Considerations

In general, the indications for proceeding with surgery need to be more carefully considered in eyes with uveitis. Because of the possible increased risks, surgery is frequently not justified for eyes with slightly reduced – but functional – vision which are not in danger of visual loss from a surgically repairable problem. On the other hand, we have observed many eyes in which surgery was delayed for so long that potentially treatable problems became hidden behind cataract or vitreous haze. Vision was then permanently lost because of the attempt to use surgery conservatively; vision might have been saved if the inflammatory disease and its complications could have been directly viewed and treated. There is a tendency to not operate on an eye with dense opacities and light perception vision while unseen and untreated inflammation causes further permanent damage to the retina. In the end the eye does go blind, but there was a point in the course of the disease at which the potential benefits of surgical intervention, even if low, outweighed the resultant risk of nonintervention. In addition to visual rehabilitation, surgical procedures in eyes with uveitis are performed as part of a diagnostic evaluation. The relative risks versus the potential benefits are considered here.

Except for the few situations in which it is necessary to operate on an acutely inflamed eye, surgery should be performed when the inflammation is quiescent. However, in some patients it is impossible to clear every cell from the anterior chamber or vitreous. Furthermore, in patients with dense cataracts and primarily vitreoretinal inflammation it is impossible accurately to assess the activity of the disease behind the opacity. Therefore, an attempt should be made to reduce the inflammation as much as possible but not to delay surgery while the patient endures a long-term visual handicap because the inflammation never becomes completely quiescent. If the inflammatory activity cannot be assessed, the eye may be treated prophylactically for a few days as though active inflammation were present.

In our opinion, corticosteroids are still the best drugs for the control of inflammation in the perioperative period. In a small study of 20 patients randomly selected to receive prednisone or ciclosporin before cataract surgery, we found that three of 10 patients treated with ciclosporin – but none of the steroid-treated patients – had severe postoperative inflammation that necessitated a change in medication to the other drug. Corticosteroids work at multiple points in the inflammatory process, as described in Chapter 7 , and are able to affect inflammation mediated by neutrophils and macrophages as well as that by lymphocytes. Therefore, they may prevent nonspecific postoperative inflammation that may occur in a previously inflamed eye as well as inhibit the more specific pathways of inflammation. We prefer to begin steroid therapy on the day before or the day of surgery if the patient is not already receiving the drug, and if there is no evidence of active inflammatory disease. Some practitioners prefer to start steroid therapy several days before the scheduled surgical intervention. If there is no strong contraindication this may not be unreasonable in some patients, although it should be remembered that the effects of the daily steroid are gone 24 hours later. Usually a dose of 1 mg/kg/day of steroid is adequate. In addition, we give a subconjunctival corticosteroid injection at the time of surgery. If the patient has a history of severe intraocular inflammation, we also give an intravenous dose of corticosteroids, such as 50–100 mg of methylprednisolone, at the time of surgery. The goal is clearly to prevent severe postoperative inflammation.

The surgeon should be aware that with the aforementioned perioperative therapy the eye may appear relatively quiet for the first few days after surgery, but the inflammation may increase a few days later. It is therefore critical to examine patients with uveitis daily during the first 5–7 days after surgery. In young patients and those with sarcoidosis we have specifically noted marked postoperative flares of inflammation accompanied by significant fibrin and protein within the eye. At times the fibrin is so dense that it mimics vitreous bands, and the retina cannot be viewed. Tissue plasminogen activator can be used to clear severe fibrin accumulation in the anterior chamber ( Fig. 8-1 ). After surgery these eyes frequently have hypotony, and although they appear to be in imminent danger of phthisis, continued treatment with steroids combined with patience usually results in marked clearing within 3–5 days without apparent additional ocular damage.

A, Uveitic eye after extracapsular cataract extraction (phacoemulsification) and intraocular lens placement. The postoperative inflammatory response was severe, with the development of fibrin in the anterior chamber. B, Clinical improvement occurred 24 hours after injection with tissue plasminogen activator.

(Courtesy Rubens Belfort Jr, MD, Sao Paulo, Brazil.)

Removal of band keratopathy

Calcium hydroxyapatite accumulates in Bowman’s membrane of the cornea in some patients with sarcoidosis or juvenile rheumatoid arthritis and occasionally in younger patients with a chronic uveitis such as pars planitis. Its location may be related to the local pH of the interpalpebral cornea. It can be removed by chelation if it is located centrally and interferes with vision because of opacity or glare, or if it is the cause of recurrent corneal erosions. Although chelation with 1–2% ethylenediaminetetraacetic acid (EDTA) can be performed under local anesthesia, the age of most patients with this condition makes general anesthesia necessary. Our procedure is to use a sterile 1.7% solution of EDTA in water or saline applied to the cornea with cotton or a cellulose applicator. Gentle pressure and movement of the applicator result in removal of the corneal epithelium and gradual solubilization of the calcium deposits. This procedure can take up to 30 minutes and often must be performed before intraocular surgery. Occasionally, resistant deposits must be gently removed with a spatula or Tooke knife, but this may cause scarring of Bowman’s membrane and should be avoided in the visual axis. The eye is then patched until the epithelial defect heals. Band keratopathy has also been removed with the excimer laser ( Fig. 8-2 ). It should remembered that eyes with uveitis may have recurrences of the band even years after the removal of the calcium. Further, the procedure may very rarely induce a severe inflammatory response, as was reported in one patient.

A, Band keratopathy in 6-year-old boy with bilateral uveitis. B, Same eye 2 days after excimer laser surgery performed under topical anesthesia.

(Courtesy Rubens Belfort Jr, MD, Sao Paulo, Brazil.)

Corneal transplantation

Although keratitic precipitates involve the corneal endothelial surface, endothelial decompensation is rare in our experience in patients with chronic uveitis. Thus corneal transplantation is rarely performed in this group because of corneal edema, but it may be necessary in patients with corneoscleral melting in association with systemic inflammatory disease and intraocular inflammation, or in patients with herpetic keratouveitis. If the intraocular inflammation has been severe, there may be coexistent glaucoma or cataract. The primary goal before corneal transplantation in these patients is to control the immune process, or the newly grafted tissue will melt as well. The ongoing inflammatory response suggests that the risk for rejection of a cornea transplant in any patient with active uveitis is high because there are already large numbers of cells in the eye that could present corneal antigens to the immune system and participate in the rejection event. Certain severe manifestations of immune-mediated ocular surface disease, such as vernal keratoconjunctivitis, can lead to limbal stem cell deficiency, and before a penetrating keratoplasty is considered the ocular surface environment needs to be addressed. The treating physician should be aware of drugs being given for other indications that may induce a uveitis and risk graft survival. Richards and co-workers described such a case of a patient receiving alendronate sodium for osteoporosis. The uveitis resolved after cessation of the drug.

Cataract surgery

Much information has been acquired during recent years concerning cataract extraction in uveitic eyes. Although there is not yet an absolute consensus on the most ideal method of cataract removal, we are moving closer to that goal and some basic tenets are being established. Individual patients and various disorders often require different approaches. Summaries of procedures performed 20–30 years ago suggested that the outcome of cataract surgery in eyes with uveitis was poor, and that phthisis occurred frequently. This is clearly no longer the case in our experience, nor in that of other investigators, thanks largely to new technologies and approaches. Today, intracapsular cataract extraction is rarely if ever performed. The theoretic advantage of not leaving residual lens material in the eye seems to be outweighed by the larger incision and higher incidence of retinal detachment and cystoid macular edema. We reserve intracapsular extraction for mature lenses in eyes with phacoantigenic uveitis or phacogenic glaucoma, or in eyes with lenses that are dislocated or mobile because of loosened ciliary processes.

Extracapsular extraction, either via irrigation/aspiration or – now almost always – with phacoemulsification, is now the centerpoint of the discussion. These approaches have the advantages of smaller incisions and less manipulation of the iris. Reports support the use of phacoemulsification over an expression/irrigation system. Several studies have shown less postoperative inflammation in eyes undergoing phacoemulsification. Chee and colleagues looked at this question in uveitic eyes prospectively, comparing the two methods, and found that the eyes with phacoemulsification invariably had less inflammation. We expect to see good results for all patients with uveitis upon whom we operate. However, we do see complications. Postoperative hyphema is not uncommon in inflamed eyes. Postoperatively one might see the formation of posterior synechiae to the capsule, pupillary block glaucoma, the inability to remove vitreous debris, haze and vitreoretinal traction, and possible opacification of the posterior capsule. We have also observed patients in whom a neodynium:yttrium–aluminum–garnet (Nd:YAG) laser capsulotomy leads to an exacerbation of inflammation. Herbort reported the case of a patient in whom a severe and recurrent scleritis and uveitis developed after an Nd:YAG capsulotomy was performed 18 months after cataract extraction. Also, laser-induced iris punctures often close off readily in inflamed eyes. In all patients undergoing surgery we wish to see the ocular inflammatory process well controlled for at least 3 months (and preferably 6).

The surgeon may encounter challenging technical problems. If the pupil is less than 2–3 mm the surgeon is likely to encounter a dense fibrin membrane across it. To widen the pupillary aperture we often use disposable iris retractors in the four quadrants. In rare eyes the pupil will not dilate until this membrane is cut with a Van Ness scissors after sodium hyaluronate (Healon) has been administered. Posterior synechiae not only may be present at the pupillary border but also may extend over the entire surface of the lens. These extensive adhesions cannot be easily broken with an iris spatula, but with the aid of a sector iridectomy and pupillary membrane sectioning an adequate anterior capsulotomy can be performed. In addition, an anterior surgical approach should be avoided if there are peripheral anterior synechiae extending onto clear cornea, because the surgical incision may not enter the anterior chamber but instead go through the iris into the posterior chamber. With the use of iris retractors we have been able to perform phacoemulsification on many uveitic eyes requiring surgery. Many of these patients are young, and phacoemulsification time is quite low. Most of the nucleus can be removed by aspiration.

The surgeon should expect uveitic eyes to bleed from the conjunctiva, Tenon’s capsule, the corneoscleral wound, and the iris. Even eyes that have been quiet for several months appear to have an increased tendency to bleed during or after surgery. Bleeding can be controlled by careful cautery, by tamponade with the irrigating solution, and by patience. Entering into the anterior chamber through a scleral tunnel or clear cornea still is discussed. The clear cornea approach has a certain advantage if glaucoma surgery is contemplated.

It is imperative to restate that one needs to use any means necessary to bring the eye’s inflammatory disease under strict control. Some would mandate even the absence of every cell, which for us is sometimes most difficult to obtain, but the message is one we fully support. The disease must be controlled, and it must be controlled for an extended period – usually 3–6 months – before we would consider surgical intervention. The patient should be informed about the possible need for continued and even more aggressive immunosuppressive therapy to control the disease postoperatively. An extracapsular cataract extraction (phacoemulsification) and placement of the intraocular lens (IOL) in the bag would be our preferred approach. We would prefer to make a small as an incision as possible, and various techniques are reported in the literature. This is often followed by a curtailed anterior vitrectomy. This sort of approach is outlined by Walker and colleagues. The decision as to whether to place an IOL continues to be debated, and although there are no strict rules (how can there be in medicine?), certain guidelines have evolved. Patients with chronically active granulomatous disease seem to cause us the most concern. Iris capture ( Fig. 8-3 ) is also a real concern (but we see this less and less), and patients with posterior synechiae have a greater risk for occurrence of this problem as well as the capsule contraction syndrome postoperatively.

Uveitic eye that underwent cataract surgery and posterior chamber IOL placement. A, Inferior portion of lens has migrated anterior to lens plane. B, The whole lens is now forward. This ‘iris capture’ occurred without marked inflammatory disease.

(Courtesy Rubens Belfort Jr, MD, Sao Paulo, Brazil.)

Much interest has been shown in surface-modified lenses and the possibility of their reducing the postoperative inflammatory response. Percival and Pai placed a heparin surface-modified one-piece lens into the posterior chamber of 36 patients with cataracts associated with previous or chronic recurrent uveitis. They found that recurrent inflammatory disease was present, including an acute postoperative fibrin reaction, but the heparin coating appeared to provide a cell-free IOL surface in most eyes ( Figs 8-4 and 8-5 ). Reports suggest that heparin modification does not provide a significant benefit. Alio and colleagues, in a prospective study of cataract surgery in uveitic eyes, compared four different types of IOL. Using hydrophobic acrylic, silicone, polymethylmethacrylate (PMMA), and heparin surface-modified PMMA lenses, they concluded that the acrylic IOLs had a better visual outcome and lower complication rate, whereas both acrylic and heparin surface-modified PMMA lenses had the lowest relapse rates. Elgohary et al. found that YAG capsulotomies were more common in older patients, but there were generally fewer problems in patients receiving prophylactic systemic steroid with either plate-haptic silicone lenses or three-piece silicone lenses than with PMMA. Others have suggested the use of acrylic lenses. Angle-supported lenses have been associated with anterior and intermediate uveitedes.

When his disease was quiescent, an 11-year-old boy with anterior uveitis underwent cataract extraction and IOL placement. A, After multiple YAG capsulotomies and EDTA treatments, a persistent anterior chamber reaction required topical steroid therapy. B, After IOL removal and normalization of intraocular pressure, inflammation was absent, steroid therapy had been discontinued, and visual acuity was 20/100.

(Courtesy Rubens Belfort Jr, MD, Sao Paulo, Brazil.)

With retroillumination, cells can be seen to accumulate on the anterior surface of IOLs in uveitic eyes. Elongated cell bodies of smaller cells suggest macrophages. Larger cells with more substantial cytoplasm also appear.

The eyes of patients with juvenile idiopathic arthritis (JIA) deserve separate comment. These eyes have not been considered good candidates for IOLs. The reason for this is the potential of the lens to act as a scaffold for the development of membranes in eyes with low-grade chronic disease that is often difficult to control. We have generally not placed an IOL in children with uveitis. Lundvall and Zetterstrom reported their experience with 10 eyes in seven children. With a follow-up of 1–5 years, they reported that most eyes had visual acuities between 20/20 and 20/50. Glaucoma developed in three eyes. In a 5-year follow-up of 17 children (20 eyes) with chronic uveitis undergoing cataract extraction, an almost equal number of patients with JRA and without underwent IOL placement or pars plana removal of the lens. Of interest in this study reported by BenEzra and Cohen was the finding that 5 years after surgery, only two of nine eyes in the JRA group retained a visual acuity of 6/9 or better. Most of the other eyes had a visual acuity of 6/240. However, in the group without JRA, 11 eyes (54.5%) retained a visual acuity of 6/12 or better. A child’s inflammatory response can be exuberant, causing serious difficulties postoperatively. However, recent studies suggest that this approach is being questioned. Kotaniemi et al. reported that 64% of 36 JIA eyes receiving an implant had visual acuities of 20/40 or better. A list of recent results reported where intraocular lenses are placed is seen in Table 8-1 . Woreta et al. reported on the risk factors for ocular complications in 75 JIA patients undergoing surgery, and found that 67% had ocular complications. Eyes with more than 1+ flare, a history of previous intraocular surgery, and posterior synechiae all had a higher risk of poorer vision postoperatively. It would seem that the degree of activity, control of disease and secondary anatomic alterations due to the uveitis may be important considerations.

Results from the literature support the notion that IOL placement in uveitic eyes can result in a positive outcome. , Pleyer and colleagues reported in a 5-year series that 71% of 28 eyes receiving an IOL had a postoperative visual acuity of 0.5 or better, compared to 54% of the 24 eyes that were aphakic. Because this was not a randomized study it is difficult to compare the relative ‘success’ rates. Foster and colleagues reported that in 20 uveitic eyes extracapsular cataract extraction with posterior chamber IOL and pars plana vitrectomy, if needed, produced 20/100 or better visual acuity in 60% of the eyes operated on. Tessler and Farber, comparing IOL placement with no IOL placement in uveitic eyes, believed that IOLs appeared to be relatively safe in patients with what was described as chronic iridocyclitis. However, a comment and our interpretation of the paper would indeed suggest that the data would argue for the reverse – that is, eyes with IOLs had a poorer postoperative result. As Tessler and Farber stated, only a larger study could definitively settle this question. In a study of 89 patients (106 eyes) Hazari and collegues reported that eyes receiving an IOL had a better visual outcome than those not receiving an IOL. It should be noted that eyes that did not have an improvement in visual acuity had persistent uveitis, a problem that the practitioner must be continuously aware of. Kim and colleagues reported a recurrence of Behçet’s disease in a patient that necessitated IOL exchange and vitrectomy, with a resultant poor visual outcome. It appears that eyes with Fuchs’ heterochromic iridocyclitis do well after cataract surgery and IOL placement.

Akova and Foster, in a retrospective analysis of their patients with sarcoidosis, found that of 102 patients, 14 (21 eyes) had surgery because of visual acuity of 20/100 or less. In stressing the need for absolute control of the intraocular inflammatory disease, they found that 19 of the 21 patients received IOLs, and that 61% of the eyes had a stable visual acuity of 20/40 or better. In eyes that did not have a dramatic improvement in visual acuity, retinal pathologic conditions such as cystoid edema and epiretinal membranes, as well as glaucomatous optic nerve disease, were found.

As mentioned above, JRA is a disorder in which IOL placement has not been recommended. In a review of 16 eyes in 10 patients with JRA, Fox and colleagues found that after pars plana lensectomy and vitrectomy the short-term visual acuity was 20/70 or better in 13 eyes (81%), but a longer follow-up showed that the final visual acuity was 20/70 or better in nine of them (56%). This drop was attributed to glaucoma and retinal disease such as cystoid edema.

There are other challenges that the uveitic eye can present to the surgeon. One is operating on an eye with long-standing exudative detachments. Secchi reported good results in these eyes with total lens removal combined with anterior vitrectomy and placement of a scleral fixation lens.

Complications from the cataract surgery or sequelae from the ongoing inflammation must be expected, particularly when an IOL has been placed. ^, Posterior capsule opacification is seen frequently. When Dana and colleagues compared rates of posterior capsule opacification between 78 patients with uveitis and 106 patients without uveitis undergoing cataract surgery, the rates were comparable when an adjustment was made for the younger age of the patients with uveitis. In a report by Estafanous and colleagues of 39 uveitic eyes needing surgery, posterior capsular opacification was seen in 24 eyes (62%), with 12 of these (31%) needing Nd:YAG capsulotomy. Of these patients 41% had a recurrence of their disease. If cystoid macular edema was present before surgery, it will most probably be there after the surgery and will need to be addressed. Epiretinal membrane formations can be seen. Lenses do need to be explanted, even after the best preoperative consideration by the surgeon. In a report by Foster and colleagues of 19 patients whose lenses were explanted, a perilental membrane was the reason in eight, low-grade inflammation in eight, a cyclitic membrane leading to hypotony in one, retinal detachment in one, and vitreous hemorrhage in one.

Pars plana lensectomy, usually combined with vitrectomy, has been used by several authors to remove a cataract, reshape the pupil, and remove vitreous opacities and vitreoretinal traction in one operation. The operation can be performed using the standard technique: three 1-mm incisions 3 or 3.5 mm posterior to the limbus. Trauma to the trabecular meshwork is avoided. Usually the use of disposable iris retractors will provide the surgeon with an adequate view initially, until the lens has been removed. However, if this is not the case, the pupillary aperture can be enlarged with the vitrectomy instrument to be round and large enough to examine the retina after surgery. However, this technique can provoke bleeding from the edge of the iris. In addition, more conjunctiva can be left undisturbed for use in a future filtering operation. The main disadvantage is the removal of the posterior capsular barrier. Some surgeons have advocated keeping the anterior capsule intact to use as a possible support if a secondary implant is considered. We have had no experience with this approach. The possibility of retinal detachment shortly after surgery is less than 1% in our experience. This operation is difficult to perform in patients over 50 years of age because, in our experience, the lens nucleus becomes rather hard, making it more difficult to remove. It has been suggested by several authors that pars plana vitrectomy reduces or eliminates recurrences of inflammation because it removes an antigen load from the vitreous. This has not been our experience. However, removal of the vitreous can reduce the accumulation of cells and haze during an episode of inflammation, but the vision will still be affected by a new inflammatory episode. We therefore prefer to perform this operation in patients younger than 50 who have had vitreitis as part of their ocular inflammation, especially if the vitreous and retina cannot be visualized preoperatively because the pupil is small or there are other structural alterations in the anterior chamber.

Our technique for pars plana lensectomy/vitrectomy is to make three radial conjunctival incisions. These are cleaned down to bare sclera, and scleral incisions are made 3 mm posterior to the limbus. An infusing cannula is inserted in one incision but is not turned on. An ultrasonic fragmenting instrument and an irrigating needle are introduced into the lens, and the lens material is removed within the capsular bag, using both the fragmenting instrument and the vitrectomy instrument. An alternative is to place the light pipe under the lens to help visualization. Entry into the eye must be parallel to the iris or the posterior capsule will be broken prematurely. After the lens material is removed, the vitreous cutter is used to remove the lens capsule and anterior vitreous and to reshape the pupil. A vitrectomy is then performed to remove opacities and any tractional abnormalities.

Results for cataract surgery have been reported to be favorable with both an anterior chamber approach and entry through the pars plana. Results depend more on previous damage from the inflammation than from surgical complications. Praeger and colleagues reported excellent improvement after extracapsular surgery in patients with JRA. Mackensen and Loeffler reported that in 86 patients with chronic iridocyclitis who underwent extracapsular surgery, inflammation was reduced by 70% and recurrences by 94%. Vision improved in most patients. In a study by Smith and colleagues 12 patients with chronic uveitis who underwent pars plana lensectomy/vitrectomy had improved vision, but cystoid macular edema impaired the acuity in most of them. Girard and colleagues reported that 5-year remissions of uveitis occurred in all of their 23 patients, and 21 had improved vision after pars plana surgery. In the 39 patients Petrilli and coworkers studied after pars plana surgery, 23% had vision better than 20/40, and 56% had vision worse than 20/100; 18% had cystoid macular edema. Other authors have had similar experiences. ^, In our experience the average vision after surgery in patients with long-standing chronic inflammation was 20/100 ( Fig. 8-6 ). Phthisis occurred in approximately 2% of the eyes, and 90% of the patients had improved vision.

Change in visual acuity after cataract removal in patients with uveitis.

Hypotony is often mentioned as a contraindication for cataract removal in these patients. However, if the cataract is dense and vision limited to hand movements or worse, surgery by the pars plana route can be considered in eyes with intraocular pressures <3 mmHg. It can be useful in treating cases of hypotony where bands are creating traction on the ciliary body and resulting in reduced aqueous production. We have seen that removal of these tractional forces will cause an increase in intraocular pressure – not enormous, but enough to achieve a more reassuring pressure. Yu et al. reported similar results in patients with JIA.

Although phthisis may result, we have observed five eyes with preoperative hypotony in which there was an increase in intraocular pressure of several mmHg after pars plana lensectomy/vitrectomy. In many eyes with posterior inflammation there is a dense anterior hyaloid that may be associated with ciliary body traction. Removal of this traction may result in improved ciliary body function.

Glaucoma surgery

Secondary glaucoma that cannot be controlled by medical therapy is sometimes a consequence of severe anterior segment inflammation. It is important to differentiate between pupillary block glaucoma and secondary angle closure by peripheral anterior synechiae. The two processes may coexist in the eye with uveitis. Active inflammation may suppress ciliary body function and also reduce trabecular filtration. Thus intraocular pressure in the actively inflamed eye may be higher or lower than in the uninflamed eye. Although patients with long-standing chronic inflammation have poor outflow, increased inflammation will lead to reduced aqueous production and a pressure decrease. If treatment with corticosteroids is accompanied by an increase in pressure, it does not necessarily mean that the patient is a ‘steroid responder.’

It may be difficult to distinguish pupillary block glaucoma in a patient with chronic inflammation and posterior synechiae that involve the entire lens surface. These patients do not show a classic iris bombé pattern but may have a subtle shallowing of the anterior chamber with a low, peripheral iris bombé. The iris elevation may be localized only to areas where the iris and the lens are not adherent ( Fig. 8-7 ). If the diagnosis is suspected but unclear, a laser iridectomy will confirm it and also be therapeutic. If the laser iridectomy fails to lower the intraocular pressure, pupillary block can be ruled out as a mechanism.

Only gold members can continue reading. Log In or Register to continue

Share this:

• Click to share on Twitter (Opens in new window)
• Click to share on Facebook (Opens in new window)

Related

Related posts:

1. Bacterial and Fungal Diseases
2. Intermediate Uveitis
3. Vogt–Koyanagi–Harada Syndrome
4. Retinal Vasculitis

Stay updated, free articles. Join our Telegram channel

Join