May 31, 2018
6 min read

Purkinje images can be used to track pseudophakodonesis

In a trial, images documented oscillations in various IOLs.

You've successfully added to your alerts. You will receive an email when new content is published.

Click Here to Manage Email Alerts

We were unable to process your request. Please try again later. If you continue to have this issue please contact

The increasing technological advancements in managing complicated ocular conditions with newer IOL designs have demanded the understanding of the possible effects of IOL oscillations in relation to their position in vivo and their effect on optical performance. Purkinje-Sanson images are produced by the reflection of a light source in the eye, and they have distinctive clinical applications, such as the clinical Hirschberg test, keratometer, videokeratography, IOL tilt determination, dual Purkinje tracker and biometric analysis. The oscillations of the IOL implant after saccadic eye movements have been rarely quantified by Purkinje images in the literature.

Purkinje images

In this trial, Purkinje images were observed through digital slit lamp photography and a video recorder (DC-3, Topcon) by Dr. Kumar. Purkinje image 1 (PI) and Purkinje image 2 (PII) were seen overlapped, defined and bright. Purkinje image 4 (PIV) from the IOL was less intense, inverted and diffuse. The motion of the image was recorded continuously at fixed frames per second while focusing on the target illumination for about 30 seconds and also following a horizontal saccade.

The video was then streamed in the video editor software (Pinnacle Studio 15, Corel) in Windows XP, and the image frames showing PIV and PI were grabbed. The JPEG file format was followed and evaluated by ImageJ analysis ( for the difference in the position of PIV in relation to PI. After correction of geometric distortion and the preliminary edge detection, three grabbed randomized image frames of the overall 30 seconds were included. Scaling was performed for 1 mm to be equivalent to 200 pixels using the set scale option. After this, the files for the patient were selected and the difference in positions of PIV in relation to PI were calculated using the line tool (Figure 1). Measurement of the distance between PI and PIV was determined for each eye at all three time frames. The time taken for the dampening of the oscillations after the abduction saccade in seconds was also determined.

Figure 1. PIV position (red circle) in relation to PI (yellow circle) was determined at various time frames.

Source: Dhivya Ashok Kumar, MD, FICO, and Amar Agarwal, MS, FRCS, FRCOphth

Pseudophakodonesis of IOLs

Out of 127 eyes included in the trial, there were capsule-fixed posterior chamber IOLs, anterior chamber IOLs, retropupillary iris-fixated IOLs, glued transscleral-fixated IOLs and sutured scleral-fixated IOLs. There were no statistically significant differences in PIV positions at various time points in posterior chamber, anterior chamber, scleral-fixated and glued IOLs. Iris claw lenses showed a statistically significant difference in position of PIV at various time frames (P = .0418).


The median difference in the range of movements of PIV was noted to be higher for the iris claw IOL and least for the posterior chamber IOL. In comparison, there was a statistically significant difference in median PIV position among the five groups (P = .0001) (Figure 2). Comparing the difference in the movement of PIV at rest and motion, there was significant exaggeration of position of PIV noted in the iris claw IOL (P = .0395). However, there were no differences noted among the posterior chamber IOL, glued IOL, anterior chamber IOL or scleral-fixated IOL. There was spontaneous suppression of oscillations noted in the posterior chamber IOL, glued IOL and anterior chamber IOL; the iris claw and scleral-fixated IOL showed mild delay 1 to 2 seconds after motion saccade. On dividing the IOL oscillations in relation to the movement of PIV difference as less than 0.5 mm (low frequency), 0.5 mm to 1 mm (moderate) and 1 mm or greater (severe), in posterior chamber IOLs about 68% (n = 34) had movement less than 0.5 mm and only 2% (n = 1) had 1 mm or greater. Iris claw IOLs had a difference of 1 mm or greater in 55% (n = 11) and 0.5 to 1 in 30% (n = 6).

On comparison of corrected distance visual acuity (CDVA) among the different IOLs, there was a significant difference noted (P = .0001) among the IOL groups. There was no statistically significant correlation with CDVA and PIV movements among the IOLs. The normal posterior chamber IOL showed a higher CDVA as compared with the other IOLs. The mean specular count was highest in the posterior chamber IOL (2,090.7 ± 398.6 cells/mm2) followed by the scleral-fixated IOL (1,847.6 ± 490.5 cells/mm2) and glued IOL (1,716.4 ± 459.7 cells/mm2). Clinically, the central corneal thickness was highest in the anterior chamber IOL with a mean of 566.8 ± 53.1 µm, and the central foveal thickness was highest in the scleral-fixated IOL with a mean of 325.3 ± 61.7 µm. There was no significant statistical difference in IOP among the IOLs.

Clinical implications

Mönestam and colleagues have shown a 0.7% and 1.4% risk for severe and moderate pseudophakodonesis after uneventful phacoemulsification with the IOL in the capsular bag. IOL types that are placed in an abnormal position apart from the capsular bag have the propensity to produce abnormal oscillations as there is pre-existing disruption of the capsule and anterior hyaloid face. Aqueous oscillates back and forth about its normal position until the energy of the system has dissipated. Therefore, the implant directly in relation to the internal ocular fluids has a higher risk for oscillations as compared with the one that is protected by the underlying capsule. Similarly, we noted that the retropupillary iris claw IOL has shown the highest incidence of pseudophakodonesis among all the IOLs. Moreover, as it is attached to the overlying iris, there is frequent risk for pigment chafing and rarely IOL drop due to loss of grip related to loss of iris stroma in the long term.


The early clinical predicting sign of future IOL subluxation is the severity of pseudophakodonesis. Various methods have been used to document the IOL oscillations. Optical quality degradation can occur in gross oscillations of the IOL as it can induce transient micro-tilt and may induce momentary aberrations. Micro-tilts usually adjust as the IOL comes back to its original position, similar to a damped harmonic oscillation. The eyes with less than 0.5 mm of PIV differences have been classified as mild or low frequency, and the same has been noted to be high in posterior chamber IOLs. This showed that low frequency of the oscillations is not affecting the visual performance and may not cause structural abnormality. However, in iris claw IOLs, high frequency oscillations of 1 mm or greater were higher (55%), indicating greater pseudophakodonesis, which correlated with CDVA.

Glued IOL has shown pseudophakodonesis similar to the anterior chamber IOL and posterior chamber IOL and no significant difference in CDVA as compared with other IOLs. The known complications of iris claw are iris chafing, late dislocation and inflammation. Pseudophakodonesis in iris claw is the risk factor for chronic iris stromal loss and inflammation and vice versa.


Clinically, pseudophakodonesis is often observed as a quick oscillation of the PIV reflection with respect to the already stationary PI. Obtaining commercial trackers for pseudophakodonesis is unlikely for all surgeons; however, utilizing the image tracking method to calculate the incidence of pseudophakodonesis in patients for follow-up analysis is possible. The effect of retinal image movement due to IOL oscillations in secondary fixated IOLs needs further evaluation in the future.

Disclosures: Agarwal reports he is a paid consultant for Staar Surgical. Kumar reports no relevant financial disclosures.