Relationship Between Sleep Position and Glaucoma Progression

Kevin Kaplowitz; Justin Dredge; Robert Honkanen


Curr Opin Ophthalmol. 2019;30(6):484-490. 

In This Article

Treatment Options

Interventions to change the effect of sleep position on disease course have already been introduced by other fields of medicine. For example, obstructive sleep apnea has been successfully improved with numerous modalities to avoid the supine position, including alarms, vests, and special pillows.[47] Other surgical fields account for head position to reduce the risk of ocular injury, particularly in spinal surgery. Elevating the head of bed, foam masks, and forceps have all been used to attempt to reduce direct orbital pressure and congestion. Given the positional increases described above, it may not be surprising that the prevalence of glaucoma was lowest in a survey among patients reporting only 7 h of sleep nightly and highest among those who slept for more than 10 h.[1] Both traditional glaucoma treatments and changes in sleep position have been evaluated as methods of reducing positional increases in IOP.

Traditional Treatment Options

Multiple studies have investigated the effect that IOP-lowering medications have on the positional increase. One study looked at newly diagnosed cases of NTG, and measured the supine positional increase before and after administering a topical prostaglandin analogue, beta-blocker, or carbonic anhydrase inhibitor.[48] The medications had no significant effect on the positional increase. One study pretreated healthy patients with medications,[36] whereas another performed a small randomized controlled trial with levobunolol.[49] None demonstrated a significant effect on positional IOP increase. Two studies focused on the effect of argon laser trabeculoplasty on the positional increase for the supine[50] and the LDP[41] positions. Both failed to find a significant protective effect.

Surgical intervention has also been proposed as a means to lower the positional increase. One study looked at the effect of unilateral phacoemulsification in nonglaucomatous eyes on the mean positional IOP increase and found that the IOP increase from the sitting to the supine position was reduced from 2 mmHg before phacoemulsification to 0.8 mmHg 1 month postoperatively.[51] The subsequent increase from supine to LDP was reduced by 45% from 2.9 mmHg preoperatively. As with other studies in this review, the sample was relatively small (n = 29), and smaller but notable reductions were seen in the fellow, unoperated eyes.

The reduction in positional increase after medications versus trabeculectomy has been compared in two studies on open-angle glaucoma. Although one found no significant difference,[52] another compared 51 cases each that had achieved an IOP 12 mmHg or less with either medications or trabeculectomy.[53] The positional increase from sitting to LDP in the nondependent eye was 3.3 mmHg with medications versus 1 mmHg with trabeculectomy. A small study found more promising results in 13 cases of bilateral open-angle glaucoma: the positional increase from sitting to supine decreased by nearly 50% from 4.1 mmHg following trabeculectomy.[54] Similar results were seen after trabeculectomy lowered the positional increase from sitting to LDP in the nondependent eye from 3.8 to 1.7 mmHg, a 55% reduction.[55]

The effect of other types of glaucoma surgery on the positional increase has been reported. One group found a 60% lower positional increase from sitting to supine following trabeculectomy compared to canaloplasty (1.7 mmHg versus 4.3 mmHg).[56] A study on nonpenetrating deep sclerectomy found that the positional increase for sitting to supine decreased from 2.9 mmHg preoperatively to 0.6 mmHg 3 months after surgery, with similarly large reductions in the LDP.[4] Apart from medication and surgery, nonocular treatments have also been proposed.

Affecting Sleep Position

One area of active research is whether different types of pillow affect the positional increase. One study compared the positional increase from supine to the LDP for the dependent eye in healthy patients.[57] They found a similar 3 mmHg rise for both hard and soft pillows, whereas another study of 28 patients with open-angle glaucoma[29] found a smaller positional increase with hard pillows.

Apart from the softness of the pillow, some groups have compared the positional increases when creating a head of bed elevation with multiple pillows versus a single large wedge pillow or elevating the actual mattress. A study including healthy controls and patients with NTG and POAG used a wedge pillow to effectively elevate the head of bed by 20 degrees.[58] The mean overnight IOP was approximately 10% lower with inclination, with no significant difference between the high and low-tension groups. The design of the pillow may affect the neck angulation, with higher IOP found with neck extension and higher still with neck flexion.[11] This may help explain the 1–2 mmHg difference when creating a 30 degree incline head up position using multiple pillows versus directly elevating the head of bed in both open-angle glaucoma[40] and healthy controls.[59] In fact, elevating the head with multiple pillows actually increased the IOP as compared to the supine position in 35% of glaucoma cases[40] and in 25% of controls.[59] In summary, a handful of studies suggest that elevating the head of the bed by 20–30 degrees may reduce the IOP by a mean of 1.9 mmHg +/− 1 mmHg in open-angle glaucoma patients[31,40,58] and 1.4 mmHg +/− 0.9 mmHg in controls[18,58,59] as compared to the supine position. Further studies with objective monitoring could help demonstrate the overall benefit of a uniform pillow, whereas the sleep position changes throughout the night.[13]

Another proposed treatment is a sleeping mask, which could reduce direct pressure to the eye or orbit. A study using continuous IOP monitoring with a contact lens sensor and a sleeping mask during prone and LDP measurements did not find a significant reduction in limbal strain (the output of the contact lens which is thought to correspond to changes in IOP) while using the mask, although they did note less fluctuation.[45]

Future Directions

There are numerous confounding variables that may explain why the connection between favored sleeping position and glaucoma progression is not stronger. First, it is not clear how much time most people actually spend in their favored sleeping position. Video analysis suggests there are at least 13 large changes (>90 degrees) in position overnight.[60] Throughout sleep, changes in position also change the amount of contact between the orbit and the bedding surface. The timing of IOP change following a shift in position is unclear, with preliminary data showing that maintaining the LDP for 30 min was associated with an additional 0–2 mmHg IOP rise as compared to measurements at only 5 min.[8,29] Another study found that the time to peak IOP was highly variable.[28] Continuous monitoring combined with video analysis will give more complete data.

Continuous monitoring may also elucidate other variables, such as if eye deviation affects IOP.[57] When one sleep position study was repeated using continuous monitoring, they were unable to replicate their original results.[61] Another study similarly detected a 4 mmHg rise in IOP after changing from sitting to supine with a tonometer, whereas the contact lens showed no increase.[38] The fluctuation in IOP during different phases of sleep is another area of active research. For instance, continuous monitoring found that measurements during non-REM sleep were higher than REM sleep.[61] Whether different types of glaucoma have similar postural responses also remains to be seen. Other variables that can be addressed in future studies include randomization of sequence measurements.[11] Individualization of treatment may be of particular concern here, as in a large study of 1 234 patients referred to a glaucoma clinic, the mean positional increase from sitting to supine was 0.29 +/− 1.8 mmHg, whereas 30% had no positional increase and an additional 23% had an IOP decrease of 1–2 mmHg in the supine position.[42] It may be quite important to help identify the smaller number of patients who have positional increases when a decrease is expected.