Combined intravitreal bevacizumab with phacoemulsification in visually significant cataract and visually significant exudative maculopathy.

PURPOSE: We investigated the visual outcome of combined phacoemulsification with intravitreal bevacizumab, in eyes with dense cataract and visually significant exudative maculopathy.
MATERIALS AND METHODS: Prospective longitudinal pilot study of consecutive patients treated by two surgeons in 2006, using intravitreal bevacizumab at the end of phacoemulsification. The historical control group consisted of consecutive subjects with exudative maculopathy and dense cataract treated by the same surgeons with the help of phacoemulsification without intravitreal bevacizumab prior to 2006.
RESULTS: Thirty-one treated patients had the mean (SD) logMar best corrected visual acuity improving from - 1.48 (0.50) preoperatively to - 0.67 (0.38) in the first postoperative week (P < 0.001), to - 0.64 (0.40) in the first postoperative month (P < 0.001), and to - 0.62 (0.42) (P < 0.001) on the last follow-up (mean 4.2 months, range 1 - 9 months). Fourteen control patients had the mean (SD) logMar best corrected visual acuity improving from - 1.78 (0.79) preoperatively, to - 0.91 (0.53) in the first postoperative week (P < 0.001), to - 0.86 (0.45) in the first postoperative month (P < 0.001), and to - 0.90 (0.47) (P < 0.001) on the last follow- up (mean 19.6 months, range 1 - 49 months). Initial visual acuities, final visual acuities, and percentage of visual improvement at one month were all not significantly better in the intervention compared to the control group at one month. In the study group, the fovea was flattened at the one-month follow-up, by 90-diopter slit lamp examination and / or Optical coherence tomography.
CONCLUSION: The combination of intravitreal bevacizumab and phacoemulsification is beneficial for maximal visual rehabilitation in the first postoperative month.

Introduction

Cataract extraction improved, in general, both the vision and quality of life. However, the benefits and risks of phacoemulsification in eyes with concomitant exudative maculopathies, such as, wet age- related macular degeneration (ARMD), diabetic cystoid macular edema (DCME), and cystoid macular edema from central retinal vein occlusion, remain uncertain to this date. Cataract surgery could be associated with aggravation of ARMD.[1-3] Several reports warned clinicians about dissatisfied patients when surgery was performed for cataract in eyes with ARMD.[45] Seward[6] warned his patients having cataract and ARMD that there was a one-third chance that they will not feel the cataract surgery to be worthwhile. This led some clinicians to shun away from treating these cataracts. This attitude was strengthened by the population findings by Klein et al.[7] that ARMD was associated with a visual decrease for reading approximately seven lines of letters, while central cataract had a small effect on visual acuity. Likewise, for DCME, diabetic cataract extraction could exacerbate macular edema from increased postoperative angiogenic cytokines[89] and impaired blood- retinal barriers.[10-12]

The vascular endothelial growth factor (VEGF) was implicated in intraocular neovascularization associated with diabetic retinopathy and ARMD.[13-18] Rosenfeld[1314] and others[15-18] pioneered the treatment of wet ARMD, DCME, or cystoid macular edema from central retinal vein occlusion by intravitreal bevacizumab. We investigated, by treating exudative maculopathy with intravitreal bevacizumab, concomitant with phacoemulsification, in the hope of decreasing the visual disability from the maculopathy. We present a longitudinal prospective clinical series assessing the visual outcome of this combined procedure and comparing it with the retrospective historical control series operated by the same surgeons.

Materials and Methods

In a prospective manner, intravitreal bevacizumab was given at the end of phacoemulsification in consecutive patients with dense cataract and exudative maculopathy (DCME, wet ARMD from subfoveal or juxtafoveal choroidal neovascularization, cystoid macular edema from central retinal vein occlusion) treated in 2006, by two surgeons. Dense cataract referred to severe nuclear sclerosis or severe posterior capsular cataract, hampering assessment of the fovea by funduscopy, intravenous fluorescein angiography or Ocular Coherence Tomography (OCT). The best spectacle corrected Snellen visual acuities were recorded by the same examiner. The control group consisted of phacoemulsification for dense cataract (without intravitreal injection) in eyes with exudative maculopathy (DCME or wet ARMD) retrieved retrospectively and consecutively from January 1994 to December 2005, by the same surgeons. Potential benefits and risks (endophthalmitis, vitreous hemorrhage, retinal tear, retinal detachment) of the injection as well as the visual limitations of cataract surgery were thoroughly explained before signing an informed consent. The study was approved by the Institutional Review Board.

All cases in both groups were operated when the systemic conditions were controlled (blood sugar, lipid, and blood pressure). Standard phacoemulsification was done and foldable Centerflex® (Rayner, Hov, England) or Acrysof® (Alcon, Fort Worth, Texas, USA) intraocular implants were used in the study group. However, in the control group, polymethylmethacrylate 5 mm by 6 mm intraocular implants were used in half of the patients and foldable implants in the second half. At the end of the surgery, intravitreal injection of 0.1 ml (2.5 mg) of bevacizumab (Avastin, Genentech Inc., San Francisco, CA) was given through a 30-gauge needle, under a surgical microscope, superotemporally, 3 mm from the limbus to eyes in the study group. Examinations were scheduled postoperatively, on day one, and after one week, one month, three months, and six months, respectively. Slit lamp biomicroscopy of the macula was performed with the 90-diopter lens. Intravenous fluorescein angiography and macular mapping by OCT (Stratus OCT, Carl Zeiss Meditec, Dublin, California, USA) complemented the follow-up clinical examination in two-thirds of the study cases.

Statistical analyses were carried out using SPSS 13.0 software (SPSS Inc, Chicago, IL, USA); paired samples t-test was used for data analysis within the groups, and Mann- Whitney test was used for data analysis between the study and control groups. P-values less than 0.05 were considered statistically significant. In bilateral surgery, only the eye with the longest follow-up was included for statistical analyses.

Results

A total of 31 eyes of 31 patients in the study group and 14 eyes of 14 patients in the control group were available for analysis. Both the intervention and control groups were comparable with respect to gender (55% male and 64% male, respectively, P = 0.553), age (means 71 (SD 12) and 68 (SD 11.7), respectively, P = 0.206) and initial visual acuities (means 1.48 (SD 0.50) and 1.78 (SD 0.79), respectively, P = 0.129).

The intervention group included 20 eyes with ARMD, 10 eyes with DCME, and one eye with central retinal vein occlusion-related cystoid macular edema. Mean (SD) logarithm of the minimal angle of resolution (logMar) of best corrected visual acuity improved in the study group from - 1.48 (0.50) preoperatively to - 0.67 (0.38) in the first week (P < 0.001) and - 0.64 (0.40) in the first month (P < 0.001) postoperatively, and to - 0.62 (0.42) (P < 0.001) on last follow-up (mean 4.2 months, range 1-9 months) [Table 1]. The fovea was flattened on the one-week and one-month follow-up, as seen on a 90-diopter slit lamp examination and / or OCT in all eyes [Figure 1a and b]. No adverse events, such as endophthalmitis, retinal tears, retinal detachment, or thromboembolic events were observed in any patient. Four diabetic and three ARMD patients needed sub-Tenon depot corticosteroid, as 40 mg of methylprednisolone acetate (1 ml DepoMedrol, Pharmacia and Upjohn, Puurs, Belgium) in the first postoperative month, to further control the maculopathy. Five eyes with wet ARMD and three eyes with DCME needed additional intravitreal bevacizumab injections after the first postoperative month. Prior therapies to phacoemulsification included two sessions (three eyes) or one session (one eye) of photodynamic therapy with verteporfin (Visudyne, Novartis Pharmaceuticals, East Hanover, NJ, USA), transpupillary thermotherapy (two eyes), focal argon laser photocoagulation (four eyes), and intravitreal triamcilonone (six eyes).

Table 1

Clinical profile of study subjects in the intervention and control groups

Figure 1

OCT before (a) and one month after (b) phacoemulsification with intravitreal bevacizumab in a 44-year-old DCME resolved at one month, with logMar best corrected visual acuity improvement from - 1.3 to - 0.4

The historical control group consisted of four patients with wet ARMD, and 10 with DCME. Mean (SD) logMar best corrected visual acuity improved in the control group from - 1.78 (0.79) preoperatively to - 0.91 (0.53) in the first postoperative week (P < 0.001), - 0.86 (0.45) in the first postoperative month (P < 0.001), and to - 0.90 (0.47) (P < 0.001) on the last follow-up (mean 19.6 months, range 1 – 49 months) [Table 1]. Four eyes needed sub-Tenon corticosteroid injections in the first postoperative month, while four eyes needed injections after the first postoperative month. Two eyes underwent focal laser photocoagulation, and one eye needed photodynamic therapy. Prior therapies to phacoemulsification included panretinal (two eyes) and focal photocoagulation (two eyes).

Best corrected visual acuity improved in both the groups, in all patients, in the first month of follow-up [Table 2]. Using the Mann-Whitney test, percent changes in visual acuities, throughout the follow-up, between the study and the control groups were found to be greater in the intervention than in the control group, however, none of the comparisons reached statistical significance. Follow-up was longer in the control group than in the study group, but this was not significant (P = 0.17).

Table 2

Comparison of logMar best corrected visual acuity between and within study groups

Discussion

Phacoemulsification (with or without bevacizumab) improved visual acuity in all 45 eyes of 45 patients from both groups. Concomitant treatment of the maculopathy further improved the degree of visual rehabilitation, but the small number of cases in our treatment and control series did not allow the superior postoperative visual acuity in the treatment group to reach statistical significance. Moreover, many control eyes did receive treatment for their maculopathy in the perioperative period, which may have improved their visual prognosis. Another point is that the majority of controls had DCME (10 out of 14) unlike the treatment group (10 out of 31). Moreover, since the visual prognosis may be better in the eyes with DCME versus wet ARMD, this may have skewed the results in favor of the control group. Furthermore, most of the reported series discussing cataract surgery in eyes with macular degeneration dealt with an overwhelming majority of the dry form of ARMD, with few cases of exudative ARMD.[1920] Our case series dealt uniquely with the exudative form, hence the small number in the present study. Some other limitations of the study design are the historical nature of the control group, differences in intraocular designs, and differences in preoperative or postoperative therapies (additional intravitreal bevacizumab, preoperative and postoperative sub-Tenon corticosteroids, with their known effect on decreasing maculopathy,[21] and focal laser therapy).

The control group had worse initial visual acuities than the study group. Despite being non-statistically significant, this could have been due to a more adequate control of the maculopathy in the study group compared to the control group, at the time of surgery, as many eyes in the study group had received treatment ((two sessions (three eyes) or one session (one eye)) of photodynamic therapy, transpupillary thermotherapy (two eyes), focal argon laser photocoagulation (four eyes), and intravitreal triamcilonone (six eyes)) versus only a few eyes in the control group ((panretinal photocoagulation (two eyes) and focal photocoagulation (two eyes)).

In eyes with both ARMD and cataract, both diseases contribute in varying degrees to the patient's visual disability. Cataracts reduce distance visual acuity and contrast sensitivity by reducing the quality and physical contrast in the retinal image by the effect of veiling glare.[19] Patients with macular degeneration have impaired macular function from the early stages of the disease, but only lose visual acuity in the late stages from geographical atrophy or exudative complications. Monestam et al.[4] conducted a prospective, population-based study of cataract surgery outcome and found 37 of the 459 cases (8%) to be dissatisfied. Logistic regression analysis revealed that the presence of ARMD was a significant risk factor for poor patient satisfaction. Poor visual acuity after cataract surgery was found by Monestam et al.[4] in 22% of these 459 cases, mostly as a result of concurrent ARMD or diabetic retinopathy, with 4% of the subjects having no improvement in visual acuity in the operated eye. The assessment of the relative contribution of the cataract to the patient's decreased visual performance was based on the clinical judgment of the ophthalmologist assessing the patient. In our case, we had chosen to operate on eyes with concomitant exudative maculopathy only when the cataract was dense. Studies on the usefulness of the potential acuity meter had reported contradictory results in eyes with cataract and maculopathy, and it would appear that there was no single instrument that was better than an experienced clinical judgment as shown by Miller et al.[22] For example, one of our patients (N.26 Table 1.) had a preoperative logMar of - 1.6; a potential acuity meter test yielded a logMar of - 0.48; the patient achieved a logMar of - 0.18 in the first postoperative week and - 0.10 in the first postoperative month.

In diabetic eyes, it was found that extracapsular cataract extraction worsens diabetic retinopathy or maculopathy.[10-1223] Recent phacoemulsification techniques do not lead, in general, to appreciable deterioration of diabetic retinopathy,[24-26] due to shorter time of the procedure, smaller self-sealed corneal incisions, avoidance of ocular hypotony, avoidance of iris trauma (iris hooks in miotic pupils), and in-the-bag implantation of more biocompatible foldable intraocular implants. However, the natural history of patients with DCME, at the time of phacoemulsification, is recognized as being poor.[252728] Chung et al.[26] studied the effect of phacoemulsification on the progression of diabetic retinopathy, in 75 patients who had the same degree of retinopathy in both eyes, using the non-operated contralateral eye as a control. The retinopathy progressed significantly, more in the operated eye (occurring in 23 patients or 30.6% after phacoemulsification) and related to the presence of preoperative DCME and poor renal function.[26] Somaiya et al.[25] analyzed the visual outcome of phacoemulsification in 106 diabetic and 55 non-diabetic controls. Achieving a postoperative best corrected visual acuity of 6 / 12 in patients with nonproliferative diabetic retinopathy was nearly five times less likely (P = .02) and 30 times less likely (P < 0.0001) in patients with proliferative diabetic retinopathy when compared to diabetic patients without retinopathy. This provides an incentive for perioperative interventions, in an attempt to prevent any worsening of the retinopathy or macular edema.

Funatsu et al.[9] analyzed VEGF vitreous levels in 36 patients with DCME, six diabetic patients without retinopathy, and 13 patients with non-diabetic ocular disease. The vitreous levels of VEGF were significantly higher in the DCME patients than in the non-diabetic patients or diabetic patients without retinopathy. Similarly, aqueous VEGF levels were also found to be increased in AMD[29] and retinal vein occlusions.[30] Furthermore, Patel et al.[8] analyzed the aqueous level of VEGF in seven eyes of six patients with diabetic retinopathy ranging from severe non-proliferative to quiescent proliferative that had uneventful phacoemulsification with intraocular lens implant. VEGF aqueous concentration increased one day after surgery from a median baseline of 68 pg/ml (range 22-87 pg/ml) to 723 pg/ml (range 336-2071) on day one. By one month it had decreased to 179 pg/ml (range 66-811 pg/ml). These altered concentrations of angiogenic factors after cataract surgery may possibly induce subclinical or clinical worsening of diabetic maculopathy. Hence, the marked increase of aqueous VEGF in eyes with diabetic retinopathy can justify our use of a high dose of intravitreal bevacizumab as compared to other studies (2.5 mg instead of 1.25 mg). In addition to its effect on DCME, intravitreal bevacizumab has a role in the inhibition of concomitant retinal neovascularization, rubeosis iridis or neovascular glaucoma, as well as in controlling macular edema in eyes that need concomitant intraoperative or postoperative focal or pan-retinal photocoagulation.[15]

One may argue that intravitreal injections can be given before or after cataract surgery. However, the patient may be disappointed (by incomplete visual recovery with phacoemulsification alone or with the intravitreal injection alone) and decide to skip the next step in this two-stage therapy. The present combination yields more cooperative patients, according to our experience, and saves another surgical procedure. For example in Case 2 [Table 1], a patient with 6 / 60 cataract and 6 / 60 diabetic cystoid maculopathy, visual acuity would improve from finger counting to 6 / 60 with one surgery alone (phacoemulsification or intravitreal injection). With the combination therapy, visual acuity would improve to 6 / 12 in the short term follow-up.

Alternatives include phacoemulsification with sub-Tenon corticosteroid (Mansour AM, unpublished data, AAO poster 2005, Chicago, Illinois), intravitreal corticosteroid,[3132] and vitrectomy[33] Amino et al.[33] analyzed the outcome of pars plana vitrectomy combined with phacoemulsification and found the treated group to have significant improvement in visual acuity from a logMar of - 1.09 to - 0.80. Another avenue similar to our study consisted of the concomitant use of intravitreal corticosteroid. Lam et al.[31] prospectively administered 4 mg of intravitreal triamcinolone concomitant with phacoemulsification, in 19 eyes of 15 consecutive diabetic patients with cataract and clinically significant macular edema. Their results revealed significant improvement in mean Snellen visual acuity of 2.4 lines at the 6-month follow-up.[31] In a similar prospective pilot study of phacoemulsification in eyes with DCME by Habib et al.,[32] 83% of 12 patients achieved a dry fovea two weeks postoperatively, and 75% of 12 patients had visual improvement at the two-month follow-up. More recently, several investigators[34-39] have found beneficial effects in concomitant intravitreal bevacizumab at the time of phacoemulsification in exudative maculopathy from diabetes or age-related macular degeneration, without side-effects. From this small pilot study we conclude that the combination of intravitreal bevacizumab and phacoemulsification for eyes with exudative maculopathy is safe and beneficial for maximal visual rehabilitation in the first postoperative month. It is important, however, to reduce unreasonable expectations about the outcome of surgery. The present data showed that the procedure was safe as we did not encounter complications. It was also effective in improving vision. Even as the present study does not give us a precise answer as to how much the combination improved vision over cataract surgery alone, yet the combination caused flattening of the fovea anatomically. One of us (AMM) operated recently on a 77-year-old lady with very dense cataract and 6 / 120 (20 / 400) visual acuity. She had no visual improvement postoperatively. Slit lamp biomicroscopy detected mild serous elevation of the fovea that could not be appreciated prior to cataract surgery. Visual acuity improved to 6 / 9 (20 / 30) after intravitreal bevacizumab. The patient was very anxious because of her initial poor visual outcome, and this anxiety was relieved by the visual improvement following bevacizumab. Again, the combined phacoemulsification injection of bevacizumab allows maximal visual rehabilitation in the immediate postoperative period, saves a second procedure (injection),[40] and often decreases patient dissatisfaction.