Evaluation of a Training Course for Simulating Head and Neck Ultrasound-guided Core Needle Biopsy

Introduction

An accurate diagnosis is necessary to treat head and neck lesions effectively. Due to the long list of differential diagnoses of head and neck pathologies, tissue sampling can play an important role in addition to history taking, physical examination, and imaging to reach a reliable diagnosis. Standard sampling techniques include open surgical biopsy, Fine Needle Aspiration (FNA), and Core Needle Biopsy (CNB). The ideal method should provide adequate and best-quality material for diagnosis and be easy to perform with the lowest complication rate [1].

Since open surgical biopsy is an invasive method that requires general anesthesia and may increase the chance of tumor spread and recurrence, less invasive methods, such as FNA and CNB, are recommended as alternatives [2,3]. Both procedures can be performed under local anesthesia with a low complication rate. Previous studies have shown that the nondiagnostic rate of CNB is lower than that of FNA. Also, CNB has higher diagnostic accuracy than FNA, making it comparable to open surgical biopsy [2,4,5]. In 2011, a systematic review and meta-analysis reported that CNB detected malignancy in the head and neck lesions with an overall accuracy of 96% [6]. It provided a correct specific diagnosis in 87% of cases and was associated with no major complications. Similar studies have confirmed that CNB can provide a high-quality histopathological yield with high diagnostic utility for head and neck lesions [7-9]. Using Utrasound (US) guidance for CNB enables precise needle placement, reducing the risk of damaging surrounding tissue or missing the lesion [10,11].

Since US-guided CNB has become an essential diagnostic tool to evaluate head and neck lesions, we held a Head and Neck CNB Workshop for graduate clinicians to improve their skills. The goal of this study was to determine how effectively our head and neck CNB Workshop enhanced the knowledge and hands-on skills of graduate clinicians.

Materials and methods

The participants attended the head and neck CNB Workshop during the 18th International Congress of the Iranian Society of Otorhinolaryngology and Head and Neck Surgery. The CNB workshop took place at the Espinas Palace Hotel in Tehran, Iran, on Thursday, October 26, 2023. A total of 23 participants attended the workshop (Figure 1). Our training program consisted of 2 hours of theoretical followed by 2 hours of hands-on training. Before the theoretical program, participants took a baseline test consisting of 7 Multiple-Choice Questions (MCQs) designed based on expert consensus.

Figure 1: Schematic representation of the seven-step CNB workshop process: (1) Participant enrollment, (2) Pre-training assessment, (3) theoretical session, (4) hands-on session, (5) post-training assessment, (6) statistical analysis. CNB: Core Needle Biopsy; US: Ultrasound.

The first question assessed their experience, and the remaining six assessed their basic understanding of US-guided CNB. Knowledge improvement was tested using the same six MCQs immediately after the end of the hands-on training program. Additional MCQs were applied to evaluate participants’ perceptions of the program’s effectiveness in teaching US-guided CNB skills.

In the theoretical session, participants learned about one free-hand technique, different types of biopsy needles, suitable gauges of needles, and the importance of preprocedural US to pass the needle correctly through the skin and to decrease the chance of bleeding. The theoretical session also discussed the necessary ultrasound imaging information relevant to core needle biopsy.

In the hands-on session, participants were instructed to perform a US-guided core-needle biopsy on training phantoms. Simulation-based training using a phantom can improve clinicians’ competency, which increases patient safety and reduces healthcare cost [10]. To enhance the efficacy of the workshop, participants were divided into small groups, and modified Peyton’s 4-step training approach was used. Peyton’s 4-step training method consists of the following steps: Step 1: “Demonstrate”: The trainer demonstrates the skill at a normal pace without any further explanation; Step 2: “Talk the trainee through”: the trainer demonstrates the skill while explaining details of each sub-step; Step 3: “Trainee talks trainer through”: the trainee describes the sub-steps to the trainer while the trainer performs the skill for the third time, based on trainees description; Step 4: “Trainee does”: the trainee performs the skill by themselves. This training approach is suitable for a student-teacher ratio of 1:1 but since there were 23 participants in this workshop divided into small groups, modified Peyton’s 4-step training approach was applied which consists of the following parts: (A): the trainer performs steps 1 and 2 for all the trainees (B): trainee number 1 and the trainer carry out step 3 together and the other trainees observe (C): trainee number 1 performs the skill based on the instructions of trainee number 2 while other trainees observe (D): peer trainees and trainer give feedback to trainee number 1 (E) part C and D are repeated in turns until the last trainee (F) the last trainee performs step 4 and receives feedback from other trainees and trainer [13].

Training Phantom and US devices

Participants practiced the US-guided core-needle biopsy on an in-house-designed training phantom of methylene blue and agar–agar gel, with several chicken and potato pieces floating inside, mimicking the targeted tissue for biopsy (Figure 2). The presence of chicken or potato material inside the biopsy needle confirmed a correct biopsy.

Figure 2: Participants practiced the ultrasound-guided core-needle biopsy on the training phantom of methylene blue and agar–agar gel, with several chicken and potato pieces floating inside.

The ultrasonography was performed using the Alpinon E-Cube i7 Vet device with an L8-17H linear probe (Alpinon Medical Systems, Seoul, South Korea). We defined the same settings for every participant.

Baseline test

We designed seven Multiple-choice Questions (MCQs) to evaluate participants’ previous experience and knowledge of CNB. The first question assessed participants’ experience and categorized them into two groups: Group A, participants with experience performing US-guided CNB at least once, and Group B, participants without expertise. The remaining six questions evaluated their proficiency in general ultrasound knowledge, types of CNB needles and their functions, needle sizes required for CNB, the capability of assessing the sufficiency of the sample volume obtained using CNB, the “one free hand” technique in performing CNB, and the recommendations and necessary actions for patients after CNB (Table 1).

Participants utilized a 3-point Likert scale for these questions with the following response options: 1 = Not at all, 2 = To some extent, 3 = Completely.

The null hypothesis for each of the six scientific background MCQs was that the chosen answers were equal before and after the program. This null hypothesis was tested using the Wilcoxon Signed Rank Test. For the total MCQ score of participants, the null hypothesis that the total score was equal before and after the program was tested using the paired sample test. We also calculated participants’ total MCQ scores for groups A and B separately, and we used an independent samples t-test to evaluate the null hypothesis that the mean MCQ score was equal in the two groups.

The impact of variables such as gender, age, and pre-training MCQ score on the post-training score was tested using linear regressions. We performed statistical analysis using the IBM SPSS Statistics version 27.01.0 and used a significance level of 0.05 in this study.

Participants’ program evaluation

At the end of the program, the participants were asked to evaluate the program using a standardized 3-point Likert scale questionnaire (1 = not at all, 2 = to some extent, 3 = completely) on 3 questions. Also, a comment box was available for participants at the end of the questionnaire.

Results

MCQ

The MCQ score for each question increased significantly following the training program (p < 0.001), as shown in Table 1. The total MCQ score of participants ranged from 6 to 18. After the course, the mean total MCQ score increased from 9.83 to 16.87 (p < 0.001) (Table 2). An independent samples t-test revealed no significant difference in score increase between the two groups (A and B), with a p - value of 0.243. (Table 3). When comparing participants with and without prior US-CNB experience, both groups demonstrated significant improvement after training. The mean score increase did not differ significantly between Group A (experienced) and Group B (inexperienced) (p = 0.243), indicating comparable learning effects across different baseline experience levels.

We investigated the effect of each factor on the post-training total MCQ score using linear regression. The only significant finding was a positive correlation of age with post-training scores (0.134 score per year (95% CI 0.022–0.246, p = 0.022)), indicating that older participants tended to achieve higher post-training scores. However, gender and pre-training scores did not show a significant correlation (p = 0.457 and p = 0.232, respectively). Overall, while age demonstrated a significant relationship with post-training performance, the remaining predictors did not exhibit statistical significance, suggesting that other factors may contribute to variations in post-training scores.

Participants’ program evaluation

All participants completed program evaluation tests, and the results were positive with a mean score of 8.13 (95% Cl: 7.62-8.64). The first and most crucial question, “How do you rate the success of this workshop in terms of learning?” had a mean score of 2.73 (91%). The second question, “Did you earn the CNB skill and the confidence to perform it?” had a mean score of 2.6 (86%), and the third question, “Did you gain a basic understanding of CNB?” had the highest mean score of 2.87 (95%). These results illustrate that most participants found the workshop highly effective (Figure 3). 16/23 participants completed the comment box. All the comments reflected the positive results of the workshop.

Figure 3: Participants’ feedback on the effectiveness of the workshop, based on responses to a 3-point Likert scale.

Discussion

CNB is a safe, minimally invasive, and highly efficient technique for obtaining a histopathological diagnosis of head and neck masses. Proficiency in performing accurate CNB benefits both clinicians and patients by improving diagnostic yield and reducing procedural complications [14,15]. This study evaluated the efficacy of a structured head and neck CNB Workshop in enhancing participants’ theoretical knowledge and hands-on skills using an agar-agar phantom. The results demonstrated that the training program significantly enhanced participants’ knowledge and skills. Older participants tended to achieve higher post-training scores, likely attributable to their accumulated clinical experience. However, the improvement of scores did not significantly differ between participants with prior CNB experience and participants without prior CNB experience, supporting the broad applicability of this training model across varying levels of prior experience. Additionally, most attendees rated the workshop highly, particularly the hands-on component, and expressed their willingness to recommend this program and similar learning opportunities to other clinicians.

Our findings aligned with previous studies assessing simulation-based training for ultrasound-guided biopsies. Schmidt, et al. assessed the effectiveness of a breast ultrasound training program including CNB using a gelatin phantom for undergraduate medical students. Results showed this supplementary program increased students’ knowledge and enhanced their hands-on skills [16]. Cheng, et al. evaluated the efficacy of a training program using a gelatin phantom to practice head and neck US-guided procedures, including CNB, FNA, percutaneous ethanol injection, and radiofrequency ablation. This study confirmed that the program effectively informed the participants about the necessary steps and skills for the procedures [17]. These studies, like ours, highlight the value of combining theoretical instruction with practical simulation in skill acquisition.

Alternative training models for US-guided head and neck biopsies have been explored, including cadavers, mannequin simulators, and tofu-based phantoms [18-20]. However, our agar-agar phantom offers distinct advantages: it is cost-effective, easy to produce, portable, and durable under needle pressure compared to fragile alternatives like tofu. Additionally, its customizable design allows for the simulation of diverse lesion textures by embedding multiple targets, improving its versatility for training purposes [21-23].

Despite these advantages, it is important to acknowledge the limitations of our study. First, while the agar-agar model provides an accessible and practical training option, its downside is the lack of an accurate human anatomical representation and visible needle tracks after each needle pass [21]. Second, the small sample size (n = 23) may restrict the generalizability of our findings, warranting further validation in larger cohort studies.

Looking ahead, emerging technologies such as Virtual Reality (VR) and Augmented Reality (AR) may offer opportunities to further enhance the realism and immersion of simulation-based CNB training [24,25]. VR platforms can create interactive three-dimensional anatomical environments in which trainees practice needle positioning, hand–eye coordination, and probe manipulation in a highly controlled and repeatable manner. Unlike phantom models or cadavers, VR-based systems can simulate anatomies with varying degrees of deformation, providing broader exposure to clinically relevant scenarios [24]. For instance, Kyle Kleiman, et al. demonstrated that a VR breast biopsy simulator reduced both procedure time and the number of needle insertions required for successful biopsy [26]. Similarly, VR-based platforms have been developed for training in US-guided FNA of thyroid nodules, showing their adaptability across different biopsy procedures [27].

AR technologies, on the other hand, can overlay real-time guidance, anatomical landmarks, or error-correction prompts directly onto physical phantom models, thereby combining the tactile realism of hands-on training with enhanced digital visualization [28,29]. When integrated with haptic feedback, AR-based systems have been reported to create highly immersive and efficient training environments, as evidenced in US-guided percutaneous liver biopsy training [30]. Incorporating VR or AR into US-CNB training could therefore address some limitations of traditional phantoms, such as the lack of anatomical fidelity or restricted feedback on needle trajectory. While cost and accessibility remain barriers, gradual integration of these tools into structured workshops may provide a scalable pathway toward more immersive and effective training for clinicians across different experience levels [31,32].

Conclusion

In conclusion, our head and neck CNB workshop significantly improved participants’ knowledge and hands-on skills, regardless of prior experience. We recommend that training on a model should be considered for all trainees in head and neck US-guided CNB before application to patients. The agar-agar phantom proved to be a practical, low-cost training tool, though supplementary methods may be needed to address its anatomical limitations.

Ethical considerations

This study was approved by the Institutional Research Board (IRB) of Tehran University of Medical Sciences. Informed consent was obtained from all participants.

Author contributions

Farrokh Heidari: Design, Conduct, Writing – review and editing, final approval of the version to be published.

Benyamin Mousavi-asl: Design, Writing – review and editing, final approval of the version to be published.

Parnian Khamushian: Design, Writing – review and editing, final approval of the version to be published.

Hosna Razeghian: Analysis, Writing, review and editing, final approval of the version to be published.

Niloufar Saeedi: Design, Writing – review and editing, final approval of the version to be published.

Mahdieh Mohebbi: Design, Writing – review and editing, final approval of the version to be published.

Kayvan Aghazadeh: Design, Writing – review and editing, final approval of the version to be published.

Ebrahim Karimi: Design, Writing – review and editing, final approval of the version to be published.

Niloufar Shabani: Analysis, Writing – original draft, and Writing – review and editing, final approval of the version to be published.

Ethical approval

This study was approved by the Institutional Research Board (IRB) of Tehran University of Medical Sciences. Informed consent was obtained from all participants.

Data availability statement

The datasets analyzed during the current study are available from the corresponding author upon reasonable request.