USB-powered warm compresses were associated with greater short-term improvements in tear film stability and tear meniscus height than disposable self-heating masks in adults with meibomian gland dysfunction, a randomized crossover trial found. However, there was no statistically significant difference in patient-reported dry eye symptoms between the 2 treatments.
The prospective, single-center trial included 15 adults aged 18 to 40 years with dry eye disease associated with meibomian gland dysfunction who were recruited from the optometry clinic at King Saud University in Riyadh, Saudi Arabia, between July and October 2025. Eligibility required an Ocular Surface Disease Index score of at least 13 and a tear breakup time (TBUT) of 10 seconds or less in at least 1 eye. Participants were randomly assigned to receive either a USB-powered warm compress or a disposable self-heating mask for 10 minutes daily over 5 consecutive days, followed by a 2-week washout period and crossover to the alternate treatment. All 15 participants completed both intervention phases.
The protocol-designated primary outcome was noninvasive TBUT, measured using the OCULUS Keratograph 4M. Secondary outcomes included tear meniscus height, visual acuity, and patient-reported symptoms assessed using the Standard Patient Evaluation of Eye Dryness (SPEED) questionnaire. Investigators evaluated changes from baseline through Day 5 and compared outcomes between devices using repeated-measures analyses with adjustments for multiple comparisons.
Mean TBUT increased from 7.2 seconds at baseline to 9.5 seconds by Day 5 with the USB-powered device, compared with an increase from 7.3 to 7.9 seconds with the self-heating mask. The between-device difference at Day 5 was statistically significant. The pattern of improvement also differed: with the USB-powered compress, mean TBUT increased to 8.8 seconds by Day 3 and continued improving through Day 5. In contrast, the self-heating mask reached 8.3 seconds by Day 3 before declining slightly. Investigators reported a statistically significant time-by-device interaction.
Tear meniscus height showed a similar pattern. Mean measurements increased from 0.16 mm at baseline to 0.28 mm by Day 5 with the USB-powered compress, compared with an increase from 0.16 to 0.22 mm with the self-heating mask. The between-device difference was statistically significant.
Despite these differences in objective measurements, patient-reported symptoms improved modestly with both devices without a statistically significant between-device difference. Mean SPEED scores decreased by 1.5 points with the USB-powered compress and 1.2 points with the self-heating mask. By Day 5, mean scores were 5.6 and 5.9, respectively. Qualitative feedback indicated a preference for the USB-powered device because of its consistent warmth and ease of reactivation.
Visual acuity remained stable throughout both intervention phases, and no adverse events were reported. Treatment sessions were conducted under supervised clinical conditions.
The greater improvements observed with the USB-powered compress may reflect its ability to provide sustained, adjustable heat. Unlike disposable self-heating masks, which generate warmth through a chemical reaction, the electrically powered device allows temperature adjustment and continuous heating. However, the investigators did not directly measure eyelid surface temperatures during treatment, limiting their ability to confirm whether differences in thermal performance accounted for the observed outcomes.
The study had several limitations, including its small sample size, single-center design, and short treatment periods, which limited assessment of long-term effectiveness. Supervised sessions also prevented evaluation of at-home adherence. The relatively young study population limited applicability to older patients and those with more severe disease. Additionally, the authors did not perform meibography or other assessments of structural and biochemical changes. The trial was registered after recruitment began, and the final enrollment of 15 participants was below the originally planned recruitment of up to 50.
The authors noted that whether the observed improvements in tear film measurements translate into meaningful, lasting benefits for patients remains uncertain. "Further research is needed to assess long-term outcomes, gland recovery, and strategies to improve adherence and cost-effectiveness," wrote lead study author Ali M. AlSaqr, of King Saud University, Riyadh, Saudi Arabia, and colleagues.
The authors declared no competing interests.
Source: PLOS One
Editor's note: The study's published abstract reports different Day 5 tear breakup time and tear meniscus height values from those in the main results and Table 1. The corresponding author confirmed, after reviewing the raw data, that the values in Table 1 are correct. These are the figures cited in this article.
