Published in Research

How often is reoperation needed post-RD repair?

This is editorially independent content
10 min read

A recent study published in Retina evaluated more than 25,000 primary retinal detachment (RD) repairs among Medicare beneficiaries to determine how often revision was needed and which patient-, disease-, surgery- and surgeon-level factors were most closely associated with another intervention.

The findings suggest that the characteristics of the detachment itself may matter considerably more than how many RD procedures a surgeon performs.

Why do retinal detachments sometimes need a second surgery?

Rhegmatogenous RD is a potentially vision-threatening condition in which a retinal break allows fluid to pass beneath the retina, separating it from the underlying tissue.

Surgical repair is the standard treatment, with approaches including scleral buckling and pars plana vitrectomy (PPV), either alone or in combination.

Over the past several decades, clinical practice has increasingly shifted toward vitrectomy, although scleral buckling continues to play an important role in selected cases.

… however …

Recurrent or persistent detachment can require another procedure after the initial repair.

Also worth noting: Proliferative vitreoretinopathy (PVR) remains the most common cause of failure following rhegmatogenous RD repair—and can cause the retina to detach again through the formation and contraction of cellular membranes.

So with retinal attachment also serving as an important measure of surgical quality, investigators wanted to know whether differences in surgeon volume meaningfully influence the likelihood of needing another operation.

Let’s talk study details.

Researchers conducted a retrospective, multicenter cohort study using Centers for Medicare & Medicaid Services fee-for-service outpatient claims data spanning 2014 through 2023.

The primary study period included surgeries performed from 2016 through 2021, allowing investigators to follow each patient for two years after the initial procedure.

RD revision was defined as another RD surgery or pneumatic retinopexy performed in the same eye within 2 years of the primary repair.

And what was analyzed?

Several potential predictors, including:

  • Patient age and sex
  • Type of retinal detachment
  • Surgical approach
  • Surgeon RD volume
  • Surgeon graduation year
  • Practice volume
  • Teaching versus non-teaching setting
  • Urban versus rural practice setting

How were the surgeries categorized?

Surgeons were divided into three volume categories within the Medicare dataset:

  • 10 to 25 RD surgeries
  • 26 to 50 surgeries
  • +50 surgeries

Importantly: The authors noted that these figures represented only Medicare RD procedures captured by the database—not each surgeon's total real-world RD surgical volume.

How big was the Medicare cohort?

Overall: 25,456 eyes undergoing primary rhegmatogenous RD repair were included.

The mean patient age was 70.4 years, and 35.8% of eyes were from female patients.

The analysis included 1,016 surgeons:

  • 659 in the low-volume group
  • 276 in the medium-volume group
  • 81 in the high-volume group

As for surgical approach:

  • 3.8% underwent scleral buckle alone
  • 85.2% underwent vitrectomy with or without scleral buckle
  • 11.0% underwent complex vitrectomy with or without scleral buckle

Multiple-break detachments accounted for the largest specified RD category at 38.4%, followed by single-break detachments at 29.3%. Giant retinal tears (GRTs) accounted for 1.6% and total RD for 6.5% of cases.

How often did patients actually need another procedure?

Overall: 3,649 eyes, or 14.3%, required RD revision within 2 years.

And most revisions occurred relatively early.

The median time from primary surgery to revision was 73 days, with an interquartile range of 37 to 163 days. Even further: 78.2% of revisions occurred within six months of the initial operation.

Disease characteristics emerged as some of the strongest predictors.

Compare this with the reference group.

  • GRT-associated RD carried a 39% higher adjusted risk of revision (RR: 1.39; 95% confidence interval [CI]: 1.13-1.72)
  • Total RD carried a 30% higher risk (RR: 1.30; 95% CI: 1.15-1.47)
  • Complex vitrectomy with or without scleral buckle carried a 29% higher risk (RR: 1.29; 95% CI: 1.09-1.53)

Retinal dialysis was associated with an even higher relative risk (RR: 2.01), although only 33 cases were present in the entire dataset (which limited how much could be concluded from that finding).

Did a surgeon’s experience make a difference?

One of the study's more notable findings involved surgeon volume, though it wasn't significantly associated with revision risk.

Revision occurred in:

  • 14.6% of procedures in the low-volume group
  • 14.1% in the medium-volume group
  • 14.1% in the high-volume group

Tell me more.

After adjustment, neither medium nor high surgeon volume was significantly associated with lower revision risk compared with the low-volume group. Practice volume was also not a significant predictor.

The investigators performed additional sensitivity analyses, including treating surgeon volume as a continuous variable and using a multilevel regression model to account for clustering by surgeon.

  • The overall conclusions remained the same.

Go on …

Interestingly, the authors' funnel-plot analysis suggested that surgeon revision rates began clustering more closely around the overall average after approximately 50 cases.

Their interpretation? At least 50 cases may be needed before an individual surgeon's RD failure rate can be assessed with enough precision for meaningful benchmarking or comparison with peers.

Any caveats to keep in mind?

A few.

First, this was a retrospective, nonrandomized analysis of Medicare beneficiaries—meaning the study population was generally older than the overall population undergoing RD repair.

Additionally, the Medicare database captured only a portion of each surgeon's actual RD surgical volume.

The claims data also lacked granular clinical details that could influence outcomes, including:

  • Location of the detachment
  • Extent of PVR at presentation
  • Type of tamponade used
  • Whether a buckle was used during vitrectomy

Another consideration: Revision surgery was used as a surrogate for primary RD failure. Patients whose initial repair failed but who declined another procedure would therefore not have been counted as revisions, potentially underestimating the true failure rate.

And what did researchers make of the surgeon-volume finding?

The authors suggested that the lack of a meaningful association between surgeon volume and revision risk may reflect the standardized fellowship training received by U.S. retina surgeons.

They also noted that surgeon graduation year did not have a clinically meaningful effect on RD outcomes, further supporting the ability of fellowship-trained retina surgeons to manage primary rhegmatogenous detachments.

However: the findings shouldn't necessarily be extrapolated to every RD.

The investigators acknowledged that surgeon experience and volume could still influence outcomes in more complicated pathologies, including diabetic tractional RD or rhegmatogenous detachments with advanced PVR.

So, what may actually be driving surgical failure?

Perhaps one of the most important findings came from what the researchers couldn't explain.

The multivariable model accounted for only 0.7% of the observed variability in RD revision. Surgery type explained about 0.3%, RD type about 0.2% and surgeon graduation year about 0.1%.

That leaves the overwhelming majority of variation unexplained.

I’m listening …

The authors pointed toward PVR as one possible piece of that puzzle.

Their study notes that despite advances in vitreoretinal surgery, RD failure rates have not substantially improved over the past several decades, potentially because PVR is a biological process that clinicians still cannot reliably prevent or predict.

Previous research also suggests that GRT-associated detachments⁠ can carry particularly high rates of PVR and surgical complexity.

As such: The investigators suggested that future research into genetic risk profiles could potentially help identify patients particularly susceptible to developing PVR.

Take home.

In this large Medicare analysis:

  • Approximately one in seven primary rhegmatogenous retinal detachment repairs required revision within 2 years, and most repeat interventions occurred within the first 6 months.
  • Complex vitrectomy, giant retinal tears, and total RD were among the primary factors associated with increased revision risk, while surgeon and practice volume were not significant predictors.

The findings suggest that disease complexity and still-unidentified biological factors may play a larger role in RD surgical failure than surgeon case volume alone, while also highlighting PVR prediction and prevention as an important area for future investigation.