Evidence summary

Population and studies description

This interventional procedures overview is based on 3,980 people from 1 systematic review, 1 RCT, 2 prospective case series and 5 retrospective cohort studies. Of these 3,980 people, 2,178 had the procedure. This is a rapid review of the literature, and a flow chart of the complete selection process is shown in figure 1. This overview presents 9 studies as the key evidence in table 2 and table 3, and lists 24 other relevant studies in table 5.

The systematic review by Housmans et al. (2020) reported outcomes for 718 hysterectomy procedures in adult women, who were having removal of the uterus for benign gynaecological disease. There were 288 people in the vNOTES arms and 430 in the control arms. The studies included 1 RCT from Belgium, 1 retrospective cohort study from Taiwan, 2 retrospective cohort studies from South Korea, 1 retrospective cohort study from Turkey and 1 retrospective cohort study from China.

The RCT by Baekelandt et al. (2021) compared vNOTES with laparoscopy for adnexectomy in 67 adult women with an adnexal mass presumed to be benign. The study was done at a single centre in Belgium and randomly assigned 34 people to the vNOTES group and 33 to the laparoscopy group. The mean ages were 50 (plus or minus 10) and 52 (plus or minus 8.5), respectively. Follow up of 6 months captured pain and quality-of-life outcomes.

The study by Karkia et al. (2019) was a prospective case series in a UK setting. They investigated the safety and efficacy of hysterectomy and adnexectomy using vNOTES. The population included 33 women with benign uterine pathology or grade 1 stage 1 endometrial cancer suitable for hysterectomy at a local unit. The patients were aged between 35 and 70 with a mean age of 50. Indications for surgery included treatment resistant dysfunctional uterine bleeding (14), atypical endometrial hyperplasia (5), BRCA positive breast cancer (2), pelvic pain (7), post-menopausal bleeding (4) and endometrial cancer stage 1 (1). Follow-up data was captured at 3 months but is not publicly available.

The study by Baekelandt and Kapurubandara (2021) is a prospective case series study that captured perioperative outcome data for 1,000 people having vNOTES for benign gynaecological conditions. 73% of procedures were hysterectomies and the rest were adnexal surgeries (18%) or other. All the procedures were done at a single centre in Belgium. The mean age of this cohort of women was 46 years.

Huang et al. (2022) is a retrospective study of patient data at a single centre in China for people who had vNOTES surgery for gynaecological conditions. The procedures included 902 adnexal surgeries, 98 myomectomies, 82 hysterectomies, 51 pelvic floor reconstruction surgeries and 14 malignant tumour surgeries. The mean ages across surgery groups ranged from 30.89 (plus or minus 6.26) to 63.55 (plus or minus 9.12) years old.

Ovarian cystectomies were the subject of the retrospective cohort study by Huang et al. (2021). They analysed 296 people, with a mean age of 30.1 (plus or minus 7.65), in need of laparoscopic surgery for ovarian cysts. This was a study in China where 86 women had vNOTES procedures and 210 were in the TU-LESS group.

Yan et al. (2022) is a retrospective study of 361 women who had either a vNOTES (n=228) or TU-LESS (129) hysterectomy at a single-centre in China. The mean ages of each group were 53.03 (plus or minus 9.44) and 54.26 (plus or minus 10.33), respectively.

Temtanakitpaisan et al. (2018) retrospectively studied the outcomes of vNOTES hysterectomy across uterine sizes for benign pelvic organ lesions of the uterus, cervix or ovaries. The procedures were done for 275 women at a single centre in Taiwan. Group 1 included 191 people with uterine weights less than 500 g, group 2 included 67 people with uterine weights of 500 g to 999 g, and group 3 included 17 people with uterine weights of 1,000 g or higher. The mean ages of the groups were 48.68 (plus or minus 6.63), 47.22 (plus or minus 3.81) and 46.53 (plus or minus 2.96), respectively.

The study by Kaya et al. (2022) was a cross-sectional of surgical outcomes for vNOTES compared with laparoscopic hysterectomy for benign gynaecological conditions in women with obesity. The study was done in Turkey and the cohort (n=83) consisted of adult women with a BMI of 30 kg/m² or greater. The mean age in the laparoscopy arm was 49 years (range 40 to 71) and the mean age in the vNOTES group was 52 (range 40 to 74).

Table 2 presents study details.

Figure 1 Flow chart of study selection

Table 2 Study details

Study no.	First author, date country	Patients (female)	Age (range)	Study design	Inclusion criteria	Intervention	Follow up
1	Housmans, 2020 Belgium	n=718 (n=288 in vNOTES arms)	Studies varied between 18 and 70 years	Systematic review	RCTs, CCTs, prospective or retrospective cohort studies that directly compares vNOTES to laparoscopy. Adult women having removal of the uterus for benign gynaecological disease. Exclusions: Genital prolapse Gynaecological malignancy	vNOTES hysterectomy compared with laparoscopic hysterectomy	NR
2	Baekelandt, 2021 Belgium	n=67 (n=34 in vNOTES arm)	Mean= 52 (plus or minus 11)	Randomised controlled trial	Women who were not pregnant, and were sexually active with an intact uterus and without obliteration of the pouch of Douglas scheduled to have removal of a benign adnexal mass. Exclusions: History of rectal surgery Suspected rectovaginal endometriosis Suspected malignancy Pelvic inflammatory disease Active lower genital tract infection	vNOTES adnexectomy compared with laparoscopic adnexectomy	6 months pain/QoL
3	Karkia, 2019 UK	n=33	Mean = 50 (35 to 75)	Prospective case series	Women with benign uterine pathology or grade 1 stage 1 endometrial cancer suitable for hysterectomy at a local unit. Exclusions: History of surgery to the rectovaginal pouch History of rectovaginal endometriosis Two or more caesarean sections Uterine prolapse	vNOTES hysterectomy and adnexectomy	NR
4	Baekelandt, 2020 Belgium	n=1,000	Mean= 46 (22 to 83)	Prospective case series	Women with benign gynaecological conditions. Exclusions: History of rectal surgery History of pelvic radiotherapy Suspected rectovaginal endometriosis Suspected malignancy Pelvic inflammatory disease Active lower genital tract infection	vNOTES hysterectomy (n=730) vNOTES other (n=270)	NR
5	Huang, 2022 China	n=1,147	Mean = 30.89 (plus or minus 6.26) to 63.55 (plus or minus 9.12)	Retrospective cohort	Women with gynaecological conditions. Exclusions: Unstable vital signs with intolerability of the procedure Acute infection Preoperative DVT or hypercoagulability Liver or kidney dysfunction Mental illness Other conditions that rendered the patient unable to tolerate laparoscopy (for example, severe cardiopulmonary disease) History of rectal surgery Suspected rectovaginal septum endometriosis or severe adhesions Virginity Pregnancy	vNOTES hysterectomy, adnexal surgery, myomectomy, pelvic floor reconstruction or malignant tumour surgery	NR
6	Huang, 2021 China	n=296	Mean = 30.1 (plus or minus 7.65)	Retrospective cohort	In need of laparoscopic surgery for unilateral ovarian cysts Stable vital signs and laparoscopic surgery can be tolerated Low probability of malignancy Exclusions: History of rectal surgery Suspected of rectovaginal septum endometriosis, tumors or severe adhesions Virginity Pregnancy	vNOTES ovarian cystectomy (n=86) compared with TU-LESS ovarian cystectomy (n=210)	NR
7	Yan, 2022 China	n=361	Mean = 53.04 (plus or minus 9.44) to 54.26 (plus or minus 10.33)	Retrospective cohort	Woman with benign gynaecological conditions	vNOTES hysterectomy (n=232) compared with TU-LESS hysterectomy (n=129)	NR
8	Temtanakitpaisan, 2018 Taiwan	n=275	Mean= 46.53 (plus or minus 2.96) to 48.68 (plus or minus 6.63)	Retrospective cohort	Women with benign pelvic organ lesions of the uterus, cervix, or ovaries that need surgical intervention. Exclusions: Virginity History of tubo-ovarian abscess Severe endometriosis Suspected severe pelvic adhesion History of abandoned NOTES hysterectomy	vNOTES hysterectomy	NR
9	Kaya, 2022 Turkey	n=83	Mean= TLH, 49 (40 to 71) vNOTES, 52 (40 to 74)	Cross-sectional	Women who had hysterectomy for benign gynaecological conditions with a BMI of 30 kg/m² or greater. Exclusions: Active urinary tract or pelvic infections Pregnancy Endometriosis Gynaecological malignancy History of pelvic radiotherapy Trendelenburg position	vNOTES hysterectomy compared with total laparoscopic hysterectomy	NR

Table 3 Study outcomes (option 1)

First author, date	Efficacy outcomes	Safety outcomes
Housmans, 2020	Procedure success: Either none or one conversion procedures were reported across studies. Operation time: 5 out of 6 studies reported shorter operating times and 4 out of 5 were statistically significant. The mean difference was -16.73 minutes (95% CI -21.07 to -12.40) for vNOTES (p<0.00001). Length of hospital stay: 4 out of 6 studies reported statistically significant shorter stays than the control group. The mean difference was -0.58 days for vNOTES (p<0.00001). Postoperative pain: Only 1 paper reported a statistically significant difference, which was lower pain in the vNOTES group. There was no statistically significant difference in the pooled analysis. Other: The RCT reported no difference between arms for incidence and severity of dyspareunia, sexual wellbeing and quality of life.	Readmissions: 4 out of 6 studies reported on readmissions in vNOTES and control groups but there was no statistically significant difference. There was no statistically significant difference in the pooled analysis. Where studies had complication information in the vNOTES arms, 3 intraoperative, and 18 postoperative complications were reported. Intraoperative: 1 case of bleeding and 2 cases of bladder trauma. There was no statistically significant difference in intraoperative complications in the pooled analysis. Postoperative: 10 transfusions, 5 fevers, 1 reintervention for bleeding, 1 infected hematoma and 1 case of suspected DVT. The odds of postoperative infection (fever or PID) were lower in the vNOTES groups (OR=0.41, 95% CI 0.17 to 0.99) compared with the control groups (p=0.05).
Baekelandt, 2021	Procedure success: Zero conversions were needed in either arm (vNOTES or laparoscopy). Operation time: The vNOTES procedure was on average 15 minutes quicker (95%CI 11 to 19) (p<0.001). Length of hospital stay: There was no statistically significant difference between day-0 discharge. Postoperative pain: The mean use of analgesics in the vNOTES arm was 6 units compared with 11 units in the laparoscopy arm. The mean difference is 5 units (95% CI 2 to 8) (p<0.001). Longer term pain/QoL: The risk difference of pelvic pain at 3 months was statistically significant with fewer people reporting pain in the vNOTES group (RD=24%, 95% CI 7% to 42%) (p<0.006). The difference in median VAS score for pelvic pain at 3 months was higher in the laparoscopy group (1.5, 95% CI 0.5 to 2.4) (p<0.002). All other pain and QoL differences were not statistically significant.	Intraoperative complications: 1 case of intraperitoneal spilling in the vNOTES group and none in the laparoscopy group Postoperative complications: 4 bleeding complications in the vNOTES group and 1 in the laparoscopy group. vNOTES: 1 person needed revision and suturing and 3 had conservative treatment Laparoscopy: 1 person had conservative treatment There were no cases of postoperative infection, readmission, lasting disability or death within 6 weeks after surgery.
Karkia, 2019	Procedure success: 33 people had vNOTES hysterectomy. 32/33 people had a planned and successful adnexectomy. Zero conversions were needed. Operation time: Mean operating time was 68.5 minutes (range 43 to 110). Estimated blood loss: Mean blood loss was 269ml (range 50 to 1,200). 15% of people had an estimated blood of 500 ml or more. Length of hospital stay: On average, people stayed in hospital for 1.4 nights after the procedure (range 1 to 2). Postoperative pain: The median VAS pain score at 6 hours after operation and at discharge was 0.	Postoperative: There were no major postoperative complications reported. 2 patients needed readmission and had conservative treatment.
Baekelandt, 2020	Procedure success: 1,000 procedures were completed, including hysterectomy (73%), adnexal surgery (18%), salpingectomy (4%), ovarian cystectomy (3%), myomectomy (1%) and other indications (1%). 4 out of 1,000 procedures needed conversion (3 to conventional laparoscopy and 1 to laparotomy). Operation time: The mean operating time for the hysterectomy group was 46 minutes (range 20 to 250) and 33 minutes (range 14 to 150) in the vNOTES other group.	vNOTES hysterectomy complications in 730 procedures (38) Intraoperative (10): 9 cystotomies and 1 case of bleeding requiring transfusion. Postoperative (28): Complications included cystitis (6), haematoma (5), postoperative nausea and vomiting (2), wound infections (2), genital herpes (1), stress urinary incontinence (1), haematoma drainage (6), wound repair (4), and adhesiolysis (1). vNOTES other complications in 270 procedures (1) Postoperative (1): cystitis (1).
Huang, 2022	Adnexal surgery Includes unilateral and bilateral forms of salpingectomies, ovarian cystectomies and adnexectomies. Procedure success: 16 conversions were reported for 902 procedures. Operation time: Mean operating times across adnexal surgeries range from 67.88 (plus or minus 25.99) to 115.74 (plus or minus 40.56) minutes. Length of hospital stay: Mean hospital stays ranged from 3.26 (plus or minus 1.77) to 4.30 (plus or minus 2.68) days. Estimated blood loss: Mean blood loss across adnexal surgeries ranged from 23.47 ml (plus or minus 17.34) to 48.87 ml (plus or minus 57.53). Postoperative pain: Mean VAS pain scores ranged from 2.85 (plus or minus 0.41) to 2.97 (plus or minus 0.18) at 12 hours. At 24 hours, the range was 2.26 (plus or minus 0.75) to 2.39 (plus or minus 0.65). Myomectomy Procedure success: 2 conversions were needed out of the 98 procedures. Operation time: Mean operating time was 103.31 (plus or minus 45.09) minutes. Length of hospital stay: Mean stay was 3.92 (plus or minus 1.95) days Estimated blood loss: Mean blood loss was 81.53 ml (plus or minus 193.43). Postoperative pain: Mean VAS pain scores were 2.91 (plus or minus 0.46) and 2.35 (plus or minus 0.56) at 12 and 24 hours, respectively. Hysterectomy Procedure success: 0 conversions out of 82 procedures. Operation time: Mean operating time was 107.40 (plus or minus 39.54) minutes. Length of hospital stay: Mean stay was 4.72 (plus or minus 1.91) days. Estimated blood loss: Mean blood loss was113.29 ml (plus or minus 182.22). Postoperative pain: Mean VAS pain scores were 2.98 (plus or minus 0.44) and 2.46 (plus or minus 0.67) at 12 and 24 hours, respectively.	Complications: Adnexal surgery: Complications occurred in 21 out of 902 procedures. Myomectomy: Complications occurred in 4 out of 98 procedures. Hysterectomy: Complications occurred in 2 out of 82 hysterectomy procedures. The nature of the complications for each type of surgery is unclear. Across all 5 vNOTES surgery types, 38 complications were reported in 1,147 procedures (3.31% complication rate). Complications were reported using the Clavien-Dindo classification. 27 were grade 1, 4 were grade 2 and 7 were grade 3 (including rectal injury, bladder injury and mesh exposure).
Huang, 2021	No significant difference in operation time or estimated blood loss between vNOTES and TU-LESS arms. Procedure success: No conversions to laparoscopy or laparotomy were needed. Length of hospital stay: Mean hospital stay was significantly shorter in the vNOTES group (3.39 days plus or minus 0.67) compared with the TU-LESS group (3.73 days plus or minus 0.97) (p<0.003). Postoperative pain: Mean VAS score at 24 hours after surgery was 0.99 (plus or minus 0.80) in the vNOTES group compared with 2.35 (plus or minus 1.05) in the TU-LESS group (p<0.001). Other: Time of flatus after surgery was significantly lower in the vNOTES group (15.8 hours plus or minus 5.27) compared with the TU-LESS group (19.5 plus or minus 4.60) (p<0.001). The mean cosmetic score in the vNOTES group was higher at 21.4 (plus or minus 1.39) compared with 19.1 (plus or minus 1.94) in the TU-LESS group (p<0.001).	Complications vNOTES: 2 cases of fever (CD-1) were treated with antipyretic drugs. TU-LESS: 4 cases of fever (CD-1) were treated with antipyretic drugs and 1 case of postoperative anaemia (CD-2) was treated with blood transfusion. No complications of CD-3, -4 or -5 were observed in either procedure group.
Yan, 2022	Procedure success: 4 out of 232 vNOTES procedures needed conversion to TU-LESS because of multiple uterine myoma. 3 procedures failed because of the large size of myoma (up to 12 cm) and 1 because of adhesion between the uterus and pelvis. Operation time: Mean operation time was significantly shorter in the vNOTES group (p<0.001). vNOTES hysterectomy took, on average, 78.21 minutes (plus or minus 30.79) compared with 112.09 (plus or minus 44.05) in the TU-LESS group. Estimated blood loss: There was no statistically significant difference in median blood loss between both groups. Length of hospital stay: Median hospital stay was shorter in the vNOTES group (2.31 days plus or minus 0.69) compared with the TU-LESS group (3.77 days plus or minus 1.57) (p<0.001). Postoperative pain: 4.39% of vNOTES patients required postoperative analgesics compared with 15.50% in the TU-LESS group (p<0.001). Other: Duration of anal exhaust was statistically significant between groups. On average, this persisted for 18.80 hours (plus or minus 6.60) in the vNOTES group compared with 36.49 hours (plus or minus 13.71) in the TU-LESS group.	Complications vNOTES: No complications were seen in this group. TU-LESS: Exudation occurred in the umbilicus wound in 2 women in this group. No readmissions were needed in either group after 6 weeks.
Temtanakitpaisan, 2018	Procedure success: 2 out of 275 procedures could not be done successfully and conversions to conventional laparoscopy were needed. Operation time: The differences of operation times between groups was statistically significant (p<0.0001). Group 1's mean operating time was 76.70 minutes (SE 0.68), group 2's was 99.99 minutes (SE 1.14) and group 3's was 152.88 minutes (SE 3.37). Estimated blood loss: Mean blood loss was statistically significant between groups (p<0.0001). Mean estimated blood loss in group 1 was 180.85 ml (SE 4.61), 342.57 ml (SE 6.98) for group 2 and 532.35 ml (SE 11.85) for group 3. Length of hospital stay: Mean hospital stays in each group ranged between 1.34 to 1.47 days but this was not statistically significant.	Complications Group 1 (n=191): 4 cases of postoperative bleeding, 2 pelvic infection and 2 bladder injury. Group 2 (n=67): 1 case of pelvic infection. Group 3 (n=17): No complications reported.
Kaya, 2022	Procedure success: No conversions were needed in the vNOTES group. Operation time: Mean operating time in the vNOTES arm was 80 minutes (35 to 170) compared with 135 minutes (105 to 220) in the TLH arm – a statistically significant difference (p<0.001). Estimated blood loss:* 8 people in the TLH group and 7 in the vNOTES group had intraoperative blood loss of more than 300 ml. Length of hospital stay: Mean stay was 48 hours (48 to 96) in the TLH arm and 48 hours (24 to 96) in the vNOTES arm. Postoperative pain: Mean VAS pain score at 6 hours was 7 in TLH compared with 6 in vNOTES (p<0.01). Mean VAS pain score at 24 hours was 3 and 4, respectively (p<0.01). *Outcomes from the propensity score matched model (n=62)	Peri/postoperative blood transfusion: 6 transfusions needed in the TLH procedures and 7 in the vNOTES procedures. Not statistically significant. Organ injury: 1 case of primary bladder injury in the TLH group and no other complications reported in the vNOTES group.

First author, date

Efficacy outcomes

Safety outcomes

Housmans, 2020

Procedure success: Either none or one conversion procedures were reported across studies.

Operation time: 5 out of 6 studies reported shorter operating times and 4 out of 5 were statistically significant.

The mean difference was -16.73 minutes (95% CI -21.07 to -12.40) for vNOTES (p<0.00001).

Length of hospital stay: 4 out of 6 studies reported statistically significant shorter stays than the control group.

The mean difference was -0.58 days for vNOTES (p<0.00001).

Postoperative pain: Only 1 paper reported a statistically significant difference, which was lower pain in the vNOTES group.

There was no statistically significant difference in the pooled analysis.

Other: The RCT reported no difference between arms for incidence and severity of dyspareunia, sexual wellbeing and quality of life.

Readmissions: 4 out of 6 studies reported on readmissions in vNOTES and control groups but there was no statistically significant difference.

There was no statistically significant difference in the pooled analysis.

Where studies had complication information in the vNOTES arms, 3 intraoperative, and 18 postoperative complications were reported.

Intraoperative: 1 case of bleeding and 2 cases of bladder trauma.

There was no statistically significant difference in intraoperative complications in the pooled analysis.

Postoperative: 10 transfusions, 5 fevers, 1 reintervention for bleeding, 1 infected hematoma and 1 case of suspected DVT.

The odds of postoperative infection (fever or PID) were lower in the vNOTES groups (OR=0.41, 95% CI 0.17 to 0.99) compared with the control groups (p=0.05).

Baekelandt, 2021

Procedure success: Zero conversions were needed in either arm (vNOTES or laparoscopy).

Operation time: The vNOTES procedure was on average 15 minutes quicker (95%CI 11 to 19) (p<0.001).

Length of hospital stay: There was no statistically significant difference between day-0 discharge.

Postoperative pain: The mean use of analgesics in the vNOTES arm was 6 units compared with 11 units in the laparoscopy arm. The mean difference is 5 units (95% CI 2 to 8) (p<0.001).

Longer term pain/QoL: The risk difference of pelvic pain at 3 months was statistically significant with fewer people reporting pain in the vNOTES group (RD=24%, 95% CI 7% to 42%) (p<0.006). The difference in median VAS score for pelvic pain at 3 months was higher in the laparoscopy group (1.5, 95% CI 0.5 to 2.4) (p<0.002). All other pain and QoL differences were not statistically significant.

Intraoperative complications: 1 case of intraperitoneal spilling in the vNOTES group and none in the laparoscopy group

Postoperative complications: 4 bleeding complications in the vNOTES group and 1 in the laparoscopy group.

vNOTES: 1 person needed revision and suturing and 3 had conservative treatment
Laparoscopy: 1 person had conservative treatment

There were no cases of postoperative infection, readmission, lasting disability or death within 6 weeks after surgery.

Karkia, 2019

Procedure success: 33 people had vNOTES hysterectomy. 32/33 people had a planned and successful adnexectomy. Zero conversions were needed.

Operation time: Mean operating time was 68.5 minutes (range 43 to 110).

Estimated blood loss: Mean blood loss was 269ml (range 50 to 1,200). 15% of people had an estimated blood of 500 ml or more.

Length of hospital stay: On average, people stayed in hospital for 1.4 nights after the procedure (range 1 to 2).

Postoperative pain: The median VAS pain score at 6 hours after operation and at discharge was 0.

Postoperative: There were no major postoperative complications reported.

2 patients needed readmission and had conservative treatment.

Baekelandt, 2020

Procedure success: 1,000 procedures were completed, including hysterectomy (73%), adnexal surgery (18%), salpingectomy (4%), ovarian cystectomy (3%), myomectomy (1%) and other indications (1%).

4 out of 1,000 procedures needed conversion (3 to conventional laparoscopy and 1 to laparotomy).

Operation time: The mean operating time for the hysterectomy group was 46 minutes (range 20 to 250) and 33 minutes (range 14 to 150) in the vNOTES other group.

vNOTES hysterectomy complications in 730 procedures (38)

Intraoperative (10): 9 cystotomies and 1 case of bleeding requiring transfusion.
Postoperative (28): Complications included cystitis (6), haematoma (5), postoperative nausea and vomiting (2), wound infections (2), genital herpes (1), stress urinary incontinence (1), haematoma drainage (6), wound repair (4), and adhesiolysis (1).

vNOTES other complications in 270 procedures (1)

Postoperative (1): cystitis (1).

Huang, 2022

Adnexal surgery

Includes unilateral and bilateral forms of salpingectomies, ovarian cystectomies and adnexectomies.

Procedure success: 16 conversions were reported for 902 procedures.
Operation time: Mean operating times across adnexal surgeries range from 67.88 (plus or minus 25.99) to 115.74 (plus or minus 40.56) minutes.
Length of hospital stay: Mean hospital stays ranged from 3.26 (plus or minus 1.77) to 4.30 (plus or minus 2.68) days.
Estimated blood loss: Mean blood loss across adnexal surgeries ranged from 23.47 ml (plus or minus 17.34) to 48.87 ml (plus or minus 57.53).
Postoperative pain: Mean VAS pain scores ranged from 2.85 (plus or minus 0.41) to 2.97 (plus or minus 0.18) at 12 hours. At 24 hours, the range was 2.26 (plus or minus 0.75) to 2.39 (plus or minus 0.65).

Myomectomy

Procedure success: 2 conversions were needed out of the 98 procedures.
Operation time: Mean operating time was 103.31 (plus or minus 45.09) minutes.
Length of hospital stay: Mean stay was 3.92 (plus or minus 1.95) days
Estimated blood loss: Mean blood loss was 81.53 ml (plus or minus 193.43).
Postoperative pain: Mean VAS pain scores were 2.91 (plus or minus 0.46) and 2.35 (plus or minus 0.56) at 12 and 24 hours, respectively.

Hysterectomy

Procedure success: 0 conversions out of 82 procedures.
Operation time: Mean operating time was 107.40 (plus or minus 39.54) minutes.
Length of hospital stay: Mean stay was 4.72 (plus or minus 1.91) days.
Estimated blood loss: Mean blood loss was113.29 ml (plus or minus 182.22).
Postoperative pain: Mean VAS pain scores were 2.98 (plus or minus 0.44) and 2.46 (plus or minus 0.67) at 12 and 24 hours, respectively.

Complications:

Adnexal surgery: Complications occurred in 21 out of 902 procedures.
Myomectomy: Complications occurred in 4 out of 98 procedures.
Hysterectomy: Complications occurred in 2 out of 82 hysterectomy procedures.

The nature of the complications for each type of surgery is unclear. Across all 5 vNOTES surgery types, 38 complications were reported in 1,147 procedures (3.31% complication rate).

Complications were reported using the Clavien-Dindo classification. 27 were grade 1, 4 were grade 2 and 7 were grade 3 (including rectal injury, bladder injury and mesh exposure).

Huang, 2021

No significant difference in operation time or estimated blood loss between vNOTES and TU-LESS arms.

Procedure success: No conversions to laparoscopy or laparotomy were needed.

Length of hospital stay: Mean hospital stay was significantly shorter in the vNOTES group (3.39 days plus or minus 0.67) compared with the TU-LESS group (3.73 days plus or minus 0.97) (p<0.003).

Postoperative pain: Mean VAS score at 24 hours after surgery was 0.99 (plus or minus 0.80) in the vNOTES group compared with 2.35 (plus or minus 1.05) in the TU-LESS group (p<0.001).

Other: Time of flatus after surgery was significantly lower in the vNOTES group (15.8 hours plus or minus 5.27) compared with the TU-LESS group (19.5 plus or minus 4.60) (p<0.001). The mean cosmetic score in the vNOTES group was higher at 21.4 (plus or minus 1.39) compared with 19.1 (plus or minus 1.94) in the TU-LESS group (p<0.001).

Complications

vNOTES: 2 cases of fever (CD-1) were treated with antipyretic drugs.
TU-LESS: 4 cases of fever (CD-1) were treated with antipyretic drugs and 1 case of postoperative anaemia (CD-2) was treated with blood transfusion.

No complications of CD-3, -4 or -5 were observed in either procedure group.

Yan, 2022

Procedure success: 4 out of 232 vNOTES procedures needed conversion to TU-LESS because of multiple uterine myoma. 3 procedures failed because of the large size of myoma (up to 12 cm) and 1 because of adhesion between the uterus and pelvis.

Operation time: Mean operation time was significantly shorter in the vNOTES group (p<0.001). vNOTES hysterectomy took, on average, 78.21 minutes (plus or minus 30.79) compared with 112.09 (plus or minus 44.05) in the TU-LESS group.

Estimated blood loss: There was no statistically significant difference in median blood loss between both groups.

Length of hospital stay: Median hospital stay was shorter in the vNOTES group (2.31 days plus or minus 0.69) compared with the TU-LESS group (3.77 days plus or minus 1.57) (p<0.001).

Postoperative pain: 4.39% of vNOTES patients required postoperative analgesics compared with 15.50% in the TU-LESS group (p<0.001).

Other: Duration of anal exhaust was statistically significant between groups. On average, this persisted for 18.80 hours (plus or minus 6.60) in the vNOTES group compared with 36.49 hours (plus or minus 13.71) in the TU-LESS group.

Complications

vNOTES: No complications were seen in this group.
TU-LESS: Exudation occurred in the umbilicus wound in 2 women in this group.

No readmissions were needed in either group after 6 weeks.

Temtanakitpaisan, 2018

Procedure success: 2 out of 275 procedures could not be done successfully and conversions to conventional laparoscopy were needed.

Operation time: The differences of operation times between groups was statistically significant (p<0.0001). Group 1's mean operating time was 76.70 minutes (SE 0.68), group 2's was 99.99 minutes (SE 1.14) and group 3's was 152.88 minutes (SE 3.37).

Estimated blood loss: Mean blood loss was statistically significant between groups (p<0.0001). Mean estimated blood loss in group 1 was 180.85 ml (SE 4.61), 342.57 ml (SE 6.98) for group 2 and 532.35 ml (SE 11.85) for group 3.

Length of hospital stay: Mean hospital stays in each group ranged between 1.34 to 1.47 days but this was not statistically significant.

Complications

Group 1 (n=191): 4 cases of postoperative bleeding, 2 pelvic infection and 2 bladder injury.
Group 2 (n=67): 1 case of pelvic infection.
Group 3 (n=17): No complications reported.

Kaya, 2022

Procedure success: No conversions were needed in the vNOTES group.

*Operation time: Mean operating time in the vNOTES arm was 80 minutes (35 to 170) compared with 135 minutes (105 to 220) in the TLH arm – a statistically significant difference (p<0.001).

Estimated blood loss: 8 people in the TLH group and 7 in the vNOTES group had intraoperative blood loss of more than 300 ml.

*Length of hospital stay: Mean stay was 48 hours (48 to 96) in the TLH arm and 48 hours (24 to 96) in the vNOTES arm.

*Postoperative pain: Mean VAS pain score at 6 hours was 7 in TLH compared with 6 in vNOTES (p<0.01). Mean VAS pain score at 24 hours was 3 and 4, respectively (p<0.01).

*Outcomes from the propensity score matched model (n=62)

Peri/postoperative blood transfusion: 6 transfusions needed in the TLH procedures and 7 in the vNOTES procedures. Not statistically significant.

Organ injury: 1 case of primary bladder injury in the TLH group and no other complications reported in the vNOTES group.

Procedure technique

Of the 9 studies, 5 studies compared a gynaecological vNOTES procedure with its laparoscopic equivalent (Housmans, 2020; Baekelandt, 2021; Huang, 2021; Yan, 2022; Kaya, 2022). Karkia (2019) alone investigated vNOTES hysterectomy with adnexal surgery. The remaining 3 studies investigated vNOTES hysterectomy alone (Temtanakitpaisan, 2018), or outcomes for vNOTES hysterectomies and other vNOTES procedures (Baekelandt, 2020; Huang, 2022).

The following multiple-instrument access ports were used across studies: GelPOINT mini advanced access platform (Baekelandt, 2020; Baekelandt, 2021), GelPOINT Alexis retractor or GelPOINT vPath (Karkia, 2019; Kaya 2022), HK-TH-60.4TY (Huang, 2022; Huang, 2021; Yan, 2022). Housmans (2020) and Temtanakitpaisan (2018) described limited details of the procedure.

In some people with uterine weights greater than 1,000 g the uterus could not be removed, which was associated with BMI and limited vaginal space. Uterus retrieval can also be difficult when the shape of large uteri is disproportionate to the pelvis. The suggested way to overcome this during procedure is to opt for manual morcellation of the uterus (Temtanakitpaisan, 2018).

Efficacy

Procedure success

All 9 studies reported procedure success. There was variability across the studies in the reporting of procedure success, where some studies only had inferred procedural success (by lack of conversion) while others stated procedure success (rates) explicitly as the proportion of women who successfully had treatment with the intended approach without conversion to any other procedure or technique.

The systematic review by Housmans (2020) reported on 288 vNOTES procedures across the 6 studies and up to 1 conversion procedure per study. The case series of 1,000 people reported 4 conversion procedures: 3 to conventional laparoscopy and 1 to laparotomy (Baekelandt, 2020). Huang (2022) reported 16 conversions in 902 adnexal surgery procedures, 2 conversions for the 98 myomectomies and 0 for the hysterectomies. Yan (2022) reported 4 conversions from vNOTES hysterectomy to TU-LESS hysterectomy because of myoma (3) and adhesion (1). The study on various uteri sizes reported 2 conversions to conventional laparoscopy out of 275 procedures (Temtanakitpaisan, 2018).

The Baekelandt (2021) RCT, the Karkia (2019) study, and the study on people with obesity (Kaya, 2022) reported that 0 conversion procedures were needed.

Operation time

All 9 studies reported operation time.

Three studies reported comparisons of operating time between vNOTES and laparoscopy. Of the 6 studies in the systematic review, 4 reported shorter operative times for vNOTES hysterectomy compared with laparoscopic hysterectomy that were statistically significant (Housmans, 2020). The pooled estimate for vNOTES operation time was a mean difference of -16.73 minutes (95% CI -21.07 to -12.40; p<0.00001). The RCT by Baekelandt (2021) estimated that vNOTES adnexectomy was, on average, 15 minutes (95% CI 11 to 19; p<0.001) quicker than laparoscopy. In the obesity study (Kaya, 2022), the operation time was 80 minutes (35 to 170) in the vNOTES hysterectomy arm, compared with 135 minutes (105 to 220) in the total laparoscopic hysterectomy arm (p<0.0001).

Karkia (2019) was the UK study of hysterectomy and adnexectomy, which had a mean operating time of 68.5 minutes (range 43 to 110). In the Baekelandt (2020) case series, the mean operating time for vNOTES hysterectomy was 33 minutes (range 14 to 150) and 46 minutes (range 20 to 250) for other vNOTES procedures.

The Huang (2022) case series reported operation times for the various vNOTES procedures. Hysterectomy, adnexal and myomectomy were the 3 procedures of interest. The mean operation times for hysterectomies was 107.40 minutes (plus or minus 39.54) and 103.31 (plus or minus 45.09) for myomectomies. The mean operating times for adnexal surgeries ranged from 67.88 (plus or minus 25.99) to 115.74 (plus or minus 40.56) minutes.

Yan (2022) reported significantly shorter operation times in the vNOTES group of 78.21 minutes (plus or minus 30.79) compared with 112.09 minutes (plus or minus 44.05) in the TU-LESS group (p<0.001). Whereas Huang (2021) reported no significant difference in operation times.

Temtanakitpaisan (2018) reported statistically significant differences in operating times for vNOTES hysterectomies across the 3 uterine weight groups. The mean operating time (in minutes) for uteri size less than 500 g was 76.70, for 500 g to 999 g was 99.99 and for 1,000 g or more was 152.88.

Length of hospital stay

Eight of the studies reported length of hospital stay.

Three studies compared length of hospital stay between vNOTES and laparoscopy. In the systematic review, 4 out of 6 studies reported statistically significantly shorter stays in the vNOTES group than in the control group (Housmans 2020). The mean difference was -0.58 days for vNOTES hysterectomy compared with laparoscopic hysterectomy (p<0.00001). The RCT reported the number of people who were discharged from hospital on day 0 (Baekelandt 2021). In both procedure arms, most of the people were discharged on day 0. There were 88% of laparoscopic adnexectomy people discharged on day 0 compared with 94% in the vNOTES arm, but there was no statistically significant difference. Kaya (2022) reported no difference between length of hospital stay between vNOTES hysterectomy and TLH.

The mean hospital stay was 1.4 nights (range 1 to 2) in the Karkia (2019) paper. Hospital stays of 1.34 to 1.47 days were reported in the Temtanakitpaisan (2018) paper. There was no statistically significant difference between groups of different uterine weights.

For vNOTES hysterectomy procedures in the Huang (2022) study, the mean length of hospital stay was 4.72 days (plus or minus 1.91). Mean stay for myomectomies was 3.92 days (plus or minus 1.95) and between 3.26 (plus or minus 1.77) and 4.30 (plus or minus 2.68) across the 6 variations of adnexal procedures.

Huang (2021) reported mean hospital stays of 3.39 days (plus or minus 0.67) in the vNOTES ovarian cystectomy group compared with 3.73 days (plus or minus 0.97) in the TU-LESS group (p<0.003).

Yan (2022) reported median length of hospital stays for both hysterectomy groups. The median stay in the vNOTES group was significantly shorter at 2.31 days (plus or minus 0.69) compared with 3.77 days (plus or minus 1.57) in the TU-LESS group.

Postoperative pain

Seven studies reported postoperative pain.

Only one of the studies in the systematic review, a retrospective chart analysis, reported a statistically significant difference in postoperative pain (Housmans, 2020). The mean difference in VAS score on day 1 was 1.25 lower in the vNOTES group.

The RCT by Baekelandt (2021) captured postoperative pain by the average use of analgesics in each group. The mean analgesics use in the vNOTES group was 6 units compared with 11 in the laparoscopy group, which is a statistically significant mean difference of 5 units (95% CI 2 to 8; p<0.001). People were asked whether they experienced pelvic or vaginal pain at 3 and 6 months after operation, and were asked to provide VAS pain scores. The only statistically significant difference was at 3 months, when fewer people reported pain in the vNOTES group, representing a risk difference of 24% (95% CI 7% to 42%; p=0.006) compared with the laparoscopy group. The difference in median VAS score for pelvic pain at 3 months was higher in the laparoscopy group (median difference 1.5, 95% CI 0.5 to 2.4; p=0.002).

Statistically significant differences in mean VAS pain scores were captured at 6 and 24 hours after operation in the study on people with obesity (Kaya, 2022). At 6 hours, the mean score in the TLH group was 7 compared with 6 in vNOTES (p<0.01). At 24 hours, the respective mean VAS pain scores were 3 and 4 (p<0.01).

In the UK study for vNOTES hysterectomy and adnexectomy, the median VAS pain score at 6 hours after operation and at discharge was 0 (Karkia, 2019).

The Huang (2022) study captured VAS pain scores at 12 and 24 hours after surgery. In the hysterectomy group, the mean VAS pain scores were 2.98 (plus or minus 0.44) at 12 hours and 2.46 (plus or minus 0.67) at 24 hours. In the adnexal surgeries group, the mean pain score ranged from 2.85 (plus or minus 0.41) to 2.97 (plus or minus 0.18) at 12 hours and then ranged from 2.26 (plus or minus 0.75) to 2.39 (plus or minus 0.65) at 24 hours. In the myomectomy group, the postoperative mean VAS pain score was 2.91 (plus or minus 0.46) at 12 hours and 2.35 (plus or minus 0.56) at 24 hours.

In the Huang (2021) study, postoperative pain was captured at 24 hours after surgery using the VAS score. The mean score was reported as 0.99 (plus or minus 0.80) in the vNOTES ovarian cystectomy group and 2.35 (plus or minus 1.05) in the TU-LESS group.

In Yan (2022), postoperative pain was indicated by the number of people who needed analgesics after the hysterectomy. A statistically significant difference was seen between groups, where 4.39% of people needed analgesics in the vNOTES group compared with 15.50% in the TU-LESS group.

Quality of life

One study reported quality-of-life measures.

Quality-of-life data was captured at 3 and 6 months after operation in the Baekelandt (2021) RCT using the EQ-5D-3L. There was no statistically significant difference between the two groups at each time interval.

Safety

Readmissions

Three studies reported readmissions.

The Housmans (2020) systematic review included 4 studies that captured readmissions. One readmission occurred across the vNOTES groups (n=288) for suspicion of DVT, which needed CT angiography. There were 10 readmissions noted across the laparoscopy control groups (n=430) because vault hematoma (2), pain (2), cuff infection (1), repair of a vesicovaginal fistula (1), pulmonary embolism with ICU admission (1) and pelvic inflammatory disease (3). None of the readmission findings were significant in the individual studies.

In the Karkia (2019) paper, 2 out of 33 people had been readmitted and had conservative treatment.

No readmissions were needed in either vNOTES or TU-LESS hysterectomy groups at 6 weeks in the Yan (2022) study.

Intraoperative complications

Four studies reported intraoperative complications.

In the vNOTES arm of the systematic review, there were 3 intraoperative complications from 288 procedures, compared with 7 in the 430 laparoscopic procedures (Housmans, 2020). The vNOTES complications were because of bleeding (1) and bladder trauma (2). Baekelandt (2020) also reported 1 case of intraoperative bleeding that needed transfusion in the 730 vNOTES hysterectomies. Also, 9 cystotomies had to be done during these procedures.

One case of intraperitoneal spilling was reported in the vNOTES group of the Baekelandt (2021) RCT. No intraoperative complications were reported for the laparoscopy group.

Postoperative complications

In the vNOTES arm of the systematic review, there were 18 intraoperative complications from 288 procedures, compared with 105 in the 430 laparoscopic procedures (Housmans, 2020). The vNOTES complications were because of blood transfusions (10), fever (5), reintervention for bleeding (1), suspected DVT (1), and infected hematoma (1).

In addition to the systematic review, 4 other studies reported postoperative bleeding complications. During the Baekelandt (2021) trial, bleeding complications occurred in 4 out of 34 vNOTES adnexal procedures and in 1 out of 33 laparoscopy procedures. The person who had laparoscopy and 3 of the -people who had vNOTES had conservative treatment, while 1 person who had vNOTES needed revision and suturing. There were 4 people in the less than 500 g uterine weight group (n=191) who experienced bleeding after surgery (Temtanakitpaisan, 2018). Blood transfusion was needed in 7 out of 31 vNOTES hysterectomies in the cohort of people with obesity compared with 6 out of 31 in the laparoscopy control group (Kaya, 2022). No bleeding complications were seen in the vNOTES arm, but 1 person in the TU-LESS ovarian cystectomy needing a blood transfusion (Huang, 2021).

The Baekelandt (2020) case series reported 28 postoperative complications in the vNOTES hysterectomy procedures (n=730) and 1 case of cystitis in the vNOTES other procedures (n=270). The complications in the hysterectomy group included cystitis (6), hematoma drainage (6), hematoma (5), wound repair (4), nausea and vomiting (2), wound infections (2), genital herpes (1), stress urinary incontinence (1) and adhesiolysis (1).

Huang (2022) reported 38 complications collectively for the 5 types of vNOTES procedures (n=1,147) for an overall complication rate of 3.31%. Complications occurred in 2 out of 82 hysterectomies, 21 out of 902 adnexal surgeries and 4 out of 98 myomectomies. The 38 complications (which includes unrelated surgeries) are reported using Clavien-Dindo classifications; 27 were grade 1, 4 were grade 2 and 7 were grade 3 (including rectal injury, bladder injury and mesh exposure).

Fever was seen in 2 people in the vNOTES hysterectomy group and 4 people in the TU-LESS hysterectomy group of the Yan (2022) study. All had treatment with antipyretic drugs.

There were 3 cases of pelvic infection recorded by Temtanakitpaisan (2018). Two were in the less than 500 g uterine weight group (n=191) and 1 was in the 500 g to 999 g group (n=67).

Bladder injury was reported in 2 studies. There were 2 instances of bladder injury during the 191 vNOTES hysterectomies for the less than 500 g uterine weight group (Temtanakitpaisan, 2018). There was 1 case in the laparoscopy control group of the cohort with obesity and no other organ injury complications were reported in the vNOTES group (Kaya, 2022).

Karkia (2019) reported no major postoperative complications in their study. No complications were seen in the vNOTES arm of the Yan (2022) study, while 2 women experienced exudation in the umbilicus wound in the TU-LESS arm.

Anecdotal and theoretical adverse events

Expert advice was sought from consultants who have been nominated or ratified by their professional Society or Royal College. They were asked if they knew of any other adverse events for this procedure that they had heard about (anecdotal), which were not reported in the literature. They were also asked if they thought there were other adverse events that might possibly occur, even if they have never happened (theoretical).

They listed the following anecdotal adverse events:

Operative risks include infection, bleeding, conversion (to laparoscopy or laparotomy), cystotomy and organ injuries. Rates of injuries to the bladder, ureter and bowel are estimated at 1 to 2 per 100, 1 per 100 and 1 per 100, respectively.
Hip, pelvic or vaginal pain, and vaginal wound complications or dehiscence.
Dyspareunia and impaired sexual function (both estimated to be uncommon), or problems with urinary retention.
Deep venous thrombosis or pulmonary embolus.

They listed the following theoretical adverse events:

Reduced ability to control unexpected haemorrhage.
Difficulties from unexpected adhesions.
Rectal injury from inadvertent placement of laparoscope in women with obliteration of pouch of Douglas.
Risk of endometrial cancer cells spilled from the cervix.
Risk of vesicovaginal fistula (very rare).

Thirteen professional expert questionnaires for this procedure were submitted. Find full details of what the professional experts said about the procedure in the specialist advice questionnaires for this procedure.

Validity and generalisability

Nine studies were included in the key evidence summary, including 1 systematic review, 1 RCT, 2 prospective case series and 5 retrospective cohort studies. Research was done in various countries worldwide (Belgium, UK, China, Taiwan, Turkey and South Korea), but only 1 was from a UK setting.
The systematic review contained mostly retrospective comparative studies and only 1 single-blinded, non-inferiority RCT. The smallest study had a sample of 48 people. The RCT was assessed with the Cochrane RoB2 tool and deemed to have low risk of bias across the board. The remaining papers' risk was assessed with the ROBINS-I tool and were all deemed to have moderate overall risk of bias. Mean ages in studies were quite representative with all but the Kaya paper under 50.
The main direction of the efficacy and safety data across the 34 studies was either positive or comparable to conventional laparoscopic hysterectomy.
The mean age across the key evidence (excluding the systematic review) tended to be around 50 or lower.
Sample sizes varied considerably – from 29 people to 1,147 across studies. The only prospective UK analysis had only 33 people.
There was variability across the studies in the outcomes they reported. For example, some studies only had inferred procedural success (by lack of conversion) while others stated procedure success (rates) explicitly.
The studies were lacking long-term follow up. The only longer-term outcomes that were captured were self-reported pain or quality of life (up to 6 months) (Baekelandt, 2021).
Mostly consistent, but some variability, in patient inclusion and exclusion criteria.
Some variability in the procedure technique (where described) and access ports were often devices by 1 company (Applied Medical) or tended to be single-use, constructed-for-purpose ports or made by other companies (Beijing Aerospace Kadi Technology Development Institute).
Some study authors are prominent (featured often) in the literature for this procedure and were involved in multiple studies that make up this overview.
Disclosures included Jan Baekelandt (a prominent study author), Supuni Kapurubandara who have provided services for/been associated with Applied Medical.
Ongoing trials include NCT04886791, NCT05150275, NCT04324034, NCT05031182, KCT0004605.

How are you taking part in this consultation?

Interventional procedure overview of vaginal transluminal endoscopic hysterectomy and adnexal surgery for benign gynaecological conditions