Better one or two? A systematic review of portable automated refractors

Author: Samanta A, Shetty A, Nelson PC.

Geographical coverage: Unclear

Sector: Biomedical

Sub-sector: Treatment

Equity focus: None – most of studies considered population 18+

Study population: Devices for establishing refraction, comparing portable automated refractors to a range of other devices.

Review type: Other review

Quantitative synthesis method: Narrative review

Qualitative synthesis method: Not applicable

Background: More than 400 million people suffer from visual impairment globally, with more than half due to uncorrected refractive error. Autorefraction (AR) is the most common examination performed prior to prescribing glasses. As technology advances, so has the accuracy and number of portable autorefractors available. Portable technology has become acutely important with the coronavirus pandemic and the conversion of in-person clinical evaluations to remote telemedicine encounters, with patients and providers wanting to do as much as possible remotely.

Objectives: To conduct a systematic literature review of the accuracy and effectiveness of available portable automated refractors compared to the current standard of care, subjective refraction (SR).

Main findings:

Overall, findings showed numerous portable autorefractors available, but few have been compared against SR. Quicksee and SVOne are the most accurate and patient-preferred devices. Quicksee was the most accurate, and it performed clinically the same as SR in some reports.

A total of 12 articles comparing a portable AR device to at least one form of SR were included in the review.

The authors reviewed various portable AR devices and found that while the Netra was the least expensive, it had the worst performance due to long testing times, low patient preference, and large standard deviation in accuracy. The Retinomax had the fastest testing speed, but it was costly and did not provide the most accurate measure. The Quicksee and SVOne were identified as the most cost-efficient, accurate, and preferred by patients. The Quicksee, in particular, was the most accurate of the portable devices reviewed and performed clinically the same as Standard Refraction (SR) in some reports. However, the authors noted that the quality of studies varied, with Durr and Agarwal’s study of the Quicksee being of the highest quality.

The author identified limitations of included studies and highlight some implications for future research i.e. devices need to be tested with pediatrics population, or with patients who have experienced previous eye surgeries.

Methodology:

Inclusion criteria was not stated explicitly in the review but can be discerned from the reported results. Inclusion criteria consisted of studies comparing a portable AR device to at least one form of SR. Exclusion criteria included all commentaries, responses and letters to the editor; studies pertaining to hand-held video refractors, photo-screening and paediatric amblyopia screening. Only studies in English were included.

A systematic literature search was performed in April 2020 using PubMed, EMBASE and ClinicalTrials.gov for publications reporting on portable autorefractors. A search of recent publications was performed, including terms “autorefractor” OR “automated refraction” OR “auto refraction” and “portable” OR “hand-held” OR “smartphone” for all searchable categories.

Pertinent information from each study, such as refraction time and differences in spherical equivalent refraction, were compiled and summarised. Studies were evaluated, and the GRADE system used for categorising the relative strength of each. Analysis involved narrative review, although no details are included on how this was conducted in practice.

Applicability/external validity: The authors note that devices do not consider all populations, such as children and those who have undergone previous eye surgeries.

Geographic focus: No geographic restriction to search was noted; however, it is not possible from the article to discern whether any of the included studies were conducted in low- and middle-income countries.

Summary of quality assessment:

The review provides limited methodological information, making it difficult to evaluate the robustness of the reported evidence. The study selection process has several limitations. The search was confined to English-language sources, and there’s no indication that unpublished materials or references from included studies were considered. The selection of articles for inclusion appears to have been conducted by a single reviewer. The quality assessment of individual studies is not tabulated, although it is occasionally mentioned in the text. The analysis was a narrative review, but there’s no evidence that more than one reviewer extracted the data. Due to these factors, we have low confidence in the review’s findings.