📄 medicine

Stromal Mechanical Anisotropy of the SMILE Lenticule Shows No Detectable Association With Postoperative Refractive Error

This study demonstrates that despite the reproducible meridional mechanical anisotropy of SMILE lenticules, there is no detectable association between this biomechanical property and postoperative refractive or astigmatic errors, indicating that lenticule-derived metrics offer little value for refining individualized refractive treatment planning.

Original authors: Pengfei Han, Peng Chen, Li Zhang, Yanli Su, Weiyi Chen, Xiaona Li, Hongwei Qin, Qizhi Zhou

Published 2026-06-24

📖 5 min read🧠 Deep dive

CC BY 4.0

Original authors: Pengfei Han, Peng Chen, Li Zhang, Yanli Su, Weiyi Chen, Xiaona Li, Hongwei Qin, Qizhi Zhou

Original paper licensed under CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/). ✨ This is an AI-generated explanation of the paper below. It is not written or endorsed by the authors. For technical accuracy, refer to the original paper. Read full disclaimer

The Big Question: Can We "Tune" Eye Surgery Like a Guitar?

Imagine your cornea (the clear front window of your eye) is like a piece of fabric. When you perform laser eye surgery (specifically a procedure called SMILE), the surgeon cuts out a tiny, lens-shaped piece of that fabric to fix your vision.

For a long time, scientists wondered: Is this fabric "stiff" in some directions and "loose" in others?

Think of a woven blanket. If you pull it from the top to the bottom, it might feel tight. If you pull it from side to side, it might feel a bit stretchier. This is called mechanical anisotropy (a fancy way of saying "directional stiffness").

The big hope was that if we knew exactly how stiff the fabric was in different directions, we could tweak the surgery to perfectly cancel out any leftover blur or astigmatism. It's like tuning a guitar: if you know exactly how tight the strings are, you can adjust the pegs to get the perfect note.

The Experiment: Measuring the "Fabric"

The researchers took 279 patients who had SMILE surgery.

The Cut: During the surgery, they removed the tiny lens-shaped piece of tissue (the "lenticule").
The Test: Instead of throwing it away, they took this tissue to a lab. They stretched it in two directions (up-down and side-to-side) to measure exactly how stiff it was.
The Prediction: They calculated a "stiffness score" based on these measurements.
The Reality Check: Three months later, they checked the patients' vision to see if their eyes were perfectly clear or if there was any leftover blur (refractive error).

The Results: The "Tuning" Didn't Work

Here is the surprising finding: The stiffness of the tissue had absolutely nothing to do with the final vision.

The "Guitar String" Theory Failed: Even though the tissue was indeed stiffer in some directions than others (the "fabric" was real), this difference did not predict whether the patient would end up with perfect vision or a little bit of blur.
The "Negative Control" (The Safety Check): To make sure they weren't making a mistake, the researchers looked at a "wrong" direction. If the stiffness really mattered, it should only affect vision in the direction it was measured. It should not affect vision in a diagonal direction.
- The Analogy: Imagine checking if a car's engine noise predicts the speed of the tires. If you check the speed of the wipers instead, and find no connection, you know your test is working correctly.
- The Result: They found no connection in the "wrong" direction either. This confirmed that the lack of connection in the "right" direction wasn't a fluke; it was a genuine "no effect."

Why Didn't It Work? (The "Why" Behind the "What")

The paper suggests a few reasons why knowing the tissue's stiffness didn't help predict the vision:

The "Leftover" Matters More: The surgery removes a piece of the cornea, but the vision is determined by the remaining cornea (the "residual bed"). The piece they tested (the lenticule) is like a scrap of fabric; the remaining cornea is the whole shirt. The scrap's stiffness doesn't necessarily tell you how the whole shirt will behave after the cut.
The Body Heals: After surgery, the eye heals, and the skin (epithelium) on top of the cornea reshapes itself. This healing process might smooth out any tiny differences caused by the tissue's stiffness, effectively "hiding" the effect.
The Surgery is Already Great: The SMILE surgery was already incredibly accurate. 99% of patients were within a very small margin of error. It's like trying to find a needle in a haystack that is already empty; there was almost no "error" left for the stiffness to explain.

The Bottom Line

The Takeaway:
While the cornea does have directional stiffness (it's not perfectly uniform), measuring this stiffness in the removed tissue does not help surgeons predict or improve the final vision for this specific type of surgery.

What This Means for the Future (According to the Paper):

Don't use it for "Tuning": We shouldn't try to use these stiffness measurements to change the laser settings for individual patients to get better vision. The paper says this approach doesn't work.
Keep it for "Safety": The paper states that the real value of measuring corneal mechanics is still safety. It helps doctors spot eyes that are too weak and at risk of bulging out later (a condition called ectasia). It's a tool for screening (deciding who can have surgery), not for designing (deciding how to do the surgery).

In short: The cornea has a "personality" (stiffness), but that personality doesn't seem to influence the final result of the SMILE surgery. The surgery is already so good that the tissue's quirks don't make a noticeable difference.

Technical Summary: Stromal Mechanical Anisotropy of the SMILE Lenticule and Postoperative Refractive Error

Problem Statement
Corneal biomechanics is a critical factor in keratorefractive surgery, traditionally utilized for preoperative risk stratification to screen for ectasia. A more speculative clinical aspiration exists: that preoperative mechanical measurements could be used to individualize ablation parameters to improve refractive accuracy. While the corneal stroma is known to be mechanically anisotropic (stiffness varies by meridian), it remains unestablished whether this meridional anisotropy biases postoperative refractive or astigmatic outcomes. Small-incision lenticule extraction (SMILE) offers a unique opportunity to test this premise because the procedure removes an intact intrastromal lenticule that can be mechanically tested ex vivo. The central question is whether the directional stiffness of the extracted lenticule correlates with axis-specific astigmatic error, thereby validating the use of biomechanical metrics for individualized treatment design.

Methodology
The study analyzed 279 right eyes from patients undergoing SMILE for myopia and myopic astigmatism, with a target of emmetropia. The study design incorporated a rigorous negative control strategy to distinguish true biological signals from spurious correlations.

Mechanical Testing: Excised lenticules were marked at the 12 o'clock position to register orientation with corneal meridians. They underwent planar biaxial tensile testing at physiological intraocular pressure to determine the tangent modulus along the superior–inferior (SI, 90°) and nasal–temporal (NT, 180°) meridians.
- Anisotropy Index ( $A_{mech}$ ): Calculated as $(ESI - ENT) / (ESI + ENT)$. This metric serves as a mechanical analogue to the $J_0$ astigmatic power vector component.
- Mean Stiffness ( $\bar{E}$ ): Calculated as the average of the two meridional moduli.
Refractive Assessment: Postoperative refractive error was evaluated at 3 months. Refractions were decomposed into Thibos power vectors: Spherical-equivalent error ( $\Delta M$ ), on-axis astigmatic error ( $\Delta J_0$ ), and oblique astigmatic error ( $\Delta J_{45}$ ).
Statistical Design:
- Primary Outcome: Association between $A_{mech}$ and $\Delta J_0$ (co-axial error).
- Negative Control: Association between $A_{mech}$ and $\Delta J_{45}$ (oblique error). Since the mechanical measurement contains no information about the 45°/135° axis, a true meridional effect should not project onto this component. A null result here guards against systematic biases like axis misregistration.
- Secondary Outcome: Association between mean modulus ( $\bar{E}$ ) and spherical-equivalent error ( $\Delta M$ ).
- Analyses utilized Spearman correlation and multivariable linear regression, with the study powered to detect correlations of $|\rho| \ge 0.17$ .

Key Results

Refractive Outcomes: The surgery demonstrated high predictability. The mean spherical-equivalent error was $+0.29 \pm 0.39$ D, with 99.3% of eyes within $\pm 1.00$ D of the target. The attempted-versus-achieved analysis yielded an $R^2$ of 0.918.
Mechanical Anisotropy vs. Astigmatism: There was no detectable association between the lenticule's mechanical anisotropy ( $A_{mech}$ ) and the on-axis astigmatic error ( $\Delta J_0$ ) ( $\rho = -0.054, P = 0.365$ ).
Negative Control Validation: The association between $A_{mech}$ and the oblique error ( $\Delta J_{45}$ ) was also null ( $\rho = -0.033, P = 0.587$ ). This confirms that the lack of correlation on the primary axis is a genuine absence of effect rather than a masking artifact.
Stiffness vs. Spherical Error: The mean tangent modulus showed only a borderline univariate association with spherical-equivalent error ( $\rho = +0.117, P = 0.050$ ), which disappeared in multivariable models ( $P = 0.231$ ) when adjusted for preoperative spherical equivalent. The preoperative spherical equivalent remained the sole independent predictor of postoperative error.
Subgroup Analysis: Sensitivity analysis in eyes with higher preoperative astigmatism showed a consistent but non-significant negative trend, suggesting that any potential effect is likely small, non-linear, or specific to high-astigmatism cases.

Key Contributions

Direct Empirical Test: This study provides the first direct clinical test of the premise that corneal mechanical properties drive refractive outcomes in SMILE, moving beyond computational simulations to measured tissue mechanics.
Negative Control Rigor: By utilizing the oblique astigmatic component ( $\Delta J_{45}$ ) as a pre-specified negative control, the study robustly validates that the null findings on the primary axis are not due to measurement error or coordinate misalignment.
Clarification of Clinical Utility: The data delineates the specific role of biomechanics in SMILE, distinguishing between its established value in safety screening and its lack of utility in refining refractive accuracy.

Significance and Claims
The paper concludes that while SMILE lenticules exhibit reproducible meridional mechanical anisotropy, this property does not translate into detectable axis-specific or magnitude-related refractive errors. The concordant null results for both the primary astigmatic outcome and the negative control argue that this reflects a genuine absence of effect rather than limited sensitivity.

Consequently, the authors assert that lenticule-derived biomechanical metrics offer little scope for refining refractive accuracy through individualized ablation planning. The principal value of corneal biomechanical assessment in this context remains preoperative ectasia risk stratification (patient selection) rather than procedure design (treatment customization). The authors note that these conclusions apply specifically to metrics derived from the excised lenticule and do not necessarily extend to all forms of corneal biomechanical measurement, particularly those assessing the residual stromal bed in vivo. The high inherent predictability of SMILE leaves minimal residual error variance for mechanical predictors to explain.