A seroma consists of an accumulation of serous fluid in a dead space created by dissection. Seroma formation is particularly frequent after open abdominoplasty or endoscopic-assisted diastasis recti (DR) correction by subcutaneous access; the fluid collection is located between the skin flap and the anterior sheath. The supposed causes of seroma formation are the creation of a dead space by extended dissection, disruption of blood and lymphatic vessels, and an inflammatory response as well as the shearing effect of movements between the skin flap and the anterior sheath [1]. Some risk factors are mentioned in the literature, such as male sex [2] and high BMI [1, 3].

In most cases, seromas are considered minor complications—or even not accounted for as such—and are not reported. Nevertheless, seromas can also cause discomfort and require frequent outpatient visits and punctures; in rare instances, they can result in severe complications requiring reoperation, such as the formation of a pseudobursa or infection.

The incidence of seroma formation after DR correction varies across studies. According to a review on different techniques for DR repair [4], the percentage of seroma formation varies from 0 to 30% for open surgery techniques and from 3 to 27% for endoscopic techniques.

Quilting is one of the diverse methods proposed to reduce seroma incidence, and it has been shown to be effective in open abdominoplasty [5, 6]. We have adapted the quilting procedure used in open abdominoplasty to the endoscopic-assisted bilayer suturing technique introduced by Ngo et al. [7]. The aim of this study was to describe the endoscopic-assisted technique of quilting and assess its efficacy by comparing two groups of patients who underwent the same technique of repair before and after introducing the quilting procedure.

Methods

Study design

This study was based on the data of a continuous series of patients who underwent surgery for concomitant ventral hernia (VH) and DR at the same institution by two expert surgeons using the same technique of endoscopic-assisted subcutaneous suturing repair.

All consecutive non-selected patients who underwent surgery from 2019/05/20 to 2023/10/31 were included. This retrospective study was based on data that were prospectively registered in the French Club Hernie database, which includes all cases of repair of any type of abdominal wall dehiscence, without any exclusion.

The study included two groups of patients: Group 1 included patients who underwent surgery before the introduction of the quilting method (up to July 2022) and Group 2 included patients for whom quilting was applied (from July 2022). Seromas were classified into two types: (1) Type A included seromas that were spontaneously resorbable and seromas that were drained by a single puncture and (2) Type B included seromas that required two or more punctures and complicated cases requiring surgical intervention (Fig. 1).

Fig. 1
figure 1

Flowchart: patient allocation

Full size image

Surgical techniques

Group 1

The operation was performed according to the technique described by Ngo et al. [7].

The first step was performed through a short periumbilical skin incision using a 6-cm Alexis retractor. The umbilical hernia was reduced, and the orifice was closed by a transverse continuous suture of slowly absorbable material. The medial edges of both rectus muscles were sutured together 5–6 cm above and 5–6 cm below the umbilicus. Then, two 3.5-mm or 5-mm trocars were introduced on both sides of the Alexis retractor, the laparoscopic cap and the endoscope were placed on the retractor, and the endoscopic step was carried out. The insufflation pressure was 12 mmHg at dissection, and it was reduced to 4 mmHg for suturing.

Extended preaponeurotic dissection and approximation of the entire length of the medial edges of both rectus muscles by continuous suturing of 0 barbed slowly absorbable material (Stratafix, Ethicon) joining the previous suture line were carried out.

Group 2

The same technique of repair was performed, and it was completed with quilting sutures of 3/0 resorbable material (Vicryl, Ethicon).

The quilting stitches took a good depth of the subcutaneous tissue, although avoiding taking the dermis.

At the endoscopic step, 2–3 stitches were placed on the midline above the umbilicus (Fig. 2), and 1–2 stitches were placed on both sides at the lateral border of the rectus muscle (Fig. 3). Tying the knots was facilitated by the low insufflation pressure and by pushing down the skin flap with the finger.

Fig. 2
figure 2

Quilting, endoscopic view: midline quilting stitches, the upper already tied, and the lower being placed

Full size image
Fig. 3
figure 3

Quilting, endoscopic view: midline stitches tied and left lateral stitch being placed

Full size image

After removal of the cap of the retractor, four more stitches were placed via the direct approach at the four cardinal points to form a circle around the umbilicus.

After removing the retractor, four additional cardinal stitches were placed within the previous circle.

Then, the umbilicus was fixed by one or two stitches. The positions of the quilting stitches are shown in Fig. 4.

Fig. 4
figure 4

Positioning of quilting stitches, the two lateral upper stitches (in dotted line) are optional

Full size image

In both groups, no drains were used, no concomitant procedures were performed, and abdominal binding was applied for 3 to 4 weeks.

Collected data

The data collected included patient demographics, history of pregnancies for women, history of parietal surgery (abdominal scar, umbilical scar, liposuction, abdominoplasty), anticoagulant/antiaggregant therapy, VH hernia type and diameter, inter-rectus distance (IRD) at rest and on exertion, type of hospitalization, operation duration, postoperative pain at rest and on exertion assessed by visual analog scale (VAS), paracetamol and Class 2 analgesic consumption, time to resume daily activity and time off work, type of seroma, and other postoperative complications.

Seroma diagnosis and management

Using the above technique, the diagnosis of seroma was easy since fluid collection occurs in the subcutaneous space. Therefore, the diagnosis was made at the first postoperative visit, 8–10 days after the operation, by clinical examination, without resorting to imaging. It was based on the perception of a renitent swelling at clinical examination or a fluid collection that moved depending on the patient’s position. It was our policy to drain every significant seroma via needle aspiration. Puncturing and draining are indeed considered advised treatment options in cases of significant fluid accumulation [8]. The puncture was performed by the surgeon at the outpatient visit in a sterile setting, except in the few cases where the surgeon expected spontaneous resorption due to the small volume of the seroma; the decision was left to the surgeon’s discretion. Every patient with a seroma was advised to return 10 days later for control. The puncture was reiterated as many times as necessary to speed up adhesion of the pouch walls.

Endpoints

Comparison of the global incidence of seromas and the incidence of Type A and Type B seromas between Group 1 and Group 2 patients.

Statistics

Descriptive and comparison statistics were calculated using the “Statistical Package for Social Sciences” (SPSS) software. For quantitative data, means are presented with their standard errors (SE), which allows direct estimation of their confidence intervals. Comparisons of the means of the quantitative data between Group 1 and Group 2 were drawn by Student’s t test, with a risk of 5%. The distributions of qualitative and categorical data were compared between the two groups using the Pearson Chi-square test, with a risk of 5%.

Ethics

The Club Hernie Registry complies with the RGPD and the French Commission Nationale de l’informatique et des Libertés (CNIL: MR004 N° 2,212,908), and the present study received an IRB agreement, N° COS-RGDS-2023-06-004, from IRB00010835.

Results

The study was based on the data of 176 consecutive patients categorized into two groups: Group 1 included 102 patients who underwent surgery without quilting (control group) and Group 2 included 74 patients in whom the quilting procedure was applied (interventional group).

The mean age was 43.6 (18–82) years, and the mean BMI was 22.74 (15.94–41.02). The mean hernia diameter was 1.6 cm (1–3). The mean IRDs at rest and on exertion were 6.55 (3–20) cm and 3.85 (2–10) cm, respectively.

The results of the comparisons between the two groups are given in Table 1. The characteristics of the patients included in Group 1 and Group 2 were comparable, except for the IRD at rest, which was significantly greater in Group 2.

Table 1 Comparison between Group 1 and Group 2
Full size table

The global percentage of seromas was 24.4%. The percentage of seromas of any type was significantly greater in Group 1 (27.5%) than in Group 2 (20.3%). The percentage of Type B seromas was greater in Group 1 (19.6%) than in Group 2 (5.4%), when the percentage of Type A seromas was greater in Group 2 (14.9) than in Group 1 (7.9%). These differences were significant (p = 0.014).

The mean operation duration was significantly longer in Group 2 than in Group 1 (83.9 mn vs. 69.9 mn; p = 0.001). The mean postoperative pain assessed by the VAS was greater in Group 2 than in Group 1 at rest (1.85 vs. 1.25; p = 0.005) and during exertion (3.95 vs. 3.18; p = 0.007). Paracetamol consumption was greater in Group 2 than in Group 1 (4.03 vs. 3.21 days; p = 0.029), but the consumption of Class 2 analgesics was not different.

The time to resume daily activities and the time off work were not significantly different.

Two postoperative hematomas, one in each group, were drained by needle puncture. Four complicated cases requiring reoperation (included in Type B) were observed in Group 1: three cases of persistent seroma treated by surgical drainage under general anesthesia and one case of encapsulation in a patient who missed postoperative visits because of COVID-19 containment measures.

Only 14 patients (8%) did not attend the 3-month medical visit. Among the 162 patients (92%) who were observed, there were no seromas and one residual hematoma that was punctured (100 mL). This patient was one of the two patients who had a hematoma drained by puncture at the postoperative visit.

Discussion

The results suggest that the quilting procedure adapted to the endoscopic-assisted suturing technique for the repair of concomitant VH and DR can minimize the global incidence of seromas and the incidence of cases that require two or more punctures as well as complicated cases requiring surgical intervention (Type B). Conversely, there were more Type A seromas (spontaneously resorbable and requiring a single puncture) in the quilting group than in the control group. In other words, in addition to minimizing the global incidence of seromas, quilting results in a shift from severe cases to mild cases. Moreover, it should be noted that the four patients with complications who required reoperation were all included in Group 1.

In patients who underwent the quilting procedure, the operation duration was significantly longer (83.9 mn vs. 69.9 mn; p = 0.001), postoperative pain assessed by the VAS score was greater, and the duration of consumption of paracetamol (but not Class 2 analgesics) was longer.

However, although the differences were statistically significant, they may not be considered clinically relevant due to the low level of pain in both groups at rest (1.85 vs. 1.25), with exertion (3.95 vs. 3.18), as well as the low paracetamol consumption (4.03 vs. 3.21 days).

The differences between the two groups in the time to resume daily activity and the time off from work, as well as the type of hospitalization, were indeed not significantly different, which means that adding the quilting procedure does not truly impact patient comfort.

The incidence of seroma formation varies considerably across studies. With the use of endoscopic-assisted and totally endoscopic techniques for DR correction via a subcutaneous approach, the incidence of seroma formation varied from 2.5 to 28.6% [7, 9,10,11,12,13] and even 81% in one study [14]. These large differences between similar techniques suggest that different teams do not talk about the same things. In fact, a systematic review of the incidence of seroma after ventral hernia repair revealed a difference between “significant seromas,” which represent actual complications, and “incidental seromas,” which disappear spontaneously [15]. According to the first Italian Consensus Conference on Laparoscopic ventral incisional hernia repair, only cases that are clinically evident and persist for more than 2 months meet the definition of seroma [16]. According to Dey, seromas should be considered complications only if they need any type of intervention [17]. In our study, all types of seromas were considered, from simple spontaneous resolution cases to complicated cases requiring reoperation, thus providing a realistic overview of this issue.

Different methods of seroma prevention following abdominoplasty have been proposed.

Drains have been used for a long time, but they are a cause of discomfort and pain as well as a risk of septic contamination; they are not well adapted to day-case surgery, and they have not proven effective in all cases, especially as seroma formation can occur 8–10 days after surgery [18]. We do not use drains because we share the same view. The effectiveness of glue has not been proven according to a systematic review [19]. Three methods have been shown to be effective, including preservation of the fascia of Scarpa [8], injection of 12% NaCl in the operative field through drains at the end of the operation to stimulate fibrotic reactions [20] and quilting. Quilting was first described by Baroudi and Ferreira in 1988 [21], and it was subsequently used by many other authors. The method consists of placing stitches between the subcutaneous layer and the aponeurotic layer to obliterate the dead space and avoid shear movements between both anatomic planes. The progressive tension suture (PTS), used for abdominoplasty, is a variant of quilting that has been shown to be effective in a large series [5] as well as in a comparative study [6]. Given the good results provided by quilting in previous publications, we decided to adapt the method to the endoscopic-assisted procedure, and the results of the present study are encouraging. The global number of seromas, as well as their severity, were indeed reduced without any inconvenience except for the slightly longer operation time. Only four complicated cases requiring reoperation for surgical drainage or decapsulation occurred in the series, and all of these occurred in patients who did not undergo the quilting procedure.

From a technical point of view, when performing quilting, it is important that the number of stitches is sufficient and that the stitches are distributed on the entire surface of the dissection site. In most cases, 6–7 stitches were placed on the midline and 6–8 were placed laterally (Fig. 4). Tying the knots is facilitated by the low pressure of insufflation (4 mmHg) and by pushing down the skin flap with the finger. Although most authors recommend taking only the fascia of Scarpa to prevent dimpling, it is our practice to take full thickness of the subcutaneous tissue while avoiding taking the skin. A dimpling of the skin is usual expected as a result, but it disappears in 1 or 2 months when the absorbable sutures are resorbed. The patients were duly informed of this issue.

We have not applied either of the existing classifications to categorize seromas in this series.

The classification proposed by Morales Conde [22], though very exhaustive, was indeed not well adapted to our practice. For instance, Type 0a indicates “neither clinical nor radiological seroma,” in other words, no seroma; Types IIIa, IIIb, and IIIc correspond to patients’ complaints that should probably be alleviated by puncture, but in our practice, these types of cases were not observed because our policy is to perform needle aspiration without delay. The classification system proposed by Haskins et al. [23] is excellent for quantifying wound events following VH repair, but it is not well suited for describing different types of seromas.

In the present study, we classified seromas into two groups based on their severity in accordance with the Clavien‒Dindo Classification (CDC) [24], considering that the CDC is indeed recognized worldwide, with the number of citations increasing from 1 in 2009 to 524 in 2017 [25]. Seromas that resorbed spontaneously and seromas that required a single puncture were grouped as Type A, corresponding to CDC Grade I (deviation of the normal postoperative course), because we believe that a single puncture simply performed at the surgeon’s office at the first postoperative visit, without having to refer the patient to the imaging department, is not severe enough to be included in the same category as cases requiring many visits and punctures as well as cases requiring reoperation, which were included in Type B, corresponding to CDC Grade III (surgical, endoscopic, or radiological intervention). We did not observe any cases corresponding to CDC Grade II (requiring pharmacological treatment) or, fortunately, Grade IV (life-threatening complications), which is a level that would be highly unlikely in this kind of surgery.

A possible limitation of using the surgical technique applied in the series is not linked to the quilting procedure but to the technique of repairing the abdominal wall. The quilting does indeed simply consist of an adjunct aimed at reducing seroma incidence, which can be applied when large subcutaneous dissection is performed. Therefore, it may be expected that this technique is applicable to different endoscopic techniques involving subcutaneous approaches, including prosthetic techniques.

The main weaknesses of the study are its nonrandomized nature, the fact that the numbers of patients in both groups were different and also that the patients were operated on at two consecutive periods of time. Nevertheless, the data were prospectively collected, the global study period was short, the basic technique of repair was the same, and comparisons between both procedures took place during two consecutive periods.

Moreover, there were no significant differences between the groups of patients concerning the type or size of hernia, the type of hospitalization, the number of pregnancies, or the history of surgery on the abdominal wall.

The main strengths of the study are that it was carried out on a continuous series of non-selected consecutive patients operated on with the same technique of repair performed by the same two experienced surgeons and that no patient or seroma was excluded.

Conclusion

The quilting technique, adapted to the endoscopic-assisted bilayer suturing technique for the repair of combined VH and DR, can significantly reduce the incidence and severity of postoperative seromas.

Considering that seroma formation is essentially related to subcutaneous dissection, it may be expected that the endoscopic-assisted quilting procedure might be useful for reducing seroma formation after other types of endoscopic techniques for repairing the abdominal wall involving extended subcutaneous dissection.