Tải bản đầy đủ - 0 (trang)
5 Pre-Descemet’s Endothelial Keratoplasty (PDEK)

5 Pre-Descemet’s Endothelial Keratoplasty (PDEK)

Tải bản đầy đủ - 0trang


C.S. Sáles et al.

Fig. 9.8 Surgical photo of

draining trypan blue from

the donor cap

More forbidding is the exchange of trypan blue for BSS prior to loading the

injector. Since the tissue cannot be seen in a sea of dark blue, care must be taken

when wicking off the fluid with a sponge. To increase the margin of safety, we add

a few drops of BSS to the pool of trypan blue to visualize the tissue’s orientation.

We also pre-moisten and ring-out the sponges because it makes drainage slower and

more controlled. Draining the trypan is then accomplished by wicking fluid from

either end of the tissue’s scroll, but not from the side of the scroll. This minimizes

the tissue from “jumping” toward the sponge (Fig. 9.8). If the tissue does happen to

adhere to a surgical sponge, it is advisable to have a pool of BSS at the ready. The

tissue will float off of the sponge once it is lowered into the pool of BSS.


Tissue Insertion

Choosing an Insertion Device: The Straiko Injector

The surgeon’s choice of a tissue insertion device may be one of the most important

determinants of perioperative DMEK complications. Our complication rates

dropped precipitously after introducing the Straiko injector as the backbone of our

standardized DMEK technique. Prior to that, we were using a number of modified

plastic IOL injectors to deliver the tissue, none with great success. The Straiko

injector is comprised primarily of a modified Jones Tube that has been reshaped and

polished by the Gunther Weiss Scientific Glass Blowing Company (Portland, OR)

to resemble the profile of a glass dropper – tapered at its orifice and more bulbous at

its midsection. The modified glass tube is then coupled to a 3-cc syringe with a short

piece of 14-French nasogastric tubing and used off-label for DMEK surgery.

An ideal tissue insertion system is both easy to use and completely atraumatic to

the endothelium. Although many iterations have been used by others [2, 23–26], there

are traits of our system that may make it a better option. Designed with these two ideals in mind, the Straiko injector is made of glass, which, unlike plastic, does not

adhere to DMEK tissue and is very nearly a no-touch system. It safely enshrouds the

tissue in BSS and moves it in and out of the injector with atraumatic venturi currents,

which are easy to modulate with a syringe. A further margin of safety is added by the

injector’s bulbous base, which has the effect of slowing the tissue during aspiration


Descemet Membrane Endothelial Keratoplasty (DMEK)


Weaker Venturi Currents

Stronger Venturi Currents

Fig. 9.9 Surgical photo of the Straiko injector with a schematic overlay of currents in the tapered

tip vs. the widened base

before it can be inadvertently sucked into the syringe. As the tissue travels into the

bulbous portion of the injector, it gradually decelerates in the same way a child gently

comes to a stop in the collecting pool at the end of a waterslide (Fig. 9.9).

Preparing the Injector

The primary concern for the DMEK surgeon is to prepare the injector so that it can

transfer donor tissue into the anterior chamber without any unwanted viscoelastic.

It is imperative that any viscoelastic that may have been retained in the injector during sizing of the corneal wound be completely rinsed from the system because it can

impede unscrolling of the graft in the anterior chamber. Air bubbles can also interfere with unscrolling and should be purged from the system as well.

Testing the Injector

If available, excess tissue removed from the donor tissue is perfectly suited for testing

the injector. Mark Terry has popularized these otherwise wasted pieces of donor tissue

as his prized “space monkeys.” “Before NASA sent a man to the moon,” he says, “it

tested its spaceship with space monkeys (Abel and Baker). So before you send your

DMEK tissue into the eye, test your injector with space monkeys of your own!”

A glass petri dish is the best place to practice loading the injector under the operating microscope. During these trial runs, the surgeon should confirm that the plunger

is positioned in the middle of its travel, or “half-mast”, and that it moves freely. Very

short, staccato pulses on the plunger should translate into direct movement of the

tissue into and out of the injector. If the action of the injector feels sluggish, check

for leaks and remove any dead space between the base of the tube and the syringe.


C.S. Sáles et al.

Fig. 9.10 Ergonomics

photo of holding the

syringe of the Straiko


It is important to establish a good feeling for what kind of inputs the injector

needs to generate sufficient currents to move the tissue. By committing to memory

what these inputs feel like, the surgeon is better positioned to not only replicate

them in the eye but, of equal importance, to know when the upper limits of normal

inputs have been exceeded.

Developing a tactile familiarity with the injector’s delicate inputs comes with

experience, but hand position on the injector can help improve one’s proprioception.

We hold the injector in our dominant hand like a pencil. The plunger is then operated with the nondominant hand by pinching the shaft of the plunger between the

index finger and thumb like throwing a dart. This provides much more control over

the injector than is possible with one’s thumb on the syringe’s plunger because the

interosseous muscles are much more dexterous than the thenar eminence (Fig. 9.10).

Loading the Injector

After the tissue has been stained and the injector has been tested, purged of viscoelastic,

and positioned to half-mast, the surgeon is ready to load his donor tissue. The first step

to loading the injector is positioning the graft in the donor rim so that it is (1) parallel to

the injector’s long axis and (2) floating in an ample pool of BSS. The easiest and safest

method for positioning the graft into an ergonomic position for loading the injector is

by rotating the entire corneoscleral rim. Drops of BSS and drainage with a sponge can

also be used to coax the tissue onto the slope of the donor rim so that there is space

beneath the scroll’s proximal end at the apex of the donor cornea for the injector’s beveled tip (Fig. 9.11). When loading the injector, it is better to have a deep pool of BSS

than a shallow one. Like the tide that has gone out from beneath an anchored boat, the

DMEK scroll will run “aground” and no longer float into the injector if the donor rim

is not filled with fluid. This can be easily addressed by replenishing BSS as it is sucked

into the injector, but requires an assistant. Care must also be taken not to overflow the

donor rim, which can easily send the graft “overboard.”


Descemet Membrane Endothelial Keratoplasty (DMEK)


Fig. 9.11 Surgical photo

of loading the Straiko

injector in the donor cap

Fig. 9.12 Surgical photo of

a DMEK scroll low in the


The most important consideration for loading the injector is the proximity of its

orifice to the DMEK scroll. Ideally, the proximal end of the scroll should almost be

in contact with the injector’s tapered tip and the tip should be bevel up. Here, the

venturi currents are strongest, and only staccato pulses on the syringe are needed to

make the tissue move. An injector that is too far from the tissue will only generate

currents that can move the graft with large excursions on the syringe. This usually

results in draining the donor rim of BSS, collapsing the scroll, and running it

aground before it has had a chance to be aspirated into the injector. Once the tissue

is safely in the injector, the surgeon should inspect its relative position to the orifice

prior to injecting it into the anterior chamber. Ideal positioning is just inside of the

orifice because it confers the smallest volume of BSS needed to generate a current

to float the tissue into the anterior chamber (Fig. 9.12), but not too close to the tip so

as to prevent the tissue from being sucked-out during injector insertion.

Attempting to inject tissue from the device’s base is fraught with danger. By its

design, small movements of the syringe confer little to no movement of the tissue in

the injector’s bulbous base. However, the injector’s “doldrums” can be overcome with

large, fast excursions on the syringe, which will overpressurize the anterior chamber.

If the tissue is high up in the device’s base, it is always better to reposition it closer to

the orifice by submersing the injector into a glass petri dish filled with BSS and applying gentle pulses to the syringe until the tissue moves to the desired location. Gently

tapping the glass tube can also help coax the tissue into better position.


C.S. Sáles et al.

Repositioning can be avoided by loading the injector correctly the first time.

Small actions on the syringe move smaller volumes of fluid at a slower velocity and

have the tendency to make the scroll swim slowly into the injector so that its position can be controlled. Large actions on the syringe will tend to accelerate the tissue

well above the injector’s tip until it comes to rest in the safety pool of the device’s


Occasionally, air can be inadvertently sucked into the injector when loading the

graft. When this occurs, the safest method to purge the system is to tap the bubbles

into the injector’s base by holding the injector upright. Bubbles residing high up in

the device usually will not travel with the graft so long as only small volumes of

fluid are injected into the anterior chamber.

Injecting the Tissue

Understanding the fluid dynamics of injecting DMEK tissue into the anterior chamber is essential to preventing the dreaded complication of graft ejection. Fluid currents and pressure gradients govern the tissue’s movement out of the injector and

into the anterior chamber, whereas volume, the anterior chamber’s compliance, and

pressure gradients govern whether the tissue remains in the anterior chamber after

the initial injection.

Depression of the syringe’s plunger generates a transient rise in pressure and

flow in the injector, which create venturi currents that float the tissue into the anterior chamber. During injection, the anterior chamber grows in volume, depending

on how much fluid was used to inject the tissue. Because the anterior chamber is

very compliant due to its mobile iris-lens diaphragm and elastic cornea, it can

accommodate this additional fluid before rising in pressure. After the surgeon stops

depressing the syringe, pressure equilibrates between the anterior chamber and

injector, and the tissue stops moving.

However, if the volume of injected BSS exceeds the anterior chamber’s compliance, the chamber’s pressure will increase. In turn, wherever there is an adjacent

area of lower pressure, a gradient will form and fluid will flow. If the surgeon is

lucky, such a gradient will cause the tissue to reflux safely back into the injector.

Otherwise, the tissue may be ejected from the eye through either the main wound or

one of the paracenteses.

The challenge for the DMEK surgeon is to know when undesirable pressure

gradients exist before problems arise from them. This can be difficult because pressure gradients, especially small ones, cannot be seen until it is too late. Pressure

gradients must therefore be anticipated and managed prior to becoming manifest.

In general, they are avoided by using small-volume injections.

When is it important to depressurize the chamber during injection? Almost

always! Using one of the paracenteses, the surgeon should depressurize the chamber

before and/or after anytime fluid is injected into it. Depressurize the chamber before

inserting the injector, then again after the tissue is in the anterior chamber, then yet

again before any further injection of BSS is used to disengage the tissue from the

injector, and, of course, once more prior to withdrawing the injector (Fig. 9.13).


Descemet Membrane Endothelial Keratoplasty (DMEK)


Fig. 9.13 Surgical photo

of burping a paracentesis

during tissue injection

Fig. 9.14 Surgical photo

of the DMEK scroll

perpendicular to the wound

prior to removal of the


Other techniques that help prevent tissue ejection from the anterior chamber

are the pivot and trap maneuvers. After injection, the tissue scroll is usually parallel to the long axis of the injector and can sometimes still be in contact with its

orifice. To prevent the tissue from following the injector out of the eye, the surgeon should aim to create a margin of distance between the injector and the tissue. An effective method for accomplishing this is by slightly pivoting the injector

within the wound so that fluid bursts can be directed almost tangential to the

proximal aspect of the scroll. This will cause the scroll to rotate about the pupillary axis. Repeating this maneuver with intervening burps of the paracenteses can

bring the tissue around until it is perpendicular to the injector (Fig. 9.14). Once

the tissue is in this orientation, there is enough space between the injector and

scroll to safely depress the anterior aspect of the main wound against the iris

plane without harming the donor endothelium. Depressing the anterior lip of the

wound with a cannula as the injector is removed adds a further margin of safety

by ensuring that there is minimal to no fluid loss from the eye during this delicate

step (Fig. 9.15).

Whether to position the injector bevel up or bevel down in the wound is a matter of preference for the surgeon, but the design of the injector’s bevel makes

bevel down the most ergonomic for injector removal. There are advantages of

using the Veldman Venn, discussed later in this chapter, to distinguish which side

of the graft is up, and then positioning the injector in the eye accordingly, irre-


C.S. Sáles et al.

Fig. 9.15 Surgical photo

of depressing the anterior

lip of the main wound with

a cannula while removing

the injector

spective of the bevel. But by design, the bevel-down configuration allows for

easier injection and a safer egress from the eye.

The final step of tissue injection is securing the main wound. One suture is sufficient, but whatever is needed to secure the wound is fine. It is important not to gape

the incision when passing the suture needle, because this can cause a pressure gradient, outflow of aqueous, and potential loss of the tissue.


Unscrolling the Tissue

Numerous papers have been published detailing different techniques for unscrolling

DMEK tissue. All of these surgical maneuvers have proven to be effective in the

hands of their proponents [1, 2, 23, 25, 27]. But to date, no technique has been

shown to be superior in preventing endothelial damage [27].

A unique feature of DMEK surgery is that the unscrolling process can be as variable as the tissue being implanted. Even two consecutive cases using tissue from the

same donor can demand slightly different approaches. This is generally not the case

in DSAEK, which is much more rote, even in eyes with anterior segment comorbidities. The DMEK surgeon must be adaptable in a way cornea surgeons did not

have to be in the past.

We have adopted and developed a series of no-touch tapping and fluid-egress

techniques using two 27-gauge cannulas to indent the cornea and burp the incisions,

which we have affectionately termed the “dance moves” of the “DMEK dance.” We

prefer this approach to the double-bubble techniques popularized by Melles [2],

Kruse [23], as well as Price [24] because we have found them to be the most adaptable to the idiosyncrasies of every DMEK tissue. Yet, on occasion, we may employ

“bubble bumps” or the double bubble because they are best suited for the situation.

In our experience, being familiar and facile with different unscrolling techniques

only helps the DMEK surgeon.

Given the potpourri of unscrolling techniques that are available, we have tailored

this section of the chapter to focus on underlying principles rather than detail the


Descemet Membrane Endothelial Keratoplasty (DMEK)


intricacies of various maneuvers, which could probably comprise an entire textbook. If observed, these principles can help keep the DMEK surgeon out of trouble – whatever technique is used.

Determining Graft Orientation

The single greatest predictor of iatrogenic primary graft failure (I-PGF) is upsidedown orientation, which was never as much of a concern in DSAEK. Upside-down

grafts are a common pitfall because, even to the trained eye, DMEK tissue can

sometimes appear the same whether right side up or upside down. Many have struggled with this fundamental conundrum of DMEK surgery, which explains why a

number of creative solutions have been developed to prevent upside down grafts.

Moutsouris and colleagues were the first to publish a visual method for determining the orientation of the DMEK scroll in the anterior chamber [2]. Known as the

Moutsouris sign, the method employs placing a cannula on top of the graft just

under the recipient cornea. By moving the cannula toward the scrolling leaflets, the

surgeon can ascertain whether the leaflets are curling over the cannula – indicating

that the scroll is right side up – or whether the leaflets are curling toward the iris,

indicating that the graft is upside down. The only consideration one must have when

using this technique is not to touch the cannula to the tissue, especially if it is upside

down, as this would damage the endothelium.

Price and colleagues popularized the slit beam method of determining scroll orientation [28]. In their approach, a handheld slit beam is used intraoperatively to

examine the scroll in the anterior chamber. If there are two arcs of light, the tissue

is scrolling toward the cornea and the graft is right side up. If there is only one arc

of light, the tissue is scrolling toward the iris and is upside down. Similarly, Straiko

has utilized a handheld slit laser beam but abandoned the practice in favor of

S-stamped tissue. Utilizing the same principles, Steven and colleagues have

employed intraoperative optical coherence tomography to visualize orientation of

the tissue [29].

Jacob has utilized the E-DMEK technique or the endoilluminator-assisted

DMEK for determining [30] (Fig. 9.16). This utilizes the vitreoretinal light pipe to

throw oblique illumination on the graft to allow its orientation to be determined.

Bachmann and colleagues developed yet another method of determining tissue

orientation comprised of punching an asymmetric scalloped pattern into the edge of

the graft [31]. Similar to an S-stamp, this method allows for intraoperative confirmation based on the orientation of the scallops.

We prefer using the Veldman Venn in conjunction with pre-S-stamped tissue to

determine the graft’s orientation. Prior to injecting the tissue, it is inspected in the

injector by rotating the syringe around its long axis (like the axle of a car). When the

“v” of the Venn moves in the same direction as the rotation, the surgeon is looking

at the graft right side up. When the “v” moves in the opposite direction, the graft is

upside down (Fig. 9.17). Knowing the scroll’s orientation relative to the bevel of the


C.S. Sáles et al.



Fig. 9.16 (a, b) E-DMEK showing the endoilluminator being used to determine orientation

Scroll based orientation techniques



Immediate pre-insertion graft orientation

Rotate Straiko-Jones tube clockwise

Correct side up

Up side down

Fig. 9.17 Illustration of the Veldman Venn (Courtesy of Peter Veldman, Massachusetts Eye and

Ear Infirmary, Boston, MA)

injector permits the surgeon to position the injector, and hopefully the scroll, in a

known orientation before depressing the plunger. However, because the scroll can

rotate at any time during the injection, this method is not foolproof.

The Veldman Venn, Montsouris sign, and slit lamp techniques are only effective

when the tissue is in a particular configuration. The Venn works wonderfully provided that the scroll does not unknowingly roll upside down. The Montsouris sign


Descemet Membrane Endothelial Keratoplasty (DMEK)


Fig. 9.18 Surgical photo of a graft with an S-stamp and illustration of the Bachmann asymmetric

edge crescents

requires that the tissue be in a double scroll, which is not always the case. Although

not mandatory, a double scroll is also the optimal configuration for using a slit

beam. By contrast, the S-stamp and Bachmann techniques provide visual confirmation immediately before the tissue is apposed against the stroma with a bubble.

Whatever conformation the tissue undergoes between injection and its final positioning with a bubble is immaterial (Fig. 9.18).

Managing the Anterior Chamber

Attentive titration of the anterior chamber’s depth and pressure is vital to the success of unscrolling any DMEK tissue. DMEK tissue assumes a scrolled configuration when it is given enough space to roll itself up. The surgeon capitalizes on this

phenomenon when loading the Straiko injector, but he is battling it in the next phase

of the surgery.

A chamber that is low in pressure is always safer than a chamber that is high in

pressure. As has been discussed at length earlier in this chapter, high pressure confers the risk of tissue ejection from inadvertent egresses of fluid from the paracenteses or main wound. The lower limit of what is a desirable pressure varies between

eyes, but in general, the pressure should not be lowered to the point of making the

cornea lose its convex shape or to the point of flattening the anterior chamber. Both

of these scenarios will make unscrolling more difficult and potentially harm the

endothelium, especially in the setting of a posterior chamber IOL.

A shallow chamber is the surgeon’s third hand in DMEK surgery because it

facilitates trapping the tissue in an opened configuration between the iris and cornea. If the chamber is too deep, no matter what technique is used, the tissue will


C.S. Sáles et al.

Fig. 9.19 Surgical photo

of a concave cornea in a

vitrectomized eye

Fig. 9.20 Surgical photo

of using a Cindy sweeper

to depress the equator

have a tendency to rescroll during the process of opening it. This is especially true

of tightly scrolled tissue, which is more frequent among younger donors.

Despite the surgeon’s best efforts to screen for risk factors that may make the

chamber difficult to modulate (i.e., prior vitrectomy, vitreous syneresis, high myopia), he can sometimes be surprised intraoperatively by an eye whose chamber

does not respond to the usual maneuvers. In these cases, the chamber tends to flipflop between either being the appropriate depth but too hypotonous or being too

deep. These cases are challenging no matter what the approach, but additional

maneuvers can be employed to coax the chamber into a better configuration

(Fig. 9.19).

In highly myopic eyes, the safest method for balancing the pressure and depth of

the anterior chamber is to depress the equator with a finger or a blunt surgical instrument (e.g., a Cindy sweeper) to create posterior pressure that plateaus the iris diaphragm anteriorly (Fig. 9.20).

Reconfiguring the Tissue

Certain tissue configurations are easier to manage than others. The tricorne hat and

double-scroll configurations are perhaps the easiest to open. By contrast, the 60:40

Tài liệu bạn tìm kiếm đã sẵn sàng tải về

5 Pre-Descemet’s Endothelial Keratoplasty (PDEK)

Tải bản đầy đủ ngay(0 tr)