In-bag dislocation of intraocular lens in patients with uveitis: a case series

We described five late in-bag PCIOL dislocations in patients with underlying uveitis. Tables 1 and 2 summarised their key clinical features, management and subsequent outcomes. The age of patients at the time of cataract operation ranged from 39 to 67 years (mean: 50 years; median: 44 years). Four out of the five patients were male. There were two cases of multifocal choroiditis, two treated herpetic viral retinitis and uveitis, and two Fuchs’ heterochromic iridocyclitis which were also treated with IVTA injections. All patients had in-bag dislocation of PCIOL in their right eyes. The best vision ranged from counting fingers to 6/18 preoperatively and postoperatively varied from 6/36 to 6/6. There were four Alcon SN60WF PCIOLs, and it was unknown for case 4. Circular capsulorhexis and divide-and-conquer phacoemulsification technique was performed in four eyes, and in case 4, the surgical technique was not known. Case 2 had a dense cortical cataract, and case 3 was found to have a loose capsular bag intraoperatively. Case 4 had intermittent inflammation, and case 5 had severe macular oedema postoperatively. Both case 2 and case 5 received multiple IVTA injections. In this retrospective series, there was no information on the cataract density or the duration of the cataract surgeries; however, those patients were generally younger and had less dense cataracts compared to age-related cataract patients, and they also require lower power IOLs.

The time from initial cataract surgery to in-bag dislocation of PCIOL ranged from 48 to 165 months (mean: 81 months and median: 64 months). This is similar to the time frame reported by the largest series of pseudoexfoliation patients - mean 86 months, range 57 to 115 months [11]. The vision at time of dislocation ranged from light perception to 6/7.5. Case 1 with vision of light perception was managed conservatively due to his poor prognosis, and the remaining four cases had additional surgeries. The surgical approach to the dislocated IOL consisted of pars plana vitrectomy with secondary ACIOL in one eye, pars plana vitrectomy with no secondary ACIOL in one eye, and anterior approach lens explantation and secondary ACIOL in two eyes. The explanted IOLs were not sent for histology, and case 4 was found to have a macroscopically fibrosed lens capsule. The final vision after the additional surgeries ranged from 6/38 to 6/6⁻². At 6-month follow-up, half of the patients had improved vision and were comparable to their best vision post-initial cataract surgery - case 2 final vision was 6/7.5, better than 6/12 post-cataract surgery; and case 4 final vision was 6/6⁻², similar to 6/6 post-cataract surgery. Of the half with worsening vision at 6-month follow-up, case 3 vision was 6/12 compared to 6/6 post-cataract surgery, but was better than pre-cataract vision of 6/18; case 5 follow-up vision was 6/36 compared to 6/9 post-cataract surgery, but was better than pre-cataract vision of counting fingers. Overall, final visual outcomes of the five cases at 6-month review were equal or better than their pre-cataract vision. This is consistent with the available literature on described cases with in-bag IOL dislocation, which suggested that prognosis of eyes with this syndrome with any method of management is quite good with most eyes regain their preoperative vision [3].

The exact incidence of late in-bag IOL dislocation following uneventful cataract surgery is unknown. As our series included patients who had their initial surgery elsewhere, we are not able to provide a denominator or a rate of late in-bag IOL dislocation. The cause of in-bag IOL dislocation is thought to be multifactorial, and 90% of reviewed cases are associated with predisposing factors - pseudoexfoliation, uveitis, trauma, vitrectomy and increased axial length [9,10,12]. There are separate reports of intracapsular dislocation of crystalline lens in uveitis patients [14,15]. In an experimental rabbit model, inflammation of the ciliary body leads to fibrin and leukocyte deposition which destabilised the zonular insertion sites [16]. In addition, several reported cases identified sphincter effect of capsular fibrosis causing contraction around an intact capsulorhexis [11] which could occur as early as 3 months post-cataract surgery, and profound shrinkage was reported in cases with uveitis [9]. It was suspected that IVTA injections and topical steroid use could weaken the zonules, as two of the five reported cases had multiple IVTA injections. However, there is limited evidence in the available literature to substantiate such theory nor there has been reported in the diabetic maculopathy cohort requiring multiple IVTA injections to have an increased rate of capsular dislocation. We postulate that in uveitis patients, the underlying inflammation would and the subsequent treatment with topical or intravitreal steroid may weaken the zonules; in addition, the younger age at the time of initial cataract surgery and the uveitis may likely contribute to profound capsular shrinkage. The centripetal force exerted by the capsular contraction adds further strain to the already weakened zonules and results in premature failure and dehiscence of the suspensory apparatus.

Several preventative measures have been described in the literature - an appropriately sized capsulorhexis or the use of capsular tension ring could reduce capsular contraction, therefore delaying the onset of IOL dislocation; tailored phacoemulsification could preserve zonular integrity by using chopping techniques and aspiration of cortex directed in a tangential orientation rather than perpendicular to the zonules [17]. Some literature also suggests that surgeons must look for phacodonesis and in cases of clinically evident zonular instability, to consider the possibility of a primary sutured IOL or endocapsular, scleral-fixated tension rings [8]. In this retrospective case series, preoperative phacodonesis and postoperative IOL stability was not documented in all cases, and this may warrant further study to compare these clinical findings between uveitis-induced and age-related cataract patients.

The management of dislocated IOLs depends on clinical features and surgeon’s preference. Current literatures have described several methods when approaching management of dislocated IOLs. Oshika et al. described a simple fixation technique using a double-armed polypropylene suture passing one over and one under the haptic and out through the sulcus via a corneal stab incision [18]. An alternative method of fixation of the IOL to the iris with modified McCannel suturing has also been considered [19,20]. Although iris and ciliary body fixation of subluxated IOLs has the theoretical advantage of moving the IOL out of the anterior chamber, there is still significant uveal contact. In addition, pars plana approach is the only technique available for explantation of IOLs dislocated significantly posteriorly into the vitreous cavity. There is insufficient evidence to support the superiority of sclera- or iris-sutured PCIOLs over ACIOLs as they are associated with similar rates of coreal oedema, glaucoma and cystoid macular oedema as open-loop ACIOLs [21]. Gimbel et al. suggest that ACIOLs may not be well tolerated in the eyes with uveitis as they may exacerbate the associated disease, compromising the anatomy and function of the anterior segment. In one eye with iritis, implantation of ACIOL resulted in severe reaction and final vision was counting fingers [3,10]. In contrast, whilst our numbers are small, in our case series, all three patients who received ACIOLs had stable uveitis status at the final review, and all had vision better than their vision prior to their initial cataract surgery and better than their vision post-dislocation.