Case 7. The frontal view.
Case 1. Slight pupillary transformation.
Hara T, Hara T. Ten-Year Results of Anterior Chamber Fixation of the Posterior ChamberIntraocular Lens. Arch Ophthalmol. 2004;122(8):1112-1116. doi:10.1001/archopht.122.8.1112
To report the outcomes in eyes in which a Sinskey-type conventionalposterior chamber intraocular lens (PC-IOL) was fixed in the anterior chamberand followed up for an average of 10 years.
Twenty-one eyes were included in the study. The IOLs were conventionalhard PC-IOLs, 13.0 mm in overall length with 6.0-mm polymethylmethacrylateoptic and polyvinylidene fluoride loops that tilted 10° anteriorly.
At implantation the corneal endothelial cell density in 13 of 21 eyeswas 382 to 1580 cells/mm2 owing to the previous implantation ofiris-clip or iridocapsular IOLs. The overall postoperative corneal endothelialcell loss was 30.9%. Six of 14 eyes with a preoperative corneal endothelialcell density less than 2000 cells/mm2 developed bullous keratopathy.Although serious complications did not occur in 7 eyes with a cell densityexceeding 2000 cells/mm2, an average corneal endothelial cell lossof 26.5% still occurred. Except for 2 eyes, all loops were fixed at the scleralspur and a slight pupillary transformation occurred in only 1 eye.
Although this procedure is easy to perform and pupillary transformationdoes not occur, it is not recommended for aphakic eyes owing to the high cornealendothelial cell loss after an average 10-year follow-up.
It is widely accepted that ciliary sulcus fixation is the method ofchoice for secondary intraocular lens (IOL) fixation in aphakic eyes withsufficient capsular support. In aphakic eyes with insufficient capsular support,the alternative methods are posterior chamber (PC) fixation of a PC-IOL witha scleral or iris suture, or anterior chamber (AC) fixation of an AC-IOL.It seems to be generally accepted that PC-IOL implantation with a scleralsuture provides stable results. However, complications such as IOL tilt, decentration,hyphema, and others were reported.1 After thedislocation of 7 IOLs secured with scleral sutures 7 years after surgery,Kim et al2 reported that patients should bewarned that their transscleral sutured IOL might subluxate 6 to 7 years postoperatively.In addition, the surgical technique is complicated and time consuming. Also,good results of iris sutured PC-IOLs were reported3 inthe eye without capsular support. However, to pursue the simpler method, wedid not try it this time. Thus, we tried a simpler and safer method.
There are 3 types of AC-IOL fixation: (1) AC angle fixation of an AC-IOLwith rigid or semirigid loops, (2) AC angle fixation of an AC-IOL with flexibleopen loops, and (3) iris fixation of an iris claw lens. Although good resultshave been reported,4 an IOL with rigid or semirigidloops requires careful selection of the lens size, and complications includingpostoperative inflammation, pupillary transformation, glaucoma, and loss ofcorneal endothelial cells were reported.5 In1989, we had a case in which a normally fixated PC-IOL dislocated into thevitreous cavity. The IOL was pulled from the vitreous and placed in the ACto await subsequent PC fixation with scleral suturing. However, the PC-IOLplaced in the AC was stable. After following up the patient's condition forseveral months, we decided to pursue the current method of AC fixation ofthe conventional PC-IOL. If the results of this method are stable, it maybecome an acceptable procedure because the IOL is one-size-fits-all and thesurgical method is easy.
In 2003, Drolsum6 reported good resultswith implantation of a flexible open-loop AC-IOL after 7 years of follow-up.Budo et al7 also reported good results withan iris-fixated AC-IOL. However, when the present study began, these 2 methodshad not yet been definitely accepted. In 1992, we reported the results fromthe first 10 eyes (9 patients) that had undergone AC implantation of a PC-IOL(average follow-up, 17 months).8 Because clinicallyrelevant complications were not observed and the mean (SD) corneal endothelialcell loss was 4.9% (2.4%), the results looked promising.
We were not the first to attempt this procedure. Girard et al9 and Nashiro et al10 performeda similar procedure and reported good postoperative outcomes. However, Girardet al reported results after only 11 months of follow-up, and Nashiro et alafter 4 to 7 months of follow-up. To determine the long-term results of thenew procedure, we report the outcomes of 21 eyes with an average 10-year follow-up.
Twenty-one aphakic eyes of 20 patients (15 females and 5 males) wereincluded in this study. The patients underwent surgery for implantation ofa conventional PC-IOL in the AC at the Hara Eye Hospital, Utsunomiya, Japan,from November 20, 1989, to June 2, 1992. The mean (SD) patient age at thetime of surgery was 70 (8.8) years (age range, 48-83 years). Table 1 gives the demographic data.
Our selection of subjects was biased toward significant previous endothelialtrauma. Of the 21 eyes, 13 eyes had previously been implanted with eithera Binkhorst iris-clip 4-loop IOL (B4-IOL) or a Binkhorst iridocapsular 2-loopIOL (B2-IOL). Because of the continuous corneal endothelial cell loss, thoseIOLs were removed immediately before the present procedure was performed.Five patients had undergone intracapsular cataract extraction without subsequentIOL implantation; and in 3 eyes the PC-IOL subluxated into the vitreous cavity.All patients were followed up until December 30, 2001; the longest follow-upwas 12 years and 1 month and the shortest 9 years and 6 months (mean [SD],10 years and 6 months [8 months]).
The results from the initial 10 eyes (9 patients) at an average 17 monthsof follow-up (range, 10-22 months) were reported in Japanese.8 Becauseof senile dementia, 3 patients could not be followed up after about 5 years,and those eyes were excluded from this study. Following a full explanation,all patients provided informed consent.
Preoperative and postoperative examinations included a determinationof visual acuity (VA), tonometry, gonioscopy, laser flare cell measurementwith the laser flare cell meter (Kowa Co, Tokyo, Japan), and keratometry.A corneal endothelial study of the central cornea was done using a contactspecular microscope (Konan Co, Osaka, Japan). Cell density (cells per squaremillimeter) was calculated automatically by a computerized analyzer attachedto the microscope. Follow-up examinations were conducted daily until the secondpostoperative week and then monthly until 1 year after surgery. Additionalexaminations were performed annually. Patients who had complications wereexamined more frequently as necessary.
In cases 1 to 20, local anesthesia was induced by retrobulbar injectionof 2% lidocaine hydrochloride (Xylocaine). In case 21, anesthesia was inducedwith topical 4% lidocaine eyedrops. Except for pupillary control, all preoperativeand postoperative steps were identical to those used in standard cataractIOL surgery.11
In eyes with a B2-IOL or a B4-IOL and those with a PC-IOL that subluxatedinto the vitreous, mydriasis was achieved using 0.5% tropicamide and 0.5%phenylephrine hydrochloride (Mydrin P). In aphakic eyes, 1% pilocarpine (Sanpilo)was used to maintain miosis.
The B2-IOLs and B4-IOLs were removed through a 6-mm scleral incision.To maintain sufficient AC depth, 1% sodium hyaluronate (Healon) was used.When vitreous prolapse occurred, an anterior vitrectomy with the use of scissorsor vitrectomy machine was performed; 1% acetylcholine (Ovisot) then was injectedinto the AC. After confirming sufficient miosis, a PC-IOL was inserted intothe AC by sliding it on the lens glide. The PC-IOL was inserted at the positionwhere the loops tilted anteriorly, similar to when it is implanted into thePC. After inserting the inferior loop, the lens was rotated clockwise andfinally fixed horizontally. No eyes developed hyphema. After rotation, a peripheraliridectomy was performed in one area. Before the incision was closed completelywith a 10-0 nylon suture, thin forceps were inserted into the AC and slippedunder the IOL to grasp the iris near the pupil and pull it toward the center;this maneuver prevents both lens loops from tugging the iris root. The spatulathen was inserted beneath the IOL for the final iris manipulation. These procedureswere useful to prevent future pupillary transformation. Finally, the viscoelasticmaterial in the AC was slowly exchanged with balanced salt solution.
The PC-IOL used in this study was a conventional Sinskey-type IOL (MeniconCo, Nagoya, Japan) with 6-mm polymethylmethacrylate optic and J-shaped polyvinylidenefluoride loops, 0.14 mm in diameter, which angled 10° anteriorly. Thetotal length of the IOL was 13.0 mm. After surgery, 1% pilocarpine eyedropswere instilled 6 times daily for 1 month, 3 times daily for the next 6 months,and then this therapy was discontinued. Because we feared postoperative iritisdue to intraoperative iris manipulation and postoperative use of pilocarpineeyedrops to prevent the postoperative pupillary transformation in all cases,and B2-IOL or B4-IOL removal in some cases, we prescribed a higher dose ofa systemic corticosteroid than that used in routine cataract surgery. Thiswas combined with topical 0.1% dexamethasone methasulfobenzoate sodium eyedrops(DM-solon; Nihon Tenganyaku Kenkyusho Co, Nagoya) 6 times daily between 1day preoperatively and 2 weeks postoperatively. Oral medications included30 mg of prednisolone (Predonisolon; Sanwa kagaku Co, Nagoya) from 1 day preoperativelyuntil 3 days postoperatively; this was then decreased at 4-day intervals to20, 10, and 5 mg, and then this treatment was discontinued. A nonsteroidalanti-inflammatory drug was then prescribed for 2 weeks and this treatmentwas discontinued. Antibiotics eyedrops were prescribed 6 times daily from1 day preoperatively to 7 days postoperatively that contained 0.5% erythromycinlactobionate and 0.5% colistin sodium methanesulfonate (Ecolicin ophthalmicsolution; Santen Pharmaceutical Co, Osaka). Kitasamicin (Leucomycin; AsahiKasei Co, Osaka), 1000 mg/d, was used orally from 1 day to 7 days postoperatively.Acetazolamide (Diamox; Wyeth Lederle Japan Co, Tokyo), 250 mg also was prescribedfrom 1 day until 3 days postoperatively.
The mean (SD) preoperative and postoperative corneal endothelial celldensities of all eyes were 1607 (985.4) cells/mm2 and 1105 (786.3)cells/mm2, respectively. The difference was statistically significant(P<.05). The percentage of postoperative cellloss in all eyes was 30.9% (30.10%). When the analysis was limited to 7 eyeswith preoperative corneal cell density exceeding 2000 cells/mm2,the postoperative cell loss remained 26.5%.
Compared with the preoperative best-corrected VA (BCVA), 7 eyes (33%)had the same VA level or 1 line of improvement or decrease on the eye chart,6 eyes (29%) had an improvement of 2 lines or more, and 8 eyes (38%) had adecrease of 2 lines or more. Of the 8 eyes with a decreased BCVA, the decreasewas the result of bullous keratopathy (BKP) in 5 eyes and development of amacular hole, progressive diabetic retinopathy, and retinal detachment in1 eye each.
Except for 1 eye with preoperative glaucoma (case 2), the intraocularpressure (IOP) was within the normal range in all eyes.
Except for 2 eyes (cases 2 and 4) with marked BKP that did not allowus sufficient observation of the angle, we confirmed that both loops of theremaining 19 IOLs (38 loops) were firmly fixed at the scleral spur; 31 loops(81.6%) had no fibrous adhesion (Figure 1). The other 7 loops (18.4%) had mild fibrous adhesion involvingthe loop genu and the iris root. Four of these loops were bilateral loopsof 2 lenses.
Six eyes developed BKP. The mean (SD) corneal endothelial cell densityin these 6 eyes before PC-IOL implantation in the AC was 932 (614.6) cells/mm2 (range, 381-1932 cells/mm2). Most of these eyes had receivedeither a B2-IOL or a B4-IOL. In 15 eyes in which BKP did not develop, themean (SD) preoperative corneal endothelial cell density was 1878 (989.4) cells/mm2. The preoperative difference between the 2 groups was statisticallysignificant (P<.05). Slight pupillary transformationdeveloped in only 1 eye (4.8%) (Figure 2).
No eyes developed clinically relevant inflammation. The mean (SD) amountof AC flare measured in 19 eyes at the last examination was 13.0 (6.92) photoncounts per millisecond.
There were 10 patients (10 eyes) in which a PC-IOL was implanted inthe AC in 1 eye and a PC-IOL in the PC in the second eye (Table 2). The mean postoperative percentage of cell loss in eyeswith a PC-IOL fixated in the AC was significantly higher than in those eyesin which the IOL was fixed in the PC (P<.05).Although there was no significant difference, the postoperative AC flare washigher in association with the PC-IOL with AC fixation.
The procedure under discussion has the following advantages: (1) itis simpler and easier than PC fixation with scleral sutures, and (2) IOL fixationis secure and causes no complications that commonly occur with AC-IOL fixationsuch as pupillary transformation,12- 14 inflammationat the clinical level, or glaucoma.
However, the major disadvantage is the reduction in the corneal endothelialcell density. One reason is the low preoperative corneal cell density (meanvalue, 1607 cells/mm2). Almost half of the eyes had undergone aprevious surgery that could most likely cause the endothelium to be subjectto damage. However, when the analysis was limited to 7 eyes with a preoperativecorneal cell density exceeding 2000 cells/mm2, the postoperativecell loss was still 26.5%. This tendency was also found in eyes with a contralateralcontrol procedure when a PC-IOL had been implanted in the PC. A significantlyhigher cell loss was found with AC implantation of the PC-IOL.
Except for the preoperative low corneal endothelial cell count, is thereanother cause of the severe postoperative cell loss? The location of the IOLoptic can be excluded. Because the loops of the IOL bend 10° anteriorly,the optic is located at the farthest distance possible from the corneal endothelium.Although the genu of most loops were at the scleral spur without any apparentsign of inflammation, pupillary transformation, or glaucoma, we suspectedthat the peripheral corneal endothelium was damaged mechanically by the loop.Another potential cause could be iris chaffing by the loop of the IOL. Asthe pupil moves, the iris surface could rub the loop and this could producelow-grade inflammation. The damage was slight (with a low laser flare photometryvalue) but might have been sufficiently continuous to cause marked cell lossafter 10 years. Regarding the direction of the loop, initially we positionedthe IOL with the loops posteriorly, similar to other AC-IOLs. However, thefixation of the loop genu was unstable, and the distance between the cornealendothelium and the IOL optic was too narrow. Thereafter, we began using thepresent position. Still, the results were unsatisfactory. One factor in ourstudy is the preoperative low corneal endothelial cell density due to B4-IOLor B2-IOL implantation in almost half the patients. It may be generally understoodthat even if a new procedure is useful, damaged corneal endothelium will besubject to more damage. Initially we thought this may be the case. However,when we observed the stable condition of the temporally placed PC-IOL in theAC, as stated initially, we hoped that the procedure might be useful. We thenexplanted the B4-IOLs or B2-IOLs from the eyes with decreasing cell densityand performed the present method. In the early report of the present procedureat 1 year 5 months' follow-up, the cell loss was 4.9%.8 However,at 10 years 6 months' follow-up, it increased to 30.9%. This clearly indicatesthat to judge the efficacy of a new procedure, long-term observation is indispensable.If stable results at least 10 years postoperatively are desired, this procedurecannot be recommended for secondary IOL implantation.
Correspondence: Tsutomu Hara, MD, Hara Eye Hospital, 1-11 Nishi 1-Chome,Utsunomiya-shi, Tochigi, 320-0861 Japan (email@example.com).
Submitted for publication February 17, 2003; final revision receivedDecember 18, 2003; accepted January 1, 2004.