The femtosecond roundtable

Five experts discuss femto: the upsides, the downsides, patient selection, even its future.

Uday Devgan, MD: Welcome. We’ve had quite a few years now of femtosecond cataract surgery. In my center in Beverly Hills, we now have two of these machines, but there’s a wide range of usage. Some surgeons use it nearly 100% of the time and some are closer to zero, like me. This roundtable will be a great opportunity to get different perspectives from noted experts who have a varying rate of use of the femtosecond lasers. I think Kendall and Vance use the femto a lot more, and I think Dr. Daoud and I are using it very infrequently.

Kendall Donaldson, MD: I do about 30% femto.

Yassine J. Daoud, MD: I started getting interested in the femto because my chairman told me to. But I’m running my own prospective clinical study in the femtosecond that is about to finish. So I use it frequently, but not 30% of my volume.

Dr. Devgan: Why is that?

Dr. Daoud: If the patient asks for it and is a good candidate, I use it. If they [enter] my study, I use it because then they don’t have to pay for it. Otherwise, I haven’t noticed that it leads to significantly better outcomes than those reached with my hands, and I have found it hard to justify offering it to patients with an upcharge without, so far, what I believe a clinically significant difference between the two technologies.

I am interested in it from the research perspective. [But] use in the patient population with increased cost, that’s my ethical and clinical dilemma.

Dr. Devgan: In our center we have six operating rooms. The femtos we have are major platforms. Initially I used one quite a bit and enjoyed using it. We have two femtosecond lasers, which I have enjoyed using, but I don’t see any difference in refractive outcomes. The limbal relaxing incisions are certainly better than when doing them by diamond or hand, but if the patient has more than 1D of cylinder I’m going to do toric lens anyway. And a toric lens for 1.5D or above is much better than any LRI whether you make it with a femto or with any other device.

I also use it in select cases where there is a clinical benefit, such as a patient with an intumescent white cataract. With the femto I can get a round ‘rhexis in about one second without the risk of run-out and the Argentinian flag sign. There’s an upside to doing femto then, and I recommend it.

Otherwise, for very loose zonules I may use it, but for 95% of my cases I’m not using it. The population in Beverly Hills can afford it and if they really want it, I’m happy to do it for them. But, I tell them that in most cases they likely won’t see a tangible benefit from it.

And the last thing is reduction in phaco energy. I found that certainly you could use the laser to help divide a nucleus into small pieces so you use less phaco energy, but it’s just a minor improvement of phaco chop. We have also reduced ultrasound energy dramatically by going from continuous phaco to the pulse-and-burst modes. I’m already putting in such a miniscule amount of ultrasound energy with phaco power modulations and phaco chop, so in my hands adding the femtosecond laser it reduces it by an insignificant amount.

Dr. Daoud: Even more important than the ultrasound, as you alluded to, is that we have gotten much better with our modulation of the ultrasound energy, which has improved the health of the cornea, the level of the edema and the endothelial cell density.

Dr. Devgan: Let’s hear from the others.

Vance Thompson, MD: I use [femto] mainly for a refractive result. In my practice, 40% of my patients choose premium cataract surgery, but only after a discussion on what sort of vision they want. I ask them, “Do you want to do a lot with glasses or a lot without glasses after cataract surgery.” In those 40% looking for the sort of vision that a multifocal or an extended depth of focus implant will provide, I typically use the femtosecond laser. I’m talking more about vision, so I really don’t get into laser discussions.

The reason the laser gives me confidence is I’m going for an OCT-guided capsulotomy that is going to be centered on the apex of the lens that’s going to allow me 360° of capsular overlap. So when I get capsular contraction, I minimize the lens tilt thus inducing less higher-order aberrations. I’m going for a refractive result, and I’m getting more confidence when I’m going in there with the laser. The therapeutic side is one thing, but on the refractive side I do feel it has helped.

It has also taught me a lot, having used the femtosecond laser basically since the beginning. I wasn’t maybe as good at performing manual capsulorhexis using the Purkinje method, but what I learned from using the laser and the femtosecond approach on my capsulotomy has taught me how to be better with my manual capsulotomies. I work hard marking the cornea to center on the Purkinje image to really try to get 360° of capsular overlap, and I know my microscope well. But I can’t make it as round and perfect with consistent capsular overlap as with an OCT-guided capsulorhexis.

Dr. Donaldson: I tend to agree with Vance. We enjoy using the femtosecond lasers. We have the LenSx (Alcon) the Victus (B&L) and the Catalys laser (AMO), and we’ve teamed up with Rob Weinstock to get his LensAR (Lensar) data just to compare all four laser platforms. We’ve been using the femtosecond lasers since 2012.

Basically, I tell patients that we can put less energy into the eye, and I think that information has been well established on every laser platform. Between 58% and 99% less energy is put in the eye, and we’ve been able to substantiate that over and over again with many publications.

Depicted are the clinical settings Dr. Berdahl uses to make corneal incisions with his LenSx laser. These settings are not translatable from laser to laser.

Also, as Vance mentioned, I also feel the LRIs are more accurate with the laser. We don’t have solid data except in PKs: laser-created LRIs compared to manual LRIs, but we’re still working on collecting additional data. But I feel much more comfortable with my LRIs now and with astigmatism correction with the laser.

I think femtosecond laser cataract surgery can benefit the patient in many instances. Some were mentioned like the white cataract, dense cataract, Fuchs’ dystrophy and patients with a low anterior chamber depth. If we have less space to work in, if we have to manipulate the nuclear material less, it’s going to be advantageous to the patient. I think these are all situations where the femtosecond laser can be a great tool for patients. Additionally, all of my premium patients (toric and presbyopia correcting lenses) always are done with the femtosecond laser. We don’t charge for the laser, but it’s a tool that is part of our upgraded packages.

John Berdahl, MD: I’ll start off with a very overarching comment, that femtosecond laser is a modest improvement to an already very good surgery. I don’t think that it is an absolutely game-changing technology or that it is superior in every situation. But I do believe that the reproducibility it brings to the surgery can make large-scale outcomes more predictable.

I do wish that there was more good evidence either supporting or denying femto, and so I’m really glad Yassine is doing a well-designed study to help us understand the role it plays.


OM: Is there anything that can be done to the laser itself that would change anybody’s mind in terms of buying it?

Dr. Thompson: I think it would help a lot of surgeons if it were less expensive and could consistently be integrated into our cataract surgery. For us, it’s in a different room. I don’t mind the hassle of having the patient go and stop in there for the femto before they go into the operating room for the intraocular work. But I think a lot of doctors would be more interested if it were less expensive and more integrated into our cataract surgery routine.

Dr. Berdahl: I would have it be part of the microscope if I could have a dream.

Dr. Daoud: You stole my idea.

Dr. Berdahl: That’s why I wanted to go first. Don’t steal my idea, Yassine.

Dr. Daoud: You said it much more succinctly than I would have, so that’s perfect.

Dr. Donaldson: But it also needs to be integrated into a preoperative plan and then the postoperative nomogram needs to be created for feedback. And so this whole femtosecond application of the patient’s treatment would be planned ahead of time, and the nomogram would be modified continuously for that particular surgeon. So everything would be so optimized and so mechanized that the laser would follow through with the plan that was created. The laser would just be a tool for the whole integration process — preoperative, intraoperative, postoperative — and it would have a complete feedback loop.

Dr. Daoud: Absolutely. I think especially for the LRI, having an optical image beforehand with the suite and then transferring that directly, like we do the WaveScan (AMO) for the LASIK patients or refractive patients, we already have an IRIS registration kind of program. If we can transfer that to the laser that is mounted or part of the microscope suite and then potentially combine it with modulation of the refractive outcomes, that would make the laser a lot sweeter to me.

The other issue is probably increasing usability of the laser from both a corneal perspective as well as cataract. Having the laser do refractive and cataract surgery may make it a bit more palatable for the surgeon if they’re going to be spending that much money on the laser. I think it’s a matter of time before the laser itself gets cheaper; its current cost seems to be a main hurdle to its widespread use.

As far as the speed, the laser part of the procedure, we already have brought it down to a sub-two-minute procedure, and, if everything goes smoothly, it could be one minute, one minute and 15 seconds. But sometimes it can be a lot longer than that. And the question is, can you make it any shorter? I’ve noticed, for example, there has to be modulation of the energy to improve the irrigation/aspiration cortical removal because, to me, the cortical removal has been a little bit more difficult with the femtosecond laser than the regular phaco. And the wound construction, we are actually playing with the software, trying to find the best angles for the main corneal incision that is always peripheral, self-sealing and with good morphology. So I’m sure that will be coming down the pipeline.

Dr. Devgan: The one thing that is unfortunate is the very high cost of the machine and the cost per use. And unlike the days of buying a YAG laser and using it to your heart’s content at no additional cost, we will be paying a per-eye fee for each of these new laser platforms.

Dr. Donaldson: I think companies are more willing to bargain and to create packages for individuals where they structure your payments over a long period of time.

Dr. Devgan: I agree. There’s certainly a balance involved in this. There are a couple of health systems that pay for every patient to have free femto for all cataract cases. And maybe that’s the way things will end up going in some situations. But, as we see declining reimbursement for medicine, I think we need to figure out ways of either cutting costs or pushing the cost onto the patients, which is what we’re doing now.

What we haven’t talked about is the varying use among surgeons based upon how they market femto in their practice. Some surgeons may market it as decreasing the energy in the eye. I know one surgeon who says, if it were my eye I’d do it. If you have the means, please, pay for it and if you don’t, don’t worry, I’ll do it with forceps. So that’s a reasonable option as well.

Another point to discuss: There’s a varying use based upon surgeon comfort, skillset and even experience. Some of my residents who are fresh out of training are so happy to do femto because they don’t have that stress about “Will I be able to get a chop on this nucleus?” It’s already chopped for you and even the capsulorhexis is already done.

And similarly, I think in our surgery center we even have an established surgeon who never learned to do a ‘rhexis by hand. That’s just the way he was taught. Or another surgeon who only phacos out the nucleus with a one-handed phaco technique, which is more common than we think. And now all of a sudden, he can have a chopped nucleus and have an easier time.

Dr. Donaldson: So you’re implying, Uday, that it’s much easier to have the femtosecond laser pre-cut the nucleus and do those other steps, your capsulorhexis and all that for the surgeon. [Your observations are] consistent with Jose de la Cruz’s study of residents who started to do femto before traditional phaco surgery, which found it was actually safer in their hands.1

Dr. Devgan: Actually, it can be but I think in my own hands, I can have a more radialized capsulorhexis in a femto case than I would in a nonfemto case, and the studies have borne that out as well in terms of strength of the ‘rhexis. I don’t see an issue with any kind of overlap compared to femto or not and I’ve had zero difference in refractive outcome.

Dr. Donaldson: I’m not seeing a difference in refractive outcomes either, but we’re still analyzing outcomes.

Dr. Devgan: I think Dr. Daoud and myself have both realized that there’s no difference in even the phaco energy. I realize some studies say it’s 60% or 90% less energy, but in terms of my practice, there’s no difference. It is very much surgeon and technique dependent.

Dr. Daoud: I guess the real question is the difference between a slightly decentered capsulotomy versus a round capsulotomy. Have any of us noticed significant predictability, refractive predictability that is different in the femto group than the conventional phaco?

Dr. Berdahl: After we got the device, five years ago, we had our surgeons look at what we call a RELACS (Refractive Laser Assisted Cataract Surgery) package, which is femto plus aberrometry. We looked at our outcomes compared to our standard outcomes and it took our outcomes from mid to low 70%+/-.5 to 85%+/-.5. We had somebody else look at that data so we weren’t biased toward wanting the femto to work. That made me say that at least our package is improving our predictability. I can’t say that it was femto or aberrometry, but the package is making us a bit better.

Dr. Daoud: John, did you use the aberrometry with and without the femto for comparison, or just the package total aberrometry and laser versus purely complete conventional?

Dr. Berdahl: Yeah, aberrometry was only used in the laser group and so it’s really not a pure comparison of femto. But we don’t really market femto to our patients. We just talk about what the patient cares about, which is how they want to see. And I think one of the best evidences to show we believe in it is, whether we use the femto or not, is that we don’t do it à la carte. Some patients are good for it. Some patients aren’t, and the vast majority of them we do it on, that cost comes out of our pocket. So we have skin in that game and so I’ll stand by that it’s a modest improvement over an already good surgery.

Let me state the obvious: cataract surgery is the most common surgery in the United States and one of the most successful and most life-changing; we owe it to a problem as big as cataract to try and keep moving that bar higher.

A better corneal incision

Here is a real pearl — how to change the settings on a femtosecond laser to achieve better corneal incisions. John Berdahl, MD, divulged his secret. (To see the settings, please see image on page 25.)

Dr. Devgan: Someone has to figure out how to make the incision a little bit right with the femto. What’s the secret?

Dr. Berdahl: Yeah, I was so underimpressed with femto incisions so we had our engineers come in. We decreased the spot separation, we decreased the energy. We had less power but closer spots. The incisions opened easier and sealed much better. The end result was it took about twice as long to make the incision as it did before, going from 1.5 seconds to 3 seconds. I thought there’s no way I’d tolerate that extra 1.5 seconds. But we get a really good incision now, consistently, and I can’t remember the last time I had to suture one. Now my femto incisions are equal to my manual ones and I use them routinely. They almost never leak, and the true triplanar architecture is good. I think it’s a good example of working with the clinical team to optimize how the laser can perform.

Dr. Donaldson: Two things, John. One, if you compare femtosecond cases to traditional cases, a lot of times this is a skewed comparison because the patients that can afford the laser package, even though we’re not really charging for laser but we’re charging for an upgrade package (including astigmatism or presbyopia correction with aberrometry), generally they’re healthier patients, they don’t have many confounding other issues such as a tube or a trab or something like that, other surgeries that they’ve had in the past, or a terrible diabetic retinopathy or macular problems. So those complicated patients are getting traditional surgery. They are two distinctly different groups of patients there. So, a lot of times it’s very difficult to compare patients unless they are part of a prospective randomized trial.

Dr. Devgan: What do you think of a phaco incision created by femtosecond laser?

Dr. Donaldson: It’s bad.

Dr. Thompson: I like my incisions in the limbus, and with laser incisions it has to be more clear corneal. I have gravitated back to my manual limbus-based incisions.

Dr. Daoud: There are two things. Number one, studies have shown that there’s probably an apoptotic thermal effect that is killing the corneal endothelial cells in the process by creating corneal incisions with a femtosecond laser.

I worry a lot more about my femtosecond-created, corneal incisions than I do about my hand-created blade incisions. Many times I have to put a suture for the femtosecond-created incisions. And it’s not as predictable as my hand-created incisions, where I want to put it right at the surgical limbus, because, if the patient has vessels or a neovascularization or vascularization of the cornea or something, the laser doesn’t read the margin very well and you could be too anterior or too posterior.

Dr. Thompson: I would agree. I have a tendency to really favor femto for the lens and capsule work, and for limbal-based surgery I still do my manual incisions.

But I also wanted to add that early on we experienced a few radial tears, but with modern-day femto that’s less of an issue. And we actually studied this issue since we’re also involved in the Zepto (Mynosys) research. Zepto uses a disposable handpiece with a soft collapsible tip and circular nitinol cutting element to make a perfectly round capsulotomy. We just finished the FDA monitored trial; it is not FDA approved yet.

We compared capsulotomy edge tear strength with Zepto, versus femto, versus manual and the Zepto was stronger than both femto and manual, but femto ended up being stronger than manual in our particular trial and that was published in Ophthalmology.2 So, I don’t think that we can say current femto has an edge issue compared to manual.

Also, there are multiple publications out there that say IOL tilt and decentration occurs more when there’s not capsular overlap. And there have been multiple publications that have shown that a femto-created continuous curvilinear capsulorhexis is a more stable refractive result with less IOL tilt and decentration than a manual capsulorhexis. Some great reports on that.

And so again, where I favor femto is in the lens-related work in going for a refractive outcome. And I think it’s most likely true for monofocal implants also. Induced high-order aberrations from lens tilt are just not as perceptible in monofocal implants as they are with multifocal or toric technology.

Dr. Donaldson: I would have to agree with Vance. Over the last five years of doing femto cataract surgery, the incidence of capsular tears has gone down. The publications that showed a higher incidence of anterior capsular tears were all in the first two to three years after its FDA approval. If you look at the data just over the last two years, no study shows that the anterior capsular tear rate is higher in femto. And we certainly are seeing that our anterior capsules are just fine now. Additionally, the laser software has improved through the years and the imaging systems are now better able to detect the limbus. Thus, we are able to make our incisions more peripheral allowing them to seal better than they did early on. Although this has improved, I think, as Dr. Daoud alluded to, there is still room for improvement with regard to the primary corneal incision.

Dr. Daoud: I think part of that is because we’re using less energy and have better and faster software to create the capsulotomy. If you use less energy, the ridged border is less prominent.

I agree that we have rarely had an anterior capsular tear in the past nearly four years. We realized that most of us haven’t changed and the anterior capsulotomy radialization rate is insignificant.

OM: Does anybody care to say anything else? I think we’ve had a wonderful discussion.

For the residents, some found the femto easier to work with and they were more comfortable with it. Is it better for them to learn with the femto, or should they be able to do things manually as well? What should come first here?

Dr. Devgan: I attend about 500 resident cases every year. And I think it’s really important to have the manual skills here. We only let the residents use a femto for cataract at the second half of their senior year residency and only for a few cases. There are many, many cases where you really can’t use a femto, especially small pupil cases where the iris blocks laser access and white, opaque cataracts that prevent laser light delivery. Tonight I’m calling from our UCLA-affiliated county hospital where we can’t afford a half-million-dollar laser and spend $300 or $400 an eye for these patients to have a femto procedure. I think our residents only learn initially with forceps manually and then as seniors they get to do a handful of cases of femto.

What about for your eyes, Doctor?

Dr. Devgan: Okay, for your eyes, for your phaco, do you want the femto? And for all parts of the surgery, or just for rhexis, LRI, or what?

Dr. Thompson: If I had 2+ NS or more, I’d want to have my nucleus pre-treated with the laser, and no matter what, I’d want my capsulorhexis performed with the laser.

Dr. Donaldson: I agree with Vance. I definitely want my capsulotomy and my lens prefragmentation and whatever LRIs I need done with the laser.

Dr. Berdahl: I would want it done by the femto for lens, capsulotomy, and AKs and I am fine with my primary incisions being made with femto as well, but I don’t think that they’re any better than manual. But, and I didn’t comment before, I do use the femto to make corneal incisions, but it took a lot of time to get it right.

Dr. Daoud: I agree with John except that I don’t want my LRIs with the femto because a lot of my patients have LRIs with the femto and are having significantly more dry eyes than the ones that don’t, so I would try toric if at all. But I wouldn’t mind having the femto for the capsulotomy and for the lens fragmentation. I think it’s definitely better.

Dr. Devgan: Great input, team. For my eyes, I would find the right surgeon and put my trust in his hands, laser or not. And I’d make sure that I ended up mildly myopic with the highest image quality.

Dr. Donaldson: I agree. Traditional phacoemulsification is not going away. It always will have a place, especially small pupils. That’s really going to be the basis of cataract surgery, even if someday femto becomes more popular. Right now, it’s only 5% to 7% of cataract surgery in the United States. Even though we don’t know what the future of femto will be, traditional phacoemulsification, which is a wonderful surgery, will always be the basis of what we’re doing here.

Dr. Daoud: If you actually compare head-to-head femto early days versus phaco early days, femto is light years ahead of the outcome of phaco. But, as John said, we are already at such an incredible high standard with the phaco that femto is having an uphill battle to show a significant difference.

As the volume of cataracts coming our way is likely to double in the next decade or so, and the numbers of ophthalmologists we are training are not increasing, the throughput has to increase. The more automation happens, the more efficiency happens, it’s more likely there will be more adoption of the laser. I think that’s probably where we are heading, so I think femto is the surgery of the future.

Dr. Donaldson: I did a study that showed my surgeon-induced astigmatism with the laser was equivalent to my manual incisions, and so I did all laser incisions for over a year. I still like my manual incisions better, though we showed that there was no difference in the astigmatism; we couldn’t find a statistical difference. OM


  1. Cortina M, Jain S, Ho J, Prickett A, De La Cruz J. A reduction in the femtosecond cataract learning curve: Initial resident experience performing cataract surgery with and without femtosecond laser. Presented at ESCRS meeting August 10, 2013.
  2. Thompson VM, Berdahl JP, Solano JM, Chang DF. Comparison of manual, femtosecond laser, and precision pulse capsulotomy edge tear strength in paired human cadaver eyes. Ophthalmology. 2016;123:265-274.