Gasless vNOTES vs. traditional vNOTES for benign gynecological disease: a randomized controlled clinical trial

Abstract

Background

Gasless transvaginal natural orifice transluminal endoscopic surgery (G-vNOTES) can avoid complications related to pneumoperitoneum, but there is limited research on G-vNOTES. Here, we aimed to compare the hemodynamic profiles and outcomes of G-vNOTES with traditional vNOTES (T-vNOTES) in the treatment of patients with benign gynecologic disease.

Methods

A total of 120 patients with benign gynecologic disease were randomly assigned to G-vNOTES (n = 60) or traditional vNOTES (n = 60). The primary outcome was vital sign at different time points. Secondary outcomes included conversion rate, surgical time, anesthesia time, the usage of anesthetics, estimated intraoperative blood loss, visual analogue scale (VAS) score for abdominal and shoulder pain and postoperative nausea and vomiting (PONV) at 2 and 24 h, intraoperative and postoperative complications, time to first anal exhaust, eating, and getting out of bed after surgery, and length of postoperative hospital stay. Multi-level model analysis was used for intraoperative hemodynamic indicators.

Results

There was no significant difference between the two groups at the baseline level. The results of the multilevel model indicate that there is no difference in intraoperative hemodynamic performance between the G-vNOTE group and the T-vNOTES group. The conversion rate in the G-vNOTES group was higher than that in the T-vNOTES group (16.95% vs. 5.26%, p = 0.046). No significant differences were observed in other areas.

Conclusions

This study did not find advantages of gasless vNOTES in intraoperative hemodynamic fluctuations. The surgical conversion rate of the G-vNOTES group is higher than that of the T-vNOTES group, which may be related to poor surgical field exposure in the G-vNOTES group, making it more suitable for experienced and confident surgeons.

Introduction

Following the application of vNOTES for cholecystectomy, appendectomy and nephrectomy, vNOTES was firstly used for gynecologic surgery in 2012 [1]. Due to gynecologist’s familiarity with the vaginal approach and the advantages of vNOTES in reducing postoperative pain, speeding up postoperative recovery and excellent cosmetic results, the number of gynecological vNOTES surgeries is increasing rapidly worldwide and its safety and feasibility have been preliminarily demonstrated [2, 3, 4]. Now, vNOTES has been used in gynecological surgery for ovarian cystectomy, myomectomy, adnexal surgery, hysterectomy, uterosacral ligament suspension, sacrocolgapexy, pelvic lymphadenectomy, retroperitoneal pelvic and paraaortic lymphadenectomy for endometrial cancer, radical hysterectomy for cervical cancer, etc [5, 6, 7, 8, 9, 10, 11, 12, 13, 14].

The first step in conventional laparoscopic surgery is to establish a pneumoperitoneum to facilitate adequate working and observation space [15]. CO₂ absorption leads to hypercapnia、acidosis、postoperative pain, and is also associated with various cardiopulmonary complications such as tachycardia, arrhythmias, and pulmonary oedema [16]. Gasless laparoscopic surgery can avoid related problems, but it may lead to higher conversion rates [17]. Some researchers have explored the application of gasless vNOTES in gynecology [18, 19, 20], including robot-assisted surgery, but there are few clinical trials comparing traditional vNOTES with gasless vNOTES. This prospective randomized controlled study aims to compare the intraoperative hemodynamic profiles and perioperative indicators of patients with benign gynecologic disease between the two methods (T-vNOTES vs. G-vNOTES). We hypothesize that compared with the T-vNOTES group, the G-vNOTES group will perform better in terms of intraoperative hemodynamics and postoperative pain while ensuring surgical safety.

Methods

Trial design

This is a prospective, single-blind, single-center randomized controlled clinical trial conducted in Chengdu Women’s and Children’s Central Hospital, lasting from October 2022 to December 2023. This study was approved by the Ethics Review Committee of Chengdu Women’s and Children’s Central Hospital on September 27, 2022 [IRB approved number: 2022(112)]. This trial was registered at https://www.chictr.org.cn/showproj.html?proj=182441 on 17/10/2022 with the registration number ChiCTR2200064779. The protocol of this trial has been published in BMC Anesthesiology [21] in 2023. The whole trial was carried out smoothly in accordance with the requirements of the randomized controlled trial (RCT) during the research process, and there was no methodological change. The results of this RCT were reported according to the Consolidated Standard of Reporting Trials (CONSORT) 2010 checklist of information.

Participants

The inclusion criteria were patients with benign gynecological diseases with a body mass index (BMI) of 18–25 kg/m², aged 18–60 years, who were to undergo vNOTES surgery under general anesthesia, ASA status I-III, and voluntary participation. Exclusion criteria were no sexual activity, pregnancy or lactation, surgery time greater than 3 h, history of 2 or more abdominal surgeries, suspected infection or malignancy or rectovaginal endometriosis, participation in other clinical trials within 3 months. During the implementation of the trial, if the patient requests to withdraw from the trial, it can be terminated at any time. Each patient enrolled in this trial was given written informed consent after signing.

Interventions

Both groups received the same general anesthesia method. Anesthesia induction was performed using propofol (2 mg/kg), sulfentanyl (0.3 µg/kg) and cisatracurium (0.15 mg/kg). Adjust sevoflurane to maintain anesthesia and achieve the target bispectral index (BIS) between 40 and 60 during surgery. There was a study indicating that spinal anesthesia is also safe when applied to vNOTES [22]. However, considering that the vast majority of gynecological surgeries in clinical practice adopt general anesthesia, general anesthesia was also chosen for this trial. During surgery, PetCO₂ was maintained between 30 and 45 mmHg by adjusting PEEP and tidal volume. All surgeries are performed by 3 gynecologists with more than 3 years of experience in vNOTES surgery. The patients took Trendelenburg position, and underwent surgical disinfection according to the standard of vaginal surgery. Urinary catheterization 10 min before surgery, and immediately remove the catheter after surgery. All patients receive prophylactic use of 24-hour antibiotics. Patients in the T-vNOTES group established pneumoperitoneum with CO₂ and maintained pneumoperitoneum pressure of 12–14 mmHg during surgery. Use conventional laparoscopic instruments to complete the surgery according to the conventional surgical procedure. For patients in the G-vNOTES group, a steel needle with a diameter of 1.2 mm is used to puncture the abdominal wall horizontally 3–5 cm below the navel (Fig. 1). The steel needle is then fixed to the steel needle grip above the abdominal wall to create a surgical operating space. The entire surgical process does not use CO₂. Other surgical procedures are the same as T-vNOTES.

Fig. 1

G-vNOTES group surgical device

Outcomes

The primary outcomes were hemodynamic indicators such as diastolic blood pressure (DBP), systolic blood pressure (SBP), heart rate (HR), SpO₂ and airway pressure at different time points during surgery. Measured time includes pre-anesthesia (T0), at the completion of tracheal intubation (T1), 10 min after endotracheal intubation (T2), at the start of surgery (T3), 10 min after the start of surgery (T4), 10 min before the end of surgery (T5), at the end of surgery (T6), at the time of tracheal extubation (T7) and leaving the operating room (T8). Secondary outcomes included time to surgery, time to anaesthesia, intraoperative blood loss, conversion rate, anesthetic consumption, VAS score of abdominal pain, VAS score of shoulder pain and VAS score of PONV at 2 and 24 h postoperatively, postoperative and intraoperative complications, time to first anal exhaust. After the surgery, doctors all encouraged patients to eat and get out of bed for activities as early as possible, but did not specify that it must be completed within a certain number of hours. The patients were instructed to get out of bed or eat according to their own situations, and the time was recorded truthfully. Conversion rate is defined as the patient not receiving the surgical modality which they are assigned. It includes the transfer to transumbilical laparoscopic single-site surgery (LESS) or traditional multiport laparoscopic surgery or open surgery or pneumoperitoneal laparoscopy due to bleeding, pelvic adhesions, etc. It also includes cases where the surgeon reassessed the patient’s vaginal condition and the size or location of the lesion after anesthesia, and decides to adopt another surgical method. In the protocol, it was mentioned that a secondary outcome measure was the administration of vasopressor drugs from the start of general anesthesia induction to 15 min after endotracheal intubation. However, during this study, only one patient experienced this situation, so we removed this observation indicator.

Sample size

Due to the pilot trial and statistical calculations, considering the dropout rate of 20%, we planned to include 60 patients in each group for a total of 120 patients. The allocation ratio between the two groups of patients is 1:1.

Randomisation

A researcher generated a random number sequence using the random number method before the start of the trial. Odd numbers were assigned to the G-vNOTES group, and even numbers were assigned to the T-vNOTES group. After obtaining written informed consent from the patient, the grouping results are placed in an envelope and handed over to the surgeon to begin the surgery, ensuring that postoperative follow-up and statistical analysts are not aware of the grouping information.

Statistical methods

In the protocol, we planned to use R language for data statistics, but in the actual process, we found that SAS is more convenient. Therefore, in this trial, SAS (version 9.4 for Windows, SAS Institute, Inc., Cary, NC, USA) was used for data statistical analysis. Qualitative data is described using n (%). The chi-square test and Fisher’s exact probability method were used for inter group comparison. For data that follow a normal distribution, $\:\overline{x}\pm\:s$ description is used, and t-test is used for inter group comparison. For quantitative data that do not conform to normal distribution, M (P₂₅,P₇₅) is used to describe, and rank-sum test is used for inter group comparison. This study repeated measurements of BIS, airway pressure, PetCO₂, tidal volume, HR, SBP, DBP and SpO₂ at different time points during surgery. Considering the intra-group correlation of repeated measurement data, a multi-level model was used for modeling. Among them, individuals are measured as 2 units horizontally, and each time point is measured as 1 unit horizontally. Taking SBP as an example, the corresponding model can be expressed as:

$$\begin{gathered}{\text{SB}}{{\text{P}}_{ij}} = {\beta _{00}} + {\beta _{01}}{\text{Treatmen}}{{\text{t}}_j} \hfill \\\,\,\,\,\,\,\,\,\,\,\, + {\beta _{10}}{t_{ij}} + {\beta _{11}}{\text{Treatmen}}{{\text{t}}_j}{t_{ij}} \hfill \\\,\,\,\,\,\,\,\,\,\,\, + \sum\nolimits_1^m {{\beta _m}} {x_j}_m + ({\mu _{0j}} + {\mu _{1j}}{t_{ij}} + {e_{ij}}) \hfill \\ \end{gathered} $$

Among them, i represents level 1 unit (time); J represents level 2 units (individuals); Treatment represents T-vNOTES/G-vNOTES.

Results

A total of 235 patients who planned to undergo vNOTES surgery were screened, including 9 refused to participate in this trial and 106 did not meet the inclusion criteria. Finally, 120 patients were included, which was randomly divided into T-vNOTES group (n = 60) or G-vNOTES group (n = 60) (Fig. 2). One patient with T-vNOTES was excluded because the operation time exceeded 3 h. Two patients in the T-vNOTES group were excluded due to lack of postoperative follow-up. In G-vNOTES group, one patient was excluded due to loss during postoperative follow-up.

Fig. 2

A summary of the study

Table 1 summarized the baseline characteristics of the enrolled patients. There were no statistically significant differences between two groups in terms of age, BMI, number of vaginal births, number of abdominal surgeries, ASA grade and inverted bed angle. The G-vNOTES group consisted of 10 cases of hysterectomy (with or without adnexectomy), 33 cases of fallopian tube surgery, 14 cases of ovarian cystectomy, and 2 cases of adnexectomy. The T-vNOTES group included 10 cases of hysterectomy (with or without adnexectomy), 35 cases of fallopian tube surgery, and 12 cases of ovarian cystectomy. There was no statistically significant difference in the surgical composition ratio between the two groups (p > 0.05).

Table 1 The patients’ characteristics

One patient initially receiving T-vNOTES was converted to LESS due to severe pelvic adhesions. Two patients who initially received T-vNOTES did not undergo vNOTES and went directly to LESS because the decision of the surgeon after re-evaluating the vaginal condition after anesthesia. The conversion rate of the T-vNOTES group was 5.26%. In G-vNOTES group, one patient was converted to T-vNOTES due to hemorrhage, one patient was converted to T-vNOTES due to severe pelvic adhesions, and one patient was injected with CO₂ into the abdominal cavity for pelvic exploration due to poor exposure of the surgical field during surgery. Because the surgeon re-evaluated the vaginal condition and condition after anesthesia, 3 patients were converted to T-vNOTES and 4 patients were transferred to LESS. The conversion rate of the G-vNOTES group was 16.95%. The conversion rate of G-vNOTES group was higher than that of T-vNOTES group (p = 0.046).

Table 2 summarized the main outcomes of this study. This trial adopted per-protocol analysis. We analyzed the main outcomes after excluding cases that lost follow-up, one case with surgery time exceeding 3 h and transferred cases (G-vNOTES group n = 49, T-vNOTES group n = 54). The airway pressure in the G-vNOTES group was lower than that in the T-vNOTES group at 10 min after the start of surgery (p = 0.009) and 10 min before the end of surgery (p = 0.039) (Fig. 3).

Table 2 A summary of the main outcomes

Fig. 3

T-vNOTES group and G-vNOTES group airway pressure

The estimation results of the null model showed that the (intraclass correlation coefficient, ICC) of HR, DBP, SBP, SpO₂ and airway pressure were 21.68%、14.39%、11.63%、0.32% and 40.36%, respectively. This meant that 21.68%, 14.39%, 11.63%, and 40.36% of the variation in HR, DBP, SBP, and airway pressure came from repeated measurements of the same individual at different time points, indicating the presence of a two-level structure in the data. The estimation results of the multilevel model showed that there is no statistically significant difference between the T-vNOTES group and the G-vNOTES group in terms of HR, DBP, SBP, SpO₂ and airway pressure (Table 3).

Table 3 Multilevel model results

Table 4 summarized the other outcomes. There is no statistically significant difference between the G-vNOTES group and the T-vNOTES group in terms of surgical time, anesthesia time, intraoperative blood loss, anesthesia consumption, VAS and PONV scores at 2 and 24 h post surgery, postoperative and intraoperative complications, time to first anal exhaust, eating, and getting out of bed after surgery.

Table 4 A summary of the other outcomes

One patient in the T-vNOTES group underwent debridement and suturing on the 5th day after surgery due to vaginal incision infection, and the wound recovered well. One patient in the G-vNOTES group was complicated with postoperative acute pelvic inflammation and was discharged after 14 days of antibiotic treatment. No other postoperative complications were observed.

Discussion

To our knowledge, this is the first RCT trial that compared the surgical treatment of benign gynecological diseases using traditional pneumoperitoneum vNOTES and gasless vNOTES. In our trial, G-vNOTES did not show significant advantages in hemodynamics such as intraoperative blood pressure fluctuations, heart rate, SpO₂, and airway pressure. At the same time, G-vNOTES and T-vNOTES had similar manifestations in terms of operation time, surgical blood loss, postoperative abdominal pain, shoulder pain, nausea and vomiting, postoperative hospital stay, time to get out of bed, and time to anal exhaust. In previous studies, some cardiopulmonary changes (increased peak airway pressure, decreased cardiac output, pulmonary compliance) occurred during pneumoperitoneum [23]. Conventional vNOTES uses CO₂ to form an artificial pneumoperitoneum to provide a field of view for the operation. Increased intra-abdominal pressure, compression of abdominal organs, increased blood volume flowing back to the heart, and increased central venous pressure (CVP). Microcirculatory perfusion index is significantly related to CVP [24]. CO₂ pneumoperitoneum causes the rising of diaphragm, increasement of intrathoracic pressure and ventricular afterload. Animal models have shown that during intra-abdominal injection of CO_2, pulmonary artery pressure increases to two times the baseline level [25]. Patients with coronary artery disease, chronic hypertension, and other conditions are extremely sensitive to changes in left ventricular afterload [26]. CO₂ pneumoperitoneum can also slow blood flow, increase blood viscosity, and cause microtears in the vascular endothelium, which promotes the formation of venous thrombosis [27]. On the other hand, CO₂ pneumoperitoneum causes hypercapnia, acidosis, and a 20% decrease in stroke volume at normal blood volume, and this more pronounced with moderate blood loss [28]. In our trial, the estimation results of the multilevel model showed that there is no statistically significant difference between the T-vNOTES group and the G-vNOTES group in terms of HR, DBP, SBP, SpO₂ and airway pressure. People with normal heart function have good tolerance to hemodynamic changes during pneumoperitoneum [26]. The patients participating in this trial are aged 19–60 years old, with a BMI of 18.25–24.97 kg/m², ASA grade I-II, and no serious complications. For patients with good basic conditions, it is not easy to obtain inter group differences. The negative result of this experiment may also be related to the shorter surgical time (average surgical time of 63–68 min) and the frequency of data collection.

A systematic review suggested that there was moderate to strong evidence to support a similar safety profile in low intra-abdominal pressure during laparoscopic surgery compared with conventional abdominal pressure laparoscopic surgery, with lower pain scores, lower rates of mild postoperative complications, lower rates of PONV, and shorter hospital stays [29]. We did not find similar results in this trial, and we suspect that it may be related to the shorter operation time in this trial. Although there was no statistically significant difference in PONV at 24 h postoperatively(p = 0.068), it was seen that the G-vNOTES group appeared to have some advantages which might be related to the lower amount of anesthetic consumed. The first postoperative exhaust median time in the G-vNOTES group was 9 (IQR7,15) hours verses 13 (IQR8,17) hours in the T-vNOTES group (p = 0.13). The first postoperative meal median time was 10 (IQR7,18) hours, and 14.5 (IQR9,18) hours in the T-vNOTES group (p = 0.149). G-vNOTES seems to be more conducive to rapid postoperative recovery.

Many studies on vNOTES have mentioned the advantages of vNOTES in avoiding abdominal incisions and their complications [30, 31, 32]. In our trial, there was one postoperative infection complication in each group, and both recovered well after anti-inflammatory therapy, but this means that the incision healing and infection prevention of the vNOTES procedure deserve more attention.

According to our definition of the conversion rate (not performed as planned) prior to the start of the trial, the conversion rate was 5.26% in the T-vNOTES group and 16.95% in the G-vNOTES group. Most transfers are caused by reassessment by surgeons after anesthesia, which is more pronounced in the G-VNOTES group (3.51% vs. 11.86%), and mostly occur in relatively inexperienced doctors. The transfer in this situation did not cause additional harm to the patient. If this factor is excluded, the rate of T-vNOTES group and G-vNOTES group conversion due to factors such as adhesion or bleeding are 1.75% and 5.08%. It can be seen that sufficient and accurate preoperative evaluation is crucial. One patient in the G-vNOTES group was transferred to pneumoperitoneal laparoscopic exploration at the end of the operation, and the other patient switched to traditional vNOTES due to intraoperative bleeding, which showed that the surgical field exposure in the G-vNOTES group was worse than that in the T-vNOTES group. Taking G-vNOTES is a matter of caution for less experienced physicians or patients with a higher risk of bleeding. A study on total hysterectomy using vNOTES revealed that 2 cases (1.9%) were converted from vNOTES to laparoscopic or open-abdominal surgeries, and 2 cases (5.56%) experienced bladder injuries [33]. For patients suspected of severe pelvic adhesions, caution should be exercised in any form of vNOTES.

This trial has some limitations. Firstly, due to cost constraints and inadequate outcome measures, this trial did not measure the patient’s blood pH, blood carbonate content, pulmonary artery pressure, cardiac stroke volume and other indicators, so the hemodynamic monitoring is not perfect, and more monitoring tools are needed to address this aspect. Secondly, this trial included a variety of surgical methods for benign gynecological diseases, including ovarian cystectomy, salpingectomy, hysterectomy, and various surgical procedures. Although it was verified that there was no statistical difference in the proportion of surgical composition between the two groups, it may still lead to errors. Thirdly, this trial was not performed by the same surgeon. Although each surgeon had more than 3 years of experience in vNOTES, we observed during the implementation of the trial that the exposure of G-vNOTES did not affect experienced surgeons, but had a certain impact on those who were relatively less experienced. Fourthly, the conversion rate of this trial is relatively high, which is closely related to insufficient preoperative evaluation, which cannot fully reflect the safety of the surgery. Finally, this trial was a single-center trial with a total of 120 cases, which was statistically significant, but still needed to be validated by a larger sample.

Notwithstanding the valuable insights gleaned from our present study, it is patently clear that the full potential and feasibility of transvaginal natural orifice transluminal endoscopic surgery (vNOTES) remain in need of further exploration. To comprehensively validate the efficacy and safety of vNOTES, a greater number of large-scale, meticulously designed RCTs are acutely warranted. Future RCTs should center on comparing vNOTES with traditional surgical approaches across a broad spectrum of outcomes. These outcomes encompass, but are not restricted to, operative duration, complication incidence, long - term recurrence rates, and the quality of life of patients. Such investigations will not only elucidate the advantages and disadvantages of vNOTES with greater precision but also furnish more robust evidence - based support for its clinical application. Consequently, this will facilitate the more extensive adoption and advancement of this innovative surgical technique within the realm of gynecology.

Although we obtained some relevant data on vNOTES in this trial, the full potential and feasibility of vNOTES still require further exploration. There is a need for more large-scale and meticulously designed RCTs. Future RCTs should focus on comparing the outcomes of vNOTES with those of traditional surgical approaches. These outcomes include, but are not limited to, operative time, complication incidence, long-term recurrence rate, and patient quality of life. Such studies will not only clarify the advantages and disadvantages of vNOTES more accurately but also provide more robust evidence - based support for its clinical application, thus promoting the wider adoption and development of this innovative surgical technique in the field of gynecology.

Conclusions

In patients with better basic conditions, for surgeries with shorter surgical time, there is no significant difference in intraoperative blood pressure, heart rate, airway pressure, and SpO₂ between G-vNOTES and T-vNOTES. G-vNOTES has a better trend in PONV, postoperative first exhaust time, and postoperative first meal time, but the conversion rate is higher. Due to its relatively poor visual field exposure, G-vNOTES is more suitable for experienced and confident surgeons. Further trials are needed to verify the benefit of G-vNOTES in haemodynamically unstable patients.