57

Cyclodestruction

Leonard K. Seibold, MD; Joel S. Schuman, MD, FACS; and Mina Pantcheva, MD

In general, cyclodestructive surgery is a procedure of last resort. It is used to lower intraocular pressure (IOP) by decreasing aqueous production. Glaucoma occasionally is refractory to conventional medical and surgical treatment and requires destruction of the ciliary body to bring the IOP to a level adequate for the preservation of optic nerve tissue and vision. Filtration surgery is usually preferable to cyclodestructive procedures, as filtration surgery attacks the root problem in most cases of glaucoma, inadequate aqueous outflow, while cyclodestructive procedures tend to have more complications than filtering surgery and do not treat the source of the problem. Several techniques have been used to destroy the ciliary processes with a gradual evolution toward less destructive and more targeted approaches.

HISTORICAL PERSPECTIVE

The first reports of destruction of the ciliary body in order to lower IOP were by Weve¹ in 1933, using diathermy. Penetrating diathermy was used by Vogt² in 1936 and was reported again by Stocker in 1945,³ by Lachman and Rochwell⁴ in 1953, and by Nesterov and Egorov⁵ in 1983. The principle was the ablation of the ciliary body by the heat caused by the diathermy procedure.⁶ A report by Walton and Grant⁷ in 1970 revealed low success rates for this procedure, as well as significant hypotony, which led to the gradual abandonment of cyclodiathermy. Nonpenetrating diathermy has been investigated by several investigators; however, both penetrating and nonpenetrating diathermy appear to cause significant scleral damage in addition to the effect on the ciliary body.^4,8–14 Haik and colleagues¹⁵ used beta irradiation to induce ciliary ablation, but this was cataractogenic. Cyclo-electrolysis was described in 1949 by Berens and associates,¹⁶ but did not gain favor, as the technique did not represent a significant advantage over cyclodiathermy.⁶

Freezing to destroy the ciliary body was suggested in 1950 by Bietti,¹⁷ and, in 1964, Polack and de Roetth¹⁸ and McLean and Lincoff¹⁹ reported on the use of this technique in rabbits and in humans. This procedure was called cyclocryotherapy (CCT), and de Roetth^20,21 later showed long-term effectiveness of CCT in lowering IOP. CCT was adopted quickly, as it was felt to be less destructive and more predictable than penetrating cyclodiathermy.²² Bellows and Grant^23,24 reported that CCT was especially useful in advanced, inadequately controlled glaucoma, particularly in aphakic eyes. The procedure had limited success in treating neovascular glaucoma; however, this glaucoma subtype was notoriously difficult to manage with any means, and reports of IOP control using CCT were variable.^25,26 A major complication noted with CCT was phthisis, rates of which varied from 0% to 12% or more. Vision loss was also a significant complication, with rates up to 67% of treated eyes.^23–29 The use of xenon arc panretinal photocoagulation in combination with CCT resulted in good IOP control and no phthisis after 2.5 years, according to Nissen and colleagues³⁰; however, there was no control group in this study.

Ultrasound has also been used to cause focal ciliary body destruction since its description by Purnell and associates³¹ in 1964 for that purpose. Perhaps the greatest proponent of this technique has been Coleman, who has reported significant pressure lowering with ultrasound, proposing a mechanism of ciliary body destruction, thinning of scleral collagen, and separation of the ciliary body from the sclera.32–35 Several reports have supported these findings, both clinically³⁶ and in animal models.^37–39

Finally, Freyler and Scheimbauer⁴⁰ noted that partial cyclectomy could control end-stage glaucoma in some cases, although the phthisis rate was 10% to 15%.

CYCLOPHOTOCOAGULATION

The first form of cyclophotocoagulation (CPC) was performed not with a laser but with the xenon arc photocoagulator. Transscleral xenon arc CPC was reported by Weekers and colleagues⁴¹ in 1961. Unfortunately, this technique offered no clear advantage over other treatment modalities, and it was not until 1971 that the use of the laser for CPC was introduced. Lee and Pomerantzeff⁴² suggested transpupillary CPC, but Shields²² showed that this procedure produced only limited success, due to the small amount of ciliary process that could be visualized through the pupil. Shields^22,43 also found that the argon laser could be used for endoscopic cyclophotocoagulation (ECP) with good results. Uram^44–46 has shown excellent results for ECP using a diode laser coupled to a fiber-optic endoscope.

A less invasive means of CPC is via the transscleral route, and in 1972, Beckman and colleagues⁴⁷ described transscleral CPC using a ruby laser. In a 10-year follow-up of 241 patients treated with transscleral CPC with the ruby laser, 62% were found to have an IOP of 5 to 22 mm Hg.^48,49 The greatest success was in aphakic eyes, and the highest phthisis rates were in neovascular glaucoma patients (10% compared to 7% overall and a 17% incidence of hypotony overall).

Despite good results with CPC, the cyclodestructive treatment of choice was CCT for many years. This was due to the limited availability of the ruby laser. However in recent years, the excellent results shown for neodymium:yttrium-aluminum-garnet (Nd:YAG) and diode laser CPC have revolutionized this therapeutic modality. The Nd:YAG and diode lasers have several advantages over the ruby and neodymium lasers. Longer wavelengths provide good scleral penetration with less backscatter than shorter wavelengths.^50–52 Both the Nd:YAG and diode lasers have excellent absorption by melanin within the ciliary body. Nd:YAG and diode lasers are present in many medical centers, and contact Nd:YAG lasers may be found in many general hospitals, as they have several uses outside of ophthalmology. Diode lasers can be used as photocoagulators for laser trabeculoplasty,^53–55 laser iridectomy,^56–58 retinal photocoagulation, endophotocoagulation, and transscleral retinopexy.^59–68 Additionally, diode lasers are compact, portable, lightweight, air or electrically cooled, and use standard current. Finally, the Nd:YAG and diode lasers offer an inexpensive means by which to perform CPC compared to the ruby laser.

INDICATIONS AND CASE SELECTION FOR TRANSSCLERAL LASER CYCLOPHOTOCOAGULATION

Laser CPC is a destructive procedure. It is typically a therapeutic modality of last resort, useful in advanced glaucoma, in which the IOP is uncontrolled despite maximum tolerated medical treatment, in eyes that have failed filtering surgery or are likely to fail future filtering surgery. This includes eyes in which filtering surgery has a high risk, such as aphakic glaucoma, neovascular glaucoma, glaucoma after penetrating keratoplasty, or in eyes with very low visual potential. CPC destroys tissue, lowering IOP by the destruction of the ciliary body, decreasing the amount of aqueous humor produced. Hypotony can result from overly aggressive treatment, because inadequate functional ciliary body remains. Conversely, an insufficient treatment will not produce adequate reduction of IOP. The therapeutic window is relatively narrow in cyclodestructive procedures, especially in light of the fact that aqueous outflow tends to be compromised in eyes undergoing this type of surgery.

Other possible risks of cyclodestructive surgery include pain, inflammation, hypotony, macular edema, vitreous hemorrhage, and phthisis. While CPC may produce fewer side effects than other cyclodestructive procedures, these risks persist and tend to make CPC a treatment of last resort in the management of advanced glaucoma.

A patient to undergo CPC should have IOP that is uncontrolled despite maximum tolerated medical treatment and one of the following:

• Failed prior filtration surgery or the expectation that further glaucoma filtering surgery will fail
  
• Glaucoma that is likely to fail filtering surgery (neovascular, inflammatory, post-penetrating keratoplasty, post-scleral buckling) or is a high risk for complications of filtering surgery (vitrectomized, aphakic eye)
  
• Poor visual acuity, such that the eye is being treated with CPC to maintain comfort and prevent visual or globe loss
  
• The patient is not a surgical candidate for filtering surgery for general medical reasons

TRANSSCLERAL LASER CYCLOPHOTOCOAGULATION

Transscleral laser CPC may be performed either with a fiber-optic probe touching the eye (contact) or through the air (noncontact).

Contact transscleral Nd:YAG laser cyclophotocoagulation (CYC) was first reported by Brancato and colleagues69 in 1987. They studied the effects of CYC on chinchilla rabbits. Federman and colleagues,⁷⁰ that same year, as well as Peyman and colleagues⁷¹ in 1987 and Schubert and Federman^72,73 in 1989, in addition to Latina and coauthors in 1989,⁷⁴ each studied CYC lesions in rabbit ciliary bodies. The lesions appeared on postoperative day 1 to be 1-mm white well-demarcated spots.^69,74 Thermal injury occurred in the ciliary body stroma, with coagulation necrosis of the ciliary pigmented and nonpigmented epithelia; there was no significant scleral damage.^69,74 Four-week follow-up demonstrated atrophy, fusion, and fibrosis of the ciliary processes.^73,74 In some cases, the ciliary process was covered with a fibrous membrane, with abnormal ciliary epithelium, pigment epithelial proliferation, and focal epithelial interruptions.^69,73,74 Areas of scleral compression and hyper-cellularity occurred, but there was no scleral thermal damage present.⁷³ Epithelial regeneration following CYC did not occur, according to van der Zypen and colleagues,^75,76 indicating that the vascular network remained atrophic in the ciliary body. Similar lesions have been noted using the diode laser in the contact mode in rabbit eyes.^77,78

Pulsed CYC results in lesions similar to noncontact transscleral Nd:YAG laser cyclophotocoagulation (NCYC).⁷⁹ The coagulative lesions seen with CYC do not occur in this setting; rather, blistering lesions are noted.

CYC performed in human cadaver eyes causes coagulation to occur in the ciliary body, with disruption of the ciliary pigmented and nonpigmented epithelium.⁸⁰ Allingham and colleagues⁸⁰ showed that optimal probe placement is with the anterior edge of the contact probe 0.5 to 1 mm posterior to the limbus. More posterior placement results in lesions in the pars plana, while more anterior placement causes iris burns. The optimal power setting was found to be 7 to 9 watts, with 5 watts creating only minimal coagulation necrosis of the ciliary body and 11 watts causing a striking loss of anatomic integrity of the ciliary process. No scleral damage was noted at this power setting. Exposure time was optimized at 0.7 seconds.⁸⁰ In vivo studies by Brancato and colleagues⁸¹ showed in eyes destined for enucleation due to choroidal melanoma that only 2 J of energy were needed to create lesions in the ciliary body using placement similar to that described above; however, in another paper, Brancato and colleagues⁸² showed that higher energy levels produce better clinical results.

Contact diode laser cyclophotocoagulation (CDC) in human cadaver eyes induced similar epithelial coagulation necrosis and thermal coagulation of the ciliary stroma and vasculature. Energy levels were 3 to 5 J.⁸³

Clinical Studies

In 116 eyes of 114 patients followed for a minimum of 1 year after treatment with CYC, Schuman and colleagues⁸⁴ in 1992 found that 75% of eyes achieved an average final IOP of 25 mm Hg or less, 66% were 22 mm Hg or less, and more than 50% of the eyes treated achieved a final IOP of 19 mm Hg or less (Table 57-1). The major complications noted were hypotony and visual loss. Nine eyes (8%) had a final IOP less than 3 mm Hg, with an IOP of 0 mm Hg in 6 eyes. Seventeen of 36 eyes (47%) with an initial visual acuity of 20/200 or better lost 2 or more Snellen lines. Nineteen eyes, all with an initial visual acuity of counting fingers or worse, progressed to no light perception. Retreatment was required in 31 eyes (27%).⁸⁴ The findings were similar to that group’s earlier report⁸⁵ in 1990, studying 160 treatments in 140 eyes of 136 patients. The interesting finding in comparing these 2 studies was that the success rates remained similar, but the rate of complications increased over time.^84,85

Brancato and colleagues⁸² reported on CYC in 23 patients, using less energy and fewer spots. The success rate in this case was somewhat lower than that reported by Schuman et al,⁸⁴ and retreatment was required in 57% of eyes.

The contact diode laser has been used clinically for CPC and was approved for this procedure by the US Food and Drug Administration in 1994. Gaasterland and colleagues⁸⁶ performed a multicenter clinical study of 30 eyes in 30 patients and noted results nearly identical to those encountered with CYC. The treatment parameters for CDC, however, called for less energy and fewer spots than CYC, and the fact that similar clinical results could be achieved using these parameters suggested that CDC resulted in a larger area of tissue destruction per spot. Further, Carassa and colleagues,87 using CDC, noted a 50% IOP lowering in 12 eyes of 12 patients, with 16 spots over 360 degrees, using 2.5 watts for 1.5 s (3.75 J).

Noncontact Cyclophotocoagulation

Laboratory Studies

NCYC preceded CYC historically; however, it has proven to be a modality resulting in more inflammation and pain postoperatively than CYC. Therefore, CYC is generally preferred to NCYC in clinical practice. Wilensky and colleagues⁸⁸ first reported NCYC in rabbits in 1985. Reports^52,89–94 on histopathology of the laser lesions and use of the technique in patients followed shortly after.

Fankhauser and colleagues⁹⁵ demonstrated NCYC lesions in the ciliary body of human cadaver eyes using 6 to 7 J aiming the spot 0.5 to 1 mm posterior to the limbus with maximum defocusing and a 20-msec pulse. Hampton and Shields⁹⁶ showed that optimum parameters for treatment were an energy level of 8 J delivered 1.5 mm posterior to the limbus with maximum defocusing. They showed histologic damage to the iris when the beam was aimed more anteriorly and pars plana lesions when the beam was aimed more posteriorly. No differences were seen when the energy was delivered tangentially vs perpendicular to the sclera.

Later studies^97,98 noted that a specialized contact lens could increase the efficiency and accuracy of laser delivery. Hampton and Shields⁹⁶ found that tissue damage with NCYC differed from that seen with CYC: the destruction of the ciliary epithelium was accompanied by the creation of a blister-like space, and coagulation necrosis was not present. Similar findings have been seen in vivo in blind human eyes treated with NCYC just prior to enucleation for pain. In these cases, the blister-like space was filled with eosinophilic material.⁹⁹

NCYC in rabbits resulted in the acute destruction of the ciliary epithelium and the associated vessels, with subsequent atrophy of the ciliary processes 1 to 2 months after injury.^52,89,90 The laser energy was absorbed by melanin; no histopathologic changes were seen in albino rabbits.¹⁰⁰

Unlike the effects seen with NCYC, noncontact transscleral diode laser cyclophotocoagulation (NCDC) resulted in a thermal ciliary body lesion.¹⁰¹ Hennis and colleagues¹⁰² demonstrated thermal effects in the ciliary body with as little as 900 mJ energy with NCDC, using a spot size of 100 to 500 μm, with the beam 0.5 mm posterior to the limbus, defocused 1 mm posteriorly. In a study comparing CDC and NCDC in living rabbits using the same laser for both treatments, Shepps and colleagues¹⁰¹ reported that 50% more energy was needed to create the same lesion with NCDC as compared to CDC. This is consistent with the findings on scleral light scattering presented by Vogel and colleagues.⁵¹ In comparing NCDC to NCYC and CCT in cadaver eyes, Assia and colleagues¹⁰³ found that all were effective in ciliary body destruction, and none damaged the crystalline lens.

Clinical Studies

Cyrlin and colleagues⁹³ reported in 1985 that 60% to 70% of patients treated with NCYC had an IOP between 5 and 22 mm Hg at 6 to 12 months follow-up, using 8 J of energy, with 32 spots. In 1986, Schwartz and Moster¹⁰⁴ found that 20 of 29 treated eyes (69%) had final IOPs of 22 mm Hg or less and noted 1 questionable case of phthisis, with an average 32-week follow-up. Devenyi and colleagues⁹⁴ used 40 spots at 1.8 to 3.0 J for 20 msec placed 2 to 3 mm posterior to the limbus with maximum defocusing and found that 11 of 24 eyes developed an IOP of 21 mm Hg or less and that neovascular glaucoma eyes did somewhat worse than others. While this study,⁹⁴ with 8.8 months’ average follow-up, had no cases of phthisis, in a subsequent study on the same patient population, Trope and Ma¹⁰⁵ reported a phthisis rate of 10.7% and a 30% rate of visual loss, when the patients were followed for an average of 21.9 months.

Klapper and colleagues¹⁰⁶ reported that 86% of treated eyes achieved an IOP between 5 and 22 mm Hg, with a mean decrease in IOP of 68%, with an average follow-up of 6 months. This study used 32 spots at 3.5 to 4.5 J for 20 msec placed 2 to 3 mm posterior to the limbus with maximum defocusing.

Hampton and colleagues¹⁰⁷ reported on 100 NCYC treatments and noted severe pain in 13.5%, severe inflammation in 28%, a retreatment rate of 25%, and visual loss in 50%. Wright and colleagues¹⁰⁸ noted that a majority of 35 eyes treated required further intervention, or lost vision if followed long enough, and that 31% of eyes lost 2 or more lines of vision or light perception over the 3-year follow-up period of the study.

Hennis and Stewart¹⁰⁹ found a 30% reduction in IOP over a 6-month follow-up with a single NCDC treatment in 14 eyes, using 1.2 watts for 0.99 s (1.2 J) with 40 to 45 applications of a 100-μm spot placed 1 mm posterior to the limbus and defocused 1 mm posteriorly. Similar to NCYC, conjunctival burns occurred with NCDC, as well as anterior chamber inflammation.

Several complications have been described after NCYC. Maus and Katz¹¹⁰ found severe hypotony, flat anterior chamber, and serous choroidal detachment after NCYC in 3 patients, all of whom had previous filtering surgery; 2 were Black patients. Fiore and colleagues¹¹¹ reported focal scleral thinning following NCYC in one patient. Hardten and Brown¹¹² reported malignant glaucoma following NCYC in one eye, as did Wand, Schuman, and Puliafito.¹¹³ Blomquist and colleagues¹¹⁴ showed that NCYC could damage intraocular lens haptics in cadaver eyes at energy levels higher than those used clinically; Lim and colleagues¹¹⁵ failed to damage intraocular lens haptics with CYC. Subretinal fibrosis and lens subluxation, while reported with CCT,116,117 have not been found with Nd:YAG or CDC.

Sympathetic ophthalmia is a controversial finding in association with CPC. Reports of this complication have been in eyes having previous trauma or other ocular procedures,^118,119 and the relationship between CPC and sympathetic ophthalmia is uncertain.¹²⁰ Sympathetic ophthalmia has been described following CCT,¹²¹ as well as after helium ion irradiation followed by CCT.¹²²

Cyclophotocoagulation Versus Other Cyclodestructive Procedures

As a treatment of last resort, CCT was an effective means to control IOP in eyes refractory to all other forms of treatment. Bellows and Grant showed that the IOP could be lowered in advanced, uncontrolled glaucoma, especially in aphakia, with CCT.^23,24 Nevertheless, phthisis occurred in up to 12% of patients and visual loss in two-thirds of CCT treated eyes; this relegated CCT to the role of a treatment of last resort.^23–27 Transscleral laser CPC destroys less tissue than CCT, with similar effects on IOP in rabbits, as demonstrated by Higginbotham et al.¹²³ NCYC may be as effective as CCT clinically, but produces fewer complications, as shown by Suzuki and colleagues.¹²⁴ Further, NCYC was demonstrated in an uncontrolled retrospective study,¹²⁵ to be superior to drainage implant surgery. Noureddin and colleagues¹²⁵ found that while both procedures lowered IOP, serious complications such as retinal detachment, expulsive hemorrhage, phthisis, and endophthalmitis were much more common in the tube-treated group. In addition, the incidence of visual loss among tube-treated eyes was higher than in eyes treated with NCYC. Both groups had similar postoperative IOPs reported.¹²⁵

Contact Versus Noncontact Cyclophotocoagulation

There are several differences between CYC and NCYC. CYC is performed with the patient supine, while NCYC is done with the patient at the slit lamp. The contact laser is portable and can be used in remote locations, such as the operating room with the patient under general anesthesia, with relative ease. NCYC uses more energy, and significantly more power, than CYC and results in more inflammation and pain postoperatively than CYC.^{52,82,85,94,104,106,107}

CYC, NCYC, CDC, and NCDC all center the laser beam approximately 1.5 mm posterior to the limbus.^52,80,83,96 With the laser delivered via fiber optic, as in CYC and CDC, the anterior edge of the probe is offset from the limbus to achieve beam placement. With the noncontact techniques, the laser is aimed approximately 1.5 mm posterior to the limbus. Schuman and colleauges^84,85,126 did not find any clinical differences in CYC treatments performed with the beam centered 1.5, 2.0, or 2.5 mm posterior to the limbus. Schubert,¹²⁷ however, showed that NCYC and CYC lesions performed with the beam centered 3.0 mm posterior to the limbus caused lesions in the pars plana. He demonstrated that these lesions resulted in an increase in outflow facility,¹²⁸ and this report was supported by Pham-Duy, who also showed that CCT increased outflow facility while transiently reducing aqueous flow.¹²⁹ These studies, however, were limited by blood-aqueous barrier disruption following CCT. Higginbotham and colleagues¹³⁰ demonstrated ciliary epithelial destruction in proportion to the degree of CCT, with consequent proportional effects on IOP and aqueous flow; however, outflow facility was not studied. CYC has been shown to increase uveoscleral outflow in rabbits, possibly by an effect mediated by prostaglandins.¹³¹ In work by Schmidt and colleagues,¹³¹ when IOP returned to normal, uveoscleral flow returned to baseline as well.

Contact CPC appears to have several advantages over the noncontact modality. A longer exposure time is used for CYC as compared to NCYC. The histopathological change in CYC is coagulative necrosis, whereas explosive damage is seen with NCYC.^73,80,96,97 With regard to the diode laser, both modalities use long exposure times; however, 50% more energy is required with NCDC to achieve the same lesion seen with CDC.¹⁰¹ In comparing the diode and Nd:YAG lasers, the diode laser has the advantage of greater absorption by melanin and nearly equivalent scleral transmission when the sclera is compressed as it is for contact CPC.⁵¹ The diode laser offers greater convenience and economy compared to the Nd:YAG laser. Anecdotally, there is less pain and inflammation with CYC as opposed to NCYC, which is most likely related to the histopathological changes noted above. The CYC unit is portable, is present in many general hospitals, and is therefore more available than the NCYC unit. The diode unit, with its multitude of uses in ophthalmology, affords even greater flexibility to the ophthalmologist, as well as potentially greater availability. Any of the above CPC techniques offer a titratable treatment that is generally well tolerated by the appropriately selected patient.

Glaucoma Subtype and the Pigment Effect

Glaucoma subtype may be important in the response to CPC. Neovascular glaucoma, difficult to treat with any modality, responds least well of the glaucoma subtypes to cyclodestruction,^23,26 and eyes with neovascular glaucoma have the most severe and greatest number of complications (inflammation, pain, and visual loss). In comparison to other glaucoma subtypes, neovascular glaucoma eyes have less IOP response to CPC as well, although the difference in IOP response does not achieve statistical significance.^84,85

While aphakic eyes respond better to CCT than phakic eyes, this is not seen with CYC.^24,84 Aphakic eyes do have less inflammation than others. CPC seems to be an effective means of controlling IOP in eyes following penetrating keratoplasty.85,123–134 Intraocular pigment is a significant factor in the response to cyclodestruction and in subsequent complications. This has been seen with CCT, as noted by de Roetth,²¹ and Cantor et al¹⁰⁰ noted this in studying NCYC in pigmented and albino rabbits. Schubert and Federman⁷³ reported this phenomenon in relation to CPC, as well. Coleman and colleagues¹³⁵ found the pigment effect in cadaver eyes and in monkeys, and Schuman and colleagues⁸⁵ described this finding in relation to CYC. Eyes with more pigment may require less energy for the same level of cyclodestruction than less heavily pigmented eyes. The pigment acts as a chromophore, absorbing the light energy and converting it to heat, with greater tissue damage at the same energy level in eyes with more pigment.

Stay updated, free articles. Join our Telegram channel

Join