figure a

Introduction

The use of telemedicine (TM) in the USA increased dramatically in response to the COVID-19 public health emergency [1]. One application of TM is virtual supervision (VS), where the supervisor and supervisee who are not in the same physical location interact for an episode of patient care via synchronous audio and/or video modalities [2, 3].Virtual supervision can facilitate cost-effective mentoring by specialists from remote locations and is also useful as a teaching tool in graduate medical education (GME) [4].

The information available in the literature on VS in ophthalmology is not well described. This scoping review aims to better understand the evidence and potential role for VS in ophthalmic practice and education.

Methods

In consultation with a reference librarian, a literature search strategy was developed in accordance with Preferred Reporting Items for Systematic reviews and Meta-Analyses extension for Scoping Reviews (PRISMA-ScR) [5]. We searched six databases (PubMed, Embase, Web of Science, CINAHL, PsycINFO, and ERIC) for articles published between January 1950 and August 2022. Our search included database-specific thesaurus terms such as medical subject headings (MeSH) and Emtree, as well as keywords related to VS in ophthalmology (Table 1).

Table 1 Search strategy
Full size table

The inclusion criteria were full-text articles published in an English-language peer-reviewed journal. We included studies that involved VS between physicians or between physicians and trainees in ophthalmology. We excluded studies in which the supervisor and supervisee were in the same physical location and studies in which clinical supervision was carried out retrospectively. We also excluded studies that included supervisors who were not ophthalmology physicians.

Eligible studies were de-duplicated in EndNote (Clarivate Analytics, Philadelphia, PA) and imported into the systematic review software Covidence (Melbourne, Victoria, Australia) for screening, full-text review, and data extraction. The screening and selection process is displayed in a PRISMA flowchart (Fig. 1). Two investigators (CK and CS) independently conducted title/abstract screening, full-text review, and data extraction in Covidence. Disagreements were resolved by the senior investigator (PBG).

Fig. 1
figure 1

PRISMA* flow diagram. PRISMA-ScR, Preferred Reporting Items for Systematic reviews and Meta-Analyses extension for Scoping Reviews

Full size image

A data template was developed in Covidence to extract relevant information, including year of publication, location, study design, characteristics of study participants, sample size, and any type of outcome. Two investigators (CK and CS) independently appraised the methodological quality of the studies using the Mixed Methods Appraisal Tool (MMAT), and disagreements were resolved by the senior investigator (PBG). The Mixed Methods Appraisal Tool (MMAT) appraises studies based on five questions assessing the sampling strategy, outcome measurement, confounders, and statistical analysis of the study [6]. All studies were scored on a scale of 1 to 5 based on the MMAT.

Results

Study characteristics

The initial database search yielded a total of 2700 articles. Following duplicate removal, title and abstract screening, and full-text review, seven articles were included in our qualitative synthesis (Fig. 1): four were case reports or case series (57%) [7,8,9,10], two were cohort studies (29%) [11, 12], and one was a cross-sectional study (14%) (Table 2) [13]. The publication year of the articles ranged from 1998 to 2022. Most studies were conducted in the USA. Other studies were conducted in Finland [13], Israel [11], the UK [12], and the Philippines [9]. Three studies were prospective in design (43%) [8,9,10,11, 13], while other studies were retrospective (14%) [12] or both (14%) [7]. The objective of the studies ranged from determining the feasibility of a teleconference device (43%) [9, 11, 13], describing the use of VS in an emergency department (ED) or rural hospital (43%) [7, 8, 10], or determining whether fellows could safely operate independently with remote supervision (14%) [12].

Table 2 Characteristics of included studies
Full size table

A total of 619 patients and 367 supervisees were included (Table 3). Two studies included supervisees who were physicians, including an ophthalmic surgeon from a foreign country (14%) [9] and a general practitioner in a rural hospital (14%) [8]. Five studies included supervisees who were medical trainees, such as ophthalmology residents (43%) [7, 11, 13], vitreoretinal fellows (14%) [12], and emergency medicine (EM) residents (14%) [10].

Table 3 Detailed characteristics of included studies
Full size table

Study setting varied between the ED [5, 7, 10, 11], operating room (OR) [9, 12], rural hospital [8], and eye clinic [13]. Virtual supervision was used for a clinical examination in three studies (43%) [7, 11, 13], surgical procedure in three studies (43%) [9, 10, 12], and in-office procedure in one study (14%) [8]. For studies involving a surgical or in-office procedure, VS was used to teach surgeons endoscopic laser-assisted dacryocystorhinostomy (ELA-DCR) [9], perform orbital decompression on a patient with extensive facial trauma [10], remove a corneal foreign body and rust ring in a rural hospital [8], and supervise retina fellows repairing macula-on rhegmatogenous retinal detachments (RRD) during bank holidays and weekends [12].

Devices used for VS included an optical coherence tomography (OCT)/fundus camera [7], slit lamp camera [8, 11], network camera [13], and ceiling-mounted camera (Table 4) [10]. Synchronous communication between the supervisor and supervisee was facilitated through various modalities such as email [11], telephone [11], video conferencing software [8, 9], and local area network (LAN) [13].

Table 4 Interventions, outcomes, and MMAT rating of studies
Full size table

Four studies reported successful transmission of real-time video of surgical or in-office procedures [8,9,10, 12, 13]. In three studies on VS for clinical examinations, supervisees held synchronous communication with their supervisors to reach an agreement in diagnosis [7, 11, 13]. Studies noted that VS was helpful for diagnosing ocular surface, anterior segment, and macular diseases but difficult to use when diagnosing vitreous and peripheral retinal conditions [11, 13].

One study noted that the slit lamp camera was sufficient to provide high enough real-time image quality for procedures [8], while another study using a network camera noted decreased transmission rate due to electronic traffic [13]. Several studies also noted machine malfunction, image artifacts, insufficient image resolution, and lighting problems as factors that decreased the effectiveness of VS [7, 11, 13].

One study reported patient outcomes related to VS and found no significant difference in 6-month retinal re-detachment rate when comparing remotely supervised to unsupervised fellows performing RRD repairs on bank holidays and weekends [12].

Only one study reported educational outcomes. In this study, four out of eight residents stated that VS helped them diagnose patients more accurately, as they could synchronously share the images with the supervisor and talk through their thought processes [7].

Critical appraisal of included studies

The MMAT was used to critically appraise the quality of the studies. Limitations were noted in outcome measurement [8, 9], statistical analysis [8,9,10], sampling strategy [11, 13], and inclusion of confounding factors [11, 13]. Overall, two studies scored a 5/5 [7, 12], one study scored a 4/5 [10], and four studies scored a 3/5 [8, 9, 11, 13].

Discussion

There is a limited evidence base on VS in ophthalmology. The few available studies suggest that VS is technologically feasible, may provide a positive experience for both supervisees and patients, and can permit synchronous communication and transmission of clinical data, which can be used to formulate diagnosis and management plans or learn new surgical skills.

Virtual supervision can improve diagnostic accuracy by allowing fast, synchronous communication between supervisor and trainee. In one study, ophthalmology residents (training level unspecified) used VS to consult their supervisors regarding complicated cases during night shifts in the ED [11]. The residents communicated with their supervisors via telephone and emailed images and videos captured by a miniature slit lamp camera. The authors found high agreement in diagnosis made during the night shift and the on-site examination by the supervisor the next day. In another study, first-year ophthalmology residents sent OCT/fundus photographs to the supervisor to synchronously discuss a differential diagnosis [7]. The use of the device did not correlate with a longer duration of visit in the ED.

Virtual supervision can also impact ophthalmic surgical GME. In one study, retina fellows who were deemed ready to operate independently by their supervisors performed surgery while supervisors provided synchronous feedback from a remote location during holidays and weekends [10, 12]. This readiness was assessed on a case-by-case basis by the supervisors after 2–3 months of supervised training based on a review of recorded surgeries and didactics. The study included only primary uncomplicated macula-on RRDs; patients with more complex retinal detachments were excluded. The study reported no statistical difference in the 6-month retinal re-detachment rate between off-hour cases utilizing VS and off-hour cases without any supervision. The authors did not mention if there were any adverse events, difficulties with VS, or a safety plan if a significant complication were to occur during the surgery.

In addition, ophthalmologists can use VS to assist physicians in areas with limited access to ophthalmic care or continuing medical education (CME). In one study, a general practitioner in a remote area provided time-sensitive ophthalmic procedures with the assistance of synchronous communication with ophthalmologists [8]. In another study, ophthalmologists in the USA were able to teach a new procedure to ophthalmologists in the Philippines who had the necessary equipment but not the surgical expertise [9].

This review suggests that various methods can be used to ensure high image and video quality during VS, although challenges remain. In one study, patients were instructed to keep voluntary movements to a minimum and to fixate on a target, and a designated remote operator kept the slit lamp image focused [7]. Despite these methods, there were technological glitches, image artifacts, and internet connection issues. Most studies reported that quality of images and videos were adequate for clinical diagnosis and procedure guidance, although in some cases, supervisees needed to spend a significant amount of time to find the optimal setup [11, 13]. Furthermore, the blue filter on the slit lamp with fluorescein was particularly difficult to visualize during VS due to degradation of the video quality [7].

We identified several gaps in the literature on the use of VS in ophthalmology. First, most studies on ophthalmic GME lacked standardized curricula to implement the interventions and methodology to assess the outcomes. Surveys were often used to assess feasibility and usability of VS; only one study collected data from all supervisors, supervisees, and patients [11]. Furthermore, no studies included control groups, which makes it difficult to compare the effectiveness of VS versus direct supervision in GME. Second, most studies focused on feasibility of technology for VS rather than on clinical or educational outcomes. One study that assessed educational outcomes relied on self-reported data without a comparison group [14]. Third, there was significant heterogeneity in the types of supervisees such as residents, fellows, general practitioners, and physicians. Supervisees have varying needs based on their background and training experience, which makes it difficult to generalize the results [15].

This review has several limitations. First, we did not include studies published in gray literature or in languages other than English. Second, the reproducibility of MMAT ratings is limited by the authors’ judgments about the quality of the study design.

In conclusion, this scoping review highlights the potential utility of VS in ophthalmology and identifies areas for future research on this topic. Recent technological advancements allow supervisors to guide other physicians and trainees through VS, with potentially improved efficiency and collaboration between different institutions and countries. Future studies should employ larger sample sizes and more rigorous methodology to better define the safety and efficacy of VS in ophthalmic practice and education [16].