Aims To investigate macular pigment optical density (MPOD) during follow-up of sealed macular holes and to study correlations of MPOD with progressive changes in spectral-domain optical coherence tomography (SD-OCT) and functional results.
Methods Consecutive patients (n=18) who had undergone successful vitrectomies for idiopathic macular holes were evaluated postoperatively at 1, 3, 6 and 9 months. At each follow-up visit, MPOD was measured with a modified confocal scanning laser ophthalmoscope and the outer retina evaluated by SD-OCT. The changes of MPOD postoperatively and the relationship of MPOD and SD-OCT findings to best corrected visual acuity were examined.
Results MPOD did not change significantly throughout follow-up, from 0.49±0.22 (mean±SD) at month 1 to 0.42±0.18 at month 9. There was a tendency towards a significant association between amount of MPOD and recovery of external limiting membrane during follow-up (p=0.068). Best corrected visual acuity increased significantly from 0.24±0.12 before surgery to 0.65±0.25 at month 9. Recovery of the ellipsoid zone determined most of visual acuity improvement (p=0.024). MPOD was not associated with visual acuity changes (p=0.394).
Conclusions Revisualisation of macular pigment after successful macular hole surgery is not associated with improved visual acuity and may merely be an accompanying sign of the reapposition of the edges of the hole.
Statistics from Altmetric.com
If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.
Today macular hole (MH) surgery results in successful hole closure and significant visual improvement in more than 85% of cases.1 However, postoperative visual acuity (VA) may occasionally be poor despite anatomic closure.2 ,3 Clinical tools such as optical coherence tomography (OCT), fundus autofluorescence (FAF), and microperimetry have been used to correlate examination results with the postoperative best corrected visual acuity (BCVA) to better predict the eventual visual function.4–11 Many studies using OCT images have shown that disruption of the photoreceptor inner segment/outer segment (IS/OS) junction in the fovea5–8 and an intact external limiting membrane (ELM)7 ,9–11 correlated with the final BCVA.
Re-accumulation of macular pigment has been demonstrated with Raman spectroscopy after successful MH surgery,12 and von Rückmann et al13 reported that FAF of the MHs due to the lack of overlying masking macular pigment was no longer visible following successful surgical repair. Recently, the amount of macular pigment has been used to evaluate the postoperative macula.4 ,14 ,15 However, these studies used qualitative FAF grading for the determination of pigment,4 ,14 ,15 did not perform serial FAF examinations, and partly failed to show a direct association between final VA and FAF patterns.4 ,15 Additionally, conflicting data were reported regarding the correlation between ELM integrity and FAF patterns.14 ,15 Macular pigment is located mainly in the Henle fibre layer,16 which is the axonal part of the outer plexiform layer, and re-accumulation following successful MH surgery might simply be due to the normal reapposition of the edges of the hole allowing visualisation of the macular pigment that was continuously present in the retracted edges. Alternatively, we hypothesise that the accumulation of macular pigment after MH surgery is the result of the renewed ability of retinal cells, that is, retinal pigment epithelium and Müller cells, to capture and stabilise lutein and zeaxanthine, which are key elements in the synthesis of the macular pigment. This would indicate a physiological recovery that could ultimately influence the functional outcome.
To test these hypotheses in this study, we made serial measurements to determine the relationships among macular pigment optical density (MPOD), restoration of ELM and ellipsoid zone17 (EZ) lines on OCT images, and BCVA after the successful repair of idiopathic MHs.
We studied one eye in each of 18 consecutive patients (10 women and 8 men; mean age, 71 years; range, 49–84 years) who underwent vitrectomy for an idiopathic MH at Sacco University Hospital, Milan, Italy, from February 2011 to January 2014. Both biomicroscopy and spectral-domain OCT (SD-OCT, Spectralis HRA+OCT, Heidelberg Engineering GmbH, Heidelberg, Germany) were used preoperatively to classify the MH stage. The size was determined by measuring the largest diameter in FAF imaging.18 Eyes with reopened MHs, MHs with retinal detachment, MHs caused by high myopia, or other macular diseases were excluded from this study.
A 25-gauge pars plana vitrectomy with internal limiting membrane peeling was performed in all cases by two of the authors (FB, MC). Brilliant Blue G (Brilliant Peel, Fluoron, Geuder AG, Heidelberg, Germany) was used for staining the internal limiting membrane. An air/fluid exchange with subsequent injection of 20% sulfur hexafluoride (SF6) gas completed each surgical procedure. After surgery, strict face-down positioning was ordered for 3–5 days. Three patients were already pseudophakic prior to surgery, whereas cataract surgery combined with vitrectomy was carried out in the others during the main surgical intervention.
Follow-up examinations were scheduled at 1 day, 1 week, and 1, 3, 6 and 9 months (T1, T3, T6 and T9) after the operation. All patients completed the T9 follow-up visit. BCVA measurements (Snellen charts), SD-OCT and FAF imaging of the macula (Spectralis HRA+OCT), and MPOD measurements were performed at T1 and thereafter.
The Spectralis HRA+OCT provides up to 40 000 A-scans/s with an axial digital resolution of 3.9 µm in tissue and a transversal digital resolution of up to 5 µm (high-resolution mode) by using a superluminescence diode at 870 nm central wavelength. Using an active eye-tracking technology, the system automatically follows eye movements and locks each OCT B-scan to the fundus image. The eye-tracking technology also permits precise scanning of the same location over consecutive visits. For the purpose of this study, the first complete volume scan acquired at the preoperative visit was set as a reference scan. For MH analysis, raster scans of 30×15° and consisting of 37–73 line scans were performed. The spacing between the B-scans ranged from 62 to 122 µm. For each eye, the outer retina was analysed at each follow-up visit by the senior author (FB). Presence or absence (even a small, focal disruption was considered an absence) of normally hyperreflective EZ17 (previously termed the IS/OS junction) and ELM lines were recorded.
MPOD measurements were performed using a two-wavelength autofluorescence technique that has been extensively described elsewhere.19–21 Briefly, it is based on the autofluorescence generated by lipofuscin, which is located in the retinal pigment epithelial cells. This fluorescence is emitted in the 520 to 800 nm spectral range with peak emission at 620–630 nm,22 and it can be excited in vivo by light between 400 and 570 nm. In the fovea, because the excitation spectrum of foveal lipofuscin is strongly influenced by the absorption spectrum of macular pigment, light absorbed by the carotenoids results in a central area of reduced lipofuscin fluorescence. In measuring MPOD, the autofluorescence method compares results from the region of maximum MPOD overlying the fovea to an area that is several degrees eccentric to the fovea and which has no optically appreciable macular pigment. The analysis uses two excitation wavelengths that are differentially absorbed by the macular pigment, that is, 488 nm that is well absorbed and 514 nm that is minimally absorbed. Quantitative imaging was performed with a modified scanning laser ophthalmoscope (HRA1, Heidelberg Retinal Angiographer, Heidelberg Engineering GmbH, Heidelberg, Germany). Dedicated software digitally subtracts the averaged images at the two wavelengths and uses a greyscale index of intensity to create a map of MPOD.
Pupils were always dilated before imaging (tropicamide 1% and phenylephrine 2.5% drops), and each eye was tested two times. While focusing the scanning laser ophthalmoscope on the macular region, rapid sequences of 30° images were captured at 488 and 514 nm. MPOD maps were generated by digital subtraction of the log autofluorescence images, and the mean MPOD was calculated for a 1°-diameter circle centred on the fovea. The reference area without MPOD was set at 6.0° from the centre of the fovea.
This study was approved by the Institutional Review Board Committee of Milan University Medical School at Sacco Hospital. After a detailed explanation of the study, informed consent was obtained from each patient. All examinations and investigations adhered to the tenets of the Declaration of Helsinki.
For all computations, ‘R’ software was used (R: A language and environment for statistical computing; R Foundation for Statistical Computing, Vienna, Austria; http://www.R-project.org). Changes of VA over time as well as association between it, OCT findings, and MPOD were evaluated using multiple regression analysis. Continuous variables were expressed as mean±SD. p Values less than 0.05 were considered significant.
A complete model tested the effect of ELM recovery, EZ recovery and time on MPOD. The effect of time was analysed with Walter matrices23 composed of four columns, one for each time point T1, T3, T6 and T9. In this model, each column is a dummy variable that tests if the MPOD changed significantly from the previous time point (ie, T3 tests if the MPOD at 3 months is different from that at T1). Furthermore, the effect on MPOD changes by a single SD-OCT finding was tested by a reduced model after removing EZ effects that were not significant.
Using all of the measurements from the same 18 subjects potentially creates a pseudoreplication problem. To eliminate this concern, we inserted one dummy variable with the ‘patient effect’; the effect and the significance of this variable were simply ignored in the result but its presence in the model allowed us to remove the patient effect from the residual SE, thus ensuring that the correct degree of freedom value was used in each calculation.
According to Gass’ classification,24 two of the eyes had stage 2 MHs, five had stage 3 and 11 had stage 4. All MHs closed after the first surgery, as confirmed by SD-OCT. No surgical complications were recorded.
It was continuous in 8 eyes at the T1 follow-up visit and in 14 eyes at T9. No eyes revealed a continuous EZ at T1, and only one eye did at T3 and five at T9. Noteworthy, there were no eyes with a continuous EZ and an interrupted ELM anytime during follow-up (table 1).
The MPOD did not change significantly throughout follow-up, from 0.49±0.22 at month 1 to 0.42±0.18 at month 9 (figure 2). There was a tendency towards a significant association between MPOD and recovery of the ELM during follow-up in the complete model (p=0.068, figure 3). The association became significant in the reduced model (p=0.045). However, recovery of the ELM occurred in six eyes at different time points during the follow-up: 4 eyes at T3 (cases 2, 6, 12 and 15), 1 eye at T6 (case 18) and 1 eye at T9 (case 13). None of them showed a corresponding increase of MPOD (table 1).
The BCVA increased significantly from 0.24±0.12 before surgery to 0.65±0.25 at T9. The mean change in VA was 0.40±0.20. Statistical analysis of changes in the ELM, EZ and MPOD revealed that most of the VA variation during the 9-month follow-up was determined by changes in EZ (p=0.024, figure 4). When the EZ became continuous, the mean gain in VA was 0.60±0.20 as opposed to 0.33±0.15 in cases of persistent EZ disruption. Neither ELM findings (p=0.496) nor MPOD (p=0.394) were associated with VA changes.
The biological mechanisms governing retinal capture and accumulation of lutein and zeaxanthin needed to make macular pigments are still poorly understood. Transport of carotenoids to the various body tissues occurs by means of lipoproteins and uptake of plasma lipoproteins is regulated by apolipoproteins. Apolipoprotein E, which is produced in the eye by Müller cells and the retinal pigment epithelium, plays a part in lipid transportation and binding of lipoproteins to target sites within the central nervous system. It also targets the uptake of lipoproteins and carries lutein and zeaxanthin within the retina.
Neelam et al12 demonstrated that the neurosensory retina of an anatomically closed MH contained macular pigment. Müller cell gliosis is the most important factor in the healing of MHs after surgery, as determined in histopathologic specimens.25 The series of junctional complexes between Müller cells and rod and cone photoreceptor cells represent the ELM line on SD-OCT imaging. Therefore, ELM visualisation may give some insights about the status of Müller cells during the healing process of the foveal defect.
Several studies have shown the importance not only of an intact IS/OS junction in the fovea5–9 but also of a normalised ELM7 ,9 ,11 for physiological recovery and better functional outcome. Due to the extreme difficulty of visualising with SD-OCT the actual very thin layer of outer nuclei,17 the ELM has become a good parameter for assessing the degree of recovery in the photoreceptor nuclear layer, which usually precedes and ultimately leads also to the restoration of the EZ.
In this study, we investigated the quantitative changes of macular pigment and the progressive SD-OCT findings after successful surgical repair of idiopathic MH. Our hypothesis was that an increase of MPOD during follow-up, which could possibly be correlated to the recovery of the ELM line, would imply the resumption of an active transport of carotenoids during the healing process of the foveal defect.
Our major finding was that the macular pigment remained stable throughout the 9-month follow-up. One likely explanation is that pigment re-accumulation occurs very early in the postoperative period, either passively or actively. We scheduled the first MPOD measurement 1 month after surgery because the two-wavelength autofluorescence technique needs clear media for the attainment of good macular images. The presence of intravitreal 20% SF6 gas used as the tamponade may last for 3 weeks and can reduce the clarity of images. If active macular pigment re-accumulation occurred in the first month after surgery, we would have missed most or all of it, along with any possible correlation with ELM recovery. It would be of interest to test this further either in patients with silicone oil or with air which would allow earlier measurements. However, there are findings that mitigate the importance of deferred imaging. Fifty-six per cent of the eyes in our series (10 out of 18) had a discontinuous ELM at the T1 follow-up visit. In six of these eyes, the ELM became continuous during follow-up and none of these six eyes showed a corresponding increase of MPOD.
Although we found a tendency towards a significant association between MPOD and recovery of ELM, 8 out of the 18 eyes already had a continuous ELM at T1. That early recovery of ELM could have biased the correlation with MPOD without signifying any real active transport of carotenoids. A more likely explanation is that macular pigment revisualisation occurred after the normal, early reapposition of the edges of the hole where xanthophyl had been retracted before surgery.
Consistent with previous reports,7 ,9 ,11 in the current study we found that the ELM became normalised before the EZ defects were resolved. In fact it was already continuous in 78% of the eyes with persistent disruption/interruption of the EZ. It seems that ELM recovery is needed throughout the affected area before regeneration of photoreceptor OS can occur.
As reported by others,5–9 we also found that recovery of the EZ determined most of the VA improvement. By contrast, MPOD did not seem to influence the final functional outcome. We have no explanations for the lack of association between ELM recovery and visual function in our series.
Our study suffers from some limitations. First, the number of patients is small. However, countering this limitation is that the same surgical technique was performed in every patient, and four follow-up visits were completed for 9 months following surgery. Furthermore, axial motion artefacts in the SD-OCT images were minimised by an active eye-tracking system that allowed the same scanning location to be imaged on each of the following visits. A second limitation might be that MPOD was examined in all patients for the first time at 1 month after surgery. As previously mentioned, that could have prevented the identification of early MPOD changes. By contrast, a strength of the study was that a quantitative objective examination technique for the determination of postoperative macular pigment was used for the first time.
In conclusion, the present study supports the concept that re-accumulation/revisualisation of macular pigment following successful MH surgery might be due to the reapposition of the edges of the hole. Changes in MPOD did not show a direct association with final VA. By contrast, and consistent with previous findings, recovery of the EZ determined most of the VA improvement.
Presented in part at the Association for Research in Vision and Ophthalmology Annual Meeting, Fort Lauderdale, FL, USA, 6–10 May 2012 and the XXIXth meeting of the Club Jules Gonin, Zurich, Switzerland, 3–6 September 2014.
Contributors Substantial contributions to the conception or design of the work: FB and MCi. Acquisition, analysis or interpretation of data: FB, EZ and ECa. Drafting the work or revising it critically for important intellectual content: FB, MCe and GS. Final approval of the version published: all authors.
Competing interests None declared.
Patient consent Obtained.
Ethics approval This Institutional Review Board Committee of Milan University Medical School at Sacco Hospital.
Provenance and peer review Not commissioned; externally peer reviewed.