Internal Reflectivity of Filtering Blebs versus Intraocular Pressure in Patients with Recent Trabeculectomy

Abstract

Purpose.: The authors conducted a morphology analysis of filtering blebs with anterior segment optical coherence tomography (AS-OCT) in patients shortly after trabeculectomy to find further criteria for success or failure of a filtering bleb.

Methods.: Thirty-seven patients (43 eyes) who had recently (<2 years) undergone trabeculectomy were measured 1 to 4 times with AS-OCT. In contrast to ultrasound biomicroscopy, AS-OCT eliminates the need for a water bath in contact with the globe. This is of particular value in a study of morphologic parameters that may be altered by contact and in the postoperative period when concerns of infection are paramount. In this study, the internal reflectivity of the fluid-filled cavity of the filtering bleb was compared with the intraocular pressure (IOP).

Results.: Mean age of the patients was 69.5 years (range, 27–84 years). Average preoperative IOP of the 43 eyes was 20.6 mm Hg (range, 10–42 mm Hg), average postoperative IOP at the time of AS-OCT measurement was 10.6 mm Hg (range, 2–40 mm Hg), and mean IOP reduction was 10.0 mm Hg (48.5%). Internal reflectivity of the fluid-filled cavity of the filtering bleb correlated very well to the IOP (R ² = 0.3762, n = 61, P < 0.0001).

Conclusions.: AS-OCT is a promising tool with which to analyze the morphology of the filtering bleb after trabeculectomy. A highly significant correlation between the internal reflectivity of the fluid-filled cavity of the filtering bleb and the IOP could be shown that may have future impact in evaluating the postoperative success or failure of filtering blebs.

Trabeculectomy is the surgical treatment of choice for patients with uncontrolled glaucoma. The procedure was introduced by Cairns in 1968. ¹ The success and outcome of trabeculectomy depends on the formation of a filtering bleb, which is an elevation of the conjunctiva over the sclerotomy site. ^{2 –5} The bleb allows the filtration of aqueous humor, thus lowering the intraocular pressure (IOP). ⁶  

Several grading systems have been proposed to classify and characterize bleb morphology, which is an indicator of bleb function and bleb-related complications and, therefore, an important clinical parameter. ^{3,5,7 –10}  

In this study we measured filtering blebs with an anterior segment (AS) optical coherence tomograph (OCT; Visante, model 1000; Carl Zeiss Meditec Inc., Dublin, CA). The principles of OCT ¹¹ and its application in imaging the AS are well described. In the context of glaucoma, ^{12 –20} AS-OCT has also been used in the assessment of anterior chamber angles. ¹⁷  

Unlike ultrasound biomicroscopy, imaging with AS-OCT requires no contact with the globe. ^{21 –27} This is an advantage in a study of morphologic parameters that may be altered by contact and during the postoperative time when infections are dreaded. 

We aimed to estimate the applicability of the AS-OCT to analyze the internal bleb structure, especially the reflectivity of the fluid-filled bleb cavity, and to show a possible correlation between AS-OCT results and clinical examination (i.e., measurement of the IOP). 

Methods

This study was approved by the local ethics committee (SPUK; University Hospital of Zurich, Zurich) and was conducted in accordance with the ethical standards of the Declaration of Helsinki. 

We included 37 glaucoma patients who had undergone trabeculectomy at the University Hospital of Zurich within the past 2 years before initiation of the study. Patients with cystic filtering blebs were excluded. Written informed consent was obtained from all patients. We examined trabeculectomized eyes with the Zeiss Visante OCT Model 1000, which is a high-resolution tomographic and biomicroscopic device specifically designed for anterior segment imaging and measurement. It has an axial resolution of 18 μm. The principle is based on low-coherence interferometry using a 1310-nm super-luminescent light-emitting diode. As does ultrasound B-scan, the Visante OCT acquires multiple A-scans and aligns them to construct two-dimensional images. Scanning of the AS structures is made according to a measuring principle similar to that used for scanning of the retina, which has been described previously. ¹¹  

Given that intraocular tissues show different optical properties, the OCT is capable of defining the internal structures (e.g., fluid collection, conjunctival/scleral tissue reflectivity) in precise detail. In the study of Izatt et al. ¹³ the usability of the OCT in the anterior eye was demonstrated in healthy subjects for the first time. Radhakrishnan et al. ¹² measured healthy subjects with a hand-held AS-OCT. Miura et al. ⁴ examined four eyes with AS-OCT, and the morphologic structure was discussed. Leung et al. ¹⁴ described the intrableb morphology with the AS-OCT in 14 patients and grouped the intrableb reflectivity into low, medium, and high. Kawana et al. ²⁸ analyzed 38 filtering blebs in 31 patients of Asian origin with 3D-CAS OCT. A significant correlation between the volume of the hyporeflective reticulate area and the IOP could be shown. ²⁸ In our study, the calculated internal reflectivity of the fluid-filled cavity of the filtering bleb was compared with the IOP obtained on the same day by applanation tonometry. All measurements were performed by an examiner masked to the study and to the patient's data. OCT scans were conducted at the same background illumination, with the window blinds completely shut and the room light turned down to the minimum with a dimmer. Acquisition of data by chart review was undertaken to document relevant clinical data, including the patient's initials, age, type of glaucoma, date of trabeculectomy, side, morphology of the bleb by slit lamp examination as last documented before OCT measurement, date of AS-OCT-measurement, IOP before surgery, IOP after surgery at AS-OCT measurement, and postoperative IOP reduction in absolute values and in percentage compared with preoperative values as last documented before trabeculectomy (Table 1). 

Table 1.

 

Data for All Included Patients

Patient Initials	Age (y)	Type of Glaucoma	Date of TE	Side	Morphology of Bleb	Date of AS-OCT	IOP Pre-op	IOP Post-op at AS-OCT	IOP Red Pre-Post TE	IOP Red Pre-Post TE (%)
BM	27	PDG	5-16-07	OD	Prominent, MC+	5-22-07	25	3	22	88
BK	55	POAG	5-16-07	OD	Flat, MC−	5-22-07	30	9	21	70
BR	76	PACG	11-22-06	OS	Flat, MC+	5-15-07	18	8	10	55
			11-22-06	OS	MC+	10-22-07	18	6	12	67
BW	73	POAG	9-26-07	OD	Prominent, MC+	10-17-07	10	9	1	10
DT	57	POAG	2-07-07	OD	Prominent, MC+	5-15-07	16	13	3	19
EM	72	PEX	10-18-06	OD	Flat	5-21-07	16	10	6	38
			11-22-06	OS	Flat	5-21-07	16	10	14	58
GM	84	POAG	6-27-07	OS	Prominent	10-18-07	14	8	6	43
GP	71	PEX	4-27-07	OD	Flat (MC)	5-2-07	36	34	2	6
			4-27-07	OD	Prominent, (MC+)	5-8-07	36	13	23	64
			4-27-07	OD	Prominent, MC+	5-15-07	36	4	32	89
			4-27-07	OD	Flat, MC?	10-17-07	36	12	24	67
HU	63	Uveitis	2-13-07	OS	Flat, MC+	5-7-07	26	18	8	31
IH	79	POAG	3-7-07	OD	Prominent, MC+	5-8-07	14	3	11	79
			3-7-07	OD	Prominent, MC+	5-10-07	14	5	9	64
			3-7-07	OD	Flat, MC+	10-22-07	14	10	4	29
KE	66	PEX	11-29-06	OD	Prominent, MC+	5-7-07	11	6	5	46
KS	63	PACG	12-11-06	OD	Flat	5-22-07	12	12	0	0
KM	79	PEX	5-11-07	OD	Prominent, MC+	5-21-07	18	6	12	67
			5-11-07	OD	Flat, MC+	10-17-07	18	12	6	33
KM	64	PEX	7-2-07	OS	Prominent, MC+	10-19-07	30	13	17	57
LT	82	POAG	11-29-06	OS	Prominent	5-15-07	22	7	15	68
			11-29-06	OS	Prominent	10-23-07	22	8	14	64
MR	78	PEX	2-27-07	OS	Prominent, MC+	5-7-07	18	6	12	67
MA	46	POAG	10-17-07	OD	Prominent, (MC+)	10-23-07	18	10	8	45
RO	78	POAG	1-3-07	OD	Prominent, (MC+)	5-21-07	16	13	3	19
RC	83	PEX	5-11-07	OD	Prominent, MC−	5-15-07	27	11	16	60
			5-11-07	OD	Prominent, MC−	10-22-07	27	12	15	56
SA	49	Uveitis	8-23-06	OS	Prominent, (MC+)	4-30-07	18	9	9	50
			8-23-06	OS	Prominent, (MC+)	5-22-07	18	18	0	0
SE	58	POAG	9-19-06	OD	Prominent, (MC+)	5-2-07	17	13	4	24
			9-20-06	OD	Prominent	5-16-07	17	10	7	41
			9-20-06	OD	Prominent	10-23-07	17	10	7	41
			2-15-06	OS	Prominent	5-2-07	16	8	8	50
			2-15-06	OS	Prominent, MC+	5-16-07	16	6	10	63
			2-15-06	OS	Prominent, MC+	10-23-07	16	8	8	50
SA	59	PEX	5-21-07	OD	Prominent, MC+	10-17-07	42	14	28	67
			7-02-07	OS	Prominent, MC+	10-17-07	30	14	16	53
SM	70	PEX	3-6-07	OS	Prominent, (MC+)	5-9-07	30	9	21	70
			3-6-07	OS	Flat	10-25-07	30	40	−10	−33
SG	82	PEX	4-4-07	OD	Prominent, MC+	4-30-07	14	8	6	43
SW	71	POAG	11-29-06	OD	Prominent, (MC+)	10-23-07	37	5	32	87
			6-29-07	OS	Prominent, (MC+)	10-23-07	20	15	5	25
SD	84	PEX	5-9-07	OS	Prominent, (MC+)	5-15-07	28	11	17	61
SE	69	POAG	8-15-06	OD	Prominent	5-7-07	20	6	14	70
			3-20-07	OS	Prominent, MC+	5-7-07	16	8	8	50
SP	74	POAG	10-15-07	OS	Prominent, (MC+)	10-23-07	21	8	13	62
SH	69	POAG	2-28-07	OD	Flat, MC−	5-8-07	20	21	−1	−5
			2-28-07	OD	Flat, MC−	10-24-07	20	18	2	14
VD	81	PEX	3-19-07	OS	Prominent, MC−	5-8-07	15	12	3	20
WG	82	PEX	5-11-07	OS	Prominent, MC+	5-15-07	20	2	18	90
WR	80	POAG	3-21-07	OD	Prominent	5-2-07	15	8	7	47
			3-21-07	OD	Prominent, (MC+)	10-19-07	15	11	4	27
WE	79	POAG	5-30-07	OS	Prominent, MC+	10-19-07	16	11	5	31
WI	75	POAG	4-4-07	OD	Prominent	5-2-07	19	15	4	21
			1-31-07	OS	Prominent, MC+	5-2-07	21	11	10	48
ZG	52	Uveitis	5-11-07	OS	Prominent, MC+	5-15-07	18	2	16	89
ZW	84	POAG	12-18-06	OS	Prominent, MC−	5-7-07	12	9	3	25
			12-18-06	OS	Prominent, MC−	10-24-07	12	12	0	0
ZR	59	Uveitis	3-14-07	OD	Prominent, (MC+)	5-2-07	18	9	9	50

 PDG, pigment dispersion glaucoma; POAG, primary open angle glaucoma; PACG, primary angle closure glaucoma; PEX, pseudoexfoliation glaucoma; uveitis, secondary glaucoma after uveitis; TE, trabeculectomy; MC+, numerous microcysts; (MC+), few microcysts; MC−, no to very few microcysts.

For the measurement, the patient was comfortably seated in front of the Visante OCT and placed his or her chin on the chinrest and the front of the head against the front head support. The patient was then asked to look down and fixate on an external fixation point. The upper lid was gently retracted by the operator to display the bleb in the superior bulbar conjunctiva as well as possible, with every effort made to avoid pressure on the globe or bleb. A real-time video integrated into the scanning panel indicated the position of the scanning location. Figure 1 shows a photograph of a fully exposed filtering bleb. The four measurement planes of the AS-OCT are shown with blue arrows. 

Figure 1.

 

Photograph showing the exposed filtering bleb while the patient is looking down and the upper lid is gently lifted by the examiner. The four standard measuring planes of the AS-OCT are marked with blue arrows.

The optical aperture of the AS-OCT was centered to the maximal elevation of the filtering bleb with the aid of a joystick. Then the four standard AS-OCT images (vertical, horizontal, oblique to the right, oblique to the left) were obtained simultaneously for each bleb. (Fig. 2). 

Figure 2.

 

The four standard AS-OCT images simultaneously by the AS-OCT. Beside the internal structures of the filtering bleb in the vertical measuring planes, the upper lid and the cornea can be seen.

Because the Visante OCT has no in-built tool to measure signal intensities, the scanned images were exported as jpg files and then imported into ImageJ software (developed by Wayne Rasband, National Institutes of Health, Bethesda, MD; available at http://rsb.info.nih.gov/ij/index.html) for further analysis. 

Results

Thirty-seven patients (43 eyes) were analyzed (21 women, 15 men). Mean age of the patients was 69.5 years (range, 27–84 years). All patients were Caucasian. Mean time to first AS-OCT measurement after trabeculectomy was 112 days (range, 4–441 days). Mean time to all AS-OCT measurements after trabeculectomy was 152 days (range, 4–615 days). Average IOP reduction at the time of the AS-OCT-measurement compared with preoperative values was 10.1 mm Hg (45.7%). Patients were examined 1 to 4 times (average, 1.42 times) with the AS-OCT in May 2007, October 2007, or both (total AS-OCT measurements, 61). 

Three parts of the fluid-filled cavity of the filtering bleb and one part of the “background noise” were marked with an ellipse, as depicted in Figure 3. Within the filtering-bleb cavity, the ellipses were drawn as large as possible without overlapping the filtering bleb wall. Figure 3 shows one patient (BM) 6 days after trabeculectomy of the right eye and shows the horizontal measuring plane of a standard AS-OCT image of the Visante OCT. Several microcysts can be seen on the surface of the conjunctiva. 

Figure 3.

 

Same patient and same OCT scan as shown in Figure 2, 6 days after trabeculectomy. The histogram of the background noise outside the filtering bleb and the three histograms of the internal reflectivity of the filtering bleb are plotted with ImageJ.

The reflectivity of the content of the ellipses was plotted as histograms with the ImageJ program. For further analysis, a mean of the three pixel density measurements within the filtering cavity was taken, and the value of the background noise was subtracted. The background noise had to be subtracted from the signal intensity of the filtering bleb because it varied considerably between patients, though the background illumination in the room was made constant by the complete closing of the window blinds and the turning down of room illumination to the minimum. A lux meter to measure room light intensity, as in the study of Wong et al., ²⁹ was not used. In patient BM, the reflectivity of the filtering bleb minus the background noise was 0.525 ([17.670 + 20.418 + 19.248]/3 − 18.587), showing, in this case, little difference between the internal reflectivity of the filtering bleb and the background noise. The corresponding eye pressure at the time of the OCT measurement was 3 mm Hg. In Figure 4 (top left), beside patient BM, three other examples are plotted with different appearances of the filtering bleb. Patient BW (Fig. 4, top right) also showed low reflectivity of the fluid-filled cavity of the filtering bleb, with a value of −0.829 compared with the background noise and an IOP of 9 mm Hg. Patient WR (Fig. 4, bottom left) showed a rather flat filtering bleb cavity. This filtering bleb had a higher internal reflectivity of 10.8. The corresponding IOP was 11 mm Hg. Patient SW (Fig. 4, bottom right) showed a prominent filtering bleb with a high internal reflectivity of the filtering bleb cavity of 4.56. At the time of AS-OCT measurement, the IOP in this patient was 15 mm Hg. 

Figure 4.

 

Four examples that show the variety of filtering bleb morphology. (A, B) The internal reflectivity of the filtering bleb minus surrounding noise is close to zero. (C, D) The internal reflectivity of the filtering bleb minus surrounding noise is 10.8 resp. 4.56.

Reproducibility of the reflectivity of the background noise and the fluid-filled cavity of the filtering bleb within the same patient and measuring session were tested with three different OCT scans for all 61 measurements. For example, in patient BM, who showed low reflectivity of the fluid-filled cavity of the bleb, scan reflectivity was as follows: first scan—19.982, 18.801, 20.226 (mean, 19.703); second scan—18.197, 18.672, 19.083 (mean, 18.651); third scan—17.670, 20.418, 19.248 (mean, 19.112). The signals of the background noise in the corresponding three scans were 18.839, 18.724, and 18.587. The mean internal reflectivity of the fluid-filled bleb cavity minus background noise was 0.864 in the first scan, −0.073 in the second scan, and 0.525 in the third scan (mean, 0.439). In patient GP, who had very high reflectivity in the fluid-filled cavity of the bleb, values were as follows: first scan—43.627, 41.283, 36.264 (mean: 40.391); second scan—41.678, 40.713, 37.731 (mean, 40.042); third scan—39.344, 36.262, 38.725 (mean, 38.110). Background noise in the three corresponding scans was 3.782, 3.718, and 3.805. Mean internal reflectivity of the fluid-filled bleb cavity minus background noise was 36.609 in the first scan, 36.324 in the second scan, and 34.305 in the third scan (mean, 35.746). As these two examples show, background noise varied considerably between patients but was constant within the measurement session. Reflectivity in different parts of the fluid-filled cavity of a bleb showed some fluctuation, but overall reproducibility was reasonably good. 

In Figure 5, all 61 measurements of the internal reflectivity of the fluid-filled filtering bleb cavities of the 37 patients are plotted against the IOP measured at the time of the AS-OCT examination by a masked investigator not involved in the study. It shows a significant correlation, with R ² = 0.3762 and P < 0.0001. To prove the stability of the data, we plotted another scatter plot without the two outliers with the highest IOP of 40 mm Hg and 34 mm Hg and without the two filtering blebs with the highest internal reflectivity of 35.75 and 28.90. These 57 measurements still showed a highly significant, though weaker, correlation, with R ² = 0.1158 and P = 0.0096. 

Figure 5.

 

Scatter plot of the internal reflectivity of the fluid-filled cavity of the filtering bleb versus the IOP at the day of the OCT measurement (n = 61).

Discussion

In the clinical setup, several subjective grading systems were used to evaluate the success or failure of a filtering bleb and to help the clinician in the follow-up period after trabeculectomy. ³ In this study, we presented a new and objective tool to guide the clinician in the postoperative period after trabeculectomy. We found the AS-OCT to be a promising method with which to analyze the morphology of the filtering bleb. The study by Ciancaglini et al. ²⁵ reports that functioning blebs presented with a low bleb wall reflectivity measured with the Visante OCT. The study by Yamamoto et al. ²¹ reported a significant correlation between the reflectivity of the fluid-filled bleb cavity measured by ultrasound biomicroscopy and IOP control. In that study, internal reflectivity was grouped into low, medium, and high. To our knowledge, our study is the first that objectively analyzes the internal reflectivity of the fluid-filled bleb cavity with the aid of drawing histograms and plotting the pixel density against IOP. A significant correlation between the internal reflectivity of the fluid-filled cavity of the filtering bleb and the IOP could be shown that even remained significant when four outlier measurements were excluded. This finding may have future impact in objectively evaluating the postoperative success or failure of filtering blebs so that postoperative management (e.g., bleb manipulation) can be adapted accordingly. 

Limitations of this pilot study are the heterogeneous group of patients (primary open angle glaucoma, pseudoexfoliation-glaucoma, primary angle closure glaucoma, pigment dispersion glaucoma, secondary glaucoma after uveitis) and the large range of measuring times and numbers of OCT measurements after trabeculectomy, attributed to the fact that the Visante OCT was available in our clinic only in May 2007 and again in October 2007. Therefore, some patients who underwent trabeculectomy before May 2007 could be measured in May and again in October, and those who underwent trabeculectomy between May 2007 and October 2007 could be measured once or twice only in October 2007. 

Further studies with a higher number of patients are needed to confirm this finding. The development of the internal reflectivity of the fluid-filled bleb cavity compared with IOP over a longer period should also be analyzed. 

In conclusion, we believe that AS-OCT has great potential for analyzing the postoperative success or failure of a filtering bleb. Because there is no contact with the eye, even measurements as early as 1 day after surgery are possible, in contrast to ultrasound biomicroscopy. 

Footnotes

 Disclosure: L. Pfenninger, None; F. Schneider, None; J. Funk, None

The authors thank Albert Ehrsam (Carl Zeiss Meditec Inc.) for his valuable instruction in the use of the Visante OCT.