Surgeon Preference and Clinical Outcome of 3D Vision Compared to 2D Vision in Laparoscopic Surgery: Systematic Review and Meta-Analysis of Randomized Trials

Abstract

INTRODUCTION

The use of laparoscopy in abdominal surgery aims to reduce the time to postoperative recovery and limit wound complications.¹ Hence, in many procedures, laparoscopy is now preferred over open surgery.^2,3 Although laparoscopy displays have improved, the loss of 3-dimensional (3D) vision is widely recognized as an important downside of laparoscopic surgery. The lack of depth perception hampers the more complex surgical steps such as intracorporeal anastomoses and other suturing.^4–11 In recent years, 3D vision laparoscopy systems have become widely available and aimed to overcome the disadvantage of 2-dimensional (2D) vision.

In 2016 and 2017, 2 systematic reviews concluded that more studies are needed to compare the clinical benefits of 3D over 2D vision, as only 3 clinical randomized controlled trials (RCTs) were available.^12,13 Retrospective comparative studies have suggested a decrease in surgical time, blood loss, complications, and length of hospital stay with 3D vision.^14–20 However, side effects of 3D vision have been considered a major hurdle during the early implementation phase.²¹ Stereopsis blindness is prevalent in 5% to 10% of the population and stereopsis declines with advancing age (1%–30%).²² Furthermore, headache and tiredness were major side effects in the active shutter glasses used in displays before 2012.²¹ It is only since the development of the high definition (HD) passive polarized display, implemented during 2012 to 2014, that a “near natural” 3D view could be reached.²¹ Systematic reviews on 3D vision during laparoscopic surgery including HD passive polarized display are lacking.

This systematic literature review and meta-analysis aim to address the added value of 3D vision compared to 2D vision among patients undergoing laparoscopic gastrointestinal (GI) surgery in terms of surgical and clinical outcomes along with surgeon preference.

We hypothesized that 3D vision, as compared to 2D vision, would reduce operative time (OT) and estimated blood loss (EBL) attributed to the enhanced depth perception.

METHODS

RESULTS

Search Results

The literature search identified 1108 studies, of which 430 were duplicates. References within the included studies were cross-examined. A total of 678 studies were screened of which 44 were assessed for eligibility, resulting in 25 RCTs to be included in the final analysis.^{10,19,26–48} (Fig. ​(Fig.11).

Open in a separate window

FIGURE 1.

Flow diagram of included studies.

Operative Time

Studies comparing 3D and 2D vision in OT (n = 24) comprised 1575 versus 1428 patients, respectively. The mean OT was shorter with 3D vision: 113 (±31) versus 122 (±39) minutes, respectively (−8.0%; SMD, −0.22; 95% CI, −0.37 to −0.06; P = 0.007; n = 3003; I² = 75%). Subgroup analysis on the OT showed significant superiority in favor of 3D laparoscopy in bariatric and general surgery with subgroup overall effect sizes of −0.74 (−16.3%; 95% CI, −1.28 to −0.21; P = 0.006; 2 studies; n = 58; I² = 0%) and −0.18 (−6.7%; 95% CI, −0.34 to −0.01; P = 0.036; 9 studies; n = 1056; I² = 41%), respectively. The outcomes are summarized in Figure ​Figure22A.

Open in a separate window

FIGURE 2.

Forest plot demonstrating the impact of 3D versus 2D vision for operative time and blood loss in minimally invasive abdominal surgery. (A) Forest plot demonstrating the significant impact of 3D vision for operative time (SMD = −0.22, P = 0.01) in subgroup bariatric (SMD = −0.74, P = 0.01)^29,36 and general surgery^{26,27,33,38,39,41,43,44,46} (SMD = −0.18, P = 0.04) and in several individual RCTs in upper GI^10,28,42,47 and colorectal surgeries.^34,40 Surgical procedure: ^aesophagectomy; ^bhiatal hernia repair; ^c(radical) gastrectomy; ^dspleen preserving splenic hilar lymphadenectomy; ^ecolorectal resection; ^ftotal mesorectal excision; ^ghemicolectomy; ^htotal colectomy; ⁱmini-gastric bypass; ^jRoux-en-Y gastric bypass; ^kcholecystectomy; ^lcholecystectomy and appendectomy; ^mtransabdominal preperitoneal inguinal hernia repair; ⁿappendectomy; ^odistal pancreatectomy. (B) Forest plot demonstrating the significant impact of 3D vision for estimated blood loss (SMD = −0.33, P = 0.06) in subgroup colorectal surgery⁴⁰ (SMD = −0.31, P = 0.09) and in several individual RCTs in upper GI surgery.^{30,32,37,42,47} Surgical procedure: ^aesophagectomy; ^b(radical) gastrectomy; ^cspleen preserving splenic hilar lymphadenectomy; ^dhemicolectomy; ^echolecystectomy; ^ftransabdominal preperitoneal inguinal hernia repair.

Individual RCTs within the specializations showed significant effect size in OT, while the group effect size was not found to be significant. In upper GI, we identified 2 RCTs on gastric surgery,^42,47 1 on esophagus surgery,²⁸ and 1 on hiatus hernia repair¹⁰ in favor of 3D vision in terms of OT. Similarly, in colorectal surgery, we identified 2 RCTs^34,40 in favor of 3D vision in terms of OT. In general surgery, RCTs on inguinal hernia (n = 1) and appendectomy (n = 1) were in favor of 3D vision in terms of OT. None of the RCTs on cholecystectomy (n = 7) and HPB surgery (n = 1) showed a significant benefit in OT of 3D vision.

Risk of Bias

In Figure ​Figure3,3, the risk of bias assessment is highlighted in the included studies. The risk of confounding bias was serious for 2 studies (8%) and for the greater part of the studies low (18/25) and unclear (4/25), accounting for 88%. Bias with respect to the selection of participants was low in the majority of the studies (88%) and unclear in the remaining 3 studies (12%). The bias in terms of classification of the interventions was serious in 2 studies and critical in 1 study (4%), while the greater portion was low (76%) and moderate (12%). Risk of bias due to missing data was critical in one study, whereas the majority were low (64%), moderate (16%), and unclear (16%). Moreover, the risk of bias with respect to deviation from intended interventions was critical in one study and the remaining were low (72%), moderate (4%), and unclear (20%). The risk of bias in the measurement of outcomes was predominately low (92%) and moderate in 2 studies. Finally, the risk of bias in the selection of the reported result was serious in 2 studies and generally low (68%) and moderate (16%). The overall risk of bias of the included studies was for the greater amount low (56%) and moderate (20%) with 4 studies having serious (16%) and 2 critical (8%) risks. For more details, see Supplemental Figure 3, see http://links.lww.com/AOSO/A320.

Open in a separate window

FIGURE 3.

Risk of bias graph of the included studies.

DISCUSSION

This systematic review and meta-analysis including 25 RCTs with 3003 patients compared surgeon preference and clinical outcome of 3D vision with those of 2D vision in patients undergoing laparoscopic GI surgery. Our findings demonstrated that 3D vision, with mainly the newer HD technology, reduced OT by 8.0%. This benefit was especially seen in patients undergoing bariatric and general surgery. In 13 of 15 RCTs, most surgeons reported a preference for 3D vision. No impact of 3D vision on other clinical outcomes was detected.

The 2 previous systematic reviews, published in 2016 and 2017, included only 2 and 3 clinical RCTs and can therefore not be compared to the present systematic review, which included 25 clinical RCTs.^12,13 Our findings regarding reduced OT confirm the findings of retrospective comparative studies, but the current study found no significant reduction in blood loss, complications, and length of hospital stay with 3D vision.^14–20 Clearly, the difference in procedures studied should be taken into account. It may well be that with future, larger RCTs, the differences observed, such as the nonsignificant reduction in blood loss, may prove to be significant. With regards to the intraoperative experience by the surgeons, some of the earlier reported complaints of 3D systems, such as dizziness, nausea, eye strain, and discomfort, were found to be associated with older 3D systems.⁶ Moreover, a comparative study demonstrated lower visual fatigue severity score in experienced surgeons using 3D vision as compared to 2D vision.⁴⁹ The new HD systems for 3D vision do not have the previously reported disadvantages.⁵⁰ The current systematic review found that in 87% (13/15) of RCTs, surgeons reported in favor of 3D vision. Another point to consider is the impact of the stereoscopic base size, which determines the depth of the system. A wider base is suggested to improve the viewer’s performance.⁵¹ One study assessing videos of 252 laparoscopic bile duct injuries found that 97% of all errors were associated with visual misperception.⁵² This finding underlines the notion that optimal vision during laparoscopic surgery may improve the quality and safety of laparoscopic surgery. In a retrospective study on laparoscopic gastrectomy, the preliminary outcome of the initial experience with 4K resolution showed comparable short-term effects to 3D vision.⁵³ Similarly, an RCT showed no difference in OT and error scores during laparoscopic cholecystectomy between HD 3D vision and 4K 2D vision.³⁹ In contrast, a simulation-based RCT with 66 participants showed better performance with 3D vision, even when compared to 4K 2D vision.⁵⁴ Notably, in the present study, RCTs comparing 3D vision to 4K 2D vision in the clinical setting showed no difference in outcomes.^39,46,48

The implementation of 3D vision in laparoscopic GI surgery has demonstrated shorter OT and favorable experience over traditional 2D vision. This may suggest a potential rise in the use of the 3D visual system in GI laparoscopy. However, the technology has been available for several years, it has not been widely adopted in the field of surgery. While some surgeons attribute this to the low display resolution of the 3D system available in their hospital settings, it is important to note that the majority of the RCTs included in the present systematic review compared the performance of 3D vision and 2D vision using HD 3D imaging displays. Another contributing factor could be ascribed to the notion that growing up with a particular technology often involves becoming accustomed to the use of it. Therefore, experienced laparoscopic surgeons may be closely accustomed to 2D vision. Consequently, it is probable that the newer generation of surgeons will be more receptive to innovative technological systems of this era, such as 3D vision. Furthermore, as “less experienced” surgeons often have longer absolute OTs, 3D vision may have a larger beneficial impact in their hands.⁴ The 2018 EAES consensus stated that 3D vision in laparoscopic surgery has advantages in reducing OTs, cognitive load, and possibly complications. However, there were 18 ongoing RCTs that were related to this topic and already registered.⁵⁵ Large surveys assessing surgeons’ perspectives and practical utilization of 3D vision remain limited, so we currently are unclear on what trends are in the adoption of 3D vision in laparoscopic surgery. Despite this, it is anticipated that the 3D vision imaging market will experience a compound annual growth rate of approximately 4.5% during a 2022 to 2027 forecast.⁵⁶ Bariatric surgery is one of the domains driving the growth of 3D vision, especially with the increasing rates of obesity. An additional contributor is an increase in reimbursement of laparoscopic surgical procedures by insurance companies. Notably, Medicare covers up to 75% of laparoscopic procedures.⁵⁶

The results of our study should be interpreted in light of some limitations. First, the heterogeneity of the included studies was high. We aimed to limit this impact through a sensitivity analysis excluding studies that contributed to heterogeneity (outside funnel plot standard error) showing no difference. Second, the reporting bias was high, as many studies reported nonparametric outcomes not suitable for meta-analysis. However, we converted the median and interquartile range to mean and SD using Hozo et al²⁵ method. Still, all of the RCTs were performed in a clinical setting, which improves the generalizability.

Our systematic review and meta-analysis suggest that 3D vision is associated with shorter OT for laparoscopic surgery and in 87% of RCTs, surgeons favored 3D over 2D vision. Future RCTs comparing 3D and 2D vision in terms of costs and in relation to robotic surgery are recommended to further validate our findings.

ACKNOWLEDGMENT

M.G.B. is the guarantor. R.A. and M.Z. drafted the manuscript. All authors contributed to the development of methodological considerations, such as selection criteria, data extraction criteria, and the risk of bias assessment strategy. All authors read, provided feedback, and approved the final manuscript.

Supplementary Material

Footnotes

REFERENCES