Defrosted versus fresh human placenta models for microvascular bypass training: A comparative study in surgical simulation

INTRODUCTION

Microsurgical techniques have revolutionized neurosurgery since the mid-20^th century, especially in the management of complex cerebrovascular disorders. As procedures such as extracranial–intracranial bypass and revascularization become more advanced, there is a growing need for neurosurgeons to develop precise microvascular anastomosis skills through effective simulation-based training.^[1]

Traditional training models, including synthetic tubes, animal vessels (e.g., chicken wings and turkey wings), and live animal surgery, each present limitations. Synthetic models lack tissue realism, while ethical, logistical, and financial concerns restrict the use of live animals.

In contrast, the human placenta has emerged as a promising alternative due to its vascular anatomy, cost-effectiveness, and accessibility. Since Goldstein (1979)^[2] first described using fresh human placenta for microsurgical training, subsequent studies have refined its preparation and validated its construct and face validity. Models developed by Zambrano and later optimized by Magaldi et al. have demonstrated the placenta’s utility not only for microvascular anastomosis but also for simulating aneurysm clipping and tumor dissection.^[2-4]

Despite these advancements, fresh placenta models have a limited shelf life (3–5 days), which presents a logistical barrier. Recent interest has turned to frozen (defrosted) placenta models as a potentially equivalent alternative, allowing for longer-term storage and more flexible training logistics.^[5]

This study aims to evaluate whether defrosted (frozen) human placenta provides comparable training quality to fresh refrigerated placenta, using a modified Northwestern objective microanastomosis assessment tool (NOMAT) scoring system and participant feedback to assess both objective performance and subjective realism.

MATERIALS AND METHODS

Placenta model preparation

This study was conducted in a neurosurgical laboratory and approved by the Institutional Research Ethics Committee (Study code: SUR-2566-0627). Twenty non-pathological, discarded human placentas were obtained from patients who tested negative for common infections and were stored frozen before preparation.

After thawing, placentas were rinsed with room-temperature water. The umbilical cord was cut at a 45° angle (7–10 cm from the placenta), and the chorioamniotic membrane was removed [Figure 1]. Umbilical vessels were cannulated using a neonatal nasogastric tube (3.5 Fr), flushed with normal saline, and then irrigated with a diluted heparin solution (0.2 mL of 10,000 IU/mL heparin in 100 mL saline) [Figure 2]. Catheters were secured using silk sutures [Figure 3].

Close

Figure 1:

Frozen human placenta with the entire of umbilical cord and remove chorioamniotic membrane and cleaning.

Export to PPT

Close

Figure 2:

Cannulation of the umbilical vessels and irrigation until a translucent coloration of the vessels by saline and heparin.

Export to PPT

Close

Figure 3:

The catheter fixation at umbilical cord with silks.

Export to PPT

Dye perfusion involved injecting carmine red into umbilical arteries and methylene blue into veins, each diluted in 50 mL saline, until full vascular pigmentation was observed [Figure 4].

Close

Figure 4:

The carmine red dyes are infused into the umbilical artery.

Export to PPT

Training workshop and validation

Participants were randomly assigned (double-blind) to train on either model. All participants performed three microvascular anastomoses: End-to-end, end-to-side, and side-to-side [Figure 5], following a live demonstration [Figure 6].

Close

Figure 5:

Overview of the laboratory and participants performing bypass and anastomosis of the frozen placenta model.

Export to PPT

Close

Figure 6:

Side to side anastomosis performed by a participant in the laboratory frozen placenta model.

Export to PPT

Each participant completed a post-training questionnaire assessing the placenta model’s realism, vessel consistency, similarity to real dissection, training value, and willingness to reuse. Objective evaluation was performed using the modified NOMAT, assessing posture, instrument handling, and anastomosis patency [Table 1].

Table 1: Survey-based questions.

Question no	Question	Possible answers
1	Do you think that the human placenta training model faithfully reproduces a possible real microsurgical scenario?	“Absolutely yes” “somewhat” “Absolutely no”
2	According to your surgical experience, is the consistency of the placental vessels comparable to the real microsurgical scenario?	“Very similar” “Quite Similar” “Different”
3	Do you think that the dissection of human placental vessels is similar to the real microsurgical scenario?
4	Do you think that practice on this type of surgical training model can improve the surgical technique and reduce errors on the patient?	“Absolutely yes” “somewhat” “Absolutely no”
5	Do you think you will reuse or propose to use of this microsurgical training model?	“Yes” “No”

RESULTS

From the initially expected target of 20 neurosurgeons, 17 had participated and completed the questionnaires. Among these, there were nine “trained” neurosurgeons and eight “untrained” neurosurgeons. In the “trained” group, four had used the frozen storage placenta, and the other five had used the cold storage placenta. In the “untrained” group, four used the frozen storage placenta, and the other four used the cold storage placenta. The assignment of neurosurgeons to the type of placenta was conducted using a double-blinded randomization technique, ensuring that participants were unaware of whether they were using a frozen storage or cold storage placenta.

DISCUSSION

Our study’s findings demonstrate that both frozen storage and cold storage placenta models are equally effective for microvascular neurosurgery training. Neither model showed a significant advantage over the other in replicating the complexities of real microvascular surgery, as evidenced by the comparable modified NOMAT scores and the participants’ subjective feedback. The lack of statistically significant differences emphasizes that either model can be used effectively to achieve similar training outcomes.

The evolution of endovascular technologies has significantly influenced the treatment of aneurysms, necessitating that neurosurgeons maintain proficiency in microsurgical anastomosis for complex cases and conditions like moyamoya disease. Traditional live animal models, offering realistic features such as pulsatile blood flow, face increasing restrictions due to ethical and cost considerations.^[6] Alternatives such as chicken and turkey wing vessels and synthetic microtubing, while accessible and inexpensive, lack the physiological accuracy needed for effective training.^[7] In contrast, human and bovine placentas have gained prominence for their anatomical fidelity and practicality, providing a rich resource for extensive practice in microvascular techniques. These placental models, reintroduced into the neurosurgical training repertoire, are celebrated for their realistic simulation capabilities, aligning with the evolving demands for ethical, cost-effective training solutions in neurosurgery.

Moreover, the human placenta, recognized for its cost-effectiveness and ethical viability, continues to be a preferred choice for training programs, circumventing the ethical concerns associated with other human or animal models.^[8] This model’s utility is further enhanced by the use of dyes that aid in distinguishing vascular structures, closely mimicking real-life surgical environments and providing trainees with a realistic and comprehensive training experience.

Despite the significant advancements and historical use of fresh placenta models in microvascular surgery training, there remains a notable gap in research regarding the use of frozen or defrosted placenta models in this context. No existing studies have specifically addressed how frozen or defrosted placenta models perform when used for microvascular training. These models could potentially offer greater flexibility in scheduling training sessions, reduce waste, and simplify the management of training resources, particularly in settings where access to fresh placentas is limited or impractical.

A notable benefit of the frozen storage placenta model is its logistical convenience. These placentas can be prepared well in advance and stored for up to 40 days, then thawed an hour before use, which streamlines the training process and reduces the need for prompt access to fresh samples. This is particularly advantageous for institutions facing limitations in immediate sample preparation or frequent access to fresh placentas. In addition, in hot and humid climates like Thailand’s, the frozen storage model provides a practical solution by allowing placentas to be kept at lower temperatures for extended periods, thus maintaining their usability for training.

Our findings suggest that training programs can maintain high-quality education using either model, offering flexibility in preparation and adapting to varying environmental conditions and institutional constraints. This supports the broader adoption of placenta-based models in neurosurgical training, providing a versatile and cost-effective approach that can be tailored to different settings. While human placenta models are generally considered ethically acceptable for simulation training, it is important to acknowledge that ethical concerns regarding the collection and use of human placental tissue may vary across institutions and countries, depending on local regulations, consent procedures, and cultural perspectives.

This study has several limitations. First, the small sample size (n = 17), falling short of the target of 33 participants, limits the statistical power and generalizability of the findings. As a result, subtle differences between frozen and refrigerated placenta models may not have been detected. Second, the absence of video recordings restricted a more thorough and objective evaluation of microsurgical techniques beyond the Modified NOMAT scoring. Although the NOMAT score demonstrated acceptable reliability (0.8), it may not fully capture nuanced surgical skills such as hand positioning and motion control. Future studies with larger sample sizes and more robust assessment tools, including surgical video analysis, are warranted to confirm and expand on these findings.

CONCLUSION

Both defrosted (frozen) and fresh refrigerated human placenta models demonstrated comparable effectiveness in microvascular neurosurgical training, with no statistically significant differences in subjective realism or objective NOMAT performance. The frozen model offers significant logistical advantages, including extended storage and flexibility in training schedules, without compromising training quality. These findings support the integration of defrosted placenta models into neurosurgical training programs, especially in settings with limited access to fresh specimens, thereby enhancing accessibility, cost-efficiency, and scalability of microsurgical education.