Introduction

Vacuum extraction is indicated when vaginal delivery needs to be assisted or accelerated in the second stage of labor (Chalmers and Chalmers 1989). The rate of instrumental vaginal delivery (ventouse and forceps) in different countries varies between 10 and 13% (Bahl et al. 2010). In the Netherland, the rate of instrumental vaginal delivery by woman delivering in a secondary hospital was 12,2% in 2013 (Brouwers et al. 2015). 

Since the development, the vacuum extractor (or ventouse) is more frequently used for assisted vaginal delivery, due to its decreased risk of maternal morbidity compared with forceps (Miksovsky and Watson 2001). More recently, the soft cup vacuum extractors were developed and introduced in a clinical setting. There is contradicting evidence regarding the benefits of soft cups in comparison with metal cups. Some literature reports present higher success rates of the metal cups (OR 1.63, 95% CI 1.17-2.28) (O'Mahony et al. 2010). However, the soft cups seem potentially safer than metal cups, especially looking at fetal morbidity as cephalhaematoma or scalp injury (O'Mahony et al. 2010). The number of studies comparing effectiveness of different kinds of soft cups is limited.

Therefore, the aim of this study was to compare three different and frequently used types of vacuum extractors, two soft cups (the Mystic pump, mushroom cup (MP) (Coopersurgical Inc, Trumbull) and the Kiwi omnicup (Clinical Innovations Inc, Salt Lake City) and the conventional metal cup (Medela Inc, MCHenry). Primary outcomes were the failure rate of the instruments. Secondary we compared the maternal and neonatal morbidity between the different instruments.

Material and methods

This retrospective research was performed in the Diakonessenhuis Utrecht, a large secondary teaching hospital. Data was collected using our validated hospital medical database. Informed consent for anonymous use of medical data was obtained from all included patients at first contact. The ethical committee of the Diakonessenhuis stated that the current study did not fall within the purview of the Medical Research Involving Human Subjects Act. Therefore, it was possible to conduct the study without approval by an accredited research ethics committee. All data were double checked by comparing them to the actual medical status. All deliveries from October 2012 till January 2014 were collected.

All deliveries were performed by trained and skilled obstetricians. Vacuum extraction was performed to accelerate delivery, when necessary for maternal or fetal wellbeing, such as non-progressive labour. Three instruments could be chosen for assisted vaginal delivery: the Mystic pump and Kiwi omnicup, both soft cups, or a conventional metal cup. The instrument of choice was based upon experience, preference and clinical assessment of the obstetrician.

Three instruments for vacuum delivery were used. The MP is a relatively recent handheld vacuum device, with a mushroom-shaped cup and a flexible stem, which flexes up to 90º in any direction to facilitate use in non-occiput anterior positions. It reaches a 580mmHg negative pressure. The kiwi omnicup is a more well-known device,  a handheld model with flexible stem and a low-profile cup, reaching a 600mmHg negative pressure. The conventional cup is a device developed in the 50’s, with a metal cup connected to a pump system, reaching an 600mmHg negative pressure.

All instruments were used according to the manufacturer's guidelines. The bladder was emptied and patient was placed in dorsal lithotomy position. Position and the station (at least Station 0) of the fetal head were assessed, after which the cup was placed as close as possible at the flexion point. After verification of the correct placement of the cup, vacuum was created gradually by increasing suction until a negative pressure of 600 mmHg for the conventional cup, 580 mmHg for the MP and 600 mmHg for the Kiwi was reached. Traction was given simultaneously with the uterus extraction and pushing of the mother. Routine use of medio lateral episiotomy was advocated (Jango et al. 2014). The maximum number of tractions was three.  After three pulls ceasarion section was strongly considered according to the guidelines of the dutch gynaecologist and obstretrian board (NVOG). Only in highly exceptional cases, more tractions were done. Extraction failed when the cup was detached 2 times or no progress was made.

In- and exclusion criteria

Data were identified and collected by one researcher (RA). In a few cases, there was a change of instrument during the vacuum extraction procedure, due to a prompt instrument defect. When there was a multiple use of instruments during the delivery, only the first used instrument is taken in to account, and noted as a failed vacuum extraction. We excluded premature deliveries (GA < 37 weeks) and multiple pregnancies.

Background and outcomes

Patient characteristics were assessed: parity, maternal age, obstetrical history, BMI, gestational age at time of delivery, birth weight and specifics about the delivery (anaesthetics, spontaneously onset or induction, oxytocin use). Characteristics of the vacuum extraction, type, position of the head, number of tractions, cup detachments and failure rate) were included. Failure rate is described as no successful delivery with de chosen instrument, leading to a caesarean delivery. Perineal lacerations, as classified by Sultan et al (Sultan and Thakar 2002) were described: moderate lacerations includes grade 1 and 2 lacerations (rupture of skin or skin and perineum) and severe laceration includes  grade 3 and 4 lacerations (perineum and anal sphincter of perineum, anal sphincter and anal mucosa). Neonatal complications includes Apgar scores, pH value umbilical cord artery, trauma and clinical admittance.

Statistics

Statistical analysis was performed using SPSS version 17.0 (SPSS Inc., Chicago, IL, USA). Results were analysed using the Chi-square test, Kruskal-Wallis test and ANOVA test. A two-sided p-value less than 0.05 was considered statistically significant.

Results

A total of 2825 deliveries were identified. After exclusion, 363 (13%) vacuum extractions were included in our analysis: 278 in the conventional metal cup group, 34 in the Kiwi group and 51 in the Mystic Pump group. In three cases, there has been a switch of vacuum extraction instrument, twice from Kiwi to conventional, once from MP to Kiwi. In all the three cases the reason for replacement by another instrument was a defect instrument, there was no suspected cephalopelvic disproportion. All three times delivery was successfully performed by the second instrument tried with. Only the vacuum extraction with the first instrument is taken in to account.

There were no significant differences in demographics (Table I) or in labor details between the three groups of the study population (Table II). Most of the included patients are nulliparous, with normal body mass indices who were in spontaneous labor.

No statistically significant differences in the failure rates of the vacuum extraction between the groups were found (Table III).  The highest failure rate is in the Kiwi group, with also the highest detachment percentage, although both not significant. The number of tractions is significantly decreased in the MP group.

There were four cases of neonatal trauma, two clavicular fractures, one scalp injury and one small subdural hematoma. All neonates recovered without any long-term injury. There were no casus of subgaleal hemorrhage or cephalohematoma. There were no significant differences in maternal and neonatal outcome variables (Table IV). 

Table 1

Demographics details of study population

Table 2

Labor details of study population

Table 3

Primary outcomes

Table 4

Secondary outcomes: maternal and neonatal morbidity

Discussion

This is the first report comparing the effectiveness of three commonly used vacuum extractors, including a relatively knew and unknown soft cup, the Mystic Pump. No significant differences are found between the three different cups concerning failure rate, cup detachments, and maternal or fetal morbidity. All of the extractors seem to be highly successful in achieving successful assisted vaginal delivery, although MP requires a lower number of tractions.

The lower amount of tractions required in het MP group is remarkable and has not been described before. This suggests that the MP is a more preferable instrument, due to its comparable success rate with significantly less traction needed. However, due to the retrospective nature of this study, we cannot rule out the influence of the station of the caput at the moment of cup application on the number of tractions or success rate.

Success rates described in this article are similar or higher compared to other literature (O'Mahony et al. 2010). Attilakos et al. and Groom et al. studied the effectiveness of the ‘standard vacuum cup’ to the newer rigid plastic cup (Kiwi omnicup in both studies)in a randomized setting (Attilakos et al. 2005;Groom et al. 2006). In the Attilakos study, among 194 deliveries, the failure rates were 34% for the ‘standard vacuum cup’ versus 21% for the kiwi omnicup (adjusted OR 2.3 (1.01–5.0))(Attilakos et al. 2005). These are higher rates than our data (3.9-14.3%), but most of the  cups were placed when the fetal caput was still high in the birth-canal, at spines (kiwi ommicup 13%, standard cup 11%) or at station spine+1 (out of three) (kiwi omnicup 61%, standard cup 55%).  

Groom et al., also noticed a significant difference in failure rate in favour of the standard vacuum cup, 30.1% (standard cup) versus 19.2% (Kiwi omnicup) (RR 1.58; 95% CI 1.10–2.24)(Groom et al. 2006), in a study including 404 patients. The higher failure rate is related to a higher number of cup detachments in the Kiwi omnicup group. Remarkable, only looking at the vacuum extractions from a station ≥ 2, in Grooms et al the failure rates are comparable (20.3 Kiwi omnicup group, 20.9 standard vacuum group, RR 0.97 (95% CI 0.51–1.84)).

Turkmen shows in a retrospective cohort study, as well as our data, no difference in maternal and fetal outcomes and failure rates between Kiwi and Metal cup (Turkmen 2014), even as Ismael et al. They have a vacuum extraction success rate of 100% in both groups (Ismail et al. 2008), possible partly by a lower overall rate of assisted vaginal delivery (8.2%) and a higher forceps rate (27%).

The only study found which included an instrument of mitysoft, as our mystic pump, was the study performed by Bothuyne-Queste (Bothuyne-Queste et al. 2009). It shows a smaller success rate of the mitysoft bellcup compared with the Kiwi omnicup (failure rates 31.3% versus 12.8%) (Bothuyne-Queste et al. 2009). Instead of a bellcup we used a mushroom cup, which could explain the improved success rate in our study. A mushroom style cups generate a greater traction force than Bellcups (Muise et al. 1993). Therefore, they are less likely to slip or release of the fetal head, which is related to failure of the vacuum extraction.

In interpreting our results, we acknowledge the fact that the retrospective design of this study did not enable us to reveal the true motivation by the clinician to choose the allocated instrument. 

There is only a small trend to a higher maternal BMI in the conventional group (p=0.06), caused by a few outlying BMI values: two women had a BMI of 50, five women had a BMI higher than 35. In labor details, there are a lot of missing data in the variable ‘caput position at moment of cup application’ (occiput anterior, occiput posterior etc.) (table II). All three vacuum extraction instruments are suitable for the different caput positions.

This study includes the MP, a new design vacuum extractor. Little research had been done with this tool and only few data is available. A variety of possible important variables were taken in to account. The well-trained obstetricians strengthen this research, especially because of the high rates of inappropriate placement of the cup mentioned in literature (Sau et al. 2004). 

Conclusion

In conclusion, the conventional metal cup, the kiwi omnicup and the mystic pump seem to be very safe instruments with high success rates in vacuum assisted vaginal delivery. MP seems to be slightly more effective in terms of number of tractions with very low failure rates as well. Randomized controlled trials and long term research should further elucidate the advantages and disadvantages of vacuum extractors, which is the aim of future research. For the moment, we advise that the instrument of first choice should be determined by the preference and experience of the obstetrician and individualized according to the current clinical situation.