Proof that Warm Compresses Work: (Though I prefer to say as warm/hot as you can stand it without hurting/burning your skin)

Before Meibography came out as a way for you to see your own meibomian glands, studies showing Warm Compresses work were limited to subjective signs and symptoms of improvement.

Meibomian gland disease (MGD) has for decades been treated in part by warm compresses. These are readily self-administered but takes time: thus most patients do not do warm compresses even if prescribed by their MD/OD.

The hypothesis was that the delivery of meibum oil out of the meibomian gland into the tear film depended on its viscosity, and this can be influenced by manipulation of lid temperature.

Goggle-based heating devices have been developed,  one containing an array of light-emitting diodes producing infra-red radiation (Mori et al., 1999) and the other using electrically-heated plates (Goto et al., 2002), but are not really used widely.


External heating (and even cooling) affects meibomian gland delivery by its effect on the viscosity and flow characteristics of the meibum oil.

The effect of cooling the lids is less obvious than that of heating. Human meibomian oil is a complex mixture of lipid classes and individual components, which as a result melts over an extended range of temperature (Tiffany, 1987, 1995). Cooling of the surface skin of the lids by 7–88 C will probably not increase the viscosity of oil within the glands to the point where little is delivered to the lid margin without undue compressive force, but there is clearly a reduction in the marginal material which is detected by meibometry. However, we cannot at this stage determine whether this reduction represents a change in oil issuing from the glands, or a chilling of material on the lid margin resulting in a lower pick-up on the meibometry tape.


How Hot Should They Be:
More on that soon: it's complicated & controversial.
Some studies say the hotter the better but trying to find good studies in good journals.

How Long Should They Be On For:
Same as above: more soon.


References:
1.

A. Nagymihályi, S. Dikstein, J.M. TiffanyThe influence of eyelid temperature on the delivery of meibomian oil

Exp. Eye Res., 78 (2004), pp. 367-370

M. Mitra, G.J. Menon, A. Casini, S. Hamada, D. Adams, C. Ricketts, E.T. Fuller, J.R. FullerTear film lipid layer thickness and ocular comfort after meibomian therapy via latent heat with a novel device in normal subjects

Eye (Lond.), 19 (2005), pp. 657-660,

4. Discussion

As the pathophysiology of MGD becomes better understood [17], [18], [19], treatments can potentially be tailored to an individual’s presentation of MGD [20], [21]. Hyperkeratinizationof the MG occurs with both increasing age [22] and MGD [17], [18], while atrophy of the gland may be more indicative of underlying MGD [23]. A wider range of effective in-house treatments has become available for MGD. For example, a single treatment of LipiFlow can improve tear breakup time at nine months [24], and improve dry eye symptoms for at least 12 months [25], while a three session course of intense pulsed light can lead to significant improvements in lipid layer thickness and tear breakup times at 45 days [26]. However, patient applied therapies remain the mainstay treatment for MGD, as cost is often a factor, and the chronic nature of the condition requires ongoing therapy.

It had been hypothesised that those with higher MG dropout (and therefore with the least potential to produce adequate meibum) might benefit most from lipid replacement, since both latent heat goggles and warm compresses, in facilitating natural meibum flow, rely on the presence of (non-atrophied) MG. However, we found no difference in the effectiveness of three commonly used home-based MGD treatments, namely liposomal spray, latent heat goggles, and warm compress, across the range of MG dropout severity. This may reflect the relatively small number of participants in the study with very severe drop out ( >75%) in whom treatment encourages the remaining MG to release sufficient meibum to improve the clinical impression of MGD. This was observed through an improvement in LLG, which corresponded to increased non-invasive tear breakup times [16] in both the mild and pronounced MGD groups, which could provide symptomatic relief [27], [28]. A poor lipid layer correlates strongly with dry eye symptoms [29], [30] therefore increasing the available lipid would be expected to be beneficial.

The results of this study would suggest that an individual’s treatment plan can be selected from amongst the three tested treatments according to patient preference, including factors such as cost or ease of use, and thereby encourage optimal treatment compliance in the home environment.

In our study, approximately half of all subjects showed an improvement in LLG of at least one grade while none showed a decrease in LLG quality following treatment. The increase in LLG across all three treatment groups suggests that the methods, whether by effecting expression of natural meibum, or by supplementing the tear fluid with artificial lipid components, contribute similarly to the natural tear lipid layer. In cases where pre-treatment non-continuous lipid layers can be made continuous post-treatment, this may decrease the rate of tear evaporation [16]. The control group ( <5% MG dropout), too, showed an increase in lipid layer thickness, as has been reported in other studies [31], [32]. Such universal improvements support the idea that there is considerable redundancy in lipid production [32], [33], via the number of concurrently active MG [34], [35]. While there must be a threshold beyond which remaining MG are unable to maintain a sufficient lipid layer [4], our results demonstrate that even individuals with >40% MG dropout can benefit from therapies which increase natural meibum release from the remaining glands.

Such a strong relationship between MG dropout and symptoms score was unexpected, as the relationship between clinically measured parameters, and dry eye symptoms, is often poor [36], [37], [38]. Due to the investigator-masked nature of this study, only objective measures of dry eye were compared between pre- and post-treatment time periods and, of the range of measures taken, lipid layer thickness and tear breakup times, both important predictors of dry eye symptoms [27], [29], showed measurable improvements. Both the single-visit, and in-clinic nature of the study limits conclusions that can be drawn about longer-term and at home benefits of the treatments, and highlights the need for future studies of increased duration that explore optimal treatment application frequency.