- Email: [email protected]
Three-dimensional Measurements of Idiopathic Macular Holes Using a Scanning Laser Tomograph
Three--dimensional Measurements of Idiopathic Macular Holes Using a Scanning Laser Tomograph Dov Weinberger, MD, 1 Hadas Stiebel, MD, 1 Dan D. Gaton, MD, 1 Ethan Priel, 2 Yuval Yassur, MD 1 Purpose: The diagnosis of macular holes is difficult. Confocal laser tomographic analysis of the retina permits the precise measurements of the macular surface. The authors used this technique to study the macular area of patients with full-thickness macular holes. The purposes of these studies is to perform three-dimensional mea surements of the macular holes and their rims and to search for a correlation between these parameters. Methods: Thirty-one eyes with idiopathic full-thickness macular holes underwent scanning of their affected macular area using the Heidelberg retinal tomograph (HRT). The authors evaluated the following parameters: area of the hole and its elevated rim, the maximal depth of the hole, and the maximal elevation of the rim at 12, 3, 6, and 9 o'clock. Results: The average hole area was 0.33 mm 2 , and the rim area average was 2.99 mm 2 . The average area of the rim was found to be 9.06 times larger than that of the hole (P < 0.001). The depth of the hole averaged 144 tlm. Conclusions: The area of the rim is usually bigger than that of the hole and in direct correlation to it. Also, the bigger the hole area, the greater its depth. The average height in each of the four quadrants (12, 3, 6, and 9 o'clock) correlates to the other quadrant heights (P < 0.001). Ophthalmology 1995;102:1445-1449
Macular holes are a common cause of decreased visual acuity. The anatomic characteristics may be difficult to define, especially when differentiation of full-thickness macular holes from the other stages of macular holes or pseudoholes is needed to decide on surgical intervention and to avoid unnecessary or incorrect surgery. A variety of instruments and means have been used to improve the diagnosis of macular holes-Amsler grid, Watzke-Allen sign, and 50-tlm argon laser aiming beam. 1- 3 Originally received: November 7, 1994. Revision accepted: June I, 1995. 1 Department of Ophthalmology, Beilinson Medical Center, Petah Tiqva and Sackler Faculty of Medicine, Tel Aviv University, Tel Aviv. 2
Mor Medical Institute, Bnei Brak, Israel. Reprint requests to Dov Weinberger, MD, Department of Ophthalmol ogy, Beilinson Medical Center, Petah Tiqva 49 100, Israel.
The main difficulty in accurately diagnosing macular holes has proven to be that of visualization, especially the intersection between the posterior vitreous face and the retinal surface. To overcome these difficulties, work has been done on an optical device designed to deliver a nar row green helium-neon laser beam, thus limiting the in tense scattering observed in conventional slit-lamp bio microscopy.4 All of these methods are either subjective tests or they were used in addition to regular slit-lamp biomicroscopy. They are highly dependent on the oper ator'sjudgment and the patient's cooperation. In addition, they require a clear medium and a dilated pupil to obtain data with good reproducibility. The use of echography in diagnosing macular hole and its clinical correlation has been suggested re cently. 5 •6 The authors found that echography is very effective in detecting the position of the posterior vit reous face, the operculum in the macular region and
1445
Ophthalmology
Volume 102, Number 10, October 1995
macular thickening. This technique correlates accu rately to the clinical features. Confocal laser tomographic analysis of the retina in eyes with macular holes and other focal macular diseases has been described in a few cases by Bartsch et al. 7 Re cently, Menezes et al 8 evaluated the reproducibility of the Heidelberg retinal tomograph (HRT) (Heidelberg Engi neering, Heidelberg, Germany, Software Version 1.11) in measuring the normal macular surface and concluded that the HRT could be used potentially to quantify small changes in retinal lesions. The HRT is a confocal scanning laser tomograph and has been commercially available since 1992. It is used primarily to evaluate progressive optic nerve head damage in glaucoma. For the purpose of this study, the HRT was used to scan the macular area of eyes with macular holes. We used this technique to study the topography ofthe macula and surrounding retinal surface in a series of patients with the diagnosis of full-thickness macular holes. These mea surements were used to create a three-dimensional map of the various parameters of the macular holes and their elevated rims, to analyze the results, and to correlate these parameters. The technology and technique that we used are presented.
Materials and Methods Thirty-one eyes of 31 patients (21 women, 10 men; mean age, 67 years) with idiopathic full-thickness macular holes were included in the study. Eighteen unaffected fellow eyes were used as a control group. The diagnosis of macular hole was made via biomi croscopy with a 78-diopter or Goldmann three-mirror lens, indirect ophthalmoscopy, and fluorescein angiog raphy. A macular hole was diagnosed when there was a full-thickness, sharp-edged retinal defect with elevated rims in the foveal area associated with a hyperfluorescent window defect on fluorescein angiography. The HRT was used to scan the macular area of both eyes. The scan angle size was 10° of the retina, and the center of the scanned region was the fovea. Heidelberg retinal tomograph uses a near-infrared laser (670 nm) to scan the posterior pole. It is table-mounted, easy to use, and manipulated similarly to a fundus camera. The retinal images are viewed on a high-resolution color video mon itor. The HRT software creates a three-dimensional image by aligning 32 consecutive, two-dimensional retinal im ages scanned by the laser. Each two-dimensional image is made up of 256 X 256 pixels scanned along the x and the y axes. The scanner then moves the laser head 50 80 Jlm posteriorly to the previously scanned plane. Thirty two transverse images are acquired along the z axis over a section of retina varying between 0.5 and 4.00 mm deep. The HRT's confocal scanning laser technology relies on an extremely shallow depth of field to record only the image at the focal plane while disregarding any infor mation outside of the focal plane. Thus, the height mea surement reproducibility ("resolution" along the z axis)
1446
is approximately 30 Jlm. The optical resolution along the x and y axes is approximately 10 Jlm. All the maculae were scanned at a scan depth of 2.0 mm. Acquisition time for a 32-image series was 1.6 sec onds, after which time the operator could evaluate the quality ofeach series by scrutinizing fixation, illumination, and scan depth. Processing and storage of the acquired data (by a 485 OX IBM computer) for each series took less than I minute. The topographic shape of the scanned area then could be displayed and analyzed in several ways. Topographic data, in numeric form, were summarized in a 16 X 16 square matrix, each square representing the average height value (in microns) of an additional 16 X 16 pixels (Fig 1). All height and depth values were cal culated relative to the focal plane which was the flat retina surrounding the elevated region around the macula. This method corresponds to that previously described by oth ers.9,JO In addition, it was possible to evaluate the topographic height variations of a given area interactively by placing a movable x axis across the image and manipulating a mouse or the keyboard. Figures 2 and 3 represent the height variations (along a chosen x axis) ofa normal mac ula and an eye with a macular hole, respectively. Figure 4 represents a three-dimensional topographic map of the macular hole and the elevated rim. To better detect the extreme limits of the sensory ele vation surrounding the macular hole, we used the red green stereoscopic viewing option available in the HRT. By wearing special glasses with separate red-green lenses and viewing the screen made up of two slightly offset red green images, we were able to locate the exact border of the elevation. By drawing a contour line along this border, the area ofthe elevated retina and the hole was calculated. The reference level used was that of the contour line (flat retina). The measurements were taken four times by the same observer and the same instrument. The fluctuations were
Figure 1. Topographic map in numeric form superimposed on a macula image. Each number, in microns, corresponds to the average height value of 256 (16 X 16) pixels. Positive values represent the depth of the hole.
Weinberger et al · 3D Measurements of Macular Hole
Figure 2. Height variations of a normal macula along a given x axis.
minimized by using the average values detected in the four measurements. In the four consecutive measurements, the average variability of the hole depth, hole and rim area, and the height of the rim was 58 11m, and in 72% of the patients the average variability was less than 50 11m. These findings correlated to the reproducibility of measurements found by Dreher et al. 11 We evaluated each hole according to the following pa rameters: 1. Area of the hole, in millimeters squared; 2. Area of the rim, in millimeters squared; 3. Area of the hole and the elevated rim, in millimeters squared; 4. Average depth of the hole, in microns; 5. The deepest point found in the hole, in microns; and 6. Elevation ofthe rim at 12, 3, 6, and 9 o'clock hours, in microns.
Figure 3. Height variations of a macular hole and adjacent retina along a given x axis. Notice the significant depression in the hole area and the elevation of the rim.
the hole plus the elevated rim ranged from 1.1 to 6.5 mm 2 (average, 3.3 mm 2 ). The maximum hole depth ranged from 30 to 300 11m (average, 144 11m). The depth of the hole's base ranged from 20 to 242 11m (average, 103 11m). The height of the elevation in the various quadrants 12, 3, 6, and 9 o'clock is summarized in Table 1. Detailed measurements of the 31 macular holes can be seen in Table 1. In the 18 control eyes (the unaffected fellow eye of the same patient with the macular hole), a normal flat reti'nal surface was seen. The foveal pit could not always be ob served accurately on the video monitor at the time of examination, but in all eyes a shallow depression of the macular area, ranging from 20 to 45 11m, was noted.
These measurements enabled us to assess the various relations among the measured parameters: 1. The ratio between the rim area and the area of the hole; 2. The ratio between the area ofthe hole and its depth; 3. The relation of heights along the rim circumference; and 4. The possible correlation among the hole area, rim area, and hole depth, and the level of significance of this correlation.
Results The area ofthe hole ranged from 0.10 to 0.811 mm 2 (av erage, 0.33 mm 2). The area of the elevated rim ranged from 0.97 to 6.07 mm 2 (average, 2.99 mm 2 ). The area of
Figure 4. Three-dimensional topographic map of the macular hole and the elevated rim. Notice the blood vessel striature on the flat retina around the hole.
1447
Ophthalmology
Volume 102, Number 10, October 1995
Table 1. Measurement Size of Variant Parameters of Macular Hole and Elevated Rim of 31 Patients Parameters
Maximum
Minimum
Average
Area of hole Area of rim (mm2) Area of hole and rim (mm2) Average hole depth (~tm) Maximum hole depth (~tm) Rim elevation (~tm)
0.81 6.07 6.5 242 300
0.10 0.97 1.158 20 30
0.33 2.99 3.3 103 144
570 287
32 27 23 25
175 160 159 170
(mm2)
12
3
6 9
800
530
Statistical correlation analysis (according to Pearson's correlation coefficient) showed the following results: l. The area of the rim is usually bigger than the area of the hole (average rim/hole area, 9.06 mm 2 ) (P < 0.001); 2. The area of the rim correlates to the area of the hole: if the hole area is bigger, the rim area will be bigger (P < 0.00 l ); 3. The area of the hole area correlates to the depth of the hole: if the hole area is bigger, the depth of the hole will be bigger (P < 0.06); 4. The average height in each of the four quadrants 12, 3, 6, and 9 o'clock correlates to the average of other quadrant heights (P < 0.00 l ).
Discussion The detection of macular holes, especially in their early stages of development, is complex due to difficulties in accurately viewing and measuring the surface topography of the macular area. Recently, vitrectomy with removal ofthe posterior vitreous face, with or without use oftrans forming growth factor-/3 2, has been suggested as a treat ment for macular hole, underscoring the need for an in strument that can detect flattening of the elevated rim of the hole and prove the closure ofthe hole. 12 •13 Currently available slit-lamp biomicroscopy, using a bright light source or laser beam, 4 is a nonquantitative and subjective method of examination. Echographic techniques using the perpendicular axis (z axis) usually are not sufficiently sensitive due to the limitation of resolution to 200 ~tm in standardized A and B scans. 7 The confocal laser imaging technology is sensitive enough to detect elevation and depression surface topographic changes of approximately 30 to 50 ~tm. The evaluation of 32 transverse optical sec tion images at consecutive height planes (50-80 ~tm each) and the measurement of 65,636 pixels (these measure ments can be obtained in eyes with undilated pupils and a moderate amount of lenticular opacity) can give the surgeon some of the necessary details about the surface of the macular region, such as size and average depth of
1448
the hole, extent and elevation of the surround cuff, and the presence of preretinal membranes and vitreoretinal adhesions if they are thicker than the resolving power of the instrument. Regarding information on macular holes in their early stages, the measurements are reliable only if there is a hole formation (stage 2-4, according to Gass's classification 14) and if the hole is deeper than 30 to 50 ~tm and the diameter is larger than 10 to 20 ~tm. The results of our study showed that the average area of the hole was approximately 0.33 mm 2 , which is the average area of the fovea, but the average area of the el evated rim surrounding the hole was 2.99 mm 2 , which is 9.06 times larger than the hole itself. This means that the visual potential ofthe patient was compromised by a much larger area of damage than that of the hole itself. The average depth of the hole was 103 ~tm, and the average maximum depth was l44~tm. This measurement, which was taken from the surface of the retina (the ref erence level), shows that the hole penetrates only into the retinal tissue and that the deeper surfaces are not regular. This information also was obtained histologically by Frangieh et al, 15 who demonstrated that minor hyperplasia and irregularity of the retinal pigment epithelium could be seen in the area of the macular hole. The average width of the retina in the macular region was measured using Shahidi et al's 16 method. They showed that the retinal thickness was 180 ~tm in the foveola and 300 to 360 ~tm in the macula. The depth of the holes that we measured was within this thickness of the retinal tissue. The fact that the elevation in the four quadrants of the rim was relatively uniform seems to indicate that gravi tational forces do not play a role in the subretinal fluid accumulation and that the traction forces of the posterior vitreous face on the different quadrants of the rim are relatively equal. In the current series, we examined patients with full thickness macular holes, yet this technique also may be applicable in evaluating early stages of this disease or other diseases with sensory retinal elevation such as central se rous choroidopathy. The advantages oflaser scanning tomography are its accuracy and reproducibility. This technique can quantify, register, and analyze macular topographic
Weinberger et al · 3D Measurements of Macular Hole changes occurring in various macular diseases over a period of time.
References I. Amsler M. Quantitative and qualitative vision. Trans Ophthalmol Soc UK 1949;69:397-402. 2. Watzke RC, Allen L. Subjective slitbeam sign for macular disease. Am J Ophthalmol 1969;68 :449-53. 3. Martinez J, Smiddy WE, Kim J, Gass JDM. Differentiating macular holes from macular pseudoholes. Am J Ophthalmol 1994;117:762-7. 4. Ogura Y, Shahidi M, Mori MT, et al. Improved visualization of macular hole lesions with laser biomicroscopy. Arch Ophthalmol 1991;109:957-61. 5. Van Newkirk MR, Gass JDM, Callanan D, eta!. Follow up and ultrasonographic examination of patients with mac ular pseudo-operculum. Am J Ophthalmol 1994; 117: 13-8. 6. Dugel PU, Smiddy WE, Byrne SF, eta!. Macular hole syn drome. Echographic findings with clinical correlation. Oph thalmology 1994;101:815-21. 7. Bartsch D-U, Intaglietta M, Bille JF, eta!. Confocal laser tomographic analysis ofthe retina in eyes with macular hole formation and other focal macular diseases. Am J Ophthal mol 1989; 108:277-87.
8. Menezes A V, Giunta M, Chisholm L, et al. Reproducibility of topographic measurements of the macula with a scanning laser ophthalmoscope. Ophthalmology 1995;102:230-5. 9. Lusky M, Bosem ME, Weinreb RN. Reproducibility of optic nerve head topography measurements in eyes with undilated pupils. J Glaucoma 1993;2: 104-9. I0. Mikelberg FS, Wijsman K, Schulzer M. Reproducibility of topographic parameters obtained with the Heidelberg retina tomograph. J Glaucoma 1993;2:101-3. 11. Dreher AW, Tso PC, Weinreb RN. Reproducibility ofto pographic measurements of the normal and glaucomatous optic nerve head with the laser tomographic scanner. Am J Ophthalmol 1991;111 :221-9. 12. Kelly NE, Wendel RT. Vitreous surgery for idiopathic mac ular hole. Results of a pilot study. Arch Ophthalmol 1991; I 09:654-9. 13. Glaser BM, Michels RG, Kuppermann BD, et a!. Trans forming growth factor-/3 2 for the treatment of full-thickness macular hole. A prospective randomized study. Ophthal mology 1992;99: 1162-73. 14. Gass DM. Idiopathic senile macular hole: its early stages and pathogenesis. Arch Ophthalmol 1988;106:629-39. 15 . Frangieh GT, Green WR, Engel HM. A histological study of macular cyst and holes. Retina 198 I; I :3 11-6. 16. Shahidi M, Seimer R, Mori M. Topograph of the retinal thickness in normal subjects. Ophthalmology 1990;97: 1120 24.
144~
Macular holes are a common cause of decreased visual acuity. The anatomic characteristics may be difficult to define, especially when differentiation of full-thickness macular holes from the other stages of macular holes or pseudoholes is needed to decide on surgical intervention and to avoid unnecessary or incorrect surgery. A variety of instruments and means have been used to improve the diagnosis of macular holes-Amsler grid, Watzke-Allen sign, and 50-tlm argon laser aiming beam. 1- 3 Originally received: November 7, 1994. Revision accepted: June I, 1995. 1 Department of Ophthalmology, Beilinson Medical Center, Petah Tiqva and Sackler Faculty of Medicine, Tel Aviv University, Tel Aviv. 2
Mor Medical Institute, Bnei Brak, Israel. Reprint requests to Dov Weinberger, MD, Department of Ophthalmol ogy, Beilinson Medical Center, Petah Tiqva 49 100, Israel.
The main difficulty in accurately diagnosing macular holes has proven to be that of visualization, especially the intersection between the posterior vitreous face and the retinal surface. To overcome these difficulties, work has been done on an optical device designed to deliver a nar row green helium-neon laser beam, thus limiting the in tense scattering observed in conventional slit-lamp bio microscopy.4 All of these methods are either subjective tests or they were used in addition to regular slit-lamp biomicroscopy. They are highly dependent on the oper ator'sjudgment and the patient's cooperation. In addition, they require a clear medium and a dilated pupil to obtain data with good reproducibility. The use of echography in diagnosing macular hole and its clinical correlation has been suggested re cently. 5 •6 The authors found that echography is very effective in detecting the position of the posterior vit reous face, the operculum in the macular region and
1445
Ophthalmology
Volume 102, Number 10, October 1995
macular thickening. This technique correlates accu rately to the clinical features. Confocal laser tomographic analysis of the retina in eyes with macular holes and other focal macular diseases has been described in a few cases by Bartsch et al. 7 Re cently, Menezes et al 8 evaluated the reproducibility of the Heidelberg retinal tomograph (HRT) (Heidelberg Engi neering, Heidelberg, Germany, Software Version 1.11) in measuring the normal macular surface and concluded that the HRT could be used potentially to quantify small changes in retinal lesions. The HRT is a confocal scanning laser tomograph and has been commercially available since 1992. It is used primarily to evaluate progressive optic nerve head damage in glaucoma. For the purpose of this study, the HRT was used to scan the macular area of eyes with macular holes. We used this technique to study the topography ofthe macula and surrounding retinal surface in a series of patients with the diagnosis of full-thickness macular holes. These mea surements were used to create a three-dimensional map of the various parameters of the macular holes and their elevated rims, to analyze the results, and to correlate these parameters. The technology and technique that we used are presented.
Materials and Methods Thirty-one eyes of 31 patients (21 women, 10 men; mean age, 67 years) with idiopathic full-thickness macular holes were included in the study. Eighteen unaffected fellow eyes were used as a control group. The diagnosis of macular hole was made via biomi croscopy with a 78-diopter or Goldmann three-mirror lens, indirect ophthalmoscopy, and fluorescein angiog raphy. A macular hole was diagnosed when there was a full-thickness, sharp-edged retinal defect with elevated rims in the foveal area associated with a hyperfluorescent window defect on fluorescein angiography. The HRT was used to scan the macular area of both eyes. The scan angle size was 10° of the retina, and the center of the scanned region was the fovea. Heidelberg retinal tomograph uses a near-infrared laser (670 nm) to scan the posterior pole. It is table-mounted, easy to use, and manipulated similarly to a fundus camera. The retinal images are viewed on a high-resolution color video mon itor. The HRT software creates a three-dimensional image by aligning 32 consecutive, two-dimensional retinal im ages scanned by the laser. Each two-dimensional image is made up of 256 X 256 pixels scanned along the x and the y axes. The scanner then moves the laser head 50 80 Jlm posteriorly to the previously scanned plane. Thirty two transverse images are acquired along the z axis over a section of retina varying between 0.5 and 4.00 mm deep. The HRT's confocal scanning laser technology relies on an extremely shallow depth of field to record only the image at the focal plane while disregarding any infor mation outside of the focal plane. Thus, the height mea surement reproducibility ("resolution" along the z axis)
1446
is approximately 30 Jlm. The optical resolution along the x and y axes is approximately 10 Jlm. All the maculae were scanned at a scan depth of 2.0 mm. Acquisition time for a 32-image series was 1.6 sec onds, after which time the operator could evaluate the quality ofeach series by scrutinizing fixation, illumination, and scan depth. Processing and storage of the acquired data (by a 485 OX IBM computer) for each series took less than I minute. The topographic shape of the scanned area then could be displayed and analyzed in several ways. Topographic data, in numeric form, were summarized in a 16 X 16 square matrix, each square representing the average height value (in microns) of an additional 16 X 16 pixels (Fig 1). All height and depth values were cal culated relative to the focal plane which was the flat retina surrounding the elevated region around the macula. This method corresponds to that previously described by oth ers.9,JO In addition, it was possible to evaluate the topographic height variations of a given area interactively by placing a movable x axis across the image and manipulating a mouse or the keyboard. Figures 2 and 3 represent the height variations (along a chosen x axis) ofa normal mac ula and an eye with a macular hole, respectively. Figure 4 represents a three-dimensional topographic map of the macular hole and the elevated rim. To better detect the extreme limits of the sensory ele vation surrounding the macular hole, we used the red green stereoscopic viewing option available in the HRT. By wearing special glasses with separate red-green lenses and viewing the screen made up of two slightly offset red green images, we were able to locate the exact border of the elevation. By drawing a contour line along this border, the area ofthe elevated retina and the hole was calculated. The reference level used was that of the contour line (flat retina). The measurements were taken four times by the same observer and the same instrument. The fluctuations were
Figure 1. Topographic map in numeric form superimposed on a macula image. Each number, in microns, corresponds to the average height value of 256 (16 X 16) pixels. Positive values represent the depth of the hole.
Weinberger et al · 3D Measurements of Macular Hole
Figure 2. Height variations of a normal macula along a given x axis.
minimized by using the average values detected in the four measurements. In the four consecutive measurements, the average variability of the hole depth, hole and rim area, and the height of the rim was 58 11m, and in 72% of the patients the average variability was less than 50 11m. These findings correlated to the reproducibility of measurements found by Dreher et al. 11 We evaluated each hole according to the following pa rameters: 1. Area of the hole, in millimeters squared; 2. Area of the rim, in millimeters squared; 3. Area of the hole and the elevated rim, in millimeters squared; 4. Average depth of the hole, in microns; 5. The deepest point found in the hole, in microns; and 6. Elevation ofthe rim at 12, 3, 6, and 9 o'clock hours, in microns.
Figure 3. Height variations of a macular hole and adjacent retina along a given x axis. Notice the significant depression in the hole area and the elevation of the rim.
the hole plus the elevated rim ranged from 1.1 to 6.5 mm 2 (average, 3.3 mm 2 ). The maximum hole depth ranged from 30 to 300 11m (average, 144 11m). The depth of the hole's base ranged from 20 to 242 11m (average, 103 11m). The height of the elevation in the various quadrants 12, 3, 6, and 9 o'clock is summarized in Table 1. Detailed measurements of the 31 macular holes can be seen in Table 1. In the 18 control eyes (the unaffected fellow eye of the same patient with the macular hole), a normal flat reti'nal surface was seen. The foveal pit could not always be ob served accurately on the video monitor at the time of examination, but in all eyes a shallow depression of the macular area, ranging from 20 to 45 11m, was noted.
These measurements enabled us to assess the various relations among the measured parameters: 1. The ratio between the rim area and the area of the hole; 2. The ratio between the area ofthe hole and its depth; 3. The relation of heights along the rim circumference; and 4. The possible correlation among the hole area, rim area, and hole depth, and the level of significance of this correlation.
Results The area ofthe hole ranged from 0.10 to 0.811 mm 2 (av erage, 0.33 mm 2). The area of the elevated rim ranged from 0.97 to 6.07 mm 2 (average, 2.99 mm 2 ). The area of
Figure 4. Three-dimensional topographic map of the macular hole and the elevated rim. Notice the blood vessel striature on the flat retina around the hole.
1447
Ophthalmology
Volume 102, Number 10, October 1995
Table 1. Measurement Size of Variant Parameters of Macular Hole and Elevated Rim of 31 Patients Parameters
Maximum
Minimum
Average
Area of hole Area of rim (mm2) Area of hole and rim (mm2) Average hole depth (~tm) Maximum hole depth (~tm) Rim elevation (~tm)
0.81 6.07 6.5 242 300
0.10 0.97 1.158 20 30
0.33 2.99 3.3 103 144
570 287
32 27 23 25
175 160 159 170
(mm2)
12
3
6 9
800
530
Statistical correlation analysis (according to Pearson's correlation coefficient) showed the following results: l. The area of the rim is usually bigger than the area of the hole (average rim/hole area, 9.06 mm 2 ) (P < 0.001); 2. The area of the rim correlates to the area of the hole: if the hole area is bigger, the rim area will be bigger (P < 0.00 l ); 3. The area of the hole area correlates to the depth of the hole: if the hole area is bigger, the depth of the hole will be bigger (P < 0.06); 4. The average height in each of the four quadrants 12, 3, 6, and 9 o'clock correlates to the average of other quadrant heights (P < 0.00 l ).
Discussion The detection of macular holes, especially in their early stages of development, is complex due to difficulties in accurately viewing and measuring the surface topography of the macular area. Recently, vitrectomy with removal ofthe posterior vitreous face, with or without use oftrans forming growth factor-/3 2, has been suggested as a treat ment for macular hole, underscoring the need for an in strument that can detect flattening of the elevated rim of the hole and prove the closure ofthe hole. 12 •13 Currently available slit-lamp biomicroscopy, using a bright light source or laser beam, 4 is a nonquantitative and subjective method of examination. Echographic techniques using the perpendicular axis (z axis) usually are not sufficiently sensitive due to the limitation of resolution to 200 ~tm in standardized A and B scans. 7 The confocal laser imaging technology is sensitive enough to detect elevation and depression surface topographic changes of approximately 30 to 50 ~tm. The evaluation of 32 transverse optical sec tion images at consecutive height planes (50-80 ~tm each) and the measurement of 65,636 pixels (these measure ments can be obtained in eyes with undilated pupils and a moderate amount of lenticular opacity) can give the surgeon some of the necessary details about the surface of the macular region, such as size and average depth of
1448
the hole, extent and elevation of the surround cuff, and the presence of preretinal membranes and vitreoretinal adhesions if they are thicker than the resolving power of the instrument. Regarding information on macular holes in their early stages, the measurements are reliable only if there is a hole formation (stage 2-4, according to Gass's classification 14) and if the hole is deeper than 30 to 50 ~tm and the diameter is larger than 10 to 20 ~tm. The results of our study showed that the average area of the hole was approximately 0.33 mm 2 , which is the average area of the fovea, but the average area of the el evated rim surrounding the hole was 2.99 mm 2 , which is 9.06 times larger than the hole itself. This means that the visual potential ofthe patient was compromised by a much larger area of damage than that of the hole itself. The average depth of the hole was 103 ~tm, and the average maximum depth was l44~tm. This measurement, which was taken from the surface of the retina (the ref erence level), shows that the hole penetrates only into the retinal tissue and that the deeper surfaces are not regular. This information also was obtained histologically by Frangieh et al, 15 who demonstrated that minor hyperplasia and irregularity of the retinal pigment epithelium could be seen in the area of the macular hole. The average width of the retina in the macular region was measured using Shahidi et al's 16 method. They showed that the retinal thickness was 180 ~tm in the foveola and 300 to 360 ~tm in the macula. The depth of the holes that we measured was within this thickness of the retinal tissue. The fact that the elevation in the four quadrants of the rim was relatively uniform seems to indicate that gravi tational forces do not play a role in the subretinal fluid accumulation and that the traction forces of the posterior vitreous face on the different quadrants of the rim are relatively equal. In the current series, we examined patients with full thickness macular holes, yet this technique also may be applicable in evaluating early stages of this disease or other diseases with sensory retinal elevation such as central se rous choroidopathy. The advantages oflaser scanning tomography are its accuracy and reproducibility. This technique can quantify, register, and analyze macular topographic
Weinberger et al · 3D Measurements of Macular Hole changes occurring in various macular diseases over a period of time.
References I. Amsler M. Quantitative and qualitative vision. Trans Ophthalmol Soc UK 1949;69:397-402. 2. Watzke RC, Allen L. Subjective slitbeam sign for macular disease. Am J Ophthalmol 1969;68 :449-53. 3. Martinez J, Smiddy WE, Kim J, Gass JDM. Differentiating macular holes from macular pseudoholes. Am J Ophthalmol 1994;117:762-7. 4. Ogura Y, Shahidi M, Mori MT, et al. Improved visualization of macular hole lesions with laser biomicroscopy. Arch Ophthalmol 1991;109:957-61. 5. Van Newkirk MR, Gass JDM, Callanan D, eta!. Follow up and ultrasonographic examination of patients with mac ular pseudo-operculum. Am J Ophthalmol 1994; 117: 13-8. 6. Dugel PU, Smiddy WE, Byrne SF, eta!. Macular hole syn drome. Echographic findings with clinical correlation. Oph thalmology 1994;101:815-21. 7. Bartsch D-U, Intaglietta M, Bille JF, eta!. Confocal laser tomographic analysis ofthe retina in eyes with macular hole formation and other focal macular diseases. Am J Ophthal mol 1989; 108:277-87.
8. Menezes A V, Giunta M, Chisholm L, et al. Reproducibility of topographic measurements of the macula with a scanning laser ophthalmoscope. Ophthalmology 1995;102:230-5. 9. Lusky M, Bosem ME, Weinreb RN. Reproducibility of optic nerve head topography measurements in eyes with undilated pupils. J Glaucoma 1993;2: 104-9. I0. Mikelberg FS, Wijsman K, Schulzer M. Reproducibility of topographic parameters obtained with the Heidelberg retina tomograph. J Glaucoma 1993;2:101-3. 11. Dreher AW, Tso PC, Weinreb RN. Reproducibility ofto pographic measurements of the normal and glaucomatous optic nerve head with the laser tomographic scanner. Am J Ophthalmol 1991;111 :221-9. 12. Kelly NE, Wendel RT. Vitreous surgery for idiopathic mac ular hole. Results of a pilot study. Arch Ophthalmol 1991; I 09:654-9. 13. Glaser BM, Michels RG, Kuppermann BD, et a!. Trans forming growth factor-/3 2 for the treatment of full-thickness macular hole. A prospective randomized study. Ophthal mology 1992;99: 1162-73. 14. Gass DM. Idiopathic senile macular hole: its early stages and pathogenesis. Arch Ophthalmol 1988;106:629-39. 15 . Frangieh GT, Green WR, Engel HM. A histological study of macular cyst and holes. Retina 198 I; I :3 11-6. 16. Shahidi M, Seimer R, Mori M. Topograph of the retinal thickness in normal subjects. Ophthalmology 1990;97: 1120 24.
144~