Online survey on uterotomy closure techniques in caesarean section

Introduction

The caesarean section (CS) is the most common operation in women [1]. The rate has doubled worldwide between 2000 and 2015 [2]. Apart from its benefits, CS (and especially repeat CS) can cause complications in subsequent pregnancies such as uterine rupture, especially at trail of labor after caesarean section (TOLAC), and the development of placenta accreta spectrum (PAS) [3], [4]. The prevalence of PAS ranges from 0.003-0.24% for nulliparous women with no previous CS to 40% after two previous CS and placenta praevia totalis [4], [5], [6], [7]. Effective risk reduction can be achieved by reducing the number of operations on the uterus or by adequate closure technique [5, 8, 9]. Surprisingly, only 10% of the CS in Germany are performed from absolute indications [10]. Years of efforts to reduce the number of the first CSs were not successful. On contrary, there is a growing amount of CS on request without a medical indication.

One way to reduce the rising CS rates and the associated short- and long-term morbidity is to reduce the number of repeat CS and increase the number of TOLAC. This will be possible if we manage to make vaginal birth after caesarean section (VBAC) safer. Although the prevalence of uterine rupture is only 0.1–1.0%, it is associated with relevant maternal and neonatal morbidity [8, 11, 12]. The low prevalence described in the literature so far is certainly due to the high rate of planned repeat CS. A strong increase of uterine ruptures can nevertheless be expected with the globally rising CS rates as a Norwegian population-based study already has documented [2], [3].

Uterine scar healing and thickness plays a significant role for scar stability and therefore for a safe TOLAC. The influence of the uterine closure technique for the stability of the uterine scar and the risk of uterine rupture has become more and more focus of studies during the last years [9, 13–18]. Although most international societies do not recommend a specific uterine closure technique, currently the most commonly used sutures are single layer continuous, single layer locked sutures, double layer continuous and interrupted sutures (Figure 1) [19]. The British Guidelines of the National Institute for Health and Care Excellence (NICE) recommend a double layer suturing technique due to uncertain data for single layer techniques [20]. With the same argumentation, the guideline on caesarean section from the German, Swiss and Austrian societies for obstetrics and gynecology does not recommend a certain suturing technique [21].

Figure 1:

Uterine closure sutures.

Single layer continuous suture (a), single layer locked suture (b), double layer continuous suture (first suture as in (a) and above suture as in (c)), interrupted sutures (d).

Studies which compare the impact of suture techniques on the scar stability come to differing results. Some studies conclude that double layer closure leads to a thicker uterine scar in postpartum ultrasound controls – a sign for a more stable uterine scar [13, 14, 22, 23]. However, whether a sonographically “thicker” scar significantly reduces the risk of a rupture is also the subject of debate.

A Canadian multicenter case-control study with 384 women has confirmed that the risk of rupture after single layer uterine closure is significantly higher compared to a double layer closure [24]. Conversely, a meta-analysis of nine studies with 5,810 women could not show a significant difference between single layer and double layer uterine closure concerning the risk of rupture. However, single layer locked suture technique was significantly associated with a higher risk of rupture [15].

So far, limited data exists on the effect of interrupted sutures on uterine scar stability. When the Misgav Ladach method for CS gained worldwide popularity around the year 2000, this widely used technique was abandoned in favor of the single-layer locked technique [25]. However, the existing studies have shown some promising results such as in 30 primiparae, a significantly smaller defect of the uterine scar in transvaginal sonography and hysteroscopy 24 months after CS was shown when interrupted sutures had been used compared to single layer locked suture [26]. In another study, the uterine scar after interrupted sutures was even significantly thicker in hysterographic examinations three months after CS than after closure with double layer continuous sutures [27]. A retrospective study evaluated the uterine rupture rate in 1,059 pregnant women with TOLAC. Fifteen (0.8%) cases of uterine rupture were described, least likely after interrupted suture [28].

Since there is no common recommendation, we created an online questionnaire to evaluate which suture techniques are currently used in Germany and how often late complications are observed.

Materials and methods

Results

The response rate was 24.7% (160/648 hospitals with an obstetric department). The average processing time was 4 min.

German care levels have been adapted according to the definition of the American College of Obstetricians and Gynecologists (ACOG) [29], [30]. Maternal care level IV corresponds to the German perinatal center level 1 (compare Figure 2). Mostly hospitals with maternal care level I (basic care) participated in the survey (43.1%, 69/160), followed by maternal care level IV (regional perinatal health care centers) (26.3%, 42/160). The participation of hospitals with maternal care level II (specialty care) was 19.4% (21/160) and of maternal care level III (subspecialty care) 10.0% (16/160) (Figure 2). The mean annual number of births in 2018 was 1,328.7 ± 813.4 (range 235–5,400), in total 211,262 births in the participating hospitals. The mean CS rate was 27.3% ± 6.2, the mean repeat CS rate was 33.2% ± 18.1, and VBAC was performed in 46.2% ± 20.2 (Table 1). It is not clear from the data on the frequency of a VBAC whether it refers to TOLACs or successful spontaneous deliveries after CS (VBAC).

Figure 2:

Frequencies of care levels of the participating hospitals according to the classification of ACOG [29], [30].

Maternal care level IV = German perinatal care center level 1: preterm birth before 29+0 gestational age (GA), <1,250 g, triplets < 33+0 GA, severe fetal anomalies, severe maternal underlying disease; Maternal care level III = German perinatal care center level 2: preterm birth between 29+0 – 31+6 GA, 1,250–1,499 g, gestational diabetes with insulin therapy, pregnancy associated diseases (e.g. HELLP), intrauterine growth restriction (IUGR) Fetus (<3rd percentile); Maternal care level II = German hospital with perinatal focus (care level 3): preterm birth between 32+0 – 35+6 GA, ≥1,500 g, gestational diabetes with insulin therapy, IUGR Fetus (3rd–10th percentile); maternal care level I = German maternity hospital (care level 4): GA ≥ 36+0, no complications expected.

To close the uterotomy, 74.4% (119/160) used single layer continuous sutures, 16.3% (26/160) single layer locked sutures, 3.8% (6/160) interrupted sutures, 3.1% (5/160) double layer continuous sutures and 2.5% (4/160) other techniques (n=1 Lembert sutures (continuous inverted), n=2 first layer continuous, second layer z-sutures, n=1 first layer continuous, second layer double continuous sutures and single stiches) (Figures 1 and 3).

Figure 3:

Representation of the distribution of the applied suturing techniques.

In 57.5% (92/160), a standardized suturing technique was defined by the hospital. For 32.5% (52/160), the surgeons explicitly included the endometrium. While 96.9% (155/158) used the same suturing technique for repeat CS. Simultaneous tubal sterilization did not affect the suturing technique (Table 1).

In total, there were 466 observed dehiscences at planned repeat CS, 944 at emergency repeat CS and 284 uterine ruptures at VBAC in the year of reporting. This corresponds to a dehiscence rate of 7.6% for repeat CS in the interviewed collective. Almost half of the participated hospitals (44%, n=70) observed more than five dehiscences per year. Due to the study design, it was unfortunately not possible to indicate the percentage rupture rate of TOLAC or VBAC.

In total, the participating hospitals treated 660 PAS in 2018. This results in an average of 4.5 ± 6.3 (range 0–35) cases per year.

Considering the data in relation to the care levels, the following results are obtained (Table 2): Hospitals with maternal care level IV and III (German perinatal center level 1 and 2) had significantly higher birth and CS rates than level II (German hospital with perinatal focus) and I (German maternity hospital) (birth rate p<0.001, CS rate p<0.001). The rate of repeat CS and VBAC did not diverge between the care levels (repeat CS p=0.92, VBAC p=0.90). The percentages of observed uterine dehiscence did not differ significantly according to the different levels of care (Table 2). Only the absolute number of observed uterine ruptures per year was significantly different between the care levels. Maternal care levels IV had significant higher rupture rates than care level I and II (p<0.00, p=0.01). As well as the number of documented uterine dehiscences at planned and emergency repeat CS (each p<0.05). The number of PAS cases per year was significant higher in care level IV than in care level I and II (p=0.02, p<0.00). The suturing technique did not diverge significantly between the care levels (p=0.91, p=0.46).

Single layer continuous sutures were mostly applied in hospitals with a high birth rate (maternal care level IV and III). The single layer locked technique was used by hospitals with rather low birth rates. Those hospitals tend to have high repeat CS rates with 40% and low VBAC rates with 39%. The double layer continuous sutures were only used in five interviewed hospitals, but with high median birth rates. In this group, the VBAC rate was 59%. The interrupted suture technique was only applied by six of the interviewed hospitals and was mostly standardized by the respective hospital (Table 3).

Discussion

In our survey, the mean CS rate was 27.3 ± 6.2%, the mean repeat CS rate was 33.2 ± 18.1%, and VBAC was performed in 46.2 ± 20.2% of cases. Hospitals with maternal care level III and IV (German perinatal care center level 2 and 1) had significantly higher birth and CS rates than level II and I. The rate of repeat CS and VBAC did not diverge between the care levels.

In the literature, the CS rate in Germany is 32.0% with rates of planned repeat CS at 19.8% [31]. The rates in our survey are comparable with a trend to a higher repeat CS rate, especially in hospitals with maternal care level III and IV. These data show that the surveyed centers are representative of German maternal care hospitals.

The consequences of a disturbed uterine wound healing like uterine rupture and PAS is a feared complication in subsequent pregnancy and it is interesting that our results show that the used suture techniques diverge strongly between the obstetricians although there is increasing evidence that suture technique plays an important role [9, 13–15, 22–24, 26–28]. The single layer continuous suture technique is used the most (74.4%), although the only recommended suture technique from an obstetric society (NICE) is the double layer continuous technique [21]. Only 3.1% of the German hospitals surveyed comply with the recommendation from Great Britain.

Admittedly, several trials have not been able to detect significant differences in uterine rupture rates and myometrial thickness in postpartum ultrasound when comparing single and double layer continuous sutures [16, 18, 32].

Fewest of the participating hospitals use interrupted sutures to close the uterus during CS (3.1%). Interestingly, this technique has been shown to enhance wound healing in several small studies [26, 27, 33]. It appears that after the widespread adoption of the Misgav Ladach technique for CS around the year 2000, the technique of interrupted sutures has been buried in oblivion [25]. In times of rising CS rates and disproportionately rising uterine rupture rates, as documented in a Norwegian population-based study, it is important to find out whether interrupted sutures allow a better uterine healing and thus a safer vaginal delivery after CS and a lower risk for PAS in subsequent pregnancies [3], [9]. A Japanese case-control study with 98 pregnant women with placenta praevia investigated the effect of the suture technique at the prior CS on the frequency of PAS. The risk for PAS was significant lower when interrupted sutures were used compared to continuous sutures. The suture technique used on the inner wall made the difference and not if it was a single or double layer closure [9]. The average number of PAS in the surveyed hospitals is low with 4.5 cases per year. A CS with PAS should always be operated at a specialized center, since the operation can be very challenging and needs an interdisciplinary well-coordinated team. Prenatal screening for PAS to ensure referral of cases to specialized centers is essential.

In only two thirds (57.5%) of the surveyed hospitals the suture technique was defined. In general it is difficult to compare outcomes of surgical procedures if the surgical steps are not standardized [34].

Furthermore, current evidence supports the fact that uterine healing is improved and niche development is prevented by including the decidua into the suture [35], [36]. The survey showed that only 32.1% of hospitals do so.

The median uterine rupture rate at VBAC was with approximately 1 case per hospital and year a rare event and comparable to the data in the literature [8, 11, 12]. Hospitals with higher care levels (IV and III) have, in our survey, significantly more uterine ruptures at VBAC. Besides the higher birth rates, those hospitals take care of high risk pregnancies like women with twin pregnancies, pregnancies after myoma enucleation, insulin dependent gestational diabetes and often consecutive fetal macrosomia, which can contributes to the higher uterine rupture risk.

The percentage of uterine dehiscence at repeat CS in our survey is high with 7.6%. The definition of dehiscence is not clear and it is possible that data was influenced by different understanding of these terms [37], [38]. An indication of this are single extreme outliers. In this study design it was not possible to correlate the risk of uterine dehiscence or rupture with the applied suture technique, because the actual suturing technique used for the previous CS is often not known, since the women were partially operated in other hospitals and the operations techniques varies even within the clinic.

As primary CS are indicated conservatively in Germany, the most effective way to reduce the CS rate seems to be a reduction of the number of repeat CS. The use of the optimal suture technique to avoid disturbance of uterine wound healing can reduce the rate of PAS and uterine ruptures and can help to restore the confidence of a safe VBAC.

This survey is the first national questionnaire on uterine closure techniques. Currently, neither a core outcome set of questionnaires on obstetric operative management exists in the database of Core outcomes in women’s and newborn’s health (CROWN) nor is one in development. It shows that national online surveys on obstetric topics are feasible and makes it possible to get into discussion on the need to define a standardized uterine closure technique during CS.

The fact that the survey was limited to hospitals in the Federal Republic of Germany and the response rate of 24.7% limits the generalizability of the results. However, the wide share of degrees of specialization from basic care centers (level I, 43.2%) to regional perinatal health care centers (level IV, 26.3%) increase the possibility to extrapolate to the total population.

Furthermore, due to the retrospective survey design, the results are probably also subject to recall bias and reporting bias. The terms of uterine dehiscence, uterine rupture and PAS were not defined clearly and the amount of VBAC was only asked for as a percentage, so answers may be influenced by different understandings of these terms. It was also not stated whether and how often further (hemostatic) sutures, like z-sutures, were applied. It is important to emphasize, that it is not possible to make a statement about the correlation of suture technique and frequency of complications like uterine dehiscence and rupture or PAS since we had no information, if the previous CS was performed at the same hospital and which suture technique was used.

Conclusions

Although the closure technique of the uterus during CS plays an important role for scar stability, our results show that there is no uniform suturing technique in Germany. There is increasing evidence that the risk of long-term complications is associated with wound healing or suturing technique. A detailed surgery report with the used suture technique should be given to every patient to allow the obstetrician to plan the birth mode for the next pregnancy. The consent of the pregnant woman is a necessary requirement to a TOLAC. We would support the idea that women with previous operation on the uterus give birth (TOLAC or repeated CS) at regional centers like maternity care level III-IV to have experienced staff on site at all times in the event of complications and to be able to intervene immediately. Women with previous uterine operation should be standardized screened for PAS in the 2nd trimester ultrasound. We need a strategy to evaluate reliable the operation techniques at CS. A mandatory participation in a registry quality assurance based on the certification process for tumor centers would be an option.