Radial and staggered treatment patterns to correct hyperopia using noncontact holmium:YAG laser thermal keratoplasty

Radial and staggered treatment patterns to correct hyperopia using noncontact holmium:YAG laser thermal keratoplasty

articles Radial and staggered treatment patterns to correct hyperopia using noncontact holmium:YAG laser thermal keratoplasty Paolo Vinciguerra, MD, T...

2MB Sizes 0 Downloads 2 Views

articles Radial and staggered treatment patterns to correct hyperopia using noncontact holmium:YAG laser thermal keratoplasty Paolo Vinciguerra, MD, Thomas Kohnen, MD, Marco Azzolini, MD, Paola Radice, MD, Daniel Epstein, MD, PhD, Douglas D. Koch, MD ABSTRACT Purpose: To compare the effects of two treatment patterns in the correction of hyperopia by noncontact holmium:YAG laser thermal keratoplasty (LTK). Setting: Divisione Oculistica, Ospedale S Gerardo, Monza, Italy. Methods: Using two treatment patterns, we performed noncontact LTK in one session in 16 eyes of 8 patients with isometropic hyperopic refractive errors, mean preoperative subjective cycloplegic refraction was + 4.90 diopters (D) :!: 1.17 (SO). The treatment consisted of 24 spots in three concentnc rings of eight spots each, nng diameters were 6.0, 7.0, and 8.0 mm, respectively. Each spot received seven pulses of laser energy at 30 mJ/pulse. We treated one eye of each patient with a radial pattern (the spots of the three rings aligned on the eight semimeridians) and the fellow eye with a staggered pattern (the spots of the cont1guous nngs at 22.5 degrees from each other). Follow-up at 1, 15, 30, 90, 180, and 360 days included subjective cycloplegic refraction , uncorrected (UCVA) and spectaclecorrected v1sual acuity (SCVA), computerized v1deokeratography (CVK), and Scheimpflug camera examination. Results: One year postoperatively, the mean subjective cycloplegic refract1on was + 2.75 :!: 1.6 0 1n the eyes treated with the radial pattern and + 3.40 :!: 1.6 0 in those treated with the staggered pattern; the mean change in subjective cycloplegic refraction was 2.15 and 1.50 0, respectively. Mean UCVA improved by five lines in the radial group and by four lines in the staggered group. Mean SCVA returned to preoperative levels by day 15 in the radial group and at 1 year in the staggered group; at 1 year, SCVA 1mproved by one line in the radial group and rema1ned unchanged in the staggered group. No eye lost one or more lines of SCVA. Refractive astigmatism was essentially unchanged 1n both groups. Schelmpflug photography and CVK indicated larger and more uniform corrected zones in the radial group. Conclusions: Radial and staggered patterns effectively corrected low hyperopia, although both were subject to a certain amount of regression . The radial pattern produced faster postoperative recovery of SCVA and demonstrated greater refractive stability. J Cataract Refract Surg 1998; 24:21- 30 J CATARACT REFRACT SURG-VOL 24, JANUARY 1998

21

HO:YAG LTK WITH RADIAL AND STAGGERED PATTERNS

N

oncontact holmium:YAG (Ho:YAG) laser thermal keratoplasty (LTK) uses heat generated by the absorption of infrared laser energy in the cornea to thermally modifY stromal collagen and thereby change anterior corneal curvature. 1-4 When heated to approximately 60 to 75°C, corneal collagen contracts maximally to one third its original length. Higher temperatures, typically used in conventional contact thermokeratoplasty, do not produce further shrinkage but, on the contrary, cause tissue relaxation and necrosis. 5- 15 The thermal effect of the laser beam at the impact site is maximal at the surface and gradually decreases in the deeper stroma. Typically, stromal haze at the treatment site extends to 50 to 70% of the corneal thickness and has a tapered conical shape that is broadest anteriorly. 13 No clinically significant damage to the endothelial cells has been detected. 6·14 With large-diameter treatment patterns, hyperopic refractive errors can be corrected. 13·14 Although up to 2.0 diopters (D) of hyperopic correction have been reported, the maximum amount of achievable correction is unknown; the limiting factors are the amount of initial correction and the amount of long-term regression (loss of induced refractive change as a function of individual healing response).12-14,16,17 Most published studies of LTK have looked at the following treatment variables: (1) laser beam energy, (2) method of delivering the energy to the cornea (contact or noncontact), (3) distance from the optical center, and (4) number, diameter, and depth of the spotsY&-22 In this study, we evaluated a new variable: pattern of the treatment spots.

From the Divisione Oculistica, Ospedale S. Gerardo, Monza, Italy (Vinciguerra, Azzolini, Radice), Cullen Eye Institute, Baylor College of Medicine, Department of Ophthalmology, Houston, Texas, USA (Kohnen, Koch), Department of Ophthalmology, University Hospital, Uppsala, Sweden (Epstein), and johann Wolfgang Goethe-University, Department ofOphthalmology, Frankfort am Main, Germany (Kohnen). Dr. Koch is a paid consultant to Sunrise Technologies, Inc. Sunrise Technologies has underwritten clinical and experimental research work ofDr. Kohnen. None of the other authors has a proprietary interest in Sunrise Technologies or in any competing device. Reprint requests to Paolo Vinciguerra, MD, Via Ripamonti #205, 20141 Milan, Italy.

22

Collagen shrinkage in a single, thermally altered spot produces mechanical stress in the cornea, with lines of tension that develop in all directions and combine in a vectoral fashion with lines of tension produced by nearby spots. The resultant forces produced in the cornea will therefore be determined by magnitude, direction, and location of each component force. This interaction forms the lines of stress that in turn determine the changes in corneal shape. We suspected that the geometric location of spots, including orientation and distance between them, is significant.

Patients and Methods We treated 16 eyes in 8 consecutive patients with bilaterally symmetrical hyperopia. Mean age of the 6 men and 2 women was 42 years± 9.7 (SD) (range 27 to 53 years). Inclusion criteria were absence of concomitant ocular pathology, no previous ocular surgery, normal intraocular pressure, normal central corneal thickness (range 480 to 600 )..lm), refractive astigmatism of 1.0 D or less, and at least 1 year of refractive stability (::50.5 D change). The mean spherical equivalent (SE) of the pretreatment subjective cycloplegic refraction was +4.90 ± 1.17 D (range +2.75 to +6.75 D) (Table 1). Preoperative uncorrected visual acuity (UCVA), spectacle-corrected visual acuity (SCVA), and Table 1. Preoperative data of eyes treated with the radial and staggered patterns.

Patient Number

Sex, Age

SCR

UCVA

SCVA

Radial Staggered

Radial Staggered

M, 48

2.75

0.10

0.10

1.0

1.0

2

M, 53

5.25

0.01

0.01

0.9

0.8

3

M, 47

6.75

0.05

0.10

0.8

1.0

4

M, 30

4.00

0.50

0.40

1.0

1.0

5

M, 46

5.50

0.10

0.05

1.0

1.0

6

F, 49

4.25

0.10

0.05

0.9

0.6

7

M, 36

6.00

0.05

0.05

1.0

1.0

8

F, 27

4.75

0.60

0.60

1.0

1.0

Mean

42

4.90

0.19

0.17

0.95

0.93

SD

9.65

1.17

0.23

0.21

0.08

0.15

SCR = subjective cycloplegic refraction (spherical equivalent) UCVA and SCVA = uncorrected and spectacle-corrected visual acuity in decimal units SD = standard deviation

] CATARACT REFRACT SURG-VOL 24, JANUARY 1998

HO:YAG LTK WITH RADIAL AND STAGGERED PATTERNS

mean central corneal power derived from the average of the mean ring power from rings 3 through 8 using the TMS-1 computerized videokeratograph (Tomey Corp.) were similar for the two eyes of each patient. The treatments were performed using a noncontact Ho:YAG laser (Corneal Shaping System, Sunrise Technologies, Inc.) emitting pulses of infrared light (2.12 J.Un) with 160 f.!Sec pulse duration at 5 Hz repetition rate. The laser energy was delivered through a proprietary slitlamp delivery system23 simultaneously projecting up to eight laser beams in a symmetrical octagonal array; the ring pattern was adjustable for ring diameter (range 3 to 8 mm) and angular position. Each spot had a nominal diameter of 600 J.Un, an energy of 30 mJ/pulse, and a mean pulse radiant exposure of approximately 10 Jlcm 2 per pulse. The number of pulses per spot was 7; each eye received 24 spots in a pattern of three rings of eight spots each at 6, 7, and 8 mm. The sequence of treatments was inner ring first, outer ring last. In each patient, one eye was treated with a radial pattern (the radial group), in which the eight spots of the three rings were radially aligned, and the fellow eye was treated with a staggered pattern (staggered group), in which the eight spots of the middle ring were rotated 22.5 degrees to the spots of the inner and outer rings (Figure 1). A joulemeter (Molectron model200) was used to check the energy at various ring diameters and its uniformity among the spots. A topical anesthetic (Ossibuprocaine®) was applied two times at 5 minute intervals before the treatment. The lids were held open with an open-wire speculum, and a limbal suction ring was applied by the surgeon to

Staggered

Radial

.



• •• •«<• • ·.:_';'

.e

~

~

0



•e

.•



~"{

...

~ ~

,.

G

•·I

~ ·

"

Figure 1. (Vinciguerra) Radial and staggered treatment patterns for noncontact holmium:YAG laser thermal keratoplasty as performed in this study.

stabilize the eye during treatment. To minimize variability in tear film thickness and temperature conditions, the treatment was performed 3 minutes after application of the suction ring. The patient was requested to fixate on a flashing red light from a light emitting diode, which is installed coaxially within the treatment spots (identified by eight red aiming HeNe laser beams at 632.8 nm) in the Sunrise slitlamp delivery system. The correct focal position was determined by two convergent green laser beams (HeNe at 543 nm) overlapping on the corneal surface. After treatment, gentamicin eyedrops were administered and the eyes were patched for 24 hours. The patients were given four analgesic tablets to be taken as needed, one every 6 hours. To eliminate individual variables that could affect the amount of initial correction and subsequent regression, we enrolled only patients presenting with bilaterally symmetrical hyperopic refractive errors, visual acuities, and corneal topographic findings, and we treated one eye with one pattern and the fellow eye with the other pattern. We performed the bilateral treatments in the same session to assist in evaluating patients' subjective comparison of the visual outcomes from the two treatment patterns and to eliminate the possibility of mnemonic bias resulting from treating the fellow eye at a later time. We felt that bilateral treatments did not pose undue risk to patients because none of our previous patients have lost best corrected vision following noncontact Ho:YAG LTK, patients have only mild to moderate foreign-body sensation for 1 to 2 days and are fully functional by day 3; and recovery of SCVA is rapid, typically returning to within two lines of normal within 2 to 3 days. All treatments were performed with the same laser system and by the same surgeon (P.V.). Pretreatment and follow-up (1, 15, 30, 90, 180, 270, and 360 days) ocular measurements included UCVA and SCVA with the Nidek SCP 960 Chart Projector System performed by ophthalmologists who did not know the preoperative refraction; subjective cycloplegic refraction (cyclopentolate drops); slitlamp biomicroscopic evaluation of anterior and posterior segments; applanation tonometry; computerized videokeratography (CVK) (Tomey TMS-1 and EyeSys Corneal Analysis System, version 2.004, EyeSys Technologies); endothelial specular microscopy, including endothelial cell density measure-

J CATARACT REFRACT SURG-VOL 24, JANUARY 1998

23

HO:YAG LTK WITH RADIAL AND STAGGERED PATTERNS

Table 2.

Spherical equivalent of subjective cycloplegic refraction in diopters (mean ± SD) in eyes treated with the radial and staggered

patterns. Follow-up Visit Group

Preop

1 day

15 days

1 month

3 months

6 months

9 months

12 months

Radial

4.90 ± 117

-0.34 ± 1.0

0.20 ± 1.3

0.71 ± 1.2

1.90 ± 1.90

2.25 ± 1.70

2.70 ± 1.90

2.75 ± 1.60

Staggered

4.94 ±1 04

-0.34 ± 1.0

0.28 ± 1.2

1.28 ± 1.2

2.50 ± 1.57

2.70 ± 1.32

3.05 ± 1.75

3.40 ± 1.60

ments, range and mean value of cell dimensions, and cellular morphology (Tomey EL 1100); Scheimpflug camera examination (EAS 1000, Nidek Ltd.) and ultrasonic pachymetry (Pach-Pen XL, Mentor). With the Nidek EAS 1000 Scheimpflug camera, two types of images were obtained: sagittal cross section and back illumination. Using the available software for the sagittal cross-section images, we calculated the anterior and posterior corneal radii of curvature; the depth, area, and volume of the opacities at the spot sites; and the degree of opacity of the spots. From back illumination images, we evaluated alteration of the stromal transparency, lines of tension or striae connecting treatment spots, and refractive inhomogeneities. The software permitted calculation of the area affected by the treatment and for both types of images provided displays of differential images at consecutive examinations. At 1 and 12 months postoperatively, all patients were asked to answer five questions about their visual quality: With which eye do you have better vision? Do you have halos or glare? If yes, with which eye do you have more? With which eye do you have better night vision? With which eye do you have better near vision without glasses? Do you have daytime fluctuating vision? If yes, with which eye do you have more? The data were collected on a standardized form and analyzed with a statistical program (Systat 5.01 for Windows, Systat Inc.); a Student's paired t-test was used to determine the statistical significance (P < .05) between and within groups.

the spots in one or two rings was irregular and slightly elliptical, indicating a small amount of intraoperative motion. Refraction. At 1 and 15 days postoperatively, the mean SE subjective cycloplegic refraction and mean change in SE subjective cycloplegic refraction were similar in both groups; at 15 days, mean SE was +0.21 ± 1.30 D in the radial group and +0.28 ± 1.28 D in the staggered group. The rather high standard deviation reflects the wide preoperative refractive range (2.75 to 6.75 D) among the patients (Table 1) and did not substantially change from its preoperative value through the follow-up visits. Although both groups showed regression of the induced refractive change, the regression between 15 days and 1 month was more pronounced in the staggered group (Table 2; P = .04); at all follow-up visits starting at 2 weeks postoperatively, there was greater refractive correction in the radial group (Figure 2). At 1 year, the mean change m SE subjective cycloplegic refraction was 2.15 D in the radial group and 1.50 D in the staggered group; this difference of 0.65 D was statistically significant (P < .05). Furthermore, the regression from 6 to 12 months was statisti-

0 .7

*

01

.

0.5

; o•

i.

~

0 3 0 .2

Results

0 1

o - --""""' - - - -

Treatment Observations. None of the patients complained of intraoperative discomfort. Direct examination of the patients, together with videokeratographic and Scheimpflug images, showed nearly motionless treatments in most cases; in some patients, the shape of 24

pre-op

1 de

15 de

1 mo

3 mo

I mo

t mo

12 mo

Figure 2. (Vinciguerra) Differences in diopters between the mean values of the subjective cycloplegic refraction (SE) in eyes treated with the radial and staggered patterns. * = statistically significant difference.

J CATARACT REFRACT SURG-VOL 24, JANUARY 1998

HO:YAG LTK WITH RADIAL AND STAGGERED PATTERNS

Table 3. Uncorrected visual acuity (mean : +: SO) in eyes treated with the radial and staggered patterns. Follow-up VIsit Group

Preop

1 day

15 days

1 month

3 months

6 months

9 months

12 months

Radial

0.19::+::0.21

0.73 ::':: 0.19

0.78 ::':: 0.27

0.73 ::':: 0.25

0.56 ::':: 0.24

0.54 ::':: 0.17

0.52 :±: 0.21

0.55 ::':: 0.21*

Staggered

0.17 ::':: 0.20

0.57 ::':: 0.19

0.63 ::':: 0.30

0.55 ::':: 0.27

0.46 ::':: 0.26

0.44 ::':: 0.10

0.42 ::':: 0.22

0.43 ::':: 0.31*

*Statistically significant difference, P < .05

Table 4. Spectacle-corrected visual acuity (mean : +: SO) in eyes treated with the radial and staggered patterns. Follow-up VIsit Group

Preop

1 day

15 days

1 month

3 months

6 months

9 months

12 months

Radial

0.97 ::':: 0.10

0.84 ::':: 0.16*

1.00 ::':: 0.10*

0.97 ::':: 0.06

0.98 ::':: 0.09

1.00 ::':: 0.12*

1.06 ::':: 0.11

1.05 ::':: 0.08

Staggered

0.90 ::':: 0.17

0.65 ::':: 0.18*

0.86 ::':: 0.11 *

0.87 ::':: 0.13

0.94 ::':: 0.12

0.88 ::':: 0.09*

0.96 ::':: 0.18

0.91 ::':: 0.21

*Statistically significant difference, P < .05

cally significant in the staggered group (P = .01) but not in the radial group. The astigmatic component was minimally affected by the spherical LTK treatment; the absolute value of the cylindrical error changed from a mean of 0.25 0 preoperatively to 0.40 0 at the end of the 12 month follow-up. In only one eye in the radial group was induced astigmatism more than 0.50 0; it reached a maximum value of 1.25 0 at 6 months and decreased to 0.75 0 at 12 months. Visual Acuity. Uncorrected visual acuity improved immediately in all patients and was higher in the radial group at all postoperative intervals (Table 3). At 1 year, the improvement in mean UCVA was 0.36 and 0.26 0 in the radial and the staggered groups, respectively (P < .05). The SCVA showed a different trend: an initial decrease, followed by a continuous increase to mean values equal or superior to the preoperative values (Table 4). In the radial group, mean SCVA was 0.84 :::!:: 0.15 0 at 1 day postoperatively, progressed to 1.00 :::!:: 0.10 0 at day 15, and remained basically stable thereafter, ending at 1.06 :::!:: 0.08 0, which was an overall gain of 0.12 0 (more than one line) from the preoperative value (P = .04). In the staggered group, the mean SCVA at 1 day postoperatively was 0.65 :::!:: 0.18 0, remained below the preoperative value until 6 months postoperatively, and at 1 year was 0.90 :::!:: 0.20 0, which represented no gain from the preoperative values. The SCVA in the radial group was significantly higher at 1 and 15 days and 6 months (Table 4). The standard

deviations in the staggered group tended to be higher than those in the radial group, suggesting greater homogeneity in the latter group of patients. Slit/amp Examination. Immediately postoperatively, we observed central corneal steepening and, at the treatment sites, local flattening of the anterior corneal surface, whitening of the epithelium, and stromal opacities with a diameter of700 to 800 j.lm and a depth of 50 to 80% of corneal thickness. In two patients (one in the radial group and one in the staggered group), we noted folds in Oescemet's membrane, which disappeared in 1 to 2 days. Throughout the 1 year follow-up, examination of dilated eyes with retroillumination revealed the presence of internal lines within the spots; these lines had a radial orientation in the eyes treated with the radial pattern and a fan-shaped orientation, pointing toward the nearby spots, in the eyes treated with the staggered pattern (Figure 3). Stromal striae between the spots were also evident in treated corneas. In the radial group, the striae were belt-like in orientation, circularly interconnecting· the spots within the 6 and 7 mm rings; these persisted through 1 year (Figure 3, left). In the staggered group, the striae were oriented in various directions, including toward the central zone and the periphery and also circumferentially in a belt-like pattern (Figure 3, right). The striae/lines of tension in the staggered group faded in intensity and typically disappeared by 6 months after treatment, whereas the striae in the radial group were still evident at 1 year.

J CATARACT REFRACT SURG-VOL 24, JANUARY 1998

25

HO:YAG LTK WITH RADIAL AND STAGGERED PATTERNS

Figure 3. (Vinciguerra) Slitlamp photographs of Ho:YAG LTK treatments using a radial (left) and staggered (right) pattern. In the eye treated with a radial pattern, radially oriented lines of tension internal to the spots and circular belt-like lines of tension interconnecting the spots can be observed. In the eye treated with a staggered pattern, the lines of tension internal to the spots were oriented in a fan-shaped way toward the nearby spots, while the stromal striae extended in all directions, including toward the central cornea and the limbus.

Computerized Videokeratography. Ring images: In both groups, the postoperative CVK ring images generally showed a regular appearance of the central rings, which appeared to be closer to each other, indicating the steepening of the central corneal area. Mid-peripheral rings (1.5 to 3.0 mm radius from the center or 3.0 to 6.0 mm diameter) had acceptable regularity in the radial group but were typically discontinuous and irregular in the staggered group, indicating topographic distortion. In the treated area (3.0 to 4.0 mm from the center or 6.0 to 8.0 mm diameter), the CVK images were distorted, with unrecognizable keratoscopic rings in all eyes in both groups. Videokeratographic power: From the CVK data, we averaged the dioptric values for rings 3 to 8 in the Tomey unit to obtain a value for central corneal curvature (Table 5). The changes in central corneal curvature power as measured by CVK paralleled the changes in subjective cycloplegic refraction. In both groups, there was immediate postoperative steepening

of the central cornea (about 3.0 mm diameter), associated with a flattening in the periphery at the sites of the laser spots. At 1 day, the increase in central corneal power was 4.30 Din the radial group and 4.10 Din the staggered group; these values regressed to + 1.90 D in the radial group and + 1.1 0 D in the staggered group at 1 year. Between 6 and 12 months, regression of the central topographic power was 0.50 D in the radial group and 0.90 D in the staggered group. At all postoperative intervals, there was greater curvature change in the radial group; at 1 and 12 monrhs postoperatively, these changes were statistically significant (P < .05). From the CVK examinations, we evaluated a variety of other parameters and/ or observations; the most interesting were the following: Size of the corrected zone: We defined the "corrected zone" as the corneal area that, in the comparative color maps between the preoperative and postoperative examinations, demonstrates an increase in dioptric power

Table 5.

Mean keratometric value using computerized videokeratography (average of the mean ring power of rings 3 through 8 ::t: SO) of the eyes treated with the radial and staggered patterns.

Follow-up Visit Group

Preop

1 day

15 days

1 month

3 months

6 months

9 months

12 months

Radial

42.8 ::t: 0.47

47.1 ::t: 0.9

47.2 ::t: 1.4

46.9 ::t: 0.85

46.1 ::t: 1.0

45.2 ::t: 0.7

44.8 ::t: 0.6

44.7 ::t: 0.8

Staggered

42.9 ::t: 0.82

47.0 ::t: 1.1

46.9 ::t: 1.4

46.3 ::t: 1.4

45.7 ::t: 1.3

44.9 ::t: 1.2

44.4 ::t: 1.5

44.0 ::t: 1.3

26

J CATARACT REFRACT SURG-VOL 24, JANUARY 1998

HO:YAG LTK WITH RADIAL AND STAGGERED PATTERNS

Figure 4. (Vinciguerra) Computerized videokeratography of Ho:YAG LTK treatments using a radial (left) and staggered (right) pattern. In each set of images, the color maps are, as follows: preoperative, lower left; 4 to 6 weeks postoperative, upper left; difference map, right. Note in the difference maps the greater uniformity of power distribution with the radial treatment.

(steepening) and is bordered by a sharp transmon of ±7.0 D. By positioning the cursor over the margins of the corrected zone, it was possible to calculate its diameter. This parameter was determined for CVK images obtained immediately and at 6 and 12 months postoperatively. Immediately after the treatment, the size of the corrected zone was similar in the two groups, within an inner ring diameter of 6.1 :±: 0.2 mm, whereas at 6 and 12 months, the size of the corrected zone was larger in the radial group: at 6 months it was 5.8 :±: 0. 3 mm in the radial group and 5.5 :±: 0.4 mm in the staggered one (P < .05) and at 12 months, 5.7 :±: 0.4 and 5.3 :±: 0.4 mm, respectively (P < .05). Dioptric variation within the corrected zone: This parameter was defined as the difference between the minimum and maximum corneal dioptric power within the corrected zone and was determined using the cursor and software options. The values for mean and maximum dioptric variation at 1 year were lower in the radial group (means 0.9 versus 1.6 D (P < .04); maximal values: 1.8 versus 2.75 D). Site of maximal corneal dioptric power: The site of maximal dioptric power (negative asphericity) was always found in the center of the cornea in the radial group, whereas it was confined to an annular zone with a diameter of 3.0 and 5.0 mm from the center of the cornea (positive asphericity) in the staggered group up to 3 months (Figure 4). Stability of topographic patterns: Comparing the sequence of the examinations in individual patients, we

found less variation m eyes treated with the radial pattern. Scheimpjlug Camera. The sagittal cross-section images showed steepening of the anterior and posterior central corneal surfaces (Figure 5). The variation in corneal thickness, even during the immediate postoperative period, was no more than 30.0 J..lm. These images showed that the thermally affected volume at the laser treatment site extended to 60 to 80% of the corneal thickness and did not involve Descemet's membrane. The retroillumination images showed the formation of striae or lines of tension between the laser treatment sites. In the radial group, these striae were uniform and belt-like in orientation, circularly interconnecting the spots of each treatment ring and still readily visible between the two inner rings in most eyes at 1 year postoperatively (Figure 5, left). In the staggered group, the striae were oriented in various directions (interconnecting the treatment spots, toward the periphery and the center), extended at times into the optical zone, and in only a few eyes remained clearly visible at 1 year (Figure 5, right).

Endothelial Specular Microscopy, Central Ultrasonic Pachymetry, and Applanation Tonometry. None of these parameters, including morphologic assessment of the endothelium, showed any statistically significant modification during the follow-up (Table 6).

Subjective Patient Comments and Questionnaire Responses. Most patients complained of moderate pain

J CATARACT REFRACT SURG-VOL 24, JANUARY 1998

27

HO:YAG LTK WITH RADIAL AND STAGGERED PATTERNS

Figure 5. (Vinciguerra) Scheimpflug pictures of Ho:YAG LTK treatments using a radial (left) and staggered (right) pattern. In the slitlamp observations, the lines of tension interconnecting the spots are circular belt-like in the radial treatments, whereas they are extended in all directions, including toward the central cornea and the limbus, in the staggered ones.

and mild tearing and photophobia during the first 1 to 2 days postoperatively. In their responses to the questionnaire, at 1 month and 1 year, all patients indicated a superior quality of vision in the eye treated with the radial pattern. No patient complained of optical problems such as halos, glare, or fluctuating daytime vision. However, complaints of halos or ghost images were noted at 1 and 12 months for three and four eyes, respectively, with the staggered pattern and for only one eye with the radial pattern. At 1 month, four patients reported better uncorrected near vision with the staggered eye while eight noted better corrected near vision with the radial eye. At 1 year, all patients preferred their radial eye for uncorrected and corrected near vision.

Discussion This study compared two treatment patterns for noncontact Ho:YAG LTK. Based on our review of earlier reports 5•6·16•18•19 and our experience with more than 200 eyes treated to date, 1.4· 12- 14 •17 the amount of induced refractive change can be largely controlled by the number and geometric position of the spots. In this study, we found that for a three-ring treatment pattern, radial distribution of spots was superior to staggered distribution in our objective and patients' subjective findings. Analyzing the representation of the two patterns, we find that in the radial pattern the orientation of the lines of strength, resulting from the vectoral combination of lines of tension produced by the single spot, is 28

more uniform and synergistic (Figure 6, left). In the staggered pattern, the vectoral interaction between the strengths produced by the spots of different rings leads to multiple orientation of the lines of tension scattered in several directions (Figure 6, right). Clinical observations resulting from the present study seem to support this theoretical assumption. Slitlamp examination of eyes treated with the radial pattern showed radially oriented lines of tension internal to the spots and circular belt-like lines of tension interconnecting the spots, exactly as proposed by the theoretical model. This clinical observation was confirmed by the retroillumination images obtained with the Scheimpflug camera. In the staggered pattern, the lines of tension internal to the spots had a fan-shaped orientation toward the nearby spots, while the stromal striae seemed to extend in all theoretically possible directions, including toward the central optical zone and the limbus. The resulting lines of stress, determined Table 6.

Mean endothelial cell density (cells/mm 2 ±SO) (upper line), central ultrasonic pachymetry (f.lm ±SO) (lower line) in eyes treated with the radial and staggered patterns. Postoperative Group

Radial

Preoperative

1 month

2804 ± 397

2890 ± 412

540 ± 87.5 2795 ± 460

Staggered

551 ± 79.5

J CATARACT REFRACT SURG-VOL 24, JANUARY 1998

547 ± 74

12 months

2799 ± 430 544 ± 905

2769 ± 417

2780 ± 398

545 ± 80

558 ± 92

HO:YAG LTK WITH RADIAL AND STAGGERED PATTERNS

Figure 6. (Vinciguerra) Theoretical orientation of the lines of tension in the radial (left) and staggered (right) patterns.

by the vectoral combination, produced corneal steepening in the midperiphery rather than in the center, as revealed by CVK (Figure 4). This positive asphericity is presumably less desirable than physiologic negative corneal asphericity, in which the highest dioptric power is located in the corneal center. After a few months, however, the striae in the staggered group disappeared and the corneas developed a more homogeneous distribution of dioptric power, with the highest values returning in the center of the cornea. In the radial pattern, the stromal striae connecting the spots of the inner and the middle rings remained visible for a longer period, extending to the 12 month follow-up visit. We believe the different orientations in the striae explain the differences in visual acuities, the CVK changes, and the symptoms obtained with the two patterns. With the staggered pattern, the striae produced within the optical zone caused the smaller gain in UCVA (due to less uniform stromal forces giving less corneal steepening) and the greater and more prolonged decrease in SCVA (due to a more irregular corneal surface). The recovery of SCVA to preoperative values typically occurred at 3 to 6 months postoperatively, coincident with the disappearance of the striae. The better uncorrected near vision reported after 1 month by four patients in the staggered group is presumably due to the greater multifocality (greater dioptric variation in the corrected zone) in these eyes and perhaps also to the eccentric location of the site of maximal correction. With the radial pattern, on the contrary, the absence of striae in the central optical zone preserved a homogeneous central corneal surface, as demonstrated by the CVK. This explains the smaller drop in SCVA immediately postoperatively and the

return of SCVA to preoperative levels by day 15. The persistence of the belt-like striae in the radial group may also explain the greater refractive correction and the absence of regression between months 9 and 12. In the staggered group, although the stromal striae disappeared after 6 to 9 months, regression was still occurring at the end of the follow-up. Finally, the subjective impressions of the patients favored the radial pattern, especially during the first months after bilateral treatment. This is undoubtedly partly attributable to the greater correction achieved with the radial pattern, especially since all the patients were undercorrected. However, we anticipate that the quality of corrected vision was also superior in the radial eyes.

Conclusions The present clinical study indicates that the geometric distribution of the spots significantly influences the outcome of noncontact Ho:YAG LTK. In particular, the radial pattern is superior to the staggered pattern because it produces a larger and more stable refractive change, preserves physiological corneal asphericity, and provides faster functional recovery, resulting in a higher level of patient satisfaction. We believe that these findings, as already demonstrated in incisional refractive surgery, 24 could be important for other refractive surgical procedures that involve stromal tissue contraction, such as contact LTK, radiofrequency thermokeratoplasty, and other techniques based on cross-linkage of proteins. Although longer follow-up is required, the efficacy and safety of this procedure suggest it is a promising approach for the treatment of low hyperopia.

References 1. Koch DD, Berry MJ, Vassiliadis AJ, et al. Noncontact holmium:YAG laser thermal keratoplasty. In: Salz JJ, ed, Corneal Laser Surgery. St Louis, MO, Mosby-Year Book, Inc, 1995, 247-254 2. Koch DD, Patrick TD, Menefee RF, et al. Laser photothermal keratoplasty: nonhuman primate results. ARVO abstract 375-47. Invest Ophthalmol Vis Sci 1992; 33:768 3. Ren Q, Simon G, Pard JM. Noncontact laser photothermal keratoplasty. III: histological study in animal eyes. J Refract Corneal Surg 1994; 10:529-539

J CATARACT REFRACT SURG-VOL 24, JANUARY 1998

29

HO:YAG LTK WITH RADIAL AND STAGGERED PATTERNS

4. Kohnen T, Husain SE, Koch DD. Corneal topographic changes after noncontact holmium:YAG laser thermal keratoplasty to correct hyperopia. J Cataract Refract Surg 1996; 22:427-435 5. Moreira H, Campos M, Sawusch MR, et al. Holmium laser thermokeratoplasty. Ophthalmology 1993; 100: 752-761 6. Seiler T, Matallana M, Bende T. Laser thermokeratoplasty by means of a pulsed holmium:YAG laser for hyperopic correction. Refract Corneal Surg 1990; 6:335339 7. Mapstone R. Measurement of corneal temperature. Exp Eye Res 1968; 7:237-243 8. Doss JD, Albillar Jl. A technique for the selective heating of corneal stroma. Contact Intraocular Lens MedJ 1980; 6:13-17 9. AquavellaJV, Smith RS, Shaw EL. Alterations in corneal morphology following thermokeratoplasty. Arch Ophthalmol 1976; 94:2082-2085 10. Shaw EL, Gassett AR. Thermokeratoplasty (TKP) temperature profile. Invest Ophthalmol1974; 13:181186 11. Ariyasu RG, Sand B, Menefee R, et al. Holmium laser thermokeratoplasty of 10 poorly sighted eyes. J Refract Corneal Surg 1995; 11:358-365 12. Koch DD, Kohnen T, Anderson JA, et al. Histopathologic changes and wound healing response following 10-pulse noncontact holmium: YAG laser thermal keratoplasty. J Refract Surg 1996; 12:623-634 13. Koch DD, Abarca A, Villarreal R, et al. Hyperopia correction by noncontact holmium:YAG laser thermal keratoplasty: clinical study with two-year follow-up. Ophthalmology 1996; 103:731-740 14. Koch DD, Kohnen T, McDonnell PJ, et al. Hyperopia correction by noncontact holmium:YAG laser thermal keratoplasty: United States Phase IIA clinical study with

30

15.

16.

17.

18.

19. 20.

21.

22.

23.

24.

a 1-year follow-up. Ophthalmology 1996; 103:15251535 Assil KK, Quantock AJ. Wound healing in response to keratorefractive surgery. Surv Ophthalmol1993; 38:289302 Thompson VM, Seiler T, Durrie DS, Cavanaugh TB. Holmium:YAG laser thermokeratoplasty for hyperopia and astigmatism: an overview. J Refract Corneal Surg 1993; 9(suppl):S134-S137 Kohnen T, Koch DD, McDonnell PJ, et al. Noncontact holmium:YAG laser thermal keratoplasty to correct hyperopia: 18-month follow-up. Ophthalmologica 1997; 211:274-282 Durrie DS, Schumer DJ, Cavanaugh TB. Holmium: YAG laser thermokeratoplasty for hyperopia. J Refract Corneal Surg 1994; 10(suppl):S277-S280 Seiler T. Ho:YAG laser thermokeratoplasty for hyperopia. Ophthalmol Clin North Am 1992; 5(4):773-780 Koch DD, Abarca A, Menefee RF, Berry MJ. Ho:YAG laser thermal keratoplasty: in vitro experiments. ARVO abstract 2667-40. Invest Ophthalmol Vis Sci 1993; 34:1246 Thompson V Holmium:YAG laser thermokeratoplasty for correction of astigmatism. J Refract Corneal Surg 1994; 10(suppl):S293 Simon G, Ren Q, Pard JM. Noncontact laser photothermal keratoplasty. II: refractive effects and treatment parameters in cadaver eyes. J Refract Corneal Surg 1994; 10:519-528 Pard JM, Ren Q, Simon G. Noncontact laser photothermal keratoplasty. 1: biophysical principles and laser beam delivery system. J Refract Corneal Surg 1994; 10:511-518 Hanna KD, Jouve FE, Waring GO III. Preliminary computer simulation of the effects of radial keratotomy. Arch Ophthalmol1989; 107:911-918

] CATARACT REFRACT SURG-VOL 24, JANUARY 1998