Clinical translation of surgical simulated closure of a ventricular septum defect

Abstract

Objectives: To demonstrate that improvement in technical performance of congenital heart surgical trainees during ventricular septum defect (VSD) closure simulation translates to better patient outcomes.

Methods: Seven trainees were divided into 2 groups. Experienced-fellows group included 4 senior trainees who had performed >5 VSD closures. Residents group consisted of 3 residents who had never performed a VSD closure. Experienced-fellows completed 3 VSD closures on real patients as a pretest. Both groups participated in a 4-week simulation requiring each participant to complete 2 VSD closures on three-dimensional printed models per week. One month later, all trainees returned for a post-test operation in real patients. All performances were recorded, blinded and scored independently by 2 cardiac surgeons using the validated Hands-On Surgical Training-Congenital Heart Surgery (HOST-CHS). Predefined surgical outcomes were analysed.

Results: The median HOST-CHS score increased significantly from week 1 to 4 [50 (39, 58) vs 73 (65, 74), P < 0.001] during simulation. The improvement in the simulation of experienced-fellows successfully transferred to skill acquisition [HOST-CHS score 72.5 (71, 74) vs 54 (51, 60), P < 0.001], with better patients outcomes including shorter total cross-clamp time [pretest: 86 (70, 99) vs post-test: 60 (53, 64) min, P = 0.006] and reduced incidence of major patch leak requiring multiple pump runs [pretest: 4/11 vs post-test: 0/9, P = 0.043]. After simulation, the technical performance and surgical outcomes of Residents were comparable to Experienced-fellows in real patients, except for significantly longer cross-clamp time [Residents: 76.5 (71.7, 86.8) vs Experienced-fellows: 60 (53, 64) min, P = 0.002].

Conclusions: Deliberate practice using simulation translates to better performance and surgical outcomes in real patients. Residents who had never completed a VSD closure could perform the procedures just as safely and effectively as their senior colleagues following simulation.

Keywords: Congenital heart surgery; Education; Simulation; Ventricular septum defect.

Figure 1:

Three-dimensional printed Heart model before and after simulation. The views (via right atrium: A, B, D, and E; via right ventricle: C, and F) of a three-dimensional-printed heart model before (A–C) and after (D–F) simulation. AL: anterior leaflet; AV: aortic valve; PL: posterior leaflet; RA: right atrium; RV: right ventricle; SL: septal leaflet; VSD: ventricular septum defect.

Figure 2:

Surgical simulation curriculum for ventricular septum defect closure. A total of 7 trainees participated in the ventricular septum defect simulation curriculum and were divided into 2 groups. Experienced-fellows group included 4 trainees (2 junior staff and 2 fellows) who have performed at least 5 VSD closure as primary surgeon. Residents group consisted of 3 residents who had never performed a VSD closure. The curriculum commenced with a pretest requiring trainees from Experienced-fellows group to perform VSD closures in the operating room. Both groups participated in a 4-week simulation programme requiring each participant to complete 2 VSD closures on three-dimensional printed models weekly. After a 1-month delay, all trainees returned for a post-test in real patients and were re-evaluated. Pretest, simulation and post-test performances were filmed, and graded blindly and independently by 2 cardiac surgeons using Hands-On Surgical Training–Congenital Heart Surgery (HOST). The predefined surgical outcomes of patients were recorded. HOST: Hands-On Surgical Training–Congenital Heart Surgery; VSD: ventricular septum defect.

Figure 3:

The comparison of technical performance on pretest and post-test. Pretest results and post-test results for Experienced-fellows and Residents are shown for (A) VSD closure time, (B) total HOST-CHS score and the 3 holistic HOST-CHS categories of (C) Knowledge, (D) Fluency and (E) Respect. The horizontal line in the box plot indicates the mean and the box indicates the upper and lower quartiles, with the vertical lines representing the minimum and maximum values. HOST-CHS: Hands-On Surgical Training–Congenital Heart Surgery; VSD: ventricular septum defect.

Figure 4:

Deficient technical tasks on the pretest and post-test. Each dot represents the mean score of technical tasks in individual patient. Pretest results and post-test results for Experienced-fellows and Residents are shown for Knowledge tasks of (A) measuring the size and shape of the VSDs, (B) trimming patch to approximate size of VSD, (C) appropriateness of patch size; Fluency tasks of (D) even suture placement; Respect tasks of (E) avoidance of plication sutures and (F) absence of residual VSD. VSD: ventricular septum defect.

Figure 5:

Graphical display of the translatability of improvement in simulation to surgical performance of congenital heart surgical trainees and outcomes of patients undergoing ventricular septum defect closure. Study design: A prospective single-centre study. Methods: Experienced-fellows completed 3 VSD closures on real patients as a pretest. Both Experienced-fellows (n = 4) and Residents (n = 3) participated in a 4-week simulation. All trainees returned for a post-test operation in real patients (n = 15). Results: The improvement in the simulation of Experienced-fellows successfully transferred to skill acquisition (A), shorter total cross-clamp time (B) and reduced major patch leak (P = 0.043) in real patients. After simulation, the technical performance of Residents was comparable to Experienced-fellows despite longer cross-clamp time. Implication: Improvement in simulation translates to better performance and surgical outcomes in patients. HOST-CHS: Hands-On Surgical Training–Congenital Heart Surgery; VSD: ventricular septum defect.