Transcript
Barton Blackorby, MD (00:08):
Welcome to the New Retina Radio Journal Club of VBS. My name is Barton Blackorby from the Retina Institute in St. Louis. I'm joined today with Matt Starr from the Mayo Clinic.
Matthew Starr, MD (00:18):
Thanks, Barton. Happy to be here.
Barton Blackorby, MD (00:20):
And Kyle Kovacs from Cornell in New York City.
Kyle Kovacs, MD (00:23):
Thanks, Barton. Great to be here.
Barton Blackorby, MD (00:25):
Today we're going to be talking about a paper entitled Vitreous Hemorrhage Due to Posterior Vitreous Detachment: Incidence of Retinal Detachment and Spontaneous Clearance During Observation by Hilal Hasbolat and colleagues. It was published in Ophthalmology in April 2026. And Dr. Saar, do you mind giving us a quick summary of the paper?
Matthew Starr, MD (00:45):
Yeah, so this study looked at what happens when patients develop a vitreous hemorrhage from a presumed PVD and are initially managed with observation. The authors out of Denmark retrospectively followed 366 patients for at least two years after VH onset. Patients with a previously known cause of vitreous hemorrhage like PDR, RVO, AMD were excluded, as well as eyes that had previously received anti-VEGF injections or had previously undergone vitrectomy or had an intraocular surgical cause potentially as a cause of VH were excluded.
(01:15):
The authors graded VH on a scale of zero to three, zero being all clear with a mild hemorrhage, and they could see a complete view of the fundus, and these eyes were actually excluded from the study, with grade three being [inaudible 00:01:27]... other extreme being a completely no view of the fundus. The authors found that 62% of these grades one through three hemorrhages cleared spontaneously, but resolution was often slower, particularly in grade two and three hemorrhages where actually only 34% of grade three VH is cleared on their own.
(01:43):
In this study, 17% of patients went on to develop a rhegmatogenous retinal detachment after presentation, with most attachments occurring within the first three weeks after presentation and then actually plateauing thereafter. In this study, men had nearly three times the risk of retinal detachment, while older patients had a substantially lower risk. The initial density of the hemorrhage was an important predictor of how quickly it cleared, and pseudophakic patients were also less likely to clear.
(02:10):
In this study, the key clinical takeaway is observation can be successful, particularly for patients with milder hemorrhages, but close monitoring during the first few weeks is critical. For severe hemorrhages or higher-risk patients, such as young men, the author suggests that early vitrectomy should be considered rather than a prolonged period of observation.
Barton Blackorby, MD (02:28):
Dr. Starr, thank you so much for the summary of that paper.
(02:30):
Dr. Kovacs, I've got a few questions for you concerning the techniques described here. So in this paper, scleral depression was not routinely performed at this institution. Although several studies, including a prospective one by Seider and colleagues, found that in patients with PVDs, scleral indentation with indirect ophthalmoscopy identified about 97% of retinal tears versus about 85% if we just look at the slit lamp and don't indent alone, meaning that about 12% of tears were only visible if we did a scleral depressed exam.
(03:03):
If a comparable mis-rate was applied here, do you think that simply exam technique rather than the biology of the PVD may affect the 17% detachment rate that they quote in the study?
Kyle Kovacs, MD (03:16):
I think you're spot on. I don't think this reflects real-world practice, but I think it gets a little bit more nuanced because of the way that they kind of graded it or scaled the vitreous hemorrhages, grade one where there's a view of the peripheral retina, grade two where there was some obscuration of that view, and grade three where there is no view. I think we all wouldn't say, "Oh, there's some blood in the eye, let's give up on trying to do a clinical exam and scleral depress these patients when the view is somewhat hazy from blood." And so thus it doesn't really reflect real world practice, and I'm sure they're missing some of the tears that eventually became rhegmatogenous detachments in these patients.
(03:52):
The other thing though is that I think as they also discuss, it shows how much emphasis they're really putting on ultrasonography in these patients, that when there's blood, they both discuss the importance of the ultrasonography, and also highlight the variability in technique and outcomes and findings that can be found in these patients. But I would say it's a little bit faulty to be relying on ultrasonography for grade one or mild vitreous hemorrhages where there's an adequate view of the periphery for finding tears.
Matthew Starr, MD (04:23):
Doctor, I talk about that a little bit in the paper, I think, Kyle, where 88 or 89 eyes didn't actually get a B-scan initially. They don't give us the number, but they say that likely that these eyes were the ones that had grade one and they could see everything. And then they talk about too, they don't depress these eyes because there's no great view.
(04:43):
And I agree, in our practice, I see all sorts of hemorrhagic PVDs and from likelihood what they would call zeros or ones, which I'll definitely depress and you want to see everything. And if I have no view into the eye, I personally am not depressing that eye. I have to rely on my ultrasound. I think this paper is a little bit more unique is that these are pretty bad, pretty dense vitreous hemorrhages.
Kyle Kovacs, MD (05:08):
That's a really great point. This is a really skewed sampling.
Matthew Starr, MD (05:11):
I think it's really skewed. And they talk about 51 eyes undergoing a vitrectomy. And when you look at it, 15 underwent an immediate vitrectomy, 10 of which were for retinal tears that they found on presentation. And then the other five were other obscure reasons, like a blind pillow eye or personal reasons. And then the other 36 eyes that underwent a vitrectomy were essentially for not clearing VH, but they found 10, which is still... I feel like we find a lot more tears initially in hemorrhagic PVDs than 10 out of 366.
Barton Blackorby, MD (05:49):
I think both techniques, scleral depression, B-scan ultrasound are things that retina surgeons need to be very comfortable with. And obviously, one doesn't apply to every single case.
(05:59):
What I want to look at is what are our practice patterns of how often we see these patients? So this study showed that most of the RDs here occurred within the first three weeks. Other studies have shown a retinal tear, retinal detachment presenting up to about 40 days after PVD induction, that recommended about a four to six week follow-up visit. Yet others have shown that up to 55% of delayed tears occurred maybe up to about four and a half months later.
(06:29):
Given that the detachment rate in this paper front-loads within the first three weeks, what do you guys do when you see a patient like this? How often do you see them, and does it depend on your view and how dense the hemorrhage was? I'll start with you, Dr. Kovacs.
Kyle Kovacs, MD (06:45):
Well, I think the paper advocates for weekly at the beginning, but then I think it's sort of suggesting maybe we can stretch it to a month or so afterwards. I tend to follow these patients every week, at least for the first three or four weeks when they're the real dense hemorrhages that obscure a view, and we're relying really purely on ultrasound to look for concerning pathology that would prompt surgery.
(07:08):
And then there is some even risk stratification based on did the fellow eye have problems? Do the fellow eye have tears? Sometimes we can get a lot of information about where we're heading based on the fellow eye's history and appearance. Is there lattice in the other eye that's crazy everywhere that's going to prick our ears up that we're maybe in a higher risk situation with this eye? But I do at some point start to elongate those because if we're really not going for surgery, which is a separate discussion, it can start to become a burden if we're through the first say three or four weeks, so maybe every two or three-week follow-ups after that.
Barton Blackorby, MD (07:42):
How about you, Dr. Starr? How often are you seeing these patients?
Matthew Starr, MD (07:46):
Yeah, I mean I think as Kyle pointed out too, for this specific subset of I can't see anything and I'm relying on ultrasound pretty frequently, one kind of unique thing that we do for a lot of them, I'm probably pretty aggressive. We'll see them; I can't see anything; they're usually coming into the resident clinic and cursory looking at an ultrasound, double-checking their exam, their ultrasound. And probably what we're doing is having them bilaterally patch, coming back in a couple of days, see if we could clear it up a little bit, get a better view in. And if I still don't have a great view, the B-scan is still concerning; a lot of times I'll take them to surgery within a week. That's kind of what my mentor's instilled on me is if there's no view, and you have no idea why this person had a hemorrhage, and you're presuming it's a PBD, the safest thing is just to clear it up in our opinion.
(08:36):
I think if you have a pretty good idea after a couple of days, and you can, okay, maybe it's an RVO, it's NVE, or you have other kind of etiologies and you can start to carefully observe, I think it's a little bit risky, and I think this study kind of highlights that up to 20% potentially detachment rate really early on would maybe favor this early intervention. And that's what we do here.
(09:02):
I think if we can start to get a better view in, I do start to see these patients. I space it out really quickly, but I'm relying it on a view in. If I don't have a view, it's pretty aggressive follow-up, and then early one to two-week surgery.
Barton Blackorby, MD (09:19):
I agree with you. If I don't have a view, this is totally different than a management of a diabetic vitreous hemorrhage. I would rather catch this with a tear. I think all of us would definitely like to avoid going into an eye with limited view to vitreous hemorrhage and a detachment hiding underneath that somewhere. That's a hard case for anybody. So getting it out early before that happens, I fully advocate for that. So let's talk about-
Matthew Starr, MD (09:44):
We've all been there. We want to avoid it.
Barton Blackorby, MD (09:45):
Yeah.
Kyle Kovacs, MD (09:45):
Also, it's just too much blood to leave for that long when you've got a break.
Matthew Starr, MD (09:50):
Yeah.
Barton Blackorby, MD (09:50):
Yup.
Kyle Kovacs, MD (09:50):
The number of cases that had vitreous hemorrhages, the paper talks about it, but PVR risk when there's that much blood filling the eye with breaks in the retina hiding somewhere, it's a lot of blood. My mantra for my mentors was always: if you can't read the big E on the chart, that's too much blood to leave in the eye. So I think if you're sitting on that, and that's basically a grade two or grade three as we're seeing in these studies, those are the ones that are, as Matt, as you said, I think my timeline's really similar for at least pushing these patients for surgery. Not all of them necessarily here want it.
Barton Blackorby, MD (10:26):
Yeah, that vitreous hemorrhage is this PVR soup, a lot of profibrotic factors floating around in there.
(10:32):
So what about patients that are just bothered by the haze? They may be a truck driver, they need their job, they need to be getting on the road and drive, and they're saying, "Doc, please clear this out for me." Let's say they're younger or phakic. How do you tailor that conversation before jumping in to clear out an early vitreous hemorrhage? Dr. Starr, I'll start with you on that one.
Matthew Starr, MD (10:55):
Yeah, younger phakics are going to be a harder sell. I mean, you got to talk to them and counsel about the risks. If they don't have a PVD like on an ultrasound, I mean some of these are PVD related, but in the paper they had the eyes up to 28 years old, you're going to be hard-pressed to make me do an early vitrectomy on a 28-year-old male. I think that's risky. I think that if they need it for their job, or they have a porcine other eye, I'd be more inclined to get into it. Probably I'll try to be reasonable. I won't make them go to my next OR slot in three months, but within a month or something, but I'm not going to be aggressively adding them on next case, next day, next case, doing surgeries at night kind of thing.
Barton Blackorby, MD (11:41):
How about you, Dr. Kovacs?
Kyle Kovacs, MD (11:42):
I mean, some of it is just the expectation management. As you said, people want the blood gone and expect to be working the next day. And it's not like the recovery from surgery is that kind of slam dunk return to arms.
(11:56):
I've had patients that we talk about the phrasing of do we take on the known course of the recovery or the more predictable course of the recovery from the surgery and express exactly what that timeframe may look like, including gas tamponades and possible positioning depending on what you find, or do we take the unknown? How long is this going to take to clear, which is we're seeing even at a couple of months out is 40% from this paper?
(12:23):
So I think that as long as patients are comfortable with that known course and have really good insight into what that entails, I'm happy to go early. They just need to have good understanding and know that it's not cataract surgery.
Barton Blackorby, MD (12:38):
Yeah. The proper expectations, I think, for all of our surgeries, are definitely the key to a successful outcome.
(12:45):
Well, this wraps up a great discussion. Dr. Kovacs, Dr. Starr, thank you for walking me through this. I think the big takeaway for our listeners is most vitreous hemorrhages from a PVD will clear on its own, but the risk is high in those first three weeks of having a retinal detachment. So early vigilance matters, and early vitrectomy, if you're not able to adequately rule out a tear detachment, is definitely the safer bet.
(13:10):
Thanks so much for tuning in to the New Retina Radio Journal Club of VBS. We'll see you next time.
