LETTER TO THE EDITOR: Comments on “The Thickness of the Tear Film”

Dear Editor,

After consultation with one of the senior authors of a recent mini-review on the thickness of the human tear film,¹ I offer three clarifying remarks.

First, the title and content (see Table 1) of the manuscript by King-Smith et al.¹ inadvertently create the impression that there is a single precorneal tear-film thickness for humans (PrCTT). The apparent reason is that sample sizes in such measurements are small, typically several subjects. In one example where a somewhat larger subject size was employed (i.e., 24 subjects), a wide variability of tear thickness emerged, ranging from 4.2 to 24 μm.² Reporting only an average thickness over all subjects (e.g., as in entry 9 of Table 1¹) is, therefore, misleading. The theory of Wong et al.³ provides a simple explanation as to why considerable variation in PrCTT is to be expected. According to Wong et al.,³ the PrCTT is proportional to the radius of curvature of the tear meniscus, R. Clearly, the volume of tear in the tear menisci increases with increasing R (i.e., measuring R is tantamount to assessing the volume of tear present in the tear menisci²). We expect that the volume of tear in the menisci is determined primarily by the tear production rate, a factor likely to exhibit considerable variation among different subjects. In addition, the theory of Wong et al.³ predicts that the PrCTT depends on the tear surface tension and viscosity, in addition to the upper lid speed. All of these factors vary from subject to subject. Accordingly, the thickness of the human precorneal tear film should not be thought of as exhibiting a single value. How wide the thickness variation might be remains be established. This same remark also holds for the thickness of the tear film deposited over contact lenses.²

Second, King-Smith et al.,¹ amplifying the comments of Braun and Fitt,⁴ argue that evaporation from the thin McDonald-Brubaker black lines formed at the two lid menisci after each blink⁵ can lead to tear rupture there (see Fig. 5A¹) and, therefore, to low values of breakup times (BUT). Black lines are caused by a suction pressure in the lid-margin menisci; hence, they are universal for all subjects.³, ⁶ Calculations show that the tear film initially deposited directly adjacent to the menisci thins to 1 μm within less than 0.1 s.³, ⁶ At this thickness, the local film depression can appear black under fluorescein instillation without the tear film actually rupturing. Confirmatory evidence (e.g., from staining or otherwise) is lacking for repetitive breakup in the black lines.

These lines are most likely very thin regions and are black under fluorescein because of their thinness.

A related misconception is that the tear film is most vulnerable to rupture near the black lines.¹, ⁴, ⁷ However, black lines form in a narrow band, about 0.1 mm in width,⁶ and steeply approach back to the normal tear-film thickness within that width span. This means that the tear film is of normal thickness in less than 0.1 mm outside the black lines. Thus, formation of the black spots and streaks normally associated with BUT measurements should have no preference for occurring in the proximity of the black lines.

Lastly, the proposal that tear-film rupture arises from apical epithelium asperities is appealing. However, the picture painted by King-Smith et al.¹ is not complete. If such asperities are water wet, then mechanical rupture is not possible. That is, completely water-wet asperities maintain a thin film of tear held in place by repulsive thin-film forces.⁸ Thus, local tear rupture is more likely when the peaks of the asperities exhibit at least some partial nonwetting (i.e., exhibit attractive thin-film forces⁸). Local nonwetting peaks might arise from incomplete mucin spreading and/or contamination from tear lipid or from airborne debris during a prior rupture event.⁹, ¹⁰ Asperity-induced breakup does provide an explanation why rupture spots and streaks do not rapidly expand as on a smooth nonwetting surface. Unfortunately, the asperity argument demands repeated breakup at the same locations upon each blink. Although repeated rupture locations are found on some subjects,¹¹ the majority of breakup occurs randomly both in specific location and in time of appearance. Further, for both evaporative and mechanical rupture over a protuberance, the dimensions of the local corneal roughness must be quite large. To justify rupture over an asperity within a 5-s blink time and with a typical evaporation rate of 1 μm/min,¹² asperity apex heights must extend out to within 0.1 μm of the local tear-film thickness. Thus, for a PrCTT of 5 μm, the asperity height must be 4.9 μm. In addition, the lateral dimensions of the asperities must correlate with the observed sizes of black spots, about 0.5–1.0 mm. There is currently no direct evidence that the human cornea is peppered with a large population of such large asperities.

Sincerely,