A novel scoring system to reduce bias in placental microsurgery training

Resumen

Objective: The manual skills required for microsurgical interventions have reached a level of complexity that presents a significant challenge to practitioners. The assessment and validation of these skills have become a fundamental aspect of training programs, with the implicit responsibility of establishing specific and efficient methodologies, as well as the proper validation of acquired skills. It is of paramount importance that a training program is free from bias, particularly in relation to the selected simulator details. In order to provide enhanced training quality and accurate results, this paper presents a novel and logical approach to classifying a placental simulator.

Method: A standardized microsurgical training protocol, previously published in the literature, was performed with the addition of a new variable for the assessment of vascular quality and measurement of simulator difficulty level (SDL). This quality assessment was also intended to serve as a comprehensive understanding of the current difficulty level of the particular simulator, allowing the operator to adjust a proper level of caution and finesse. Subsequently, a statistical analysis was conducted to examine the impact of the SDL on the protocol main variables.

Results: Thirty-four placentas were used for sixty-three procedures. The time decreased from 42 minutes (first attempt) to 21 minutes (attempt number 47). Then a plateau level was reached with small variations in time. The mistake rate also decreased from an average of 35% major mistakes to 12.25%. Even after reaching the plateau level, both time and mistakes shown irregular and small variations.

For 21 samples graded as High difficulty (HD), the mistake score accounted 53, with an average of 2,52. For 31 samples graded as Regular difficulty (RD), accounted 24, average of 0,77. For 11 samples graded as Low difficulty (LD), accounted 6, average of 0,54. These results were confirmed after filtering for “first procedure on that simulator” only.

Addtitionally, it was found that a second attempt on the same simulator was always a more quality (less mistakes made) and faster (less time to achieve the goal) one.

Conclusion: The SDL score (assessed by measuring vascular and perivascular variables) demonstrated the anticipated concordance with the time taken and the errors committed. The SDL score and its interpretation can be effectively employed to reduce bias and justify the variance resulting from minor simulator discrepancies. 

Details about regular variations (arising from the biological substrates used as simulators and/or real surgical scenarios) should be taken into account when assessing the effectiveness of microsurgical procedures.

DOI:https://doi.org/10.56238/sevened2024.025-032