Knowledge of advanced reverse geometry lens design is the foundation for understanding orthokeratology.

With the rising incidence of myopia, strategies to slow myopia progression are becoming increasingly important for eyecare practitioners to understand and implement. Orthokeratology (ortho-k) is an excellent option for myopia control management as well as for treating refractive error in both children and adults. For practitioners interested in providing this service to patients, knowing the general anatomy of an ortho-k contact lens is essential.

While expertise in ortho-k lens design may seem more relevant for those planning to design custom lenses, a fundamental understanding is important for anyone planning to use ortho-k in their practice. This knowledge relates closely to the physiology and optics behind the process and therefore improves our troubleshooting techniques and our ability to field questions from patients, parents, and other healthcare professionals about the process.

Although advances in ortho-k have led to the creation of lenses designed to treat hyperopia and presbyopia (Herzberg, 2010), this article focuses on myopic ortho-k lens design.


The anatomy of ortho-k lenses has undergone a significant transformation since the 1960s, when the idea was first put into practice. Initially, it was observed that fitting standard rigid lenses flatter than the cornea for myopia, and steeper than the cornea for hyperopia, created a temporary reduction in uncorrected refractive error. This led to the development of “orthofocus”­—the precursor to orthokeratology—and soon after to the “recessed optic” lens, in which the optic zone was cut flatter than the periphery of the lens to improve centration (Mountford et al, 2004).

The most notable change in lens design occurred with the creation of the reverse geometry lens in the late 1980s, made possible by advancements in lathing technology. Today, all ortho-k lenses utilize reverse geometry in their designs, and the variations among brands stem primarily from different assumptions made about average corneal shape (Mountford et al, 2004). This article provides an overview of general patterns of design and will not delve into specific lens brands.


Modern ortho-k lenses have between four and six zones, described below in order from central to peripheral (Figure 1).

Figure 1. Tear profile (top) and profile view (bottom) of a reverse geometry ortho-k lens.

Base Curve (BC)/Back Optic Zone Radius (BOZR) This central curve is typically between 5.0mm and 6.8mm in diameter and overlies what becomes the treatment zone on the cornea, as observed on post-fitting topography. The BC/BOZR is chosen based on the amount of central corneal flattening desired and is therefore related to corneal curvature and the amount of myopia treatment required.

To determine the appropriate radius, many lens designs use a calculation called the Jessen Formula (Mountford, 1998). The first step in using this formula is to identify the amount of myopia correction desired, called the target Rx. This is often the spherical equivalent of the manifest refraction. The example below shows a manifest refraction of –3.00 –0.50 x 180. In this case, the target Rx is –3.25D.

The flat corneal meridian (flat K) is then identified in dioptric power. The BC/BOZR is made flatter than flat K by the target Rx (–3.25D) and an additional amount called the Jessen Factor (–1.25D). The Jessen Factor differs among lens designs, but ranges from 0.50D to 3.00D (Mountford et al, 2004).


Manifest Rx: –3.00 –0.50 x 180

Target Rx: –3.25D (spherical equivalent)

Flat K: 42.75D (7.90mm)
BC/BOZR = Target Rx (–3.25D) + Jessen Factor
(–1.25D) = 4.50D flatter than flat K
4.50D flatter than 42.75D = 38.25D (8.80mm)

The Jessen Factor is added to ensure that the desired treatment amount is achieved and that it lasts throughout the day. While a standard Jessen Factor may be adequate for some patients, Chan et al (2008) found that even flatter values may be required to achieve adequate treatment, especially in higher myopes.

The amount of desired apical clearance under this curve is commonly 5µm to 10µm; this will falsely appear as bearing on the fluorescein pattern, as it falls below the 20µm required to perceive fluorescein behind a contact lens (Young, 1988). Although this curve is usually spherical, an aspheric radius can be used to increase the amount of midperipheral clearance, thereby creating more peripheral myopic defocus when desired (Figure 2).

Figure 2. Tear profile of a lens with a spherical BC/BOZR (top) versus an aspheric BC/BOZR (bottom) of the same central value. Both lenses are fit for the same eye and degree of myopia. The aspheric BC/BOZR creates a deeper tear film reservoir.

Reverse Zone This 0.5mm to 1.0mm wide zone joins the BC/BOZR to the relief/alignment zones and is steeper than its neighboring curves. The geometry of the reverse zone is often a curve; however, splines and tangents are used in some designs. The clearance under this curve is commonly referred to as the “tear film reservoir,” and its depth is a function of the amount of myopia being corrected. With lower amounts of myopic correction, the tear film reservoir will be shallower than with higher amounts. For example, when correcting a –1.00D myope in one particular design, the tear film reservoir depth is calculated as 24µm, whereas it is 79µm for a –6.00D myope (with the same corneal shape) in the same lens design.

One of the primary functions of the reverse zone is to raise or lower the BC/BOZR to create the desired amount of apical clearance. If the reverse zone is too steep, there will be excessive apical clearance, resulting in a topographical central island. If it is too flat, the lens will land on the corneal apex and not in the corneal periphery, resulting in lens decentration and a decentered treatment pattern (Figures 3 and 4).

Figure 3. Three lenses with different reverse curve values; all other parameters are the same. The proper curvature creates ideal apical clearance (top); too steep causes excessive apical clearance (middle); and too flat causes apical bearing, preventing lens landing in the alignment zone (bottom).

Figure 4. Central island (left) and superiorly decentered treatment zone (right).

Relief Zone Not every orthokeratology design incorporates a relief zone, which aims to encourage proliferation of epithelial cells from the alignment zone toward the tear film reservoir. This may allow for more effective overall treatment in higher degrees of myopia. When present, the relief zone is 0.5mm to 0.7mm wide, with a depth of 10µm to 20µm.

Alignment Zone This 0.5mm to 1.0mm wide zone can be spherical, aspheric, or a tangent. The shape of the alignment zone is best determined by considering the amount of eccentricity present along the flat meridian of the midperipheral cornea. Figure 5 shows three corneas that have the same apical radius of curvature (R0) of 42.75D (7.90mm) with three different eccentricities. The lower the eccentricity, the steeper the alignment curve needs to be to provide alignment. The higher the eccentricity, the flatter the curve needs to be. The fit of this zone contributes most to proper lens centration (Caroline, 2001). This zone is where the lens lands on the eye, and minimal clearance (alignment) is desired. The greatest amount of epithelial thinning occurs under this curve; it is even greater than that in the central treatment zone (Figure 6).

Figure 5. Three corneas with the same R0 but different eccentricities (top). Eccentricity values can be used to determine the curvature of the alignment zone (bottom).

Figure 6. Epithelial changes following ortho-k treatment, easily identified on tangential display. Areas of lower elevation (blue) indicate epithelial thinning; higher elevation (red) indicates epithelial thickening.

Secondary/Peripheral Zone(s) In the periphery of a reverse geometry lens, two curves are commonly incorporated to create appropriate edge lift at the peripheral cornea. The secondary zone commonly has a width of 0.2mm to 0.5mm and a depth of approximately 20µm, but it is not included in all lens designs. Its function is to form a smooth transition between the alignment and peripheral zones. The peripheral zone also has a width of 0.2mm to 0.5mm, and its depth is usually between 80µm to 100µm.

Overall Lens Diameter The overall diameter of a reverse geometry ortho-k lens is often 0.8mm to 1.2mm smaller than the horizontal visible iris diameter (HVID). For example, if the corneal diameter is 12.0mm, the lens diameter should be approximately 11.0mm.

Lens Material Modern lens designs and ortho-k principles are based on the current practice of overnight orthokeratology lens wear. This requires the use of high-Dk lens materials to ensure proper oxygen transmission to the cornea. A lens transmissibility (Dk/t) of 87 or higher will limit overnight corneal swelling to levels comparable to those without a contact lens (Holden and Mertz, 1984). A study by Lum and Swarbrick (2011) found that using higher-Dk lens materials for overnight ortho-k not only served this physiological purpose, but improved the clinical efficacy of the treatment as well.


The primary measurements needed to design a modern ortho-k lens include manifest refraction, HVID, and corneal topography. Some lenses are designed simply from flat K measurements, while custom lens design software may require eccentricity and R0 as well. Perhaps in future designs, a combination of R0 and eccentricity may become more widely used compared to flat K, as it better represents corneal shape with regard to orthokeratology. The corneal curvature at the apex is more relevant than that measured by keratometry at a 3mm chord. The lens-to-cornea relationship at the corneal apex must be designed with a high level of precision to create the proper fluid forces responsible for effective treatment.

Ethnicity may also play a role in ortho-k lens design. For example, Wu et al (2017) found that the average eye shape profile of Chinese eyes is different (more asymmetric) from that of Caucasian eyes. This may relate to the observation made by some practitioners that peripheral curves often need to be steeper when designed for Asian eyes than for Caucasian eyes to achieve the same fit. Future studies may shed more light on variations in corneal shape among patients of different ethnic backgrounds in relation to contact lens design.


A toric ortho-k lens incorporates a spherical BC/BOZR and toricity across all other zones. The aim of using toricity is to improve lens centration and ensure lens landing 360º around the midperipheral cornea. This lens-to-cornea relationship is crucial for maintaining the fluid forces that drive epithelial changes in orthokeratology (Chen et al, 2012).

To determine whether a toric design is indicated, use corneal topography to evaluate the sagittal height difference between the flat and steep corneal meridians. If the difference between the two is greater than 30µm at a chord length where lens landing occurs, then lens toricity will contribute to a better fit (Kojima et al, 2016). The chord length at which the lens lands is defined by joining opposing sides of the alignment curve and will therefore vary depending on the particular lens design used, but it is typically 8.0mm to 9.0mm (Figure 7). Kojima et al (2016) illustrated that even patients who have relatively low amounts of corneal astigmatism may have height differentials that exceed 30µm, thus requiring toric ortho-k lenses (Figure 8). A toric design is more likely required for eyes with limbus-to-limbus astigmatism than for those with central astigmatism (Figure 9).

Figure 7. Ortho-k lenses “land” on the eye at the alignment curve. A corneal height differential > 30µm between the two primary meridians at this chord length (8.5mm in the above lens) suggests that lens toricity will improve landing 360º around.

Figure 8. Average corneal height differentials at three chord lengths of 114 eyes, organized by amount of corneal astigmatism.

Figure 9. A spherical ortho-k lens can land properly and center on an eye with central astigmatism (left), but not on an eye with limbus-to-limbus astigmatism (right).


The relative peripheral myopic defocus created by orthokeratology is believed to slow the progression of myopia (Kang and Swarbrick, 2011), but it can also be a source of bothersome aberrations. Therefore, different optic zone sizes should be used based on the effect desired (Figure 10). Most myopia control designs use a smaller optic zone, around 5.4mm, to ensure delivery of the peripheral optics through the pupil. Larger optic zone sizes of 6.0mm to 6.8mm work well for adults, as they limit the amount of peripheral defocus reaching the retina. Of course, pupil size will affect the amount of peripheral rays entering the eye and should be considered when initiating ortho-k treatment.

Figure 10. A myopia control design with a smaller optic zone (top) versus an adult ortho-k design with a larger optic zone (bottom).


The process of ortho-k has been rapidly improving over the course of its history, although many questions and limitations still remain. For example, why does the effect occur more quickly and last longer in some individuals than in others, even with the same degree of myopia? Much of the error and variation in efficacy of ortho-k likely stems from limitations in measurement and interpretation of corneal shape, but may also relate to the precision of lens manufacturing.

Many philosophies exist regarding the design of reverse geometry lenses, specifically of the alignment curve. This curve must align with the midperipheral cornea across all meridians, and so in designing this curve, an understanding of the precise shape of the midperipheral cornea is essential. Corneal eccentricity is an important variable to use, as it contributes significantly to ocular sagittal depth, even more so than does central corneal radius (Caroline and André, 2010). Ideally, corneal eccentricity values and R0 would be used to design ortho-k lenses, but this may only be possible with custom lens design software and would be difficult to incorporate into commercially available lenses.

This concern is further complicated by errors in eccentricity measurements from corneal topographers, arising from the corneal shape reconstruction algorithm used, the measurement accuracy, and the fixation and tear film quality (Mountford et al, 2002). Characterizing the cornea by “eccentricity” may also contribute to error in interpreting corneal flattening, as this model may not describe corneal shape completely.

The increasing sophistication of lathing techniques has made the manufacture of reverse geometry lenses possible and has contributed to the success of ortho-k thus far. However, some limits still exist in lathing technology, with a tolerance for computer numeric-controlled lathes of around ± 2µm to 10µm (Mountford et al, 2004). When this error is combined with that of corneal topography, it can clearly affect the outcome of the ortho-k process, based on the degree of precision required for lens fitting.


In a perfect world, corneal apical curvature and eccentricity would be reliably measured with complete accuracy, and lenses would be manufactured without error. Custom-designed ortho-k lenses would then theoretically deliver perfect treatment on the first attempt. We have not quite made it to this point yet, but if the field of ortho-k continues to develop and grow, as it has over the past few decades, we may find ourselves closer to this ideal situation than we might expect.

Along with improvements in corneal measurement and lathing techniques, other beneficial advancements in the future of orthokeratology may include increased myopia control efficacy, more user-friendly design software, and improved algorithms for analyzing corneal shape. Some corneal topographers already provide the option to “smooth out” minor irregularities on a topographical map, thereby converting it to a symmetric ellipsoid surface. This may improve ortho-k fitting accuracy, as lens designs are based on this model of eye shape, although future investigation is required to validate this idea.

There remains significant potential for further development of ortho-k lens design. Imagine the possibilities if trained practitioners could simply enter basic patient information—such as HVID, R0, corneal eccentricity, and manifest refractive error—into a sophisticated algorithm to determine optimum lens parameters. If these lenses then required minimal troubleshooting and parameter changes—and provided a robust reduction of myopia progression—ortho-k would become a mainstay treatment option offered to nearly all young myopic patients. This would lead to notable improvements in quality of life and ocular health for countless patients. CLS

For references, please visit and click on document #256.