Indian Journal of Ophthalmology

: 1998  |  Volume : 46  |  Issue : 3  |  Page : 139--143

Management of phacolytic glaucoma : Experience of 135 cases

A Braganza, R Thomas, T George, A Mermoud 
 Schell Eye Hospital, Vellore, India

Correspondence Address:
A Braganza
Schell Eye Hospital, Vellore


We retrospectively analyzed 135 eyes with phacolytic glaucoma. A trabeculectomy was added to standard cataract surgery if symptoms endured for more than seven days, or if preoperative control of intraocular pressure (IOP) with maximal medical treatment was inadequate. In the early postoperative period, IOP was significantly lower in the combined surgery group (89 eyes) compared to the cataract surgery group (46 eyes) (p < 0.001). At 6 months there was no difference in IOP or visual acuity between the two groups. There were no serious complications related to trabeculectomy. It is reasonable to conclude that in eyes with a long duration of phacolytic glaucoma, addition of a trabeculectomy to cataract surgery is safe, prevents postoperative rise in intraocular pressure and decreases the need for systemic hypotensive medications. A randomized trial is on to further address this question.

How to cite this article:
Braganza A, Thomas R, George T, Mermoud A. Management of phacolytic glaucoma : Experience of 135 cases.Indian J Ophthalmol 1998;46:139-143

How to cite this URL:
Braganza A, Thomas R, George T, Mermoud A. Management of phacolytic glaucoma : Experience of 135 cases. Indian J Ophthalmol [serial online] 1998 [cited 2023 Sep 25 ];46:139-143
Available from:

Full Text

Phacolytic glaucoma (PG) is characterized by an acute rise in intraocular pressure (IOP) in eyes with advanced cataracts, associated with aqueous flare and cells. The pathogenic mechanism is microleakage of high molecular weight lens proteins through an intact anterior lens capsule causing an inflammatory response and leading to obstruction of the aqueous drainage channels by proteins, protein-laden macrophages, and inflammatory debris.[1][2][3][4] No information has been reported regarding the long-term effect of this pathologic process on the trabecular meshwork (TM) and IOP. PG is usually managed by a simple lens extraction, either intracapsular (ICCE) or, as is now preferred, extracapsular (ECCE).[1, 4, 5] Such surgery undertaken within a day or two after onset of symptoms has been shown to be effective in controlling the IOP and restoring vision, but no established guidelines exist for the management of eyes in which the duration of symptoms has been long. In our experience, patients with PG present at least a week after onset of symptoms, and a 2-3 week delay is common. While a simple lens extraction should, theoretically, cure even these patients and eventually clear the TM of the proteins and/or macrophages, this may occur at the expense of a very stormy postoperative period with its attendant sequelae.[4] Though these data are only available for ICCE, ECCE may arguably aggravate (albeit minimally with modern techniques) this inflammation due to residual lens matter. Based on our prior clinical experience with PG, we have been empirically combining lens extraction with a trabeculectomy in patients with a duration of symptoms exceeding one week, or in eyes where the IOP could not be controlled with maximal medical therapy. We undertook this retrospective study to assess the effect of combining a trabeculectomy with lens extraction, and to compare it to simple lens extraction, as well as to determine the importance of complications, if any, of the additional procedure. We compared the IOP, the visual outcome and the complication rates in these two groups.

 Materials and Methods

All patients treated for PG at Schell Eye Hospital, Vellore between June 1990 and June 1993 were analyzed in the study. Patients presenting with PG following trauma or with a history of trauma and unilateral cataract were excluded as were patients with peripheral anterior chamber depth less than one-third corneal thickness. A combination of the following signs and symptoms established the diagnosis of PG: a history of pain/redness/watering; a congested eye with or without corneal edema; marked anterior chamber inflammation with minimal keratic precipitates and posterior synechiae; the presence of cholesterol-like crystals in the anterior chamber, a mature or hypermature cataract with or without soft white spots on the anterior capsule and a raised IOP. All patients underwent a complete ocular examination of both eyes preoperatively, including visual acuity, slitlamp examination, applanation tonometry, gonioscopy when possible, and fundus examination with an indirect ophthalmoscope.

The uveitis was treated with two-hourly topical betamethasone drops. Raised IOP was treated with topical timolol 0.5% two times a day, systemic acetazolamide 250 mg four times a day, and, if required, oral glycerol 60 ml three times a day and/or intravenous 20% mannitol, 5-10 ml/kg. At least 48 hours of such treatment was tried before surgery. The objective was to control the IOP and operate on a quiet eye.

Patients with a duration of symptoms 0-7 days and preoperative IOP less than 31 mmHg underwent lens extraction alone (ECCE 45 eyes, ICCE 1 eye with phacodonesis). Patients with a duration of symptoms of more than 7 days or IOP that remained uncontrolled preoperatively (IOP >30mmHg) underwent lens extraction (ECCE 82 eyes, ICCE 2 eyes with phacodonesis, ECCE+IOL 5 eyes) combined with a trabeculectomy. The choice of a cut-off period of seven days was empirical, based on prior experience with such cases in this hospital.

Cases were operated on by final-year residents and consultants. ECCE was performed in a standard manner using a limbal incision and a linear capsulotomy. Cataract surgery with a trabeculectomy was performed with a fornix-based conjunctival flap and a limbal section incorporating the trabeculectomy scleral flap (before 1992) or a clear corneal section lens extraction with a separate trabeculectomy (1992 onwards).

All patients undergoing lens extraction alone received a drop of timolol 0.5 % at the conclusion of surgery, and oral acetazolamide 250 mg every six hours postoperatively for one day. Postoperative elevated IOP was treated if >30 mmHg or if associated with corneal edema. Topical betamethasone 0.3% and cycloplegics were used in all operated eyes to control postoperative inflammation. Severe uveitis was defined as cells and flare of 3+ or more; persistent uveitis was considered to be cells and flare exceeding 1+ after 6 weeks.[7] Eyes which developed severe postoperative uveitis or exudative membranes were treated with sub-conjunctival or systemic steroids.

Postoperative follow-up examinations were performed at 1, 3 and 6 weeks, and monthly thereafter. At each visit the following parameters were recorded for the operated eye: best corrected visual acuity, IOP, slitlamp examination, indirect ophthalmoscopy, and optic disc evaluation with a 78 D lens. In eyes in which gonioscopy could not be performed preoperatively, the assessment was done at or after the 6-week visit postoperatively.

To analyze the results, patients were divided into two groups. Group I was composed of all eyes which underwent lens extraction alone, while Group II consisted of all eyes in which a trabeculectomy had been performed in addition to lens extraction. The groups were compared using multivariate analysis to allow for confounding variables (IOP and visual acuity) and the Chi square test with Yates' correction (presenting signs and complications). Pearson's coefficient of correlation was used to examine correlations where needed.


There were 135 eyes of 133 patients in the study. All patients were South Indian and lived in the same geographical area. Two patients had bilateral PG presenting on different occasions for each eye. One patient was one-eyed. The age and sex distribution for the two groups was similar [Table:1]. The follow-up response was 99.3% at 6 weeks, 88.7% at 6 months, and 3.7% at one year.

All patients presented with symptoms of slowly progressive decreased vision, followed by an acute onset of redness, pain, and watering in the affected eye. The mean duration of symptoms was 11.5 days (standard deviation [SD] 21.7, range 1-180 days, median 7 days). Group I patients had a mean duration of symptoms of 4.3 days (SD 2.2) and Group II 15.2 days (SD 25.9). Six patients (4.5%) had diabetes mellitus and four patients (3.0%) had systemic hypertension.

The visual acuities of the eyes with PG and the fellow eyes are shown in [Table:2]. Of 130 fellow eyes, 99 (76.2%) presented with a best corrected visual acuity of 20/80 or better; 57 (43.5%) eyes were phakic, and 73 (56.5%) were aphakic. The mean duration of aphakia in the fellow eye was 5.1 years (SD 3.3, range 56 days to 20 years).

The clinical findings at the initial examination are presented in [Table:3]. There was no difference in presenting signs between the two groups. Pupillary examination showed a relative afferent pupillary defect in 3 eyes (2.2%). All patients had open angles. Posterior synechiae were present in 4 eyes (3.0%). The cataracts were mature in 5 (3.7%) and hypermature in 130 eyes (96.3%). Phacodonesis was noted in 12 patients (8.9%), but only 3 of these patients required ICCE.

Mean IOP measurements for the two groups at each period of the study are presented in [Table:4]. The overall mean presenting IOP was 42.5 mm Hg (SD 11.4, range 10 to 70 mm Hg). Four patients presented with IOP within the normal range. These patients had been started on glaucoma treatment and referred for surgery from the outreach program. The mean IOP was significantly different between the two groups, higher in Group II preoperatively and lower postoperatively for each follow-up period. After discharge from the hospital, both groups had mean IOP within the normal range. There was no significant correlation (Pearson's) between duration of symptoms and the presenting IOP (r = 0.09). A weak positive correlation was found between the duration of symptoms and the best controlled pre-operative IOP (r = 0.29, p < 0.001).

Postoperatively, of 89 eyes in Group II, one (1.1%) required therapy for control of elevated IOP. In Group I, of 46 eyes 22 (47.8 %) developed a rise of IOP (>21 mm Hg) of which 16 were over 30 mm Hg. These had to be treated with at least two glaucoma drugs. The mean number of ocular hypotensive medications per patient in the first two weeks postoperatively was 1.69 for Group I and 0.04 for Group II (p < 0.001). Five patients in Group I required at least one dose of intravenous mannitol to achieve IOP control. After six weeks no patient in either group required medications for control of IOP.

Pre- and postoperative visual acuity are detailed in [Table:5]. Of 135 eyes, 115 (85.4%) achieved 20/80 or better visual acuity by six weeks postoperatively. Of the 20 patients presenting with inaccurate light projection, 15 improved to 20/60, two to 20/120, while in 3 patients the vision remained unchanged. All three eyes which did not recover any vision presented with a long period of PG (120-180 days) and were diagnosed as "absolute" glaucomas. There was no statistically significant difference in visual acuity at any stage between the two groups.

Postoperative complications for both groups are listed in [Table:6]. Severe uveitis was seen in more than 50% of eyes but was easily controlled with topical and systemic corticosteroids. Persistent uveitis was seen in only 18 (13.3%) eyes and cleared by two months in all. The vitreous hemorrhage seen in 3 eyes (2.2%), and vitreous opacities seen in 5 eyes (3.7%) cleared spontaneously[8] after 3 months of follow-up.

Complications related to the additional trabeculectomy (included in [Table:6] resolved with conservative management. Of 89 eyes, 8 (8.9%) had flat filtering blebs 6 months postoperatively, with IOP remaining in the normal range.

On examination of the discs postoperatively, the eyes which did not recover vision had optic atrophy; none of the eyes studied showed typical glaucomatous changes.


The pathogenesis of phacolytic glaucoma is an obstruction of the TM by lens proteins and/or protein-laden macrophages, a process that has been shown to be reversible by lens extraction alone.[1][2][3][4][5],[9] Epstein has experimentally demonstrated that leakage of soluble lens proteins can cause a severe obstruction to the aqueous outflow pathway. His experimental study showed that, in the enucleated eye, the obstruction of the TM by proteins cannot be relieved by vigorous anterior chamber irrigation or prolonged perfusion with mock aqueous humour.[3] Yet, clinically, in PG of a short duration, the glaucoma is seen to resolve following lens extraction alone, often with a short period of inflammation and raised IOP. In our prior clinical experience, patients with a longer duration of PG symptoms demonstrated a postoperative IOP elevation, which in severe cases was prolonged and necessitated a trabeculectomy (unpublished data). In such patients, we could reasonably assume that the TM may remain obstructed for longer than usual periods after cataract extraction. It is also possible that very prolonged PG over weeks or months may lead to irreversible TM damage. This hypotheses may explain our clinical impression that lens extraction alone was not sufficient to treat the glaucoma in patients with prolonged PG. However, no information to support this hypothesis is available and the eventual effects of the phacolytic process on the trabecular meshwork are unknown. Our centre has been routinely employing a trabeculectomy in patients with a longer duration of PG for the past 10 years. The present retrospective analysis was undertaken to evaluate this therapeutic approach.

The two groups did not differ in age, sex and race or in presenting signs and symptoms. The presenting IOP and the duration of symptoms were significantly higher in Group II. The latter is one of the criteria of division into groups. Due to effects of the higher mean presenting IOP and the longer duration of PG, Group II may represent a different subset of PG. Therefore it is possible that the two groups may not be scientifically comparable. This is an inherent defect in a retrospective study and the results must be interpreted with caution. However, if differences between the groups were to bias the results, this bias should be towards a higher IOP and a higher incidence of complications in Group II. If, as is the case, the opposite result is seen it can be safely assumed to be truly significant, both clinically and statistically.

The best controlled preoperative IOP was higher in Group II (p<0.001). This again was expected as one of the criteria for dividing the groups was uncontrollable preoperative IOP. Postoperatively, the IOP was statistically significantly lower in Group II (p<0.001). Furthermore, the good postoperative IOP control in Group II was achieved without the use of any medications, while Group I patients had received a drop of 0.5% timolol on the table at completion of surgery and were on oral acetazolamide 250 mg every 6 hours thereafter. The lower IOP in Group II is not of clinical significance after 6 weeks as both groups demonstrated IOP in the normal range and none needed glaucoma medications. There was no correlation between the duration of symptoms of PG and the level of IOP at presentation. However, there was a significant positive correlation between the duration of symptoms and the best IOP control achieved preoperatively (r = 0.29, p<0.001), indicating that the duration of the pathologic process may influence the outcome of medical treatment of elevated IOP in PG. The coefficient of correlation is, however, low.

While only one of the eyes undergoing trabeculectomy had IOP >30 mmHg in the first three days postoperatively, 16 of the simple lens extraction group had IOP >30 mmHg of which 5 were >40 mm Hg. No factor could be identified in the presenting signs or symptoms that distinguished these eyes from others. None of these eyes required subsequent trabeculectomy; however, they required prolonged hospitalization and glaucoma treatment with two or more drugs up to four weeks. The treatment included intravenous mannitol when required.

The recovery of visual acuity was uniformly good in both groups. Interestingly, of the 20 eyes with an inaccurate light projection at presentation, only 3 remained unchanged postoperatively. This is in agreement with previous studies that have shown a relatively good visual outcome in PG.[4] The addition of a trabeculectomy did not worsen the visual outcome in our series.

The complications caused by the addition of a trabeculectomy were few and were successfully managed conservatively. The addition of a trabeculectomy while safe, may theoretically, increase the postoperative inflammation. The rate of postoperative uveitis was, however, not higher in Group II. There is reported evidence that a trabeculectomy is more likely to fail in an eye with severe postoperative inflammation.[10] This may explain the bleb failure noted in 8 patients. In these patients as in the others in Group II, the trabeculectomy nevertheless seems to have played a role in achieving IOP control and avoiding glaucoma medication in the early postoperative period.

Our results show that the addition of a trabeculectomy to the surgical management of PG reduces the postoperative ocular hypertension and the need for glaucoma medications in patients with a long history of PG or in patients with preoperative uncontrolled ocular hypertension. This surgical approach protects against postoperative spikes of high IOP which may represent a fatal insult to an optic nerve already compromised by a long duration of PG. Cataract surgery itself has been shown to initially reduce IOP with a gradual long-term rise in IOP postoperatively.[11] Thus, the addition of a trabeculectomy may also be safer in a situation like ours, where more than 90% of patients are likely to be lost to follow up after 6 months. Finally, inclusion of a trabeculectomy reduces the cost of postoperative medications needed; a significant advantage in a developing country. To assess the need of a trabeculectomy in shorter durations of PG, and to identify the parameters by which a clinical decision on treatment can be made, we are now undertaking a randomized study in all patients with PG regardless of duration[12].


1Flocks M, Littwin CS, Zimmerman LE. Phacolytic glaucoma: a clinicopathologic study of one hundred thirty-eight cases of glaucoma associated with hypermature cataract. Arch Ophthalmol 1955;54:37-45.
2Rosenbaum JT, Samples JR, Seymour B, Langlois L, David L. Chemotactic activity of lens proteins and the pathogenesis of phacolytic glaucoma. Arch Ophthalmol 1987;105:1582-84.
3Epstein DL, Jedziniak JA, Grant WM.Obstruction of outflow by lens particles and by heavy-molecular-weight soluble lens proteins. Invest Ophthalmol Vis Sci 1978;17:272-77.
4Epstein DL. Lens-induced glaucoma. In: Epstein DL, Rand-Allingham R, Schuman JS, editors. Chandler & Grant's Glaucoma. Philadelphia: Lea & Febiger; 1986. p 422-30.
5Lane SS, Kopietz LA, Lindquist TD, Leavenworth N. Treatment of phacolytic glaucoma with extracapsular cataract extraction. Ophthalmology 1988;95:749-53.
6Brookes AMV, Grant G, Gillies WE. Comparison of specular microscopy and examination of aspirate in phacolytic glaucoma. Ophthalmology 1990;97:85-89.
7Kimura SJ, Thygeson P, Hogan MJ. Signs and symptoms of uveitis. I. Anterior uveitis. Am J Ophthalmol 1959;47:155-70.
8Thomas R, Braganza A, George T, Mermoud A.Vitreous opacities in phacolytic glaucoma. Ophthalmic Surg 1996;27:839-43.
9Epstein DL, Jedziniak JA, Grant WM. Identification of heavy-molecular-weight soluble protein in aqueous humor in human phacolytic glaucoma. Invest Ophthalmol Vis Sci 1978;17:272-77.
10Skuta GL, Parrish II RK. Wound healing in glaucoma filtering surgery. Surv Ophthalmol 1987;32:149.
11Bigger JF, Becker B. The effect of uncomplicated cataract extraction on glaucoma control. Trans Am Acad Ophthalmol Otolaryngol 1971;75:260-72.
12Holiday JT, Prager TC. Mean visual acuity. Am J Ophthalmol 1991;111:372-74.