MiOLO.AIPublications

    Clinical review · Post-refractive biometry

    Counting the Posterior Cornea Once: Total Keratometry, Posterior Corneal Astigmatism and the Toric IOL After Refractive Surgery

    A modern biometry printout offers three different corneal powers for the same eye. A toric calculator wants exactly one of them. Which one, why, and what happens after LASIK or PRK when the wrong one goes in.

    By Bruno Miolo, MD, M.Sc

    Cataract Surgeon

     · 24 min read

    In brief

    1. Standard keratometry converts the anterior corneal radius into a total power with a fixed index. That shortcut assumes a fixed relation between the two corneal surfaces, and myopic ablation breaks it: after surgery, total keratometry sits about 0.5 to 0.8 D below standard K on average.
    2. Most toric calculators take anterior keratometry and add their own estimate of the posterior cornea. In virgin eyes, that estimate performs as well as measured posterior values on swept-source OCT biometers and better than measured values from Scheimpflug devices.
    3. After myopic LASIK or PRK, the posterior cornea keeps the orientation and magnitude of a virgin cornea. In 2026, a dedicated comparison found that the predicted posterior term forecast the refractive astigmatism better than the measured one in the Barrett True-K Toric formula, while measured posterior data improved the spherical equivalent.
    4. Total keratometry, TCRP, TNP and TCP already contain the posterior surface. Entering any of them into a calculator that predicts the posterior cornea counts that surface twice, and feeding them to a post-refractive formula that expects standard K applies the index correction twice.
    5. The rule is short: the posterior cornea enters a calculation once. Give each calculator the K it was built for, from the device it was validated with.

    1. Introduction: Three Numbers for One Cornea

    Twenty years ago the cornea entered an intraocular lens (IOL) calculation as two numbers, K1 and K2, and nobody asked where they came from. Today a single biometry printout from a swept-source OCT biometer lists standard keratometry, posterior keratometry and total keratometry side by side, and a Scheimpflug tomographer adds simulated keratometry, true net power and total corneal refractive power in three different zones. The numbers differ from each other by amounts that matter clinically, and every one of them is called, somewhere on the page, a corneal power.

    The surgeon faces a practical question that is rarely spelled out: which of these values does the toric calculator expect? The question becomes sharper in eyes that had laser vision correction, because there the difference between standard and total keratometry stops being a rounding error. After myopic corneal refractive surgery, total keratometry from the same instrument sits about 0.78 D below standard K on average, against 0.03 D in the same eyes before surgery.1 In a series of 341 eyes with prior myopic laser vision correction, the mean gap was 0.50 D.2

    This review follows the posterior cornea through the calculation. It starts with what a keratometer really measures, explains why formulas predict the posterior surface rather than ignore it, lays out the evidence comparing predicted and measured posterior values in virgin and operated eyes, and then describes the error that ties everything together: counting the posterior cornea twice. The last sections turn the evidence into a rule the surgeon can apply at the biometer, and say which value the MIOLO calculators expect.

    2. What a Keratometer Actually Measures

    2.1 The keratometric shortcut

    No keratometer measures corneal power. It measures the radius of curvature of the anterior surface, usually from the reflection of a ring or a set of points over the central cornea, and converts that radius into dioptres with a fixed keratometric index, 1.3375 in most instruments:

    K = (nk − 1) / ranterior = 337.5 / ranterior

    r in millimetres, K in dioptres. The index 1.3375 is not the refractive index of the cornea (1.376); it is a fictitious value chosen so that the anterior radius alone yields an approximate power of the whole cornea, posterior surface included.

    The choice of index embeds an assumption: that the posterior surface has a fixed relation to the anterior one, so that its negative contribution can be absorbed into a single conversion factor. In a population study of 2,429 eyes, the mean ratio between anterior and posterior corneal radii was 1.22, and the index that reproduced the measured posterior power in each eye was 1.3273, below the Gullstrand value of 1.3315 the study tested and further still from the 1.3375 of most keratometers.3 Theoretical work has since shown that 1.3375 overestimates the total corneal power in most clinical situations,4 and direct estimation in eyes implanted with zero-power lenses returned an effective index of 1.329 to 1.331.5 In an unoperated cornea these discrepancies are small, systematic, and absorbed by the lens constants that every formula optimises on real outcomes. That is why standard keratometry works as well as it does in routine eyes.

    2.2 What myopic ablation does to the shortcut

    Myopic laser vision correction flattens the anterior surface and leaves the posterior surface essentially where it was. The ratio between the two radii rises, and the larger the correction, the larger the rise.4 A fixed index calibrated on virgin corneas now converts a flatter anterior radius into a power that still assumes the old posterior contribution, and the result is a standard K that overestimates the total corneal power. Scheimpflug analysis of 36 eyes after myopic excimer surgery showed that the average K underestimated the surgically induced refractive change, while total corneal refractive power in the central 3.0 mm zone and the 2.0 mm ring tracked it closely.6 With a swept-source OCT biometer, the change in standard K fell short of the refractive change at the corneal plane by 0.65 D, whereas the change in total keratometry followed it.1

    The consequence for lens power is well known. A formula that estimates the effective lens position from K reads a flat cornea, predicts a more anterior lens, and chooses a weaker implant. With the standard SRK/T after LASIK or PRK, the mean postoperative error was +1.82 D of hyperopia; using the pre-ablation K for the lens position estimate reduced it to +0.13 D.7 Single-K formulas would have left most post-LASIK eyes hyperopic.8 The Figure 1 sketch summarises the two ideas: the index is a shortcut across the posterior surface, and ablation changes the ground the shortcut was drawn on.

    Unoperated cornea: one measured surface, one assumed surfaceSchematic cross-section of an unoperated cornea. The keratometer measures the anterior radius and converts it with a fixed index that stands in for the posterior surface: K equals 337.5 divided by the anterior radius.aUnoperated corneaAnterior (measured)Posterior (assumed)LightK = 337.5 / ranteriorThe fixed index stands in for the posterior surfaceAfter myopic ablation: flattened anterior surface, unchanged posterior surfaceSchematic cross-section after myopic ablation. Central tissue is removed and the anterior surface is flattened while the posterior surface is unchanged, so the fixed ratio between the surfaces no longer holds and standard keratometry overestimates the total power.bAfter myopic ablationAnterior (ablated)Posterior (unchanged)Tissue removedLightStandard K overestimates the total powerThe fixed ratio between the two surfaces no longer holds
    Figure 1. Two surfaces, one keratometric number. a, A keratometer measures the anterior radius (solid line) and converts it with a fixed index that stands in for the unmeasured posterior surface (dashed line). b, Myopic ablation removes central tissue (shaded blue) and flattens the anterior surface, so the ratio between the two surfaces rises; the dotted line is the original profile and the posterior surface stays in place, so the relation the index relied on no longer holds. After myopic laser vision correction, total keratometry from a swept-source OCT biometer was on average 0.78 D below standard K, against 0.03 D before surgery.1 Proportions are exaggerated for clarity.

    3. Posterior Corneal Astigmatism: The Term the Calculator Adds

    Power is only half of the story. The posterior cornea also has astigmatism of its own, and its orientation is strikingly constant. In 715 corneas measured with a dual-Scheimpflug analyser, the mean posterior astigmatism was 0.30 D, and the steep posterior meridian was vertical in 86.6% of eyes, whereas the steep anterior meridian was vertical in only 51.9%. With age the anterior meridian drifts from vertical to horizontal; the posterior one does not.9 Because the posterior surface is a negative lens, a vertically steep posterior meridian behaves optically as against-the-rule astigmatism: it partially cancels with-the-rule anterior astigmatism and adds to against-the-rule anterior astigmatism. Anterior measurements alone underestimated the total corneal astigmatism by 0.22 D at 180 degrees, and the error exceeded 0.50 D in 5% of eyes.9

    The clinical signature followed immediately. Toric lenses selected on keratometric astigmatism overcorrected with-the-rule eyes by 0.59 D and undercorrected against-the-rule eyes by 0.32 D; when the total corneal astigmatism replaced the keratometric value, both errors lost statistical significance.10 Ray-traced total corneal power from a dual-Scheimpflug device lacked the two biases that the automated keratometer showed.11 The regression that turned this observation into everyday practice, the Abulafia-Koch formula, estimates the total corneal astigmatism from standard keratometry alone: applied to two commercial toric calculators, it reduced the centroid error in predicted residual astigmatism from 0.55 D against-the-rule to 0.05 D and raised the proportion of eyes within ±0.50 D from 30.8% to about 77%, matching the Barrett Toric Calculator, which carries its own internal prediction.12 In a comparison of six modern toric formulas in 823 eyes, the Kane formula led with 65.6% of eyes within ±0.50 D, and Barrett, Abulafia-Koch and EVO 2.0 did not differ from one another.13

    Three facts from this literature govern everything that follows. First, the posterior astigmatism is real and clinically relevant. Second, its regularity is what allows a formula to predict it from the anterior surface without measuring it. Third, and this is the point that the rest of this review turns on, the prediction is a term that the calculator adds to whatever astigmatism it receives. Figure 2 shows the arithmetic.

    With-the-rule anterior astigmatism: the posterior term applied once, and applied twiceVector sketch in the double-angle plane for with-the-rule anterior astigmatism. Row a, correct input: anterior astigmatism of 1.50 D minus a predicted posterior term of 0.30 D gives a net of 1.20 D. Row b, wrong input: the total of 1.20 D is entered and the posterior term is applied again, so 0.90 D is used and 0.30 D is lost.With-the-rule anterior astigmatismaCorrect inputbWrong inputWTRATRWTRATRanterior 1.50 Dposterior term 0.30 Dnet 1.20 Dentered total 1.20 Dposterior term again 0.30 Dused 0.90 D: 0.30 D lostAgainst-the-rule anterior astigmatism: the posterior term applied once, and applied twiceVector sketch in the double-angle plane for against-the-rule anterior astigmatism. Row a, correct input: anterior astigmatism of 1.00 D plus a predicted posterior term of 0.30 D gives a net of 1.30 D. Row b, wrong input: the total of 1.30 D is entered and the posterior term is applied again, so 1.60 D is used and 0.30 D is added.Against-the-rule anterior astigmatismaCorrect inputbWrong inputWTRATRWTRATRanterior 1.00 Dposterior term 0.30 Dnet 1.30 Dentered total 1.30 Dposterior term again 0.30 Dused 1.60 D: 0.30 D added
    Figure 2. The posterior term is added to whatever the calculator receives. Illustrative arithmetic in the double-angle plane, where with-the-rule astigmatism points left and against-the-rule points right; a vertically steep posterior cornea acts as against-the-rule astigmatism. In each panel, row a shows anterior keratometry entered, with the predicted posterior term (blue) applied once; row b shows total astigmatism entered, with the same term applied to a value that already contained the posterior surface, so the cylinder used for planning (dark red) loses or gains 0.30 D. Vectors are aligned for simplicity. The posterior term of 0.30 D is the population mean magnitude reported by Koch and colleagues,9 not the value any particular calculator uses; commercial regressions scale the term with the anterior astigmatism and other biometric variables. A 0.30 D shift at the corneal plane is close to half a toric step and moves the recommended cylinder by one step in borderline eyes.

    4. The Three Families of Corneal Power on a Printout

    Every corneal value a device can print belongs to one of three families, and the family, not the brand of the instrument, decides whether a calculator may add a posterior term to it. Table 1 organises them.

    Table 1. Three families of corneal power and how a toric calculator should treat each
    FamilyWhat it containsTypical labelsWhere it comes fromIn a calculator that predicts the posterior cornea
    Anterior keratometryAnterior radius converted with a fixed index (1.3375 in most instruments). The posterior surface is assumed, not measured.K1, K2, Km, K, SimK, standard K, SKKeratometers, optical biometers, Placido topographers, Scheimpflug and OCT tomographersYes. This is the input the calculator was built for.
    Total corneal powerBoth surfaces measured and combined, by ray tracing or thick-lens optics, over a declared zone. The posterior surface is already inside.TK (swept-source OCT biometer), TCRP and TNP (Scheimpflug), TCP (dual Scheimpflug with Placido), total corneal power (OCT tomographers)Devices that image the posterior surfaceNo. Use only in a version of the formula built for that value.
    Posterior keratometryPosterior radius converted to power; negative, close to −6 D, with its own astigmatism of about 0.3 D.PK, posterior K1 and K2, PCASwept-source OCT biometers, Scheimpflug and OCT tomographersOnly in a measured-posterior mode, together with anterior K from the same device.

    Two properties of the total-power family deserve emphasis, because both are invisible on the printout. The first is that total power depends on the zone and the optical method, so the same cornea yields different totals from different instruments. After small-incision lenticule extraction, simulated keratometry agreed between a swept-source OCT biometer and a Scheimpflug tomographer, but total keratometry from the biometer was 1.34 to 1.37 D higher than the true net power and 0.87 to 0.95 D higher than the total corneal refractive power of the tomographer, and the authors concluded that the two devices cannot be used interchangeably for total values.14 Anterior, posterior and total astigmatism were likewise not interchangeable between an OCT-based and a Scheimpflug-Placido biometer.15 Even within one Scheimpflug device, the zone changes the number: after myopic ablation, the total corneal refractive power in the 3.0 mm zone and the 2.0 mm ring tracked the induced refractive change, while the 3.0 and 4.0 mm rings overestimated it,6 and a 4.0 mm apex-centred zone predicted the change after SMILE and femtosecond LASIK with mean errors of 0.05 to 0.15 D.16 The no-history algorithm of Potvin and Hill was built specifically on the true net power in a 4.0 mm apex-centred zone,17 and paraxial Gaussian totals and ray-traced totals diverge after refractive surgery as a function of the anterior curvature and the posterior-to-anterior ratio.18 A total value is therefore not a portable number: it belongs to the instrument, the zone and the formula it was validated with.

    The second property concerns noise. Posterior measurements from rotating Scheimpflug cameras carry more measurement noise than swept-source OCT: in pseudophakic eyes, the difference vector between predicted and subjective cylinder was 0.70 D with Scheimpflug tomography against 0.43 D with OCT, and keratometry plus OCT gave the best precision.19 A measured posterior value is only as useful as the device that produced it, and the evidence in the next section separates the two.

    5. Predicted or Measured? The Evidence in Virgin Eyes

    Once biometers began to measure the posterior surface, the obvious question was whether measured posterior astigmatism should replace the predicted term. Table 2 collects the comparisons published in the last five years for unoperated eyes. They are consistent.

    Table 2. Predicted versus measured posterior corneal astigmatism in virgin eyes, and total keratometry versus anterior keratometry
    StudyEyesComparisonResult
    Yang et al., 20212079Barrett Toric with predicted vs measured PCA (SS-OCT biometer), Kane toricNo difference in absolute error; in the real-world analysis the predicted version had the best result (53.2% within 0.50 D)
    Lukewich et al., 20222124Barrett predicted vs measured PCA (SS-OCT biometer)Comparable prediction errors
    Shammas et al., 202222122OCT K with predicted PCA vs OCT K with Scheimpflug-measured PCA vs Scheimpflug SimK with measured PCAMean absolute errors 0.41, 0.45 and 0.49 D; within 0.75 D in 99%, 95% and 84%
    Reitblat et al., 20232380Barrett predicted vs measured PCA, Abulafia-Koch, KaneMeasured and predicted comparable; Kane with a slight against-the-rule error
    Wang and Koch, 202324602Barrett predicted vs measured PCA (SS-OCT biometer), non-toric eyesMeasured slightly better: vector error 0.54 vs 0.57 D; 57.6% vs 52.5% within 0.50 D
    Segers et al., 202425110Barrett estimated PCA vs measured PCA from SS-OCT and from ScheimpflugEstimated PCA had the best accuracy (0.40 D), not different from SS-OCT (0.42 D), better than Scheimpflug (0.44 D)
    Stewart et al., 202426418Barrett and EVO with predicted vs measured PCA, by orientation of the posterior steep axisEVO with predicted PCA had a smaller mean absolute error than both Barrett versions and Abulafia-Koch; measured PCA helped when the posterior steep axis was not vertical (54% vs 43% within 0.50 D)
    Yang et al., 202427146Barrett predicted vs measured PCA from SS-OCT and ScheimpflugPredicted PCA significantly better: 78.1% within 0.50 D
    Jin et al., 20252894Barrett predicted vs measured PCA from three devicesSS-OCT biometer: no difference (0.39 vs 0.38 D); Scheimpflug and anterior-segment OCT: predicted better than measured
    Levron et al., 202129237 (2:1 match)Total keratometry vs anterior keratometry, one toric lens platformEquivalent outcomes; TK showed no superiority in regular corneas
    Rémignon et al., 202330201Total astigmatism estimated by Abulafia-Koch vs measured by TK, same calculatorEstimated total had smaller errors (0.46 vs 0.50 D); TK overestimated the total in with-the-rule eyes
    Ramsauer et al., 20243123Standard K vs TK (SS-OCT biometer) vs TCRP (Scheimpflug) in one toric platformK and TK equivalent for the spherical equivalent (78% within 0.50 D each); TCRP significantly worse (39%)

    The pattern has three layers. Predicted posterior astigmatism is the validated default: in most series it performed as well as the measured value or better.20-23,25,27,28 Measured posterior values from a swept-source OCT biometer are equivalent to the prediction, with one large series of non-toric eyes showing a modest advantage for measurement.24 Measured values from Scheimpflug devices, burdened by more noise, tend to be worse than the prediction.22,25,28 The one situation in which measurement has been shown to help is the minority of eyes whose posterior steep axis is not vertical, the eyes the regression cannot anticipate,26 although a smaller series found the prediction still ahead even when the posterior steep meridian was horizontal.27 Total keratometry used as the astigmatism input did not beat anterior keratometry with a good prediction, and in with-the-rule eyes it overestimated the total.29-31

    None of this argues against measuring the posterior cornea. It argues for putting the measurement where the calculator expects it. When a calculator has no posterior model at all, as with the original manufacturer toric calculators, feeding it a ray-traced total corneal refractive power improved the vector error compared with anterior keratometry alone,32 exactly as the total-astigmatism model of Savini and Næser did.10 The posterior surface was counted once in both cases; only the place where it entered differed.

    6. After Refractive Surgery: Two Different Questions

    In an eye with prior LASIK or PRK, the posterior cornea enters the calculation twice, in two different roles, and the evidence answers each role differently. The first role is the spherical power: how much total corneal power the formula should assume when it estimates the lens power. The second is the cylinder: how much posterior astigmatism it should add to the anterior astigmatism when it chooses the toricity and the axis. Conflating the two is the commonest source of confusion in this field.

    6.1 The spherical power: measured posterior data help

    For the lens power, the no-history formulas of the last decade have progressively incorporated measured posterior curvature, and the largest datasets favour the versions that do so. Table 3 summarises them.

    Table 3. Spherical IOL power after laser vision correction: formulas with and without measured posterior data
    StudyEyesSettingMain finding
    Wang et al., 201933129Myopic, hyperopic and RK eyes; K vs TKAfter myopic LASIK/PRK, Haigis-TK (0.50 D) and Barrett True-K (0.54 D) had lower errors than Haigis (0.72 D); no differences after hyperopic LASIK/PRK or RK
    Kenny et al., 20243485Myopic LVC; biometer-embedded True-K TK, EVO 2.0, PEARL-DGS, Hoffer QSTTrue-K TK and EVO 2.0 were the only formulas with mean errors not different from zero, with performance comparable to existing formulas
    Anter et al., 202435576Myopic LVC; 11 no-history formulas with and without PK or TKVersions with PK or TK outperformed versions without; EVO 2.0-PK, PEARL-DGS-PK and True-K-TK on top
    Suraneni et al., 202636153Hyperopic LVC; 12 formulasTrue-K-TK, K6-TK and EVO 2.0-PK ranked highest; without PK or TK, EVO 2.0-K, True-K NH-K and PEARL-DGS were the most reliable
    Cooke et al., 2026 (AJO)373,738Myopic LVC; 14 formulas, multicentre databaseFormulas with posterior curvature had lower errors: K6 0.37 D, PEARL-DGS, EVO 2.0 and True-K 0.38 D, against 0.40 to 0.49 D without; the additional gain of the newest online formulas over Barrett True-K was concentrated in eyes with AL below 23.5 mm or K above 43 D and described as clinically negligible in typical eyes
    Cooke et al., 2026 (JRS)38507Post-LVC eyes with TK minus K between 0.00 and 0.06 DFormulas for hyperopic LVC using K were the most accurate regardless of reported history; K-based versions outperformed PK and TK variants in this band
    Liang et al., 202539108Myopic LVC stratified by axial lengthNo differences below 28 mm; at 28 mm or more, Haigis-TK with modified axial length and PEARL-DGS were superior, and most formulas drifted myopic in extremely long eyes

    The reading is clear but modest. Measured posterior curvature improves the spherical prediction after myopic laser vision correction by a few hundredths of a dioptre of mean absolute error in the largest series; in the same series, the further gain of the newest online formulas over Barrett True-K was concentrated in short eyes and steep corneas and described as clinically negligible in typical eyes.37 The improvement exists because a posterior measurement replaces an assumption that ablation has broken. It is not a licence to type total keratometry into any formula: each of the versions in Table 3 was built and validated for its own input, and in the ambiguous band where total and standard keratometry almost coincide, the K-based versions won.38

    6.2 The cylinder: the posterior cornea after LASIK behaves like a virgin one

    The cylinder question has a different premise. A predicted posterior term is a regression derived from unoperated corneas. If ablation altered the posterior astigmatism, or altered its relation to the anterior astigmatism beyond repair, the prediction would lose its footing in operated eyes and the measurement would have to replace it. The evidence says otherwise. After femtosecond LASIK for myopia, the central posterior curvature remained unchanged at one and six months, and the posterior astigmatism changed minimally.40 After femtosecond and microkeratome LASIK, the measured change in posterior astigmatic power was 0.11 and 0.08 D.41 In 341 eyes with prior myopic laser vision correction measured on a swept-source OCT biometer, the posterior astigmatism averaged 0.29 D and was vertically steep in 91.5% of eyes, the orientation, and close to the magnitude, reported in virgin corneas, which led the authors to conclude that the assumption built into toric formulas holds in these eyes.2

    Outcomes agree. In 56 eyes after myopic and 19 after hyperopic LASIK or PRK, toric lenses left 80% and 84% of eyes with 0.50 D or less of refractive astigmatism, using biometry from an instrument that measures only the anterior surface.42 And in 2026 a direct comparison appeared: in 182 eyes with prior myopic laser vision correction, the Barrett True-K Toric formula with predicted posterior astigmatism produced a smaller refractive astigmatism prediction error than the same formula with measured posterior astigmatism, while for the EVO Toric formula the two did not differ. Measured posterior data improved the spherical equivalent prediction of both formulas, exactly as Table 3 would lead one to expect, and the authors described its effect on the Barrett True-K cylinder prediction as detrimental.43

    Table 4. The posterior cornea and the toric lens after laser vision correction
    StudyEyesQuestionFinding
    Cao, Wang and Koch, 20204275Do toric lenses work after LASIK/PRK?80% (myopic) and 84% (hyperopic) with refractive astigmatism of 0.50 D or less
    Bohac et al., 202141200Does LASIK change posterior astigmatism?Measured changes of 0.11 and 0.08 D; too small to explain the refractive deficit
    Low et al., 202440284Is the posterior surface stable after LASIK?Central posterior curvature unchanged at 1 and 6 months; minimal change in posterior astigmatism
    Ng, Stewart and Yeo, 20252341Does post-LASIK posterior astigmatism look like a virgin one?Mean 0.29 D, vertically steep in 91.5% of eyes; the assumption in toric formulas is reinforced
    Stewart and Yeo, 202643182Predicted or measured posterior astigmatism in toric formulas after myopic LVC?Barrett True-K Toric: predicted better than measured for astigmatism (P below 0.01); EVO: no difference; measured improved the spherical equivalent of both

    The two answers are not in tension. For the lens power, measured posterior curvature repairs an index assumption that ablation broke; for the cylinder, the predicted posterior astigmatism was never resting on that assumption, because the posterior astigmatism itself did not change. A surgeon who wants the best of both uses a post-refractive formula version that consumes total or posterior keratometry for the sphere, and lets the toric algorithm keep its predicted posterior term for the cylinder. What a surgeon must not do is the subject of the next section.

    7. The Double-Counting Error

    Three errors share one mechanism: a value that already contains a correction receives the same correction again. They differ only in which correction is duplicated. Table 5 names them.

    Table 5. Three ways the posterior cornea gets counted twice
    ErrorHow it happensWhat it doesHow to avoid it
    Double posterior astigmatismTotal astigmatism (from TK, TCRP, TNP or TCP) entered as the K1 and K2 of a calculator that predicts the posterior corneaThe posterior term is applied to a value that already had it: undercorrection of with-the-rule eyes and overcorrection of against-the-rule eyes, by up to one toric step; the axis can shift as wellEnter anterior keratometry (K, SimK) into calculators that predict the posterior cornea; use total values only in a total-K or measured-posterior mode
    Double index correctionTotal keratometry entered as the K of a post-refractive no-history formula that corrects standard K internallyThe corneal power is lowered twice; the lens comes out stronger than needed by roughly the TK to K gap (about 0.5 to 0.8 D on average after myopic ablation)1,2 and the eye ends up myopicGive a K-based formula standard K; give a TK-based version TK. Never both in one calculation
    Mixed scalesA total value from one device or zone entered into a version validated on anotherOffsets of 0.87 to 1.37 D between total values of two devices after SMILE,14 and zone-dependent differences within one device6Use each total-power version only with the instrument and zone it was built for

    The first error deserves a worked example, because it is silent: the calculator accepts the numbers, the printout looks right, and the result is plausible. Take an eye with anterior keratometry of 43.00 D at 180 and 44.50 D at 90, that is, 1.50 D of with-the-rule anterior astigmatism. The posterior surface, steep vertically, contributes about 0.30 D of against-the-rule effect, so the total corneal astigmatism is close to 1.20 D. A calculator that predicts the posterior cornea receives the anterior values, subtracts its estimate, arrives at roughly 1.20 D and selects the toric cylinder accordingly. Now suppose the surgeon copies the total keratometry instead: 43.15 D and 44.35 D, say, with 1.20 D of astigmatism. The calculator subtracts its posterior estimate again and plans for 0.90 D. The eye is undercorrected by about 0.30 D at the corneal plane, which is close to half a toric step, and in a borderline eye it moves the recommendation down one step. In an against-the-rule eye the arithmetic runs the other way and the lens is one step too strong, leaving a with-the-rule residual, the flip that patients tolerate least. The bottom row of Figure 2 shows both cases.

    The second error compounds the first in post-refractive eyes. A no-history formula built for standard K estimates how much the index shortcut overstates the operated cornea and corrects for it. Total keratometry has already made that correction: that is what it is for. Entered into a K-based post-myopic formula, it lowers the corneal power twice, and since lens power moves by roughly a dioptre for each dioptre of corneal power, the implant comes out about 0.5 to 1.0 D too strong, with a myopic surprise to match. The same values entered into the TK version of the same formula give the intended result, because that version was built without the internal correction.

    The third error is the least intuitive, because it involves no calculator at all. Total corneal power is not a physical constant of the eye; it is the output of an optical model over a chosen zone. Two devices can agree on the anterior surface and disagree on the total by more than a dioptre,14 and two zones of one device can disagree after ablation by amounts that matter.6 A formula version validated with one of these values inherits its scale. Using another value in its place is a calibration error, not a measurement error, and no amount of care at the biometer repairs it.

    8. Which K Goes Where: A Practical Rule

    The decision reduces to identifying what the calculator expects, then reading the matching family off the printout. Figure 3 lays it out, and the checklist that follows adds the handful of details that the flowchart cannot hold.

    Flowchart: which corneal value belongs in which type of calculatorDecision flow with three branches chosen by the formula version: anterior K only, where the calculator predicts the posterior cornea and receives K1 and K2 and never total values; anterior plus posterior K, the measured-posterior mode with both surfaces from the same device; and total corneal power, a version built for that value with no predicted posterior term. After LASIK or PRK the same rule applies with one index correction per calculation. The posterior cornea enters once.What does this calculator expect?Read the formula version, not the device brandA. Anterior K onlyThe calculator predicts theposterior cornea itself(standard mode of most toric calculators)B. Anterior K plus posterior KMeasured-posterior mode:the prediction is switched off(offered by some calculators)C. Total corneal powerA TK, TCRP or TCP versionbuilt for that value(device-specific versions)Enter K1 and K2 with axesK, SimK or standard K, index 1.3375Never TK, TCRP, TNP or TCPthey would count the posterior twiceEnter both surfacessame device, same sessionPrefer swept-source OCTScheimpflug posterior values are noisierEnter the total valuefrom the device and zone it was built forNo predicted posterior termnot interchangeable between devicesAFTER LASIK OR PRK: THE SAME RULE, PLUS ONEA no-history formula that expects K corrects the index shortcut itself; a TK version already contains that correction.Give each version the value it was built for: one index correction and one posterior term per calculation.The posterior cornea enters once.
    Figure 3. Which corneal value belongs in which calculator. The branch is chosen by the formula version, not by the biometer: the same swept-source OCT biometer prints standard K, posterior K and total keratometry, and each belongs to a different branch. Total values are device and zone specific.6,14,15

    8.1 A checklist before pressing Calculate

    Identify the family. Find K1 and K2 on the printout and confirm they are anterior values. If the page also lists TK, TCRP, TNP, TCP or posterior K, leave those where they are unless the formula version you are using was built for them.

    Confirm the index. Calculators expect anterior keratometry converted with 1.3375. Some instruments can be configured to report with a different index; the conversion factors for 1.3375 and 1.332, that is 337.5 and 332, differ by a ratio of 1.0166, which at 43 D amounts to about 0.7 D of corneal power. A calculator that offers an index setting exists for this reason; if yours does not, make sure the biometer reports with 1.3375.

    Keep surfaces together. If a measured-posterior mode is used, take anterior and posterior values from the same device in the same session. Posterior and total values are not interchangeable across instruments,14,15 and a mixed pair is a mixed scale.

    After refractive surgery, choose the version first. Decide whether you are using a K-based or a TK-based version of the post-refractive formula, and give it that value only. For the cylinder, the predicted posterior term remains the best-supported choice after myopic laser vision correction.2,43

    Do not add what the device already did. If a calculator reports that a total or posterior value was used, do not also tick a predicted-posterior option. If two calculators disagree by a toric step on the same eye, the first suspect is the input family, not the formula.

    Repeat before you trust. When the astigmatism magnitude or axis differs between devices by more than the toric step, or the axis moves more than about ten degrees between examinations, repeat the measurement after treating the ocular surface. Vector averaging of keratometry and OCT data has given the best precision for planning.19

    9. Which K the MIOLO Calculators Expect

    The MIOLO Standard, MIOLO Toric and MIOLO TK calculators expect standard anterior keratometry. This is why the line under the keratometry fields asks for K or SimK and refuses total corneal power values (TK, TCRP, TNP and TCP).

    The same rule is enforced upstream of the form. When a biometry printout is read automatically to fill the fields, the reader is instructed to take the standard keratometry and to ignore total keratometry lines even when they appear in the same block, so that a value from the wrong family cannot enter by that route.

    We do not offer a measured-posterior or total-keratometry mode today, and the decision is deliberate. The best evidence available in 2026 shows no cylinder benefit from measured posterior astigmatism after myopic laser vision correction and a spherical benefit measured in hundredths of a dioptre in the largest series,37,43 such a mode could not yet be validated with the discipline the platform applies to everything else, and every additional field is an additional way to enter the wrong family. When measured posterior data can be taken directly from the examination file, without a surgeon having to choose a mode, and validated against postoperative refraction, the question will be revisited on those terms.

    10. Conclusion

    The posterior cornea has become measurable, and the measurement is valuable when it replaces a broken assumption: after myopic ablation, total or posterior keratometry improves the spherical power prediction in formulas built to receive it. For the cylinder the story is different, because the posterior astigmatism did not change with the ablation, and the predicted term that toric calculators derived from virgin corneas keeps its footing in operated ones. The 2026 comparison in post-LASIK eyes found predicted posterior astigmatism at least as good as measured, and better in the Barrett True-K Toric formula.

    What connects the two findings is not a preference for prediction or measurement but a rule of bookkeeping: the posterior cornea enters a calculation once. Standard keratometry belongs in calculators that predict it; total values belong in versions built for them, from the device and zone they were validated with; and no calculation should receive both. A surgeon who reads the printout by family rather than by number closes the commonest silent error in post-refractive toric planning without adding a single measurement.

    Enter the K the formula expects. The MIOLO calculators are free to use: bring a post-LASIK case with its standard keratometry, compare the plan with the tools you already trust, and draw your own conclusions.

    MIOLO TKMIOLO Toric

    About the author

    Bruno Miolo, MD, M.Sc, is an ophthalmologist and cataract surgeon and the developer of the MIOLO.AI platform.

    Conflicts of interest: Bruno Miolo is the developer of the MIOLO.AI platform. No external funding was received for this article.

    Methodological note: Figures 1 to 3 are schematic. The vectors in Figure 2 and the worked example in Section 7 use a posterior term of 0.30 D, the population mean magnitude reported by Koch and colleagues, for illustration only; commercial regressions scale the term with the anterior astigmatism and other variables. Every reference was checked against its PubMed record in September 2026, and the results quoted are those stated in the published abstracts. No patient data were used.

    How to cite: Miolo BL. Counting the Posterior Cornea Once: Total Keratometry, Posterior Corneal Astigmatism and the Toric IOL After Refractive Surgery. MIOLO.AI Blog. September 2026. Available at: https://miolo.ai/article/posterior-cornea-total-keratometry-post-lasik

    References

    1. Wang Z, Song Y, Yang W, et al. Comparing standard keratometry and total keratometry before and after myopic corneal refractive surgery with a swept-source OCT biometer. Front Med (Lausanne). 2022;9:928027. PMID 35903314
    2. Ng AW, Stewart S, Yeo TK. Patterns of corneal astigmatism in eyes post myopic laser keratorefractive surgery. Clin Ophthalmol. 2025;19:1139-1147. PMID 40190999
    3. Fam HB, Lim KL. Validity of the keratometric index: large population-based study. J Cataract Refract Surg. 2007;33(4):686-691. PMID 17397744
    4. Gatinel D, Debellemanière G, Saad A, Rampat R. Theoretical relationship between the anterior-posterior corneal curvature ratio, keratometric index, and estimated total corneal power. J Refract Surg. 2023;39(4):266-272. PMID 37040212
    5. Gatinel D, Hoffmann PC, Cooke DL, et al. Leaving trusted paths: estimating corneal keratometric index in cataract surgery eyes with zero-power implants. Graefes Arch Clin Exp Ophthalmol. 2024;262(8):2533-2539. PMID 38456928
    6. Savini G, Hoffer KJ, Carbonelli M, Barboni P. Scheimpflug analysis of corneal power changes after myopic excimer laser surgery. J Cataract Refract Surg. 2013;39(4):605-610. PMID 23465330
    7. Aramberri J. Intraocular lens power calculation after corneal refractive surgery: double-K method. J Cataract Refract Surg. 2003;29(11):2063-2068. PMID 14670413
    8. Wang L, Booth MA, Koch DD. Comparison of intraocular lens power calculation methods in eyes that have undergone LASIK. Ophthalmology. 2004;111(10):1825-1831. PMID 15465542
    9. Koch DD, Ali SF, Weikert MP, Shirayama M, Jenkins R, Wang L. Contribution of posterior corneal astigmatism to total corneal astigmatism. J Cataract Refract Surg. 2012;38(12):2080-2087. PMID 23069271
    10. Savini G, Næser K. An analysis of the factors influencing the residual refractive astigmatism after cataract surgery with toric intraocular lenses. Invest Ophthalmol Vis Sci. 2015;56(2):827-835. PMID 25587061
    11. Zhang L, Sy ME, Mai H, Yu F, Hamilton DR. Effect of posterior corneal astigmatism on refractive outcomes after toric intraocular lens implantation. J Cataract Refract Surg. 2015;41(1):84-89. PMID 25532636
    12. Abulafia A, Koch DD, Wang L, et al. New regression formula for toric intraocular lens calculations. J Cataract Refract Surg. 2016;42(5):663-671. PMID 27255241
    13. Kane JX, Connell B. A comparison of the accuracy of 6 modern toric intraocular lens formulas. Ophthalmology. 2020;127(11):1472-1486. PMID 32371252
    14. Liu Q, Yang W, Li D, et al. Total and simulated keratometry measurements using IOLMaster 700 and Pentacam AXL after small incision lenticule extraction. Int Ophthalmol. 2024;44(1):237. PMID 38902457
    15. McLintock C, Uprety S, McKelvie J. Corneal astigmatism agreement between a swept-source ocular coherence tomography and Scheimpflug-Placido based optical biometers. Int Ophthalmol. 2025;45(1):83. PMID 40045024
    16. Gyldenkerne A, Ivarsen A, Hjortdal JØ. Assessing the corneal power change after refractive surgery using Scheimpflug imaging. Ophthalmic Physiol Opt. 2015;35(3):299-307. PMID 25703372
    17. Potvin R, Hill W. New algorithm for intraocular lens power calculations after myopic laser in situ keratomileusis based on rotating Scheimpflug camera data. J Cataract Refract Surg. 2015;41(2):339-347. PMID 25661127
    18. Wang L, Mahmoud AM, Anderson BL, Koch DD, Roberts CJ. Total corneal power estimation: ray tracing method versus Gaussian optics formula. Invest Ophthalmol Vis Sci. 2011;52(3):1716-1722. PMID 21071742
    19. Hoffmann PC, Abraham M, Hirnschall N, Findl O. Prediction of residual astigmatism after cataract surgery using swept source Fourier domain optical coherence tomography. Curr Eye Res. 2014;39(12):1178-1186. PMID 25310575
    20. Yang S, Byun YS, Kim HS, Chung SH. Comparative accuracy of Barrett toric calculator with and without posterior corneal astigmatism measurements and the Kane toric formula. Am J Ophthalmol. 2021;231:48-57. PMID 34116010
    21. Lukewich MK, Murtaza F, Somani S, Tam ES, Chiu HH. Comparison of Barrett toric calculations using measured and predicted posterior corneal astigmatism in cataract surgery patients. Clin Ophthalmol. 2022;16:1739-1751. PMID 35673350
    22. Shammas HJ, Yu F, Shammas MC, Jivrajka R, Hakimeh C. Predicted vs measured posterior corneal astigmatism for toric intraocular lens calculations. J Cataract Refract Surg. 2022;48(6):690-696. PMID 34561361
    23. Reitblat O, Barnir M, Qassoom A, Levy A, Assia EI, Kleinmann G. Comparison of the Barrett toric calculator using measured and predicted posterior corneal astigmatism and the Kane and Abulafia-Koch calculators. J Cataract Refract Surg. 2023;49(7):704-710. PMID 36888558
    24. Wang L, Koch DD. Comparison of accuracy of a toric calculator with predicted vs measured posterior corneal astigmatism. J Cataract Refract Surg. 2023;49(1):29-33. PMID 35916540
    25. Segers MHM, Abulafia A, Webers VSC, et al. Accuracy of toric intraocular lens calculations using estimated versus measured posterior corneal astigmatism. Am J Ophthalmol. 2024;262:107-113. PMID 38000619
    26. Stewart S, Yeo TK, Moutari S, McNeely R, Moore JE. Accuracy of toric intraocular lens formulas with measured posterior corneal astigmatism of different orientations. Am J Ophthalmol. 2024;266:26-36. PMID 38705551
    27. Yang S, Park J, Whang WJ, Byun YS, Kim HS, Chung SH. Accuracy of toric intraocular lens calculators with predicted and measured posterior corneal astigmatism across different types of astigmatism. Ophthalmol Ther. 2024;13(7):1877-1889. PMID 38581606
    28. Jin A, Zhang J, Zhang Y, et al. Effect of posterior corneal astigmatism measured with different biometers on toric IOL power calculation. J Refract Surg. 2025;41(10):e1032-e1041. PMID 41065748
    29. Levron A, El Chehab H, Agard E, Chudzinski R, Billant J, Dot C. Impact of measured total keratometry versus anterior keratometry on the refractive outcomes of the AT TORBI 709-MP toric intraocular lens. Graefes Arch Clin Exp Ophthalmol. 2021;259(5):1199-1207. PMID 33449216
    30. Rémignon CH, Levron A, Agard E, et al. Impact of total corneal astigmatism estimated with the Abulafia-Koch formula versus measured with a SS-OCT biometer on the refractive outcomes of a toric intraocular lens in cataract surgery. J Refract Surg. 2023;39(3):171-179. PMID 36892238
    31. Ramsauer M, Luft N, Vounotrypidis E, Priglinger SG, Mayer WJ. Accuracy of toric intraocular lens power calculation depending on different keratometry values using a novel network based software platform. Front Med (Lausanne). 2024;11:1363286. PMID 38665295
    32. Davison JA, Potvin R. Refractive cylinder outcomes after calculating toric intraocular lens cylinder power using total corneal refractive power. Clin Ophthalmol. 2015;9:1511-1517. PMID 26316693
    33. Wang L, Spektor T, de Souza RG, Koch DD. Evaluation of total keratometry and its accuracy for intraocular lens power calculation in eyes after corneal refractive surgery. J Cataract Refract Surg. 2019;45(10):1416-1421. PMID 31399324
    34. Kenny PI, Kozhaya K, Truong P, Wang L, Koch DD, Weikert MP. Performance of IOL calculation formulas that use measured posterior corneal power in eyes following myopic laser vision correction. J Cataract Refract Surg. 2024;50(1):7-11. PMID 37702522
    35. Anter AM, Bleeker AR, Shammas HJ, et al. Comparison of legacy and new no-history IOL power calculation formulas in postmyopic laser vision correction eyes. Am J Ophthalmol. 2024;264:44-52. PMID 38518990
    36. Suraneni S, Bleeker AR, Anter AM, et al. Intraocular lens power calculations in posthyperopic laser vision correction eyes: comparing legacy and new no-history formulas. Am J Ophthalmol. 2026;281:118-125. PMID 40935225
    37. Cooke DL, Kaufmann CL, Englisch CN. No-history IOL power calculation after myopic laser vision correction: a multicenter retrospective database study of 14 formulas with and without posterior corneal curvature. Am J Ophthalmol. 2026;291:403-415. PMID 42556630
    38. Cooke DL, Wendelstein JA, Riaz KM. Biometry-guided IOL formula selection in post-laser vision correction eyes. J Refract Surg. 2026;42(8):e815-e823. PMID 42573221
    39. Liang J, Tan Q, Wu Z, et al. Accuracy of modern IOL power calculation formulas in eyes after myopic laser vision correction: a stratified axial length analysis. J Refract Surg. 2025;41(5):e435-e443. PMID 40340674
    40. Low A, Kadir AJ, Chow ZY, Khang TF, Singh S. Posterior corneal surface stability after femtosecond laser-assisted in situ keratomileusis in patients with myopia and myopic astigmatism. Indian J Ophthalmol. 2024;72(8):1118-1123. PMID 39078954
    41. Bohac M, Biscevic A, Gabric I, Gabric K, Shijakova V, Patel S. Predicted and measured changes in posterior corneal astigmatism after uncomplicated femtosecond assisted LASIK (FsLASIK) and microkeratome LASIK correction for myopia and low astigmatism. Semin Ophthalmol. 2021;36(8):832-838. PMID 34010104
    42. Cao D, Wang L, Koch DD. Outcome of toric intraocular lenses implanted in eyes with previous corneal refractive surgery. J Cataract Refract Surg. 2020;46(4):534-539. PMID 32271520
    43. Stewart S, Yeo TK. Accuracy of toric intraocular lens formulas in eyes with previous myopic laser vision correction. J Refract Surg. 2026;42(6):e557-e564. PMID 42261924