Materials and methods

From a 1 year retrospective analysis of medical charts, 32 cases of complicated PCCC on intact capsules were identified in a cohort of 650 patients with PCCC (5%). Complicated PCCC was defined by the peroperative observation of vitreous passage into the anterior segment, tears in the posterior capsule, uncontrolled size or location of the PCCC, or technical difficulty in PCCC performance. These 32 cases represent all the unplanned problems encountered with the posterior capsule, and they were further analysed for type of surgery, preoperative and postoperative factors, and relative risk factors for vitreous issue. Background ocular pathology (Table 1), where it was felt that visual outcome could be suboptimal, was present in 16 patients (50%). Table 2 gives age, sex, axial length, and preoperative intraocular pressure (IOP) characteristics. Only one case was carried out under local anaesthesia (via an inferotemporal peribulbar injection of 4 ml lignocaine 2%). The rest were operated under general anaesthesia. The surgeons performing the procedure were two of the authors (FV and AG). No device was used preoperatively to lower vitreous pressure.

After either corneal incision (90.6%) or fornix based flap (9.4%), an anterior circular capsulorhexis was created. This was followed by phacoemulsification (41%), nucleus reduction (28%),8Kansas technique (19%),9 or, for patients with a very soft nucleus, irrigation aspiration only (12%). Before the insertion of an intraocular lens, a hole was created with a 30 gauge needle in the posterior capsule, viscoelastic substance was injected through the hole and a PCCC was attempted. Complications appeared and were handled as discussed below. After removal of the viscoelastic, the self sealing incision was sutured, and a drop of β blocker was immediately applied. Intravenous acetazolamide was systematically administered.

Statistical analysis of the relative risk factors involved in vitreous loss was performed with the odds ratio (OR) method and the χ² test.

Results

A complete overview of the incidence of each type of complication is given in Table 3. The most frequent problem was to perform a central capsulorhexis of the optimum size (<5 mm diameter). This was not possible in 14 cases, and in 12 cases a PCCC was carried out with great difficulty because of either insufficient visibility during the procedure or an anatomically changed capsule (floppy or fibrotic in its centre).

Table 4 gives the patient characteristics and the type of procedure in cases with vitreous loss. Risk factors were found to be the presence of multiple vascular problems (OR=25, χ²=9.5, p<0.002), age over 80 years (OR=9.4, χ²=4.7, p=0.03), and manual extraction of the nucleus (OR=4.2, χ²=1.8, not statistically significant). These six cases necessitated the use of anterior dry vitrectomy before the implantation of the IOL.

Tears in the posterior capsule were created during the opening of the capsule with vitreous positive pressure (one case), during injection of viscoelastic with exaggerated pressure in the PCCC (one case), and during the IOL implantation in a non-circular posterior capsulorhexis (one case).

In the 11 cases demonstrating too large and/or eccentric PCCC, the loss of control of the size and of the location of the PCCC occurred during viscoelastic injection in eight cases (viscoelastic was injected under too much pressure or with air bubbles), as a result of positive vitreous pressure in one case, poor visibility in another case, and without identifiable reason in one case.

Finally, in 12 patients, the procedure was controlled but technically difficult. The problem involved mainly the quality of the posterior capsule which was either floppy and located very deeply in the eye or fibrotic in its centre making the control difficult since the fibrotic plaque tends to direct the size and the location of the PCCC. A poor visibility during the procedure was the consequence of a damaged cornea (two cases), myotic pupil (one case), air bubbles in viscoelastic (one case), and improper position of the coaxial light as regards the position of the PCCC (one case).

In all these cases, after a thorough checkup of the capsular stability under the iris, endocapsular implantation was performed during the initial procedure.

Table 5 details the results of the examination performed on the first postoperative day. Table 6 gives the postoperative complications related to the performance of a PCCC with a mean follow up of 13.5 months (range 4 to 25). The mean IOP was 14.1 mm Hg (range 8–24) and no case resulted in chronic glaucoma. The visual performance was good (mean visual acuity of 32/40, range 20/60–60/60). Only two patients with macular degeneration had visual acuity under 20/40. Five out of six patients, in whom vitreous was present in the anterior chamber, already had vitreous loss during the procedure. The only patient who developed this complication postoperatively had a PCCC larger than the size of the optic.

Discussion

There are currently three main strategies in the research of after-cataract prevention. The first, by analogy with the experience acquired in glaucoma, involves the application of antiproliferative drugs in the capsular bag.10 The second is oriented towards complete removal of epithelial cells from the capsule during surgery (diathermy)¹¹ or after a latent period (toxins coupled with a specific antibody or a growth factor).12 13 The third tries to preserve a clear central axis for a long period of time (IOL geometry and biochemistry,14 equatorial rings,15 removal of the central posterior capsule during the initial procedure).

Posterior capsulorhexis has been introduced in cataract surgery for paediatric cataract extraction or for extracapsular surgery complicated by capsular disruption. There are, however, no reports in the literature on the possible intraoperative or postoperative complications of these cases. We have extended the indications of this procedure to the majority of our cases presenting an intact capsule at the end of cataract extraction. Our analysis of 32 cases of complicated PCCC (due either to vitreous passage into the anterior segment, loss of control of the rhexis, or problems in obtaining a continuous circular centred rhexis) highlighted the contributing factors to the problems observed.

The only statistically significant relative risk factors for vitreous loss during the procedure were found to be systemic cardiovascular disease and age over 80 years. This is in accordance with previous studies which suggested a relation between positive vitreous pressure and coexisting vascular problems.16 Our six cases with vitreous loss necessitated the use of anterior dry vitrectomy with viscoelastics to prevent further vitreous hydration. The anterior vitrectomy was probably not complete enough since half of these cases still had vitreous in the anterior chamber the next day. Another possibility is that, even if the vitrectomy were complete, a small amount of vitreous somehow found its way towards the anterior segment through the posterior rhexis. One case in our series developed vitreous in the anterior chamber while vitreous loss was not noticed immediately after the operation. The reason was probably a PCCC size larger than the size of the optic.

The control of the PCCC was lost mainly during injection of viscoelastic under too high pressure or with air bubbles in it. The posterior capsule is very thin and exquisitely sensitive to minor variations of pressure around its plane, especially after a small hole has been created in its centre. The viscoelastic should ideally be checked for the presence of air bubbles in the syringe before insertion into the eye and it should be injected very carefully to push the anterior hyaloid membrane backwards without creating a large difference of pressure between the anterior and posterior faces of the posterior capsule.

PCCC should be performed with great care in eyes having coexistent abnormalities that preclude a clear view of the posterior capsule or in eyes in which the posterior capsule is located very deeply (at the limit of focus of the operative microscope). Patients with a myotic pupil or a damaged cornea are obviously not good candidates.

Another potential problem was occasionally the presence of a fibrous plaque on the posterior capsule which, by itself is a good indication, but which can also direct, by its position, the size and the form of the rhexis.

Different authors have expressed concern about the increased difficulty of implantation once the posterior capsule has been opened.17 PCCC performance after IOL implantation has the advantage of reducing problems with implantation on an opened capsule. This timing, in our opinion, increases overall the technical difficulty of PCCC realisation. Insertion of the IOL in the capsular bag was always possible in our patients despite the complications encountered. We performed, however, in every doubtful case, a thorough check up of the capsular bag stability with the aid of an iris retractor.

Nearly half of our patients showed a mild to moderate corneal oedema on the first postoperative day. This was probably the result of increased operation time and increased instrumentation. Only five cases had elevated IOP (over 25 mm Hg) which necessitated the administration of intravenous osmotic agents, despite the extensive use of viscoelastics for the PCCC realisation and for the management of the complications. The low number of elevated IOP is probably related to the systematic use of prophylactic β blockers and intravenous acetazolamide at the end of surgery in all our cases with a complicated PCCC, since the viscoelastic in the posterior segment was only partly removed.

Table 6 lists the postoperative complications found in our series after a mean follow up of 13.5 months. One patient developed a retinal detachment 8 months after the procedure and was successfully treated with scleral buckling. It was felt unjustifiable to perform fluorescein angiography for study purposes only on all our patients. Fluorescein angiography was performed on the first 50 consecutive patients with uncomplicated PCCC; the incidence of angiographic cystoid macular oedema was found to be 6%.18 We observed, in the complicated PCCC series, two cases of clinically significant cystoid macular oedema, which spontaneously regresses over a period of 6 months. Other authors did not mention occurrence of cystoid macular oedema in their series.19 Chambless found an incidence of 7% in phacoemulsification cases complicated by posterior capsule disruption.20

The visual prognosis after complicated PCCC can be very good; 93% of patients had corrected visual acuity ≥ 20/40. This compares favourably with 77% in Gimbel’s series,4 while half of our cases had background ocular pathology. However, a complicated PCCC in which all lens matter has already been removed is obviously an easier situation than a disruption of the posterior capsule at the beginning of the phacoemulsification.

Loss of control of the PCCC can lead to potential problems during surgery. Our analysis of 32 difficult cases suggests that old patients with serious systemic vascular disease, eyes with floppy or fibrotic capsules, and those offering us poor visibility during the procedure are not good candidates for the performance of a safe PCCC. A good knowledge of the technique, especially during the manipulation of viscoelastics around the posterior capsule is mandatory to complete the procedure with a posterior continuous circular centred capsulorhexis of the optimum size, without vitreous loss.