This article is more than 5 years old. Information may no longer be current.
BLOG: There’s nothing as boring as biometry
Click Here to Manage Email Alerts
For our blog’s first real entry, we’d like to talk about the most boring, but perhaps most basic, element of cataract surgery outcomes: biometrics and the postoperative refraction.
Biometry is the mathematical crossover to biology, in this case, to determine an accurately powered IOL to leave the patient with his or her desired postsurgical refractive outcome.
We all know that cataract surgery has a refractive impact. I’m sure we have all had patients share very interesting stories about how their neighbor no longer has to wear glasses after cataract surgery, and so they don’t think they’ll have to either. Expectations are high, and they are high because refractive outcomes with cataract surgery are very good. The vast majority of our patients end up with good refractive outcomes, to the point that not only are patients expecting this, but eye care providers are as well.
The refractive component of cataract surgery has become so ingrained in the mind of both patients and optometrists that many times the true goal of cataract surgery, removing the cataract and allowing improved corrected acuity, has been obscured and replaced with having great uncorrected acuity. And, yet, as prominent a part of patient satisfaction as postoperative refraction is, there is very little understanding of how IOL powers – and thus postoperative refractions – are arrived at in optometry.
Unfortunately, this lack of understanding isn’t a tenable arrangement, with comanaged patients becoming more and more visually demanding. It’s imperative that they hear a consistent message from both the surgery center and primary optometrist about expectations and outcomes, and to educate appropriately we need to have a good understanding of how refractive targets are derived and how the system occasionally fails.
First of all, unlike LASIK, postcataract surgery refraction is not determined by preoperative refraction. In fact, preoperative refraction has exceptionally little to do with what glasses prescription your patient ends up with postoperatively.
Instead, postoperative outcome is determined by IOL power, which is predicted by proprietary formulas that look at the patient’s anterior chamber depth, keratometry, axial length, horizontal white-to-white measures and crystalline lens thickness to try to predict where an IOL will sit in the eye (effective lens position, or ELP). Once ELP is calculated, the formula then calculates what IOL power, sitting at the predicted position, will result in the desired refraction (typically, emmetropia) given measured keratometric power and axial length.
It is entirely biometric- (read: measurement) and formula- (read: extrapolation) based. As with any directly measured or extrapolated system, error in measurement or extrapolation will lead to errors in outcome. In this sense, assuming an uneventful surgery, refractive outcomes have little to do with the actual surgery itself and a lot to do with the preoperative exam.
Of the measurements and extrapolations made in calculating IOLs, there are three that carry the most potential to lead to unexpected refractive error: the directly measured axial length of the eye, the extrapolated ELP and keratometry, which is somewhat of a hybrid between direct measurement and extrapolation
Routinely in North America, cataract surgery axial lengths are determined with optical devices such as the Lenstar (Haag-Streit) or IOLMaster (Zeiss), although, occasionally, ultrasound may be used. The number varies a bit depending on the precise axial length, but, in general, 0.3 mm of error in measurement leads to 1 D of error.
By today’s standards and patient expectations, 1 D of unanticipated error may be considered a poor outcome; this, however, represents a measurement error of only 1.25% in an average length 24.0 mm eye. So, with axial length, a little bit of error goes a long way. Fortunately, modern devices are pretty savvy instruments and will flag any highly suspicious readings. Further, the preoperative provider can often catch other errors by scrutinizing the data.
For example, eyes that shows an axial asymmetry of 0.9 mm would be expected to have 3 D of anisometropia (assuming Ks are symmetric); if this is not present in the glasses prescription, then the clinician would flag that measure as suspicious and repeat the test.
Next are K measurements, which are a surrogate for central corneal power. Although most of us assume that K measures are a direct measurement of corneal power, this is not quite true, and traditional devices that measure Ks make two assumptions that allow them to estimate central corneal power.
First, the devices (with the exception of Scheimpflug imaging and corneal optical coherence tomography) are unable to measure the central cornea and, instead, use paracentral measures to estimate what is happening centrally. Next, the cornea is a thick lens with both anterior surface power (which is positive) and back surface power (which is negative). Traditional keratometers and topographers, however, only measure the anterior corneal radius of curvature, which they use with a known ratio of anterior to posterior corneal curvature to give an estimate of corneal power.
These two assumptions – paracentral corneal power approximately equaling central corneal power and a consistent ratio between anterior and posterior corneal curvature – hold up remarkably well in a normal population. They fall apart, unfortunately, in the worst possible population: eyes that have had previous refractive surgery. Postrefractive surgery corneas throw off devices by altering both the paracentral-to-central relationship and the anterior-to-posterior relationship (with LASIK and PRK, only the anterior curvature is altered; the posterior curvature remains habitual).
The breakdown of these two assumptions in these eyes leads to a frequent over estimation of corneal power and frequent hyperopic outcomes, the magnitude of which is tied to the magnitude of the previous refractive surgery. This is exactly the worst patient population for this error to occur in. People who were motivated to get out of glasses enough to shell out money for refractive surgery are not generally going to be happy with higher prescriptions postoperatively and require explicit and direct education about the potential for refractive error.
Finally, although extrapolated ELP is not incredibly flexible in most patient’s eyes – there is a relatively limited space the IOL can move to – there is some potential that the IOL may “float” to an unanticipated position. In these cases, the ELP’s potential to generate error is robust.
For a standard 20-D IOL, a large departure in lens position of 0.5 mm would result in a 1-D error. Again, given the limited space in the posterior chamber, the vast majority of patients won’t float this much, and modern ELP formulas are very good.
The system does tend to fall apart at the margins, however. Very short eyes will notoriously end up significantly more myopic than planned, while longer eyes tend to end up more hyperopic than planned, as the formulae aren’t as effective with outliers. Therefore, patients with extremely high prescriptions due to axial issues need to be counseled appropriately.
So, while we have a good problem – refractive outcomes with modern cataract surgery are so good there is a misconception in patients' goals of cataract surgery – we also need to be aware of how and why the system occasionally breaks down.
In a coming blog post, we’ll explore how we are trying to work around some of these limitations (and separate hype from real potential) to result in even better refractive outcomes. In comanagement, consistency of message is key. For patients who are in a higher risk group for postoperative refractive error (exceptionally high prescription, previous refractive surgery, etc.) educate, educate, educate (!!!) them on risk for the very real chance that, no matter how well their neighbor saw without glasses after cataract surgery, they may need a pair.
Fun fact: In 1950, Sir Harold Ridley, an English ophthalmologist, implanted the first IOL. The patient ended up with a nearly 20-D refractive surprise. How far we have come in the last 70 years!
Schematic of relationship between K, axial length and ELP.
Source: Aaron Bronner, OD
Source: Aaron Bronner, OD