In the early period after glaucoma drainage device (GDD) surgery the intraocular pressure (IOP) classically goes through 2 phases.1 The hypotensive phase occurs immediately after surgery, lasts around 1 week and is characterized by low IOP. This is followed by the hypertensive period where the IOP tends to rise steadily, most likely owing to edema and fibrosis around the episcleral plate of the GDD. With resolution of the hypertensive phase the IOP tends to improve, hopefully leading to a prolonged period of good IOP control. The hypertensive response seems to occur more commonly after Ahmed GDD surgery than nonvalved implants, with an IOP above 21 mm Hg reported to occur in 40% to 80% of cases.1–4 Although the hypertensive phase can last as long as 6 months it is usually during the first 1 to 4 weeks, when there is intense congestion of the bleb wall, that IOP is highest.1,3
Although ocular massage is frequently used for IOP control after trabeculectomy surgery,5 its role after GDD surgery is less well defined. A retrospective study from our department concluded that digital massage might be useful after Ahmed GDD surgery.6 The aim of this prospective study was to examine the role of ocular massage in IOP control during the hypertensive phase after Ahmed GDD surgery.
This prospective study received approval from the Toronto Western Hospital Research Ethics Board. The study protocol is summarized in Figure 1. The inclusion criterion was IOP above target 1 to 8 weeks after Ahmed GDD surgery. Exclusion criteria were an obvious cause of high IOP (such as lumen blocked by blood or fibrin), tube-lens or tube-corneal touch, and the inability to massage owing to physical or mental disability. The individual target IOP was set preoperatively. After informed consent the examining doctor performed digital ocular massage (inferiorly when GDD positioned superiorly, superiorly when GDD positioned inferiorly) pushing for 1 second 10 times at the slit lamp, observing the eye for signs of anterior chamber shallowing, or tube-iris or tube-cornea apposition. The IOP was checked immediately and then 1 hour after the massage. Patients failing to achieve a 20% reduction in IOP 1 hour after massage were considered as nonresponders, commenced on glaucoma drops and played no further part in the trial. If a 20% reduction was maintained at 1 hour the patient was instructed on how to perform digital ocular massage at home, initially 2 hourly, with the frequency reduced according to response at later visits. Massage was continued for at least 6 weeks. A target IOP was set for each patient by the examining doctor before surgery, and if a 20% drop in IOP was achieved with massage but IOP remained above target glaucoma drops could be added at the examining doctor's discretion. At follow-up visits no further massage was performed by the examining doctor to allow an accurate assessment of the IOP with the patient massaging at the prescribed intervals. After the initial assessment study visits were at 2-weeks, 6-weeks, and 6-month postrecruitment.
Primary outcome measure was IOP at 1-hour postdigital massage. Secondary outcome measures were IOP and glaucoma drops at 2 weeks, 6 weeks, and 6 months after commencement of ocular massage. We also examined massage-associated complications.
All operations were performed by 1 of 2 surgeons (Y.M.B. and G.E.T.) or glaucoma fellows under their direct supervision, with a standard technique.7 A fornix-based conjunctival flap was fashioned with relaxing incisions. The Ahmed glaucoma valve was used for all surgeries (Model FP7, New World Medical Inc, Rancho Cucamonga, CA). After priming with balanced salt solution it was inserted under the conjunctiva and Tenon's capsule, and sutured to the sclera with 8-0 silk sutures at a distance of at least 8 mm from the limbus. The tube was trimmed to the correct length with a bevelled edge and then a paracentesis was created with a 23-guage needle at the limbus parallel to the iris. Viscoelastic was injected into the anterior chamber through the same needle and the tube inserted into the anterior chamber and then secured to sclera with a 10-0 nylon suture. The tube was covered with a donor corneal button dissected to half thickness, and sutured with 10-0 nylon. The conjunctiva was closed with 10-0 nylon or 8-0 polyglycolic acid, depending on surgeon preference. Topical atropine 1% and a steroid-antibiotic combination were given at the end of surgery. Topical antibiotics were used for 1 week after surgery, with topical steroids continued for 4 to 6 weeks. Patients were reviewed on postoperative day 1 and then at least weekly for 4 weeks. After this the intervals between assessments were varied according to the clinical picture.
Before the study a sample size calculation indicated that we would require 17 patients to detect a 2 mm Hg change in IOP as a result of ocular massage (standard deviation=2 mm Hg, α=0.05, β=0.9). Data were entered into Microsoft Excel and statistical analysis performed using StatPlus 2007 Professional Version 4.9.2 (Analystsoft, Vancouver). Shapiro-Wilk tests were used to test for the presence or absence of normal distribution in the data. Comparison between preoperative and postoperative data was performed using a paired t test, with a P value of ≤0.05 considered statistically significant.
Eighteen consecutive patients who met the study inclusion criteria were invited to join the study, with all patients agreeing to take part. Surgery was performed between September 2007 and March 2008. Preoperative characteristics are summarized in Table 1. Nine patients (50%) had no history of previous same-eye glaucoma surgery whereas 6 (33.3%) had undergone 1 previous trabeculectomy, 2 (11.1%) had undergone 2 trabeculectomies, and 1 patient (5.6%) had undergone both a trabeculectomy and Ahmed GDD implantation.
Mean (±standard deviation) time from surgery until IOP was above target that was 20.1 (±9.6) days, with a range of 12 to 44 days. Table 2 shows the response to ocular massage. The mean IOP dropped by 8.6 mm Hg immediately after massage (37.6%, P<0.0001). One hour later the mean IOP was 4.3 mm Hg lower than before massage (18.8%, P=0.0008). Although 17 patients (94.4%) achieved a 20% drop in IOP immediately after massage this was only sustained in 9 patients (50%) at 1 hour.
The 9 patients (50%) who failed to achieve a 20% drop in IOP with massage were considered nonresponders and commenced on glaucoma drops. These patients are excluded from further analysis. An additional patient, who achieved a 33.3% drop in IOP 1 hour after massage, felt unable to perform massage despite instruction and was therefore commenced on glaucoma drops. Figure 2 summarises the results of the remaining 8 patients (44.4%) who continued with digital ocular self-massage with the mean IOP and number of glaucoma drops at the 2-week, 6-week, and 6-month reviews. Although the time between last massage and the IOP measurement varied between examinations, all patients had performed the massage at the prescribed time points on the day of examination. All patients maintained a drop of 20% or greater at these reviews, although 1 (12.5%), 3 (37.5%), and 4 (50%) patients were using glaucoma drops at the 2-week, 6-week, and 6-month visit, respectively.
There were no massage-associated complications in this series.
Our results show that ocular massage during the hypertensive phase of Ahmed GDD surgery is an effective technique in a proportion of patients. We used a 20% reduction in IOP at 1-hour postmassage to differentiate responders from nonresponders and in a group of 18 patients who exceeded target IOP 1 to 8 weeks after surgery 50% responded to ocular massage. One patient (5.6%) responded well to massage but was unable to perform massage at home and was commenced on glaucoma drops. The remaining 8 patients (44.4%) performed regular digital massage and the 20% drop in IOP was maintained at the 2-week, 6-week, and 6-month review, although by 6 months, 4 patients (50%) required glaucoma drops to achieve target IOP.
This is, to our knowledge, the first prospective study to examine the use of ocular massage after Ahmed GDD surgery. McIlraith et al,6 at our institution, performed a retrospective analysis of the outcomes of digital massage in 20 patients who had undergone Ahmed GDD surgery. For the control group McIlraith used patients who had undergone Ahmed GDD surgery but did not develop a hypertensive phase, and concluded that there was no difference between the 2 groups in terms of IOP or glaucoma medications at 1-year postoperatively. Our results confirm some of the findings in this report but the prospective nature of our study also adds some insights into the use of ocular massage after Ahmed GDD surgery.
McIlraith et al6 reported a 40% drop in mean IOP immediately after massage with 93.8% of patients achieving at least a 20% drop in IOP. This corresponds to the 37.6% drop and 94.4% response rate we observed but on rechecking the IOP 1 hour later we discovered that in many patients the IOP response is short lived, with only 50% of patients maintaining a 20% drop. On the basis of this finding, we recommend that when patients seem to respond to ocular massage the IOP be rechecked in 1 hour to identify a transient IOP drop.
The mechanism by which ocular massage improves IOP is likely to consist of several elements. The immediate drop in IOP is simply a result of aqueous fluid being forced from the eye through the valve of the Ahmed GDD. The more sustained response may be due to the rush of aqueous fluid clearing inflammatory material and fibrin from the tube, valve, and plate area. This effect may be similar to that seen with forced tube irrigation, where the tube is cannulated and balanced salt solution injected up the tube.8 It is also possible that massage has beneficial long-term effects on IOP via its effect on bleb morphology, but any conclusions regarding this requires further investigation and longer follow-up.
Complications as a result of digital massage after trabeculectomy are uncommon but well described.7,9–11 There were no massage-associated complications in this study. Although this may be related to our small sample size visualization of the anterior chamber depth and tube during the initial massage by the examining doctor should identify most of the potential complications of ocular massage, although the possibility that a patient will be unduly vigorous when self-massaging remains.
This study has several limitations. We compared IOP before and after commencement of digital massage rather than randomizing patients to massage or no massage. Although this would have improved the methodologic quality of the study the results of the previous study mentioned above left us reluctant to deny this treatment option to patients with high IOPs after GDD surgery. In addition, the protocol allowed for additional glaucoma drops to be prescribed should IOP remain above target despite a good response to digital massage. At the 6 months review half of the 8 patients who responded and continued with massage were using drops and although it seems reasonable to conclude their response to massage was sustained during the study period it is not possible to separate the effect of the drops from the massage. In addition, there was no defined criterion for setting the target IOP that was decided by the examining doctor before surgery, and is likely to have varied according to the practice of the individual clinician. A more robust study design would have included a study protocol for target IOP but this would have reduced the ability to tailor the postoperative management to individual patients, potentially resulting in a poorer outcome. Finally, our sample size calculation was based on the primary outcome measure of the study (IOP at 1-h postmassage) and therefore our analysis of the secondary outcomes (IOP and glaucoma drops at 2-wk, 6-wk, and 6-mo postrecruitment) may not have sufficient statistical power to come to firm conclusions. It is also important to note that this study involves Ahmed valves only and the results may not be relevant to nonvalved implants, which often have aqueous flow restricted by internal stenting or external ligature during the postoperative period.
In conclusion, digital ocular massage has a useful role to play in the management of the hypertensive phase after Ahmed GDD surgery. In this series 50% of patients achieved a 20% drop in IOP with massage.
1. Nouri-Mahdavi K, Caprioli J. Evaluation of the hypertensive phase after insertion of the Ahmed glaucoma
valve. Am J Ophthalmol. 2003;136:1001–1008.
2. Hong CH, Arosemena A, Zurakowski D, et al. Glaucoma
drainage devices: a systematic literature review and current controversies. Surv Ophthalmol. 2005;50:48–60.
3. Ayyala RS, Zurakowski D, Smith JA. A Clinical study of the Ahmed glaucoma
valve implant in advanced glaucoma
. Ophthalmology. 1998;105:1968–1976.
4. Ayyala RS, Duarte JL, Sahiner N. Glaucoma
drainage devices: state of the art. Expert Rev Med Devices. 2006;3:509–521.
5. Buys YM, Trope GE. Massage
: technique and complications. In: Trope GE, ed. Glaucoma Surgery
. Boca Raton: Taylor and Francis; 2005:139–144.
6. McIlraith I, Buys YM, Campbell RJ, et al. Ocular massage
for intraocular pressure control after Ahmed valve
insertion. Can J Ophthalmol. 2008;43:48–52.
7. Segrest DR, Ellis PP. Iris incarceration associated with digital ocular massage
. Ophthalmic Surg. 1981;12:349–351.
8. Hatanaka M, Mendonça M, Babic M, et al. Forced tube irrigation (FTI) as treatment for glaucoma drainage device
surgical failure. Invest Ophthalmol Vis Sci. 2008;49:E-Abstract 1254.
9. Miller GR, Kurstin J. Ruptured filtering bleb after ocular massage
. Arch Ophthalmol. 1966;76:363.
10. Ruderman JM, Jampol LM, Krueger DM. Visual loss caused by subretinal hemorrhage and rupture of Bruch's membrane after digital ocular massage
. Am J Ophthalmol. 1988;106:493–494.
11. Baldassare RD, Brunette I, Desjardins DC, et al. Corneal ectasia secondary to excessive ocular massage
following trabeculectomy with 5-fluorouracil. Can J Ophthalmol. 1996;31:252–254.