before surgery (n = 48), these acuities were not functionally useful because they depended on disabling cylindric correction. Postoperatively, 44% of the eyes attained 20/40 acuity with well tolerated correction.

Figure 5: Distribution of near spectacle corrected visual acuities (Parinaud chart) before and after surgery (n = 48).Table 4Obtained Correction as a Function of Desired Correction (n = 47)

Figure 5: Distribution of near spectacle corrected visual acuities (Parinaud chart) before and after surgery (n = 48).

Table 4

Obtained Correction as a Function of Desired Correction (n = 47)

When necessary, rigid gas permeable contact lens wear was possible after the 2nd postoperative month.

Near Corrected Visual Acuity. Figure 5 shows the distribution of the near spectacle corrected visual acuities measured on a Parinaud chart, before and after surgery. Forty-three percent of eyes had Parinaud 2 preoperatively whereas 68% of the eyes attained this level postoperatively.

Predictability

For the 48 single procedures, the average astigmatic correction desired was 6.18 ± 1.84 D (range, 3.00 to 11.00 D). The average astigmatic correction obtained was 4.09 ± 2.22 D. (range, 0 to 13.00 D), for an average of 67.79% correction (range, 0% to 170%) compared to the desired correction. Table 4 depicts the average correction obtained as a function of the preoperative goal.

Reoperations. Undercorrected or overcorrected patients may be reoperated if they have no functionally significant amelioration in their best corrected visual acuity.

In one case, a large overcorrection was obtained using the protocol with one incision (13.00 D of correction for a desired correction of 3.00 D). This was associated with excessive postoperative gape at the incision site. The wound gape was treated with a single suture, reducing the astigmatism 7.00 D. Two transverse keratotomies placed along the steep meridian permitted a final result of 4.50 D of astigmatism (by Javal keratometer).

In two cases, a large undercorrection was noted using a pair of incisions with an optical zone of 5 mm. The addition of a second pair of incisions of optical zone 7 mm permitted keratometric reduction from 10.00 D to 2.00 D in one case, and from 8.00 D to 3.50 D in another.

Safety

Qf the 51 procedures, two were complicated by inadvertent corneal perforation repaired by placement of interrupted sutures, but healed without functional or anatomic complications.

Postoperative complications were:

1. Transient immunologic graft rejection occurred 3 months postoperatively in one case away from the incision site.

2. An episode of recurrent superficial Herpes simplex virus infection occurred superficially in one case 1 month postoperatively, in the proximity of a transverse incision, but responded well to topical therapy.

3. In one case, localized graft edema in the area of a transverse incision scar was observed and remained unchanged 6 months later.

No cases of persistent graft edema caused by immunologic rejection or endothelial failure were observed. Moreover, in no case was there worsening of the best corrected visual acuity.

Stability

In the majority of the cases, keratometry readings were stable 4 weeks after surgery. A complementary study with a longer follow-up period of all cases would be necessary to support long-term stability.

DISCUSSION

When high postkeratoplasty astigmatism persists even after all sutures are removed, one is faced with the arduous task of optical correction. If optical correction is impossible or disabling, surgical correction must be considered.

Because of the numerous factors involved, postkeratoplasty astigmatism has a complex pathogenesis which differs from one case to the other. This, in part, explains the lack of predictability of corrective surgical procedures whose applications and indications are difficult to standardize.

Nevertheless, corneal topography analysis, clinical experience, and improvements in instrumentation (ultrasound pachymeters, diamond knives, and surgical keratometers) have led in recent years to improvement of existing techniques and to their predictability.

Wedge resection is an effective procedure which does not alter graft transparency but which yields poor predictability.10,11 Relaxing incisions placed at the two poles of the steeper meridian have the advantage of avoiding the opening of the eye and of being easy to perform.6,7 Compression sutures placed along the orthogonal meridian away accentuate their corrective effect.7,15 This combined technique is complicated by poor predictability and the possibility of secondary wound shift.10,12

We preferred to perform incisions on the graft itself for many reasons. Transverse incisions lose efficiency the closer they are to the limbus. The recipient cornea is often pathological, affecting the healing process and reliable refractive effect of incisions. Moreover, alteration of the recipient cornea is not advisable in case a change of repeat graft becomes necessary.

Our preference was for transverse incisions despite the theoretic advantages of curved keratotomies.16 The former are more easily performed and reproducible with standard refractive surgery instrumentation. Radial incisions were avoided, even in case of high myopic astigmatism, so as to limit graft surgical injury caused by passing through the host-graft interface and recipient cornea.

The effectiveness of a technique is difficult to evaluate in astigmatic surgery. Astigmatic power changes should take into account possible axis shift (rotation or inversion). Thus, the distribution of astigmatism before and after surgery along with the cylindric power variation (which takes into consideration axis inversions) are of upmost importance.

It is impossible to compare the results of the different published clinical studies concerning relaxing incisions. Not only are there few series, but they also differ from each other in indications and methodology.8,11,15,17-21

The effect of transverse incisions in our series was greater than that observed by Lindstrom and Lindquist12 and Lavery and Lindstrom,12,18 who limit the use of this technique to patients with low astigmatism. In addition, to establish a valid comparison, one must consider the patient age and the nature of astigmatism.

Our results are similar to that obtained in some published series, notably Krachmer and Fenzl17 or Lindstrom and Lavery,11 with relaxing incisions in the graft-host interface. Lindstrom obtained in 12 cases a mean reduction of 4.84 D with a range of 1.75 to 10.50 D. The effect of relaxing incisions can be increased by placing compression sutures at the meridian 90° away.8,12,15

Lack of predictability characterizes all surgical methods for astigmatic correction. The protocol chosen must be in function of each particular case, with complementary corneal topography analysis. Maguire and Bourne22 demonstrated that the two steeper semimeridians do not always have the same optical power and are not necessarily separated by 180° angle. In these cases, it is logical to consider a protocol for each semimeridian with asymmetric disposition of incisions if necessary. Unfortunately, location of the two steeper semimeridians during surgery is often approximate.

The predictability of relaxing incisions seems better than that of Ruiz technique,9,12,18 abandoned in its original description, or the wedge resection of Troutman,6,8 in which greater corrections are possible.23 To improve relaxing incision predictability, their length and depth can be adjusted using intraoperative keratometry.11,12,19,21 Nonetheless, the result obtained in the operating room can vary greatly from the postoperative result after healing. In our opinion, it is preferable to undercorrect after first operation. Use of a secondary procedure such as the addition of one or two incisions poses no technical problem,24 while overcorrection is more difficult to manage.

The direct corneal effect of transverse keratotomy is flattening on the operated meridian with similar steepening of the orthogonal meridian. In our series, the mean spherical equivalent was unchanged despite frequent individual variations. This finding is in agreement with those of Lindstrom and Lindquist12 and Lindquist et al.14

It is difficult to estimate the influence of the initial corneal pathology on efficiency, predictabihty, and stability of results. Our study with a limited number of cases and relatively high frequency of keratoconus did not to permit us to establish a significant correlation.

It is also impossible for us to demonstrate a relationship between the age of the graft itself and the efficiency of astigmatic surgery. The impression of a relative decrease of effect with the time could not be confirmed.

The operative risk appears to be acceptable considering the gain in visual acuity after surgery. None of our patients experienced definitive loss of graft transparency or irreversible decrease in the corrected visual acuity.

The technique of transverse incisions in the graft to correct or reduce high postkeratoplasty astigmatism is simple, efficient, and safe. Its use should be reserved for patients who either cannot tolerate or achieve better visual acuity with cylindric correction or those intolerant to contact lens use. The lack of predictability prevents it from being an ideal procedure, but used properly, it allows improvement of visual function in the majority of the cases.

REFERENCES

1. Saragoussi JJ, Hanna K, Clay C, et al. Astigmatisme postkératoplastie et trépan de Harma: premiers résultats. Bull Soc Ophtalmol Fr. 1987;3:355-357.

2. Van Rij G, Waring GO. Configuration of Corneal Trephine Opening in a Laboratory Setting. Changes in Corneal Curvature Induced by Surgery, monograph. Rotterdam, The Netherlands: Erasmus University Rotterdam; 1987:47-63.

3. McNeil JI, Weasels IF. Adjustment of single continuous suture to control astigmatism after penetrating keratoplasty. Refract Corneal Surg. 1989;5:216-223.

4. Binder PS. The effect of suture removal on post-keratoplasty astigmatism. Am J Ophthalmol. 1988;105:637-645.

5. Harris DJ, Waring GO, Burk LL. Keratography as a guide to selective suture removal for the reduction of astigmatism after penetrating keratoplasty. Ophthalmology. 1989;96:1597-1607.

6. Troutman RC. Microsurgery of the Anterior Segment of the Eye. St Louis, Mo: CV Mosby; 1977:286.

7. Troutman RC, Swinger C Relaxing incisions for control of postoperative astigmatism following keratoplasty. Ophthalmic Surg. 1980;11:117-120.

8. Troutman RC Improved techniques in refractive surgery in astigmatism. Cornea. 1982;1:57.

9. Merck MP, Williams PA, Lindstrom RL. Trapezoidal keratotomy. A vector analysis. Ophthalmology. 1986;93:719-726.

10. Kaufman HE. Corneal transplant optics and visual disability. Refract Corneal Surg. 1989;5:213-215.

11. Lindstrom RL, Lavery GW. Correction of postkeratoplasty astigmatism. In: Sanders DR, Hoffman RF, SaIz JJ, eds. Refractive Corneal Surgery. Thorofare, NJ: 1986:215.

12. Lindstrom RL, Lindquist TD. Surgical correction of postkeratoplasty astigmatism. Cornea. 1988;7:138-148.

13. Thornton SP. Graded non intersecting transverse incisions for correction of idiopathic astigmatism. In: Sanders DR, ed. Radial Keratotomy. Thorofare, NJ: SLACK Ine; 1986:103-116.

14. Lindquist TD, Rubenstein JB, Rice SW, Williams PA, Lindstrom RL. Trapezoidal astigmatic keratotomy: quantification in human cadaver eyes. Arch Ophthalmol. 1986; 104: 1534-1539.

15. Mandel MR, Shapiro MB, Krachmer JH. Relaxing incisions with augmentation sutures for the correction of postkeratoplasty astigmatism. Am J Ophthalmol. 1987; 103:441-447.

16. Merlin U. Curved keratotomy procedure for congenital astigmatism. Journal of Refractive Surgery. 1987;3:92-97.

17. Krachmer JH, Fenzl RE. Surgical correction of high postkeratoplasty astigmatism. Arch Ophthalmol. 1980;98:1400-1402.

18. Lavery GW, Lindstrom RL. Clinical results of the Ruiz astigmatic keratotomy. Journal of Refractive Surgery. 1985;1:70-74.

19. Rowsey JJ. Corneal trauma: the prevention and correction of corneal transplant astigmatism. Refractive and Corneal Surgery. 1988;4:209-217.

20. Cohen KL, Tripoli NK, Noecker RJ. Prospective analysis of photokeratoscopy for arcuate keratotomy to reduce postkeratoplasty astigmatism Refract Corneal Surg. 1989;6:388-393.

21. Arne JL, Bourdiol AM, Colin J, et al. Chirurgie refractive cornéenne. Rapport des Sociétés d'ophtalmologie de France. 1989.

22. Maguire LJ, Bourne WM. Corneal topography of transverse keratotomies for astigmatism after penetrating keratoplasty. Am J Ophthalmol. 1989;107:323-330.

23. Burillon C, Durand L, Hachmanian KF. La résection cunéiforme, traitement correctif des astigmatismes cornéens géants. J Fr Ophtalmol. 1989;12:447-453.

24. Forstot SL. Modified relaxing incision technique for postkeratoplasty astigmatism. Cornea. 1988;7:133-137.

Table 1

Distribution of Initial Corneal Pathology Requiring Corneal Transplantation (n = 48)

Table 2

Protocol of Transverse Incisions in Grafts of ≥* 8-Millimeter Diameter

Figure 1: Biomicroscopy view of transverse incision on grafts.

10.3928/1081-597X-19920101-10

Sign up to receive

Journal E-contents