Validity and effectiveness of augmented reality in surgical education: A systematic review

BACKGROUND

Current challenges in surgical training have led to the investigation of augmented reality as a potential method of supplementary education. However, its value for this purpose remains uncertain. The aim of this study was to perform a systematic review of the published literature to evaluate the validity and effectiveness of augmented reality in surgical education, and to compare it with other simulation modalities.

METHODS

Electronic literature searches were performed in accordance with the Preferred Reporting Items for Systematic reviews and Meta-Analyses guidelines. Two authors independently extracted pertinent data and assessed study quality. The primary outcome measures of interest were the validity and effectiveness of augmented reality as an educational tool.

RESULTS

Of 6,500 articles, 24 studies met eligibility criteria for inclusion, of which 2 were randomized. Ten studies investigated validity, with 7 establishing both face and content validity and an additional 1 just content validity. Construct validity was demonstrated in 9 of 11 studies. Of the 11 studies that examined the effectiveness of augmented reality in skills acquisition, 9 demonstrated enhanced learning. Of the 5 studies in which the effectiveness of augmented reality as an educational tool was compared with other modes of simulation, augmented reality was found to be superior in 2 and equivalent in the others.

CONCLUSION

Overall, the majority, including 2 high-quality randomized controlled trials, demonstrated the validity and effectiveness of augmented reality in surgical education. However, the quality of published studies was poor with marked heterogeneity. Although these results are encouraging, additional high-quality studies, preferably in the real-life environment, are required before the widespread implementation of augmented reality within surgical curricula can be recommended.

Patient-reported and physician-recorded bowel dysfunction following colorectal resection and radical cystectomy: a prospective, comparative study

AIM

Bowel dysfunction following anterior resection (AR) is termed low anterior resection syndrome. It is unclear whether such dysfunction occurs following other bowel/pelvic operations as well. This study aimed to characterize and compare bowel dysfunction following AR, right hemicolectomy (RH) and radical cystectomy (RC).

METHOD

A prospective study of consecutive patients undergoing AR, RH and RC (2002-2012) was performed at a tertiary referral centre in Sydney, Australia. Outcome measures included (i) patient-reported satisfaction with bowel function, self-described bowel function and self-reported change in bowel function; (ii) objective assessment of bowel function using validated criteria to identify symptoms and stratify patients into those with constipation and/or faecal incontinence (FI); and (iii) health-related quality of life (SF-36v2 Health Survey).

RESULTS

Of 743 eligible patients, 70% participated [AR, n = 338, mean age 69.6 years (SD 11.9), 59% men; RH, n = 150, 75.8 years (SD 10.5), 54% men; RC, n = 34, 71.1 years (SD 14.1), 71% men]. AR patients were three times more likely to report change in bowel function post-surgery and self-judged their bowel function as abnormal more frequently (64%) than RH patients (35%) and RC patients (35%) (P < 0.01). AR patients were four times more likely to meet criteria for concomitant constipation and FI. Patients with concomitant constipation and FI had lower physical and mental SF-36v2 scores (P < 0.001).

CONCLUSION

Bowel dysfunction occurred after RH and RC but rates were higher following AR. This suggests that low anterior resection syndrome occurs due to a direct impact of partial/complete loss of the rectum rather than just due to loss of bowel length and/or the consequence(s) of pelvic dissection.

Colonic transit in patients after anterior resection: prospective, comparative study using single-photon emission CT/CT scintigraphy

BACKGROUND

Bowel dysfunction after anterior resection is well documented, but its pathophysiology remains poorly understood. No study has assessed whether postoperative variation in colonic transit contributes to symptoms. This study measured colonic transit using planar scintigraphy and single-photon emission CT (SPECT)/CT in patients after anterior resection, stratified according to postoperative bowel function.

METHODS

Symptoms were assessed using the low anterior resection syndrome (LARS) score. Following gallium-67 ingestion, scintigraphy was performed at predefined time points. Nine regions of interest were defined, and geometric centre (GC), percentage isotope retained, GC velocity index and colonic half-clearance time (T_½ ) determined. Transit parameters were compared between subgroups based on LARS score using receiver operating characteristic (ROC) curve analyses.

RESULTS

Fifty patients (37 men; median age 72·6 (range 44·4-87·7) years) underwent planar and SPECT scintigraphy. Overall, 17 patients had major and nine had minor LARS; 24 did not have LARS. There were significant differences in transit profiles between patients with major LARs and those without LARS: GCs were greater (median 5·94 (range 2·35-7·72) versus 4·30 (2·12-6·47) at 32 h; P = 0·015); the percentage retained isotope was lower (median 53·8 (range 6·5-100) versus 89·9 (38·4-100) per cent at 32 h; P = 0·002); GC velocity indices were greater (median 1·70 (range 1·18-1·92) versus 1·45 (0·98-1·80); P = 0·013); and T_½ was shorter (median 38·3 (17·0-65·0) versus 57·0 (32·1-160·0) h; P = 0·003). Percentage tracer retained at 32 h best discriminated major LARS from no LARS (area under curve (AUC) 0·828).

CONCLUSION

Patients with major LARS had accelerated colonic transit compared with those without LARS, which may help explain postoperative bowel dysfunction in this group. The percentage tracer retained at 32 h had the greatest AUC value in discriminating such patients.

Quantification and neurochemical coding of the myenteric plexus in humans: No regional variation between the distal colon and rectum

BACKGROUND

It remains unclear whether regional variation exists in the human enteric nervous system (ENS) ie, whether intrinsic innervation varies along the gut. Recent classification of gastrointestinal neuropathies has highlighted inadequacies in the quantification of the human ENS. This study used paired wholemounts to accurately quantify and neurochemically code the hindgut myenteric plexus, comparing human distal colon and rectum.

METHODS

Paired human descending colonic/rectal specimens were procured from 15 patients undergoing anterior resection. Wholemounts of myenteric plexi were triple-immunostained with anti-Hu/NOS/ChAT antibodies. Images were acquired by motorized epifluorescence microscopy, allowing assessment of ganglionic density/size, ganglionic area density, and neuronal density. 'Stretch-corrected' values were calculated using stretched/relaxed tissue dimensions.

KEY RESULTS

Tile-stitching created a collage with average area 99 300 000 μm² . Stretch-corrected ganglionic densities were similar (colon: median 510 ganglia/100 mm² [range 386-1170], rectum: 585 [307-923]; P = .99), as were average ganglionic sizes (colon: 57 593 μm² [40 301-126 579], rectum: 54 901 [38 701-90 211], P = .36). Ganglionic area density (colon: 11.92 mm² per 100 mm² [7.53-18.64], rectum: 9.84 [5.80-17.19], P = .10) and stretch-corrected neuronal densities (colon: 189 neurons/mm² [117-388], rectum: 182 [89-361], P = .31) were also similar, as were the neurochemical profiles of myenteric ganglia, with comparable proportions of NOS+ and ChAT+ neurons (P > .10).

CONCLUSIONS AND INFERENCES

This study has revealed similar neuronal and ganglionic densities and neurochemical profiles in human distal colon and rectum. Further investigation of other components of the ENS, incorporating additional immunohistochemical markers are required to confirm that there is no regional variation in the human hindgut ENS.

Brain responses to mechanical rectal stimulation in patients with faecal incontinence: an fMRI study

AIM

Continence is dependent on anorectal-brain interactions. Consequently, aberrations of the brain-gut axis may be important in the pathophysiology of faecal incontinence (FI) in certain patients. The aim of this study was to assess the feasibility of recording brain responses to rectal mechanical stimulation in patients with FI using functional magnetic resonance imaging (fMRI).

METHOD

A prospective, cohort pilot study was performed to assess brain responses during rectal stimulation in 14 patients [four men, mean (SD) age 62 (15) years]. Blood oxygen level dependent (BOLD) signals were measured by fMRI during rest and mechanical distension, involving random repetitions of isobaric phasic rectal distensions at fixed (15 and 45 mmHg) and variable (10% above sensory perception threshold) pressures.

RESULTS

Increases in BOLD signals in response to high pressure rectal distension (45 mmHg) and maximum toleration were observed in the cingulate gyrus, thalamus, insular cortex, inferior frontal gyrus, cerebellum, caudate nucleus, supramarginal gyrus, putamen and amygdala. Additionally, activation of the supplementary motor cortex and caudate nucleus with inconsistent activity in the frontal lobe was observed.

CONCLUSIONS

This study has demonstrated the feasibility of recording brain responses to rectal mechanical stimulation using fMRI in patients with FI, revealing activity in widespread areas of the brain involved in visceral sensory processing. The observed activity in the supplementary motor cortex and caudate nucleus, with relative paucity of activity in the frontal lobes, warrants investigation in future studies to determine whether aberrations in cerebral processing of rectal stimuli play a role in the pathogenesis of FI.

Acute lower gastrointestinal haemorrhage: outcomes and risk factors for intervention in 949 emergency cases

PURPOSE

Outcomes of acute lower gastrointestinal haemorrhage (ALGIH) are mostly derived from studies performed in the sub-acute/elective rather than the emergency department (ED) setting. The aims of this study were to determine the incidence and outcomes of patients presenting to a tertiary hospital ED with ALGIH and to identify associated clinicopathological risk factors.

METHOD

A retrospective observational cohort study of consecutive patients presenting with ALGIH to a tertiary hospital ED was performed. Primary outcome measures included mortality and hospital (including high dependency [HDU]) admission. Secondary outcome measures included rates of (i) blood transfusion, (ii) radiological/endoscopic investigation(s) and (iii) therapeutic intervention.

RESULTS

ALGIH accounted for 949 (512 M, mean age 62.3 years) of 130,262 (0.73%) ED presentations, of which 285 patients (30.1%) were on anti-platelet/coagulant therapy. There were five deaths (0.5%). Hospital admission was required in 498 patients (52.5%), of which 19 (3.8%) required HDU monitoring. Hospital admission was twice as likely in males and four times more likely in patients >75 years old and those taking multiple anti-platelet/coagulant therapy (P < 0.05). Blood product transfusion was required in 172 patients (34.5%), specialist investigations in 230 (46.2%) and therapeutic intervention in 51 (10.2%) (surgery in 24 [4.8%]; endoscopic haemostasis in 20 [4.0%] and angiographic embolisation in 9 [1.8%] patients).

CONCLUSION

ALGIH accounts for 1% of all ED presentations, with half requiring hospital admission. Mortality and surgical intervention rates are low and although most patients can be managed supportively, access to interventional radiology/endoscopy is important.