The Effect of Trabeculectomy on Ocular Pulse Amplitude

Abstract

Purpose.: To investigate whether trabeculectomy, besides its intraocular pressure (IOP)–lowering effect, has an effect on ocular pulse amplitude (OPA); and to determine whether OPA changes are influenced by IOP changes.

Methods.: Forty-eight patients with glaucoma (48 eyes) scheduled for unilateral first-time trabeculectomy were prospectively enrolled from October 2007 to April 2008. The eye undergoing trabeculectomy was considered the study eye, whereas the nonsurgical fellow eye was used as the control eye. OPA, IOP, blood pressure, and heart rate were measured before and 4 weeks after trabeculectomy by means of Pascal dynamic contour tonometry (DCT), Goldmann applanation tonometry (GAT), and sphygmomanometry. A regression model for repeated measures was used.

Results.: Preoperative GAT, DCT, and OPA were 20.92 ± 8.55, 21.33 ± 7.06, and 3.23 ± 1.58 mm Hg, respectively. One month after trabeculectomy, GAT, DCT, and OPA were 11.23 ± 5.03, 14.45 ± 4.79, and 2.12 ± 1.07 mm Hg, respectively. There was a significant decrease in OPA after filtration surgery in the study eye compared with the control eye (P < 0.0001). Changes in OPA were correlated with changes in IOP (Spearman ρ = 0.49, P = 0.0004). When the IOP change caused by filtration surgery was taken into account, no significant difference in effect on OPA after trabeculectomy was demonstrated in the study eye compared with the control eye (P = 0.18).

Conclusions.: OPA changes correlated strongly with IOP changes. There was no evidence of an effect of filtration surgery on OPA when the concomitant IOP decrease after trabeculectomy was taken into account.

Elevated intraocular pressure (IOP) is the most important modifiable risk factor for glaucoma. In recent years, the role of ocular blood flow in glaucoma has been recognized. Reduced ocular perfusion and pulsations have been associated with glaucoma and its progression. ^1–5  

The Pascal dynamic contour tonometer (DCT) represents a relatively new technology for direct and noninvasive IOP measurement, together with measurement of ocular pulse amplitude (OPA). It has been proposed to accurately measure true IOP irrespective of corneal properties. ^6–16 It measures IOP in a continuous way and provides a pressure curve that is synchronous with the cardiac cycle. ¹⁷ From this curve, the difference between the mean systolic and mean diastolic IOP can be determined, and the result represents the OPA. These pulsatile variations in IOP are thought to be caused by the blood volume that is pumped into the eye, mainly in the choriocapillary bed, during each cardiac cycle. 

Recent research pointed out that OPA is reduced in eyes with primary open-angle or normal-tension glaucoma compared with normal healthy control eyes. ² Conversely, higher OPA levels correlate with less severe glaucoma. ⁵ In addition, a low OPA has been demonstrated to be associated with moderate to severe glaucomatous visual field loss and may be a risk factor for the development of glaucomatous visual field defects. ³ Regarding the therapeutic options for glaucoma, the topical carbonic anhydrase inhibitor dorzolamide was found to significantly increase OPA, in addition to its IOP-lowering effect. ^18–20 This effect is in contrast with that of the β-blocker timolol, which demonstrates no effect on OPA, besides the lowering of IOP. Similar results were found by other researchers for dorzolamide and timolol when their effect on the ocular blood flow and fundus pulsation was investigated, using techniques other than DCT. ²¹  

The purpose of this study was to investigate whether filtration surgery has an impact on OPA. Trabeculectomy is still the most performed surgical procedure worldwide to reduce IOP below an individually acceptable IOP target to halt or slow down visual field deterioration in patients with glaucoma. Only one research group has reported that filtration surgery probably decreases the OPA. ^22,23 However, it was a pilot study with a small number of patients and IOP, which is a known influencing factor on OPA, was not taken into account. We hypothesized that the change in IOP after trabeculectomy may be responsible for the observed change in OPA. This study was intended to clarify whether trabeculectomy has an effect on OPA besides its IOP-lowering effect. 

Methods

This prospective interventional study was approved by the ethics committee of Leuven University Hospitals and was conducted in accordance with Good Clinical Practice and the tenets of the Declaration of Helsinki. Patients signed an informed consent form before inclusion. 

Study Design

Two study visits with a 1-month interval were planned: on the day of surgery before the trabeculectomy and 4 weeks (±1 week) after trabeculectomy. During each visit, the following examinations were performed in the same order: visual acuity (Snellen), OPA and IOP measurements, blood pressure, and heart rate measurements. During one of the visits, pachymetry and biometry were also performed. The patients were instructed to avoid caffeine intake, smoking, and exercise for 3 hours before each study visit. One eye per patient was considered to be the study eye, whereas the fellow eye was used to assess variability over time. 

Patient Eligibility

Patients scheduled for trabeculectomy between October 2007 and April 2008 were consecutively enrolled at Leuven University Hospitals. Inclusion criteria were diagnosed glaucoma, requiring first-time trabeculectomy (medically uncontrolled); willingness to sign informed consent; and ability to comply with the study requirements. All had to have either primary open-angle glaucoma or normal-tension glaucoma (see Refs. 24, 25 for criteria). Optic nerve head criteria were glaucomatous damage of the optic nerve head such as a vertical cup/disc ratio higher than 0.5, reduced retinal nerve fiber layer thickness, peripapillary atrophy, disc hemorrhage or a history of it, asymmetrical cupping between the right and left eyes greater than 0.2, nasal cupping, and violation of the inferior-superior-nasal-temporal rim thickness rule. Visual field criteria were a mean defect greater than 6 dB and defects characteristic of glaucoma, such as nasal step or Bjerrum scotomas respecting the horizontal raphe. At least three reliable visual field tests had to be available. There had to be a correlation between the obtained visual field defects and the observed optic nerve head findings. Exclusion criteria were patients with a history of ocular trauma, eye diseases that could not be accounted for by refractive error (except glaucoma and cataract), history of previous intraocular surgery such as trabeculectomy in the study eye (with exception of phacoemulsification), inability to perform Pascal DCT measurements of acceptable quality (e.g., corneal disease, eyelid spasms, nystagmus, or hypersensitivity), or change in (topical or systemic) medications ≤4 weeks before the onset of the study. Most patients were receiving medical therapy when the first OPA measurement was performed. This therapy was stable for at least 1 month before the first OPA measurement and discontinued immediately after surgery. 

Measuring Devices

OPA was measured with a Pascal tonometer (Dynamic Contour Tonometer [DCT]; Ziemer Ophthalmic Systems, Port, Switzerland), whereas IOP was measured by Goldmann applanation tonometer (GAT) and DCT. Up to three DCT measurements were performed to obtain one good-quality measurement (quality level 1–2). If after three measurements, no one measurement was of good quality, the data from the best-quality measurement was taken for analysis. In addition, the following was determined once in both eyes of each study patient during any of the study visits: axial length, corneal curvature, and anterior chamber depth by means of an optical biometry system (IOL Master; Carl Zeiss AG, Feldbach, Switzerland) ²⁶ and central corneal thickness measured with ultrasonic pachymetry (Pachmate DGH 55; DGH Technology Inc., Exton, PA) placed on the central cornea over an undilated pupil. The automatically calculated mean of 20 measurements of the central corneal thickness with a SD ≤5 μm was recorded. Blood pressure and heart rate were assessed by a calibrated sphygmomanometer (Omron IntellSense M6 [HEM-7001-E] SN: 5306618L; Omron Health Care Co., Ltd., Kyoto, Japan). ²⁷  

Surgical Technique

The used trabeculectomy technique for the study eye was a modified Cairns procedure performed by two experienced surgeons (IS, TZ). ²⁸ One hour before surgery, topical pilocarpine 2% was instilled. Anesthesia was induced by a periocular injection of lidocaine with adrenaline in combination with hyaluronidase to facilitate diffusion. A silk 8-0 corneal traction suture was placed, and a fornix-based conjunctival flap was dissected, exposing the sclera. Bipolar diathermy was sparsely applied for hemostasis. A scleral flap measuring 3 × 5 mm with side incisions at 0.5 mm from the cornea was delineated. This flap was dissected for approximately half scleral thickness. Two diagonal releasable sutures (nylon 10-0) were preinstalled at the edges of the scleral flap. An inferotemporal corneal paracentesis was made and a viscoelastic (Viscoat; Alcon Laboratories, Fort Worth, TX) was injected to maintain the anterior chamber. A punch trabeculectomy of 0.5 mm (Khaw punch; Duckworth and Kent, Hertfordshire, UK) was performed followed by an iridectomy. Subsequently, the scleral flap sutures were closed. In patients with a low target pressure, or when excessive fibrosis was anticipated (e.g., young patient age, a history of previous fibrosis occurring after trabeculectomy in the fellow eye), mitomycin C 0.2 mg/mL in sponges was applied for 1 to 3 minutes beneath the conjunctiva. Finally, the conjunctiva was closed with two nylon 10-0 sutures using a purse-string technique. A drop of atropine and tobramycin ointment was administered before patching. Postoperative treatment consisted of topical dexamethasone and tobramycin (Tobradex; Alcon Laboratories) four times daily for 8 weeks. 

Statistical Analysis

A difference of 0.5 in OPA was considered as clinically relevant based on the results of a previous reproducibility study. ²⁹ A sample size of 48 subjects was therefore needed to detect with 80% power and an α of 5%—a decrease of 20% based on log-transformed OPA values, with a common standard deviation of the log-transformed OPA of 0.7 and a correlation of 0.70 between the log-transformed OPAs measured before and after trabeculectomy. 

The OPA and IOP measurements by DCT used for statistical analysis were those that corresponded to the highest quality level. If measurements of equal quality were obtained, the first measurement was chosen. Of the 48 patients, 2 had one missing OPA measurement and 1 had two missing measurements in the control eye. A linear model for repeated measures was used to compare the pre- and postsurgical OPA changes between the study and control eyes. To meet the distributional assumptions of the linear model, we used a square root transformation for OPA. To verify whether the effect of trabeculectomy on OPA was not due to a change in IOP, we included the log-transformed IOP in the model as a predictor. The relation between IOP and OPA was allowed to be specific for each of the four time point–eye combinations. Results for the effect of trabeculectomy were compared with and without IOP correction. Note that since interest was in the comparison of changes over time, correction for baseline characteristics (e.g., axial length) was superfluous. ³⁰ The Spearman correlation was used to study associations between OPA and IOP. Visual acuity was measured with Snellen charts and converted to logarithmic minimum angle of resolution (logMAR), to enable calculation. Descriptive results are reported as the mean ± SD unless otherwise indicated. Probabilities are two-sided and considered statistically significant if <0.05 (SAS ver. 9.1; SAS Institute Inc, Cary, NC). 

Results

Patient Characteristics

Overall, the age was 66.8 ± 9.5 years. All were Caucasian; 24 (50%) were women and 24 (50%) were men. The study eye was the left eye in 24 (50%) patients and the right eye in 24 (50%). Baseline visual acuity was 0.15 ± 0.60 logMAR. During the first study visit before surgery, the systolic, diastolic blood pressure, and heart rate were 139 ± 22 mm Hg, 84 ± 10 mm Hg, and 67 ± 13 beats per minute, respectively. During the second study visit after surgery, the systolic, diastolic blood pressure, and heart rate were 148 ± 25 mm Hg, 85 ± 12 mm Hg, and 70 ± 12 beats per minute, respectively. Central corneal thickness was 539 ± 40 μm in the study eye, and spherical equivalent was 44.03 ± 1.59 D; axial length, 23.63 ± 1.19 mm; and anterior chamber depth, 3.03 ± 0.74 mm. OPA and IOP baseline features are presented in Table 1. 

Table 1.

 

OPA and IOP Baseline Characteristics before and after Surgery

Characteristic	Mean	SD	Minimum	Median	Maximum
OPA study eye pre	3.23	1.58	0.70	3.00	7.10
OPA study eye post	2.12	1.07	0.50	1.90	4.70
OPA control eye pre	2.70	1.20	0.50	2.50	5.70
OPA control eye post	2.76	1.23	0.70	2.60	5.30
IOP DCT study eye pre	21.33	7.06	11.50	19.10	52.80
IOP DCT study eye post	14.45	4.79	5.20	13.90	25.20
IOP DCT control eye pre	17.35	3.99	8.10	16.95	29.30
IOP DCT control eye post	18.15	4.87	7.60	17.50	31.80
IOP GAT study eye pre	20.92	8.55	10.00	18.50	55.00
IOP GAT study eye post	11.23	5.03	3.00	10.00	26.00
IOP GAT control eye pre	16.43	6.58	8.00	16.00	48.00
IOP GAT control eye post	14.43	4.53	3.00	14.00	30.00

 Data are expressed in mm Hg.; pre, preoperatively; post, postoperatively.

OPA and IOP Outcomes

The change in OPA in the study eye correlated with the change in IOP measured by DCT (Spearman ρ =0.34, P = 0.017) as well as by GAT (Spearman ρ = 0.49, P = 0.0004; Figure 1). Because of this correlation, a part of the change in OPA after trabeculectomy may be attributable to a change in IOP. Tables 2 and 3 indicate that without correction for IOP, there is a significant effect of trabeculectomy on OPA (P < 0.0001). Indeed, the OPA change in the study eye differed from the OPA change in the control eye, so that this change cannot be attributed to daily OPA variations. After correction for IOP (DCT or GAT), the evidence of a significant effect of trabeculectomy on the OPA disappears (P = 0.26 and P = 0.18, respectively). Note also that the magnitude of the change in OPA is markedly reduced after IOP correction. Without IOP correction, the OPA changes with 1.42 mm Hg (from 3.03 to 1.61) in the study eye (P < 0.0001). After correction for IOP (DCT or GAT), the OPA change drops to 0.33 mm Hg (P = 0.30) and 0.16 mm Hg (P = 0.62), respectively, representing 23.2% or 11.3% of the original decrease in OPA. 

Figure 1.

 

IOP plotted against OPA, separately for the pre- and postoperative measurements. IOP measured by (A) GAT and (B) DCT is shown.

Table 2.

 

Mean OPA before and after Surgery

	Without IOP Correction	With IOP Correction (DCT)	With IOP Correction (GAT)
OPA study eye pre	3.03 (2.59–3.50)	2.63 (2.18–3.11)	2.50 (2.05–3.00)
OPA study eye post	1.61 (1.30–1.97)	2.30 (1.83–2.82)	2.34 (1.85–2.90)
OPA control eye pre	2.56 (2.23–2.92)	2.54 (2.23–2.87)	2.46 (2.15–2.80)
OPA control eye post	2.65 (2.28–3.03)	2.56 (2.25–2.90)	2.69 (2.37–3.02)

 Data are expressed in mm Hg (95% Cl). Estimates are given from the model without and with correction for IOP (separately for DCT and GAT). Pre, preoperatively; post, postoperatively.

Table 3.

 

Probabilities, without and with Correction for IOP

	Without IOP Correction	With IOP Correction (DCT)	With IOP Correction (GAT)
OPA change in control eye	0.57	0.86	0.10
OPA change in study eye	<0.0001	0.30	0.62
Difference in OPA change	<0.0001	0.26	0.18

Discussion

Low OPAs have been associated with a higher severity of glaucoma and associated visual field defects. ^1–5 Previous research has demonstrated that the topical IOP-lowering drug dorzolamide increases the OPA, whereas timolol had no effect on OPA besides its IOP-lowering effect. ^18–20 This prospective study was conducted to determine whether OPA is altered by trabeculectomy, the most common surgical IOP-lowering procedure. 

Our results indicate that there was no evidence of an effect of filtration surgery on OPA, if the concomitant change in IOP was taken into account. Two papers by another group reported that OPA was decreased after trabeculectomy. ^22,23 They hypothesized that this reduction in OPA after filtration surgery could be caused by either an overall reduction in scleral rigidity, ³¹ or a changed volume-pressure relation when IOP is altered, ³² or finally the influence of postoperative wound healing. ²² However, the authors did not consider the possible influence of changes in IOP after trabeculectomy on OPA in their statistical analysis. Possible shortcomings of the aforementioned study were a small sample size (14 subjects) and the postoperative time point for OPA measurements on the first day after trabeculectomy, when IOPs are still unstable and early postoperative phenomena including inflammation may interfere with the measurements. 

Our study had an appropriate sample size and power that were calculated a priori. The OPA was measured 1 month after surgery, when the IOPs are more stable and inflammation and other early postoperative confounding factors have resolved. Most important, this study corrected for the change in IOP, as opposed to the study by Von Schulthess et al. ²² A recent study by our group and some other studies reported the correlation between IOP and OPA, demonstrating that the OPA is correlated with the IOP (Dastiridou A, et al. IOVS. 2008;49:E-Abstract 4128). ^2,16,33,34 Indeed, the present study confirms that there is a strong correlation between OPA changes and IOP changes after trabeculectomy. Therefore, we conclude that the previously reported drop in OPA after trabeculectomy by Von Schulthess et al. ²² can be almost fully attributed to the associated reduction in IOP, rather than the previously suggested hypotheses. 

The mechanisms by which OPA is influenced by IOP remain unclear. Some possible mechanisms by which the IOP levels may influence the OPA levels, and therefore also explain the observed (IOP-dependent) changes in OPA after trabeculectomy, have already been proposed. Kaufman et al. ¹⁶ demonstrated that there is a positive correlation between OPA and IOP (0.12 mm Hg/1 mm Hg of IOP; P < 0.001). Filling of the orbital vessels could cause a pulsatile protrusion of the whole bulbus, allowing for pressure wave recordings even from enucleated eye sockets. The IOP response to the increase in ocular volume during cardiac systole depends on the elastic properties of the ocular shell, a phenomenon that may explain the finding of a positive correlation between OPA and IOP. With higher IOP levels, scleral wall tensions increase, and the injection of a given volume of blood results in a distinct pressure increase rather than a further elastic extension of an already prestressed bulbus walls. In their study, Kaufman et al. found a tendency toward higher OPA in thinner and therefore potentially more elastic corneas. Moreover, OPA and IOP are not only linked by the elastic properties of the bulbus walls but also by ocular blood flow. The difference between the pressure in the ophthalmic artery and IOP represents the pressure gradient maintaining ocular perfusion. Elevated IOP levels may therefore affect pulsatile ocular blood flow, although the direction of the net effect (increase or decrease) is difficult to predict owing to regulatory mechanisms. They concluded that OPA measurements were affected by IOP. In another study, Medeiros et al. ¹² found that the difference between DCT and GAT IOP measurements was significantly influenced by corneal thickness (P < 0.001) and OPA (P = 0.004). Previous work by our group showed that OPA increases with rising IOP (slope 0.026/0.033 and P = 0.002/<0.0001 for GAT/DCT) and thus that OPA is influenced by IOP. ² The same finding was published by Kniestedt et al. ³⁴ : Increased IOP correlated significantly with large OPAs (r = 0.13, P < 0.001). We think a possible explanation for the higher OPAs obtained when IOP is high is that identical volume changes in a sphere (caused by the blood volume pumped into the eye during each cardiac cycle) can cause greater pressure changes (OPA) within this sphere when the pressure is higher. In analogy, when one makes an impression with a finger on an inflated ball, greater pressure changes are induced within that ball than when inducing the same degree of impression in a flat ball. Moreover, the properties of the sclera are likely to have an important influence on OPA. In fact, scleral properties may vary with the level of IOP, which may contribute to the fact that OPA increases with higher IOP. In this regard, Yang et al. ³¹ suggested that an overall reduction in scleral rigidity after successful filtration surgery could be a cause of the decrease in OPA. 

A first limitation of this study was the possible interference of concomitant intake of topical IOP-lowering medication when the first OPA measurement was performed. The influence of different classes of antiglaucoma medications on OPA was not within the scope of this study. Moreover, it was appropriate to include patients on various medications to make our study population representative. Because of the inhomogeneity of medication use, it would not be appropriate to perform statistical analyses on multiple subgroups with different medications. A second limitation was that in two included patients the control eye had already had a trabeculectomy in the past. We corrected for the IOP and used the other eye to take into account fluctuations in OPA. Since the control eye was only used to assess OPA fluctuations, the surgical history of the control eye is not an issue, as this eye did not undergo any intervention during the study. 

In summary, our study indicates that OPA changes correlate strongly with IOP changes and that there is no evidence of an effect of filtration surgery on the OPA, taking into account the concomitant IOP decrease. ^18–20  

Footnotes

 Disclosure: C. Breusegem, None; S. Fieuws, None; T. Zeyen, None; I. Stalmans, None