Development of a reliable surgical quality assurance tool for gastrectomy in oncological trials.

BACKGROUND: Despite its recognized importance, there is currently no reliable tool for surgical quality assurance (SQA) of gastrectomy in surgical oncology. The aim of this study was to develop an SQA tool for gastrectomy and to apply this tool within the ADDICT Trial in order to assess the extent and completeness of lymphadenectomy. METHODS: The operative steps for D1+ and D2 gastrectomy have been previously described in the literature and ADDICT trial manual. Two researchers also performed fieldwork in the UK and Japan to document key operative steps through photographs and semi-structured interviews with expert surgeons. This provided the steps that were used as the framework for the SQA tool. Sixty-two photographic cases from the ADDICT Trial were rated by three independent surgeons. Generalizability (G) theory determined inter-rater reliability. D-studies examined the effect of varying the number of assessors and photographic series they rated. Chi-square assessed intra-rater reliability, comparing how the individual assessor's responses corresponded to their global rating for extent of lymphadenectomy. RESULTS: The tool comprised 20 items, including 19 anatomical landmarks and a global rating score. Overall reliability had G-coefficient of 0.557. Internal consistency was measured with a Cronbach's alpha score of 0.869 and Chi-square confirmed intra-rater reliability for each assessor as < 0.05. CONCLUSIONS: A photographic surgical quality assurance tool is presented for gastrectomy. Using this tool, the assessor can reliably determine not only the quality but also the extent of the lymphadenectomy performed based on remaining anatomy rather than the excised specimen.

Incisional hernia repair in a high-fidelity silicone model for open retro-muscular mesh implantation with preparation of the fatty triangle: validation and educational impact study.

BACKGROUND: Incisional hernia repair requires detailed anatomic knowledge. Regarding median subxiphoidal hernias, the proper preparation of the fatty triangle is challenging. To foster proficiency-based training, a cost-efficient model for open median retromuscular mesh repair resembling the human body was developed, including the main anatomical structures related to the procedure. The aim is to create and validate a high-fidelity model on open retromuscular mesh repair suitable for "training before doing". MATERIALS AND METHODS: Different types of fabrics for imitation of connective tissue and 2-component silicones were used to construct the incisional hernia model. Sample size for validation of the model was determined by a triangular testing approach. Operations from six beginners and six experts were assessed by three blinded-raters. Reliability and construct-validity were evaluated on a behaviorally anchored rating scale (highest score: 4) for the criteria: "instrument use", "tissue handling", "near misses and errors", and "end-product quality". RESULTS: The model authentically mimicked an open median retromuscular mesh repair. Participants considered the procedure realistic. Reliability was excellent, ranging from 0.811 to 0.974 for "end-product quality", and "tissue handling" respectively. Construct-validity was confirmed with experts significantly outperforming beginners in the "use of instruments" (Mbeg. = 2.33, Mexp. = 3.94, p < 0.001), "tissue handling" (Mbeg. = 2.11, Mexp. = 3.72, p < 0.001), "near misses and errors" (Mbeg. = 2.67, Mexp. = 3.67, p < 0.001), and "end-product quality" (Mbeg. = 2.78, Mexp. = 3.72, p < 0.001). Criterion-validity revealed a paradox effect: beginners performed significantly better than experts (p < 0.05) when preparing the fatty triangle. CONCLUSIONS: The model covers all relevant aspects involved in median-open retromuscular incisional hernia mesh repair. Performance differences between beginners and experts confirm construct-validity and thereby realism of the model. It enables to efficiently improve and practice technical skills of the demanding surgery.

Validation and educational impact study of the NANEP high-fidelity simulation model for open preperitoneal mesh repair of umbilical hernia.

OBJECTIVE: The aim of the study was to develop, validate and analyze the educational impact of a high-fidelity simulation model for open preperitoneal mesh repair of an umbilical hernia. The number of surgical simulators available for training residents is limited. Primary for ethical reasons and secondary for the emerging pay-per-quality policies, practicing-on simulators rather than patients is considered gold standard. Validated full-procedural surgical models will become more and more important in training residents. Such models may assure that evidence-based standards regarding technical aspects of the procedures become integral part of the curriculum. Furthermore, they can be employed as a quality control of residents' skills (Fonseca et al. in J Surg Educ 70:129-137, 2013). METHODS: In a repeated measures design, medical students, residents in their last year of training and attending surgeons performed an open preperitoneal mesh repair on the NANEP model [NANEP stands for the German acronym Nabelhernien-Netzimplatation-Präperitonal (English: Umbilical hernia mesh implantation preperitoneal)]. Subjects were categorized as "Beginners" (internship students) or "Experts" (residents and surgeons). Content validity was analyzed by criteria of subject-matter-experts. Blinded raters assessed surgical skills by means of the Competency Assessment Tool (CAT) using the online platform "CATLIVE". Differential validity was measured by group differences. Proficiency gain was analyzed by monitoring the learning curve (Gallagher et al. in Ann Surg 241:364-372, 2005). Post-operative examination of the simulators shed light on criterion validity. RESULTS: The NANEP model-proofed content and construct-valid significant Bonferroni-corrected differences were found between beginners and experts (p < 0.05). Beginners showed a significant learning increase from the first to the second surgery (p < 0.05). Post-operative examination data confirmed criterion validity. CONCLUSION: The NANEP model is an inexpensive, simple and efficient simulation model. It has highly realistic features, it has been shown to be of high-fidelity, full-procedural and benchtop-model. The NANEP model meets the main needs of surgical educational courses at the beginning of residency.

Validation of an oncoplastic breast simulator for assessment of technical skills in wide local excision.

BACKGROUND: Simulation enables safe practice and facilitates objective assessment of technical skills. However, simulation training in breast surgery is rare and assessment remains subjective. The primary aim was to evaluate the construct validity of technical skills assessments in wide local excision (WLE). METHODS: Surgeons of different grades performed a WLE of a 25-mm palpable tumour on an in-house synthetic breast simulator. Procedures were videotaped (blinded), reviewed retrospectively, and independently rated against a procedure-specific global rating scale by two consultant breast surgeons. Specimen radiographs were obtained and the macroscopic distance from the 'tumour' edge to the resection margin was recorded in four cardinal directions. Expert consensus was used to construct an Oncoplastic Deviation Score (ODS), assigning points for excessively wide (more than 10 mm) and, conversely, close (less than 5 mm) macroscopic margins. RESULTS: Thirty-four surgeons (12 consultant surgeons, 12 specialty trainees and 10 core trainees) participated in the study. Video-based rating scores varied hierarchically with operator expertise (P < 0.050). Inter-rater reliability was excellent (α ≥ 0.80, P < 0.050 for all scales), and inter-rater agreement was moderate (κ = 0.132-0.361, P < 0.050 for all scales). Statistically significant differences were observed on pairwise comparisons between each grade of surgeon in scores for 'exposure', 'skin flap development', 'glandular remodelling', 'skin closure' and 'final product review' (P < 0.050). Consultants received significantly fewer ODS points than specialty trainees (P = 0.012) and core trainees (P = 0.028). Compared with experts (median 9.0 mm), wider margins were observed amongst specialty trainees (median 12.0 mm) and narrower margins amongst core trainees (median 7.1 mm) (P = 0.001). CONCLUSION: Video ratings of performance and a proposed ODS differentiate surgeons based on technical skills in WLE and may be useful for objective assessment of breast surgery trainees.