Outcomes of the novel Odon Device in indicated operative vaginal birth.

BACKGROUND: No new method of assisting vaginal birth has been introduced into clinical practice since the development of the vacuum extractor in the 1950s. The Odon Device is a new device that employs a circumferential air cuff over the fetal head to assist birth. In this study, the Odon Device has been used to assist vaginal birth for standard clinical indications. OBJECTIVE: This study aimed to investigate the clinical impact, safety, and acceptability of the Odon Device to women, their babies, and clinicians and to assess the feasibility of recruiting women to an interventional intrapartum research study. STUDY DESIGN: This is a nonrandomized, single-arm interventional feasibility study of the Odon Device for operative vaginal birth undertaken in a single maternity unit: Southmead Hospital, Bristol, United Kingdom. The Odon Device was used to assist birth in 40 women who required the birth to be assisted for suspected fetal compromise and/or prolonged second stage of labor. The primary clinical outcome was the proportion of births successfully assisted with the Odon Device, and the primary feasibility outcome was the proportion of eligible women who were approached and who agreed to participate. Neonatal outcome data were reviewed at day 28, and maternal outcomes were investigated up to day 90. RESULTS: Between October 2018 and January 2019, 298 of 384 approached, eligible women (77.6%) consented to participate. Of these women, 40 received the intervention—the use of the Odon Device. Birth was assisted in all cephalic (occiput anterior, occiput transverse, and occiput posterior) fetal positions, at all stations at or below the ischial spine and with or without regional analgesia. The Odon Device was effective in 19 of 40 cases (48%). Of the 40births, 21(52.5%) required additional assistance: 18 of 40 births (45%) were completed using nonrotational forceps, 1 of 40 births (3%) required rotational forceps, and 2 of 40 births (5%) required an emergency cesarean delivery. There was no serious maternal or neonatal adverse event related to the use of the device, and there was no serious adverse device effect. There were 4 devices (10%) that were ineffective because of a manufacturing fault. Furthermore, 39 of 40 women (98%) reported a high birth perception score. All practitioners were able to use the device as intended, although some steps in using the device were reported to be easier to perform (setup and deﬂation of air chamber) than others (application of the device and withdrawal of the applicator). CONCLUSION: Recruitment to an interventional study of a new device for operative vaginal birth was feasible; 78% of eligible women were willing to participate, often expressing an aspiration for an alternative to forceps and vacuum. The success rate of the Odon Device was lower than reported success rates of vacuum and forceps; however, in this study, the device had been used to assist birth for standard clinical indications. There was no signiﬁcant maternal or neonatal safety concern associated with the use of the device, although the number of births studied was small. Further feasibility study to establish iterative changes to the device, technique, and clinical indications is necessary.


Introduction
The optimal duration of the second stage of labor is contentious, 1e4 but complications in the second stage of labor remain a major cause of preventable maternal and neonatal morbidities and mortalities globally. 4 Skilled operative vaginal birth improves the outcomes for women and their babies when birth is assisted in the second stage of labor for suspected fetal compromise and/or delay 4,5 compared with cesarean delivery. Despite this fact, the rate of cesarean delivery is increasing, whereas the rate of operative vaginal birth is low or nonexistent in many health services (reported rates as low as 0.5% of births). 6 The reasons for low operative vaginal birth rates include inexperienced healthcare workers, lack of equipment, and operator and patient dislike of currently available methods. 6 A new device to assist vaginal birth provides an opportunity to improve outcomes and reinvigorate this essential life-saving skill. However, in contrast to the huge advances in medical care over the past 70 years, no new method for assisting vaginal birth has been introduced into clinical practice since the development of the vacuum extractor in the 1950s. The advantages of the Odon Device compared with the advantages of standard methods for assisting birth (ie, forceps and vacuum) are currently unknown as the device has not yet been used in clinically indicated cases. Potential advantages could include improved neonatal outcomes (eg, reduction in subaponeurotic and retinal hemorrhages, skull fracture, and facial nerve palsy), increased maternal acceptability, reduction in failed rate of operative vaginal birth, decreased risk of adverse maternal outcomes (eg, postpartum hemorrhage and vaginal tears), and ease of use (ie, the same application technique of the device for all cephalic positions). Original Research

ajog.org
The Odon Device is a new device that can be used for operative vaginal birth ( Figure 1) consisting of a plastic applicator and polyethylene sleeve. The applicator has 4 flexible spatulas that position the sleeve over the fetal head. A progress indicator allows the operator to confirm the correct depth of insertion. The sleeve contains a circumferential air chamber that is inflated around the fetal head, providing the grip for the operator to apply traction (Video). The instructions for the use of the Odon Device are shown in Supplemental Figure 1. A phase 1 study of 4 earlier versions of the Odon Device in 49 healthy volunteers (women who were about to have a spontaneous vaginal birth) demon-

AJOG at a Glance
Why was this study conducted? No new method of operative vaginal birth has been introduced since the development of the vacuum extractor in the 1950s. The Odon Device is a new type of instrument that uses a circumferential air cuff over the fetal head and offers an alternative to vacuum and forceps.
Key findings There was no serious maternal or neonatal adverse event related to the use of the Odon Device. In the first 40 clinically indicated cases ever to be conducted, the Odon Device was successful in 19 (48%), and there was a high maternal birth perception score.

What does this add to what is known?
The Odon Device currently has a lower success rate than current devices, but this should improve with technique refinements. High recruitment rates to studies of novel devices to assist birth are feasible.  strated that the device could be applied to the fetal head during the second stage of labor with no serious safety concern. 7 Animal studies were subsequently performed for safety, and an extensive program of simulation studies confirmed that the device could be reliably sited and used in occiput anterior, occiput transverse, and occiput posterior fetal positions. 8e11 The ASSIST Study used the Odon Device to assist birth for routine clinical FIGURE 2 Consent and allocation to the intervention criteria AVB, atrioventricular block. ajog.org OBSTETRICS Original Research indications: suspected fetal compromise and/or failure to progress in the second stage of labor. The study aimed to investigate the feasibility of recruitment and to investigate the efficacy, safety, and acceptability of the Odon Device to women and their babies, midwives, obstetricians, and neonatologists.

Study design
This is a nonrandomized, single-arm feasibility study of the Odon Device for 40 women who required an operative vaginal birth for a recognized clinical indication. Qualitative work that explored the experience and views of women and clinicians will be published in full, separately.

Population
Potential participants were approached at Southmead Hospital, Bristol, United Kingdom. Figure 2 outlines the eligibility criteria for initial consent and allocation to the intervention if an operative vaginal birth for a clinical indication was required for prolonged second stage of labor or presumed fetal compromise-as defined by the Royal College of Obstetricians & Gynaecologists. 12 Women were provided with study information through a verbal discussion, an information leaflet, and a video demonstrating the Odon Device.

Intervention
Odon Deviceeassisted births were conducted by 1 of 5 obstetricians who had undergone specialized training (Supplemental Figure 2). If the Odon Device was unsuccessful, the obstetrician used their clinical judgment to complete the birth using forceps, vacuum, or cesarean delivery. After use (following high-level disinfection), devices were systematically examined by the study team and manufacturer. The device was used in women who had a fully dilated cervix, with the vertex presenting at or below the ischial spines, in any position.

Outcomes
The primary feasibility outcome was the proportion of approached and eligible women agreeing to participate. The primary clinical outcome was the proportion of births successfully assisted with the Odon Device. A birth was defined as "successful" if all of the following 6 criteria were met: (1) vaginal birth assisted with the Odon Device, (2) no serious maternal adverse reaction 13 related to the use of the device, (3) no serious neonatal adverse reaction 13 related to the use of the device, (4) no serious adverse device effect, 13 Table 2).
The PPS has been previously validated for operative vaginal birth. 14 Operators were asked their perceptions of the use of the device using a simple, nonvalidated tool (Supplemental Table  2). Secondary outcomes included metrics related to study feasibility and safety (ie, maternal perineal trauma and measured blood loss, neonatal soft tissues trauma and pain, and device safety) ( Figure 3). Neonates were followed up to day 28 via notes review and mothers on days 7, 28, and 90 after birth via telephone consultation. Further details are contained in the published study protocol. 15

Sample size
Sample sizes between 24 and 50 have been recommended for feasibility studies. 16,17 A complete sample size of 40 women was fixed a priori to enable the estimation of a potential rate of successful operative vaginal birth of 80% to within a 95% confidence interval (CI) of AE12%. This sample size will demonstrate the use of a secondary instrument of 50% to within a 95% CI of AE15%.

Statistical analysis
Data were entered and stored on a bespoke study database (GeneSYS) designed and managed by the Clinical Trial and Evaluation Unit, University of Bristol, Bristol, United Kingdom. Data were analyzed using Stata (version 15.1, StataCorp, College Station, TX). Continuous variables were reported as mean and standard deviation or median and interquartile range; categorical variables were reported as frequency and percentages. Relationships among characteristics that affect the success of the Odon Device were explored using nonparametric tests, the Fisher exact test for categorical variables, and the Mann-Whitney test for continuous variables.

Results Recruitment
Women were recruited between October 2018 and January 2019 ( Figure 4). Pregnancy notes of 545 women were screened; 441 of 545 women (80.9%) were initially deemed eligible and approached. Furthermore, 57 of the approached women were then identified to be ineligible. Of the 384 women who were approached and eligible, 298 (77.6%) consented to participate should they require an operative vaginal birth ( Figure 4).
Of the 298 women who had consented, 224 (75.2%) did not have an operative vaginal birth: 139 (62.1%) had a spontaneous vaginal birth, 76 (33.9%) had a cesarean delivery, and 9 (4.0%) were not yet in labor when the recruitment target was reached. In addition, 72 women (24.2%) who had provided consent required an operative vaginal birth; the Odon Device was used in 40 cases (55.6%).

Data quality
There were minimal missing data: body mass index at 36 weeks' gestation (6 of 40, 15%); umbilical artery pH and base excess (7 of 40, 18%); umbilical vein pH and base excess (4 of 40, 10%); Neonatal Participants were deemed lost to followup if they did not respond within 4 days of the planned follow-up. Follow-up was not obtained from 3 (8%) and 4 (10%) women on day 28 and 90, respectively. A day 28 neonatal follow-up (3%) was not completed. Table 1 summarizes the demographics and birth characteristics. Furthermore, 29 women (73%) had an operative vaginal birth for presumed fetal compromise and 11 women (28%) for prolonged second stage of labor. Birth was assisted in all cephalic fetal positions, at all stations at or below the ischial spines and with or without regional analgesia.

Primary clinical outcome
The Odon Device was the only device required to assist birth in 19 of 40 cases (48%). There was no serious maternal or neonatal adverse reaction related to the use of the device during birth, and there was no serious adverse device effect. All practitioners found the device easy to use; however, 1 woman (3%) whose birth was assisted using the Odon Device reported her perception of the operative birth as poor. Therefore, by our a priori definition, the proportion of births that were "successfully" assisted with the Odon Device was 18 of 40 (45%).
Additional assistance was required to complete the birth in 21 of 40 cases (52%). Of the 40 cases, 18 (45%) were completed using nonrotational forceps, 1 (3%) required rotational forceps, and 2 (5%) required an emergency cesarean delivery (1 immediately following an unsuccessful Odon Device and 1 after a failed attempt using rotational forceps). The fetal vertex was at the level of the ischial spines in all cases requiring rotational forceps or cesarean delivery.
In addition, 39 of 40 devices (98%) were inspected after use, and 4 of 40 devices (10%) had a fault in the bulb pump mechanism, unable to inflate the air cuff. None of the faulty device had been successful, and in each case, the operator had raised concerns that the air chamber had not inflated adequately during use. Table 2 summarizes the characteristics of 36 births, which were assisted with

Clinical birth outcomes
Neonatal outcomes Table 3 summarizes neonatal outcomes. In this study, 2 infants (1 successful Odon Device and 1 failed Odon Device followed by nonrotational forceps, both cases assisted for "presumed fetal compromise") had an Apgar score of <7 at 5 minutes of life (both Apgar scores of 6). No infant was born with an Apgar score of <7 at 5 minutes where the indication to assisted birth was prolonged second stage of labor. In addition, 3 infants (8%) were admitted to the neonatal intensive care unit (NICU) following birth, all with respiratory distress. In all 3 cases, the indication for operative vaginal birth was "presumed fetal compromise." NIPS was reassuring in only 1 infant, who was admitted to the NICU, with a score of !4 indicating pain.
Most neonatal events (16 of 17) were attributed to soft tissue trauma (ie, bruise, graze, scalp injury, or facial injury) ( Table 4). There was less soft tissue trauma in infants successfully delivered with the use of the Odon Device; 3 babies (16%) whose birth was successfully assisted with the Odon Device had evidence of soft tissue trauma compared with 12 cases (60%) in which the device failed. In addition, 4 infants (10%) had a cephalohematoma diagnosed at their postnatal checkup (2 successful and 2 unsuccessful cases). No infant required phototherapy for jaundice contributed to by bruising, received a blood transfusion, had a neonatal seizure, was therapeutically cooled, was diagnosed with organ failure, or died within 28 days following birth.
Furthermore, 11 infants (28%) experienced a serious adverse event (SAE) defined as an event that required hospitalization or prolongation of hospital stay or further intervention: 3 cases of jaundice requiring phototherapy not contributed to by bruising; 2 cases of respiratory difficulties requiring NICU admission; 1 prolonged neonatal stay for intravenous antibiotics secondary to maternal sepsis in labor; 1 hospitalization for weight loss; and 1 readmission for bronchiolitis aged 10 days. None of the cases were directly attributable to the use of the Odon Device. Table 5 summarizes maternal outcomes. In this study, 36 of 40 women (90%) had a perineal tear (28 episiotomies). The rate of episiotomy in successful Odon Deviceeassisted births was the same as when forceps were used-14 of 19 cases (74%), respectively. Furthermore, 3 women (8%) sustained a third-degree perineal tear: 1 (3%) during the successful use of the Odon Device and 2 (5%) when forceps were used following a failed Odon attempt. In addition, 1 cervical tear (3%) (which did not require Demographics and characteristics of participants (continued)

Variable
Overall (n¼40) ajog.org OBSTETRICS Original Research suturing) was identified after a successful Odon Deviceeassisted delivery. The median weighed blood loss was 499 mL (25th 355 mL, 75th 810 mL). Furthermore, 2 women (5%) received a postnatal red blood cell transfusion. In both cases, the Odon Device had been unsuccessful, and nonrotational forceps were used, resulting in hemorrhage from perineal trauma.

Molding
In addition, 7 women (18%) experienced an SAE: 3 third-degree tears, 2 postpartum hemorrhages required blood transfusion, 1 postnatal infection, 1 complication from a known neurologic condition, and 1 case of fecal incontinence at 28 days after birth.

Device outcomes
In the 19 cases in which the Odon Device assisted the birth of the baby, the median time between the application of the device and time of birth was 5 minutes (25th 4, 75th 7), with a median "decision-to-delivery" interval of 11 minutes (25th 9, 75th 17). There were 21 births where the Odon Device did not assisted the birth of the baby: a manufacturing error affecting 4 devices accounted for 19% of the failed births; 3 failures (14%) occurred when the vertex was at the level of the ischial spines; further failures were attributed to (1) failure to reach "0" during the application of the device (n¼6, 29%), (2) operator learning curve (n¼3, 14%), and (3) no reason identified (n¼5, 24%).
The operator reported that the device was "easy" or "very easy" to "set up" and "deflate the air chamber" in 38 (95%) and 35 (88%) of 40 cases, respectively. The application of the device and the withdrawal of the applicator were more challenging: only 21 (53%) and 20 (50%) cases were reported as "easy" or "very easy," respectively.
The 5 operators joined the study in stages. The number of births and success rates per operator were variable

Principal findings
It is feasible to recruit women to a study investigating a novel device for operative vaginal birth. A recruitment rate of 78% is high and appears to be related to a general maternal aspiration for an alternative to forceps and vacuum (details from qualitative findings that will be published separately). The success rate of the Odon Device was lower than the reported success rates of both vacuum and forceps 12,19 ; however, the device had been used to induce birth where clinically indicated.

Results
Factors that are likely to affect the efficacy of the device assisted the clinical context, technique of application and traction, the initial uncertainty of the optimal technique in inserting and using the device (as the device was used for clinically indicated cases), and the design and functionality of the device itself. For example, it became clear after 3 births that the device is most effectively placed during a uterine contraction, in contrast to forceps and vacuum.
Although the study was not powered to look at predictors, births were more likely to be successful if the fetal vertex was lower in the birth canal or if there was minimal molding of the fetal skull or similar other assisted births. 12 The Odon Device was never successful when it was used with the vertex at the level of the ischial spines; indeed, a vertex above þ1 is a contraindication to attempting operative vaginal birth in some national guidelines. 20 The average time of decision to delivery was 11 minutes, which was faster than documented average times in the literature of 15 to 59 minutes depending on whether the birth was assisted in a labor or operating room. 21e23 The Odon Device was successfully employed across the full range of fetal positions using the same technique, which may be particularly useful for maternity settings with low operative vaginal birth rates. During the routine device inspection following birth, 4 devices (10%) were found to have a fault with the inflation mechanism. Further investigation along with the manufacturer identified a single issue related to the use of gamma radiation in the sterilization process that degraded the device components. This issue has been rectified by the manufactures and highlights the importance of postuse device inspection during feasibility studies.

Clinical implications
There was no significant maternal or neonatal safety concern associated with the use of the device, although the number of births is small. There seem to be some neonatal benefits associated with the use of the device, consistent with the findings from simulation 11 ; 16% of babies whose birth was successfully assisted with the Odon Device had evidence of soft tissue trauma compared with 60% of babies in cases where the device was unsuccessful. We recorded any soft tissue trauma, including a bruise, graze, laceration, or hematoma, whether it appeared to be related to device use or not. Evidence demonstrated that the laceration rate following operative vaginal delivery can be up to 10% 22 and instrument bruises up to 37% 24 ; however, there are few studies reporting all soft tissue injury. We had 4 cephalohematomas (10.3%), which ajog.org OBSTETRICS Original Research agrees published rates of occurrence following operative vaginal birth. 22 Clearly, the use of 2 instruments is associated with an increased risk of neonatal trauma, 25 but the trauma rates in this study were consistent with other operative vaginal birth studies in the same unit. 26 Operators used their clinical judgment to decide whether to use a second device or perform a cesarean delivery following the failure of a primary device. This study routinely used NIPS to assess infants following operative vaginal birth, and the scores have indicated reassuringly low levels of neonatal pain. Maternal outcomes were also acceptable. No woman required a blood transfusion after a successful Odon Deviceeassisted birth compared with 10% of women in unsuccessful cases. Clearly, this may reflect the use of 2 instruments 25 and/or a more difficult cohort of births. Overall, the maternal outcomes were again comparable to a previous study of operative vaginal birth in the study unit. 26 There were 3 thirddegree tears (7.5%) during the study, 1 (5.3%) from a birth successfully assisted by the Odon Device and 2 (9.5%) from births requiring a second device to assist birth. This agrees tear rates in the literature, which range from 1% to 4% for ventouse-assisted births, 22 8% to 12%    for forceps-assisted births, 22 and 17.4% for women who had births assisted with sequential devices. 25,27 This study routinely collected maternal birth perceptions as part of the assessment of operative vaginal birth. It is encouraging that 98% of women in the study rated their birth very positively.

Research implications
A future randomized controlled trial will be required to objectively compare outcomes among the Odon Device, forceps, and vacuum. However, this would require equipoise among the options for operative vaginal birth. We are confident that the success rates of the Odon Device can be improved with iterative improvements to the insertion technique, refinement of participant selection, and a more reliable device design. Further feasibility studies are currently underway in Bristol, United Kingdom (https:// www.nbt.nhs.uk/research-innovation/ our-research/current-research/womenchildrens-health-research-unit/women-4), and Besançon, France, and the success rate has markedly improved in this second phase; an interim analysis of the current ongoing studies demonstrated a success rate of 78% (66 of 85 cases). 28

Strengths and limitations
The Odon Device has been used in clinically indicated cases for operative vaginal birth. It demonstrates an innovative approach to information sharing and recruitment, with positive results. The study ensured a holistic approach when evaluating a novel device, ensuring that data from women, infants, clinicians, midwives, and the device were scrutinized. Key quantitative data were

Conclusions
Investigating a new device for operative vaginal birth is both feasible and supported by women. The potential advantages of the Odon Device (eg, possible reduction in neonatal soft tissue trauma, single mode of application irrespective of fetal position) merit further iterative exploration and investigation of the device in larger studies.

Highlights
The Odon Device may offer women an alternative instrumental birth. There was no maternal or neonatal safety concerns in the births. Recruitment rate was higher than expected. n

Operator training for the ASSIST Study
All operators undertaking operative vaginal births as part of the ASSIST Study attended a half-day training session. This included an instructional video on the use of the Odon Device, with accompanying training slides and intensive 1:1 practical teaching of how to use the Odon Device on a high-fidelity pelvic simulator. The training is based on the Royal College of Obstetricians & Gynaecologists training course for instrumental birth, ROBuST, and the findings of the extensive human factor studies conducted using the Odon Device. The half-day training sessions will be repeated if any applicable amendments are made to the study protocol or device, to ensure that all operators remain up to date. Should any operator be found during interim analyses to be performing with a success rate deemed to be insufficient, they will undergo retraining in both the Odon Device training package and ROBuST.
ROBuST, RCOG Operative Birth Simulation Training.