Herring Bone Stitch: Knitting to Secure Abdominal Wall Closure for Emergency Midline Laparotomy

Secure abdominal wall closure after any laparotomy is every surgeon’s aim. However, in spite of all scientific advancements, better suture materials and different/ modified closure techniques; 5-26% of patients develop incisional hernia (IH) after midline laparotomy [1-6]. This figure goes further up if high risk groups are analyzed separately or if the follow up is for a period longer than 3 years [7-10]. Surgeons have risen to this challenge and have tried different sutures, techniques and even prophylactic mesh placement; with varying degree of success. We hypothesized that a simple ‘herring bone’ stitch repair can provide secure abdominal wall closure and minimize the incidence of IH in patients undergoing emergency midline laparotomy.


Introduction
Secure abdominal wall closure after any laparotomy is every surgeon's aim. However, in spite of all scientific advancements, better suture materials and different/ modified closure techniques; 5-26% of patients develop incisional hernia (IH) after midline laparotomy [1][2][3][4][5][6]. This figure goes further up if high risk groups are analyzed separately or if the follow up is for a period longer than 3 years [7][8][9][10]. Surgeons have risen to this challenge and have tried different sutures, techniques and even prophylactic mesh placement; with varying degree of success. We hypothesized that a simple 'herring bone' stitch repair can provide secure abdominal wall closure and minimize the incidence of IH in patients undergoing emergency midline laparotomy.

Methods
This prospective observational 'proof of concept' study was done from March 2015 to December 2017 in a teaching hospital of Central India. Consecutive patients undergoing midline laparotomy in two different units in emergency setting were included. Patients younger than 14 years of age, those with previous midline surgical scar or incisional hernia were excluded. Control group consisted of patients who underwent standard single layer continuous closure of rectus sheath with Polypropylene no. 1 suture (Unit of DS) and Study group consisted of patients who underwent single layer continuous herring bone closure of rectus sheath with Polypropylene no. 1 suture (Unit of PA). Surgical procedures were done by Junior Consultant or Senior Registrar. The study was approved by the institutional ethical committee and due informed consent was taken from patients.

Statistics
The expected sample size was calculated a priori at 256 patients. The demographic data among the two groups were compared with the help of appropriate statistical tests: Z statistics was applied for comparison between two proportions where the sample size was large; Chi-square was applied for comparison of contingency tables (gender and diagnosis); Fischer's exact test was considered where frequency was less than 5; and Student t test was applied to compare difference between two means where the sample size was small. Data were analyzed by IBM SPSS Statistics.  The first stitch is taken from inside out then outside in so as to put the knot inside. The needle is passed 1cm from the edge to take a 1cm long bite parallel to the incision from distal to proximal direction. Next this suture goes to the opposite side across the midline to take a similar bite parallel to the previous one, again from distal to proximal direction. Similar procedure is repeated for each successive stitch (Figure 1 A-B). In this technique the suture does not cross the incision line on the peritoneal aspect; instead it crosses the midline over the anterior rectus sheath in a crisscross manner making a lattice of suture ( Figure 2).

Herringbone Technique
Patients were followed up post-operatively at regular intervals: at 1 week, at 2 weeks, at 6 weeks, at 6 months and at 1 year. Outcomes measured were surgical site infection (SSI), proline knot granuloma or sinus formation, superficial wound dehiscence (disruption of superficial wound but rectus repair intact), fascial dehiscence (complete disruption of wound with intra-abdominal contents/ intestines visible) and development of incisional hernia. Fascial dehiscence (burst abdomen) was not managed by primary repair in view of patient's emergency surgery but by secondary suturing and resulting IH was managed 6 months later when patient's condition was optimized. Only those patients were included in the analysis that completed one year of follow up.

Results
There were 112 patients in study group (mean age 46.99 years, range 18-73 years; 82 males and 30 females) and 108 in control group (mean age 36.81 years, range 21-67 years; 88 males and 20 females). Their hemodynamic status and laboratory parameters were comparable and are shown in (Table 1). CONSORT flow diagram of present study is shown as (Figure 3).  More than 80% of cases in both groups had contaminated surgical wounds as 90/ 112 (80.36%) of study group and 88/108 (81.48%) of control group had perforation (peptic or typhoid) peritonitis. Other indications for emergency laparotomy were Intestinal obstruction (16/112 in study group and 12/108 in control group) and abdominal trauma (6/112 in study group and 8/108 in control group). As expected, longer suture was required in the study group. The suture length to wound length ratio was 6.8:1 in study group and 4.3:1 in control group.

Vector physics of herring bone stitch
( Figure 4) shows a comparison of the herring bone suture geometry with the simple running suture geometry. Here, we designated the vertical axis to be parallel to the edges of the wound, and the horizontal axis runs across the cross-section of the wound. Using this formalism, we qualitatively evaluated the effective direction of the force experienced by any bite point in the tissue through which the suture is threaded. Assuming that the tension throughout the suture is constant and equal, any point in the tissue "A" will experience a force that is a vectorial combination of the tension vectors along "AB" and "AC", oriented at an angle θ with respect to each other. Due to the geometry of the herring bone pattern, the addition of these vectors results in an effective tension vector which is oriented at an angle φ with respect to the vertical axis, parallel to the wound edge. However, for the simple running suture geometry, the effective tension vector is oriented at an angle φ with respect to the horizontal axis. In the regime where 0<θ<90°, we estimate that in all cases, φ<45°. This suggests that in the herring bone suture geometry, the load will be preferentially distributed parallel to the wound boundary, whereas for simple running suture geometry, load will be directed laterally across the wound cross-section. The wound is expected to stretch laterally horizontally in (Figure 2) during any movement or breathing. For the same tension in the suture and the same angle θ between the sutures, the simple running suture geometry will provide more resistance to lateral expansion of the wound than the herring bone suture geometry. For the same nominal angle of θ_simple= θ_herring=45°, the resistance ratio can be as much as 2.414, or equivalently, the resistance experienced by the tissue can be up to 140% greater in the simple running geometry compared to the herring bone geometry. Based on this analysis, the herring bone suture is less likely to cut through the tissue during lateral expansion than the simple running suture.

Discussion
A midline incision is frequently used in abdominal surgery as it providesa relatively quick and wide access to the abdominal cavity with minimal damage to the muscle, nerve and bloodvessels as these sructures do not cross the midline. Risk factors for development of IH after midline laparotomy are well documented: male gender, age > 60 years, preoperative anemia, BMI>25, non-elective admission, co-morbidities, previous laparotomy, bowel surgery, creation of an ostomy, type of suture, suture to wound length ratio < 4, wound infection, wound dehiscence, postoperative pulmonary problems, postoperative abdominal distension/ intestinal obstruction, postoperative catecholamine-therapy and disturbed wound healing [7,[11][12][13][14][15][16][17]. Incisional hernia risk prediction models have been constructed by statistically combining these risk factors [18][19][20][21].
Most risk factors are patient-related and surgeons do not have any control over them. This knowledge has affected the attitudes of surgeons and has given rise to different practices (types of suture and suturing techniques) to minimize IH. These include: debates about the optimum suture length, continuous Vs interrupted sutures, absorbable Vs slow absorbable Vs non-absorbable sutures, etc; but consensus among surgeons was slow to arrive [22][23][24]. It was left to the landmark systemic review and meta-analysis by Diener et al. to conclude that using a continuous (vs. interrupted) technique (OR: 0.59; P=0.001) with slowly absorbable (vs. rapidabsorbable) suture material (OR: 0.65; P=0.009) was best in the elective setting; and no further trials should be conducted for evaluation of technique and available materials for 'elective' midline abdominal fascial closure [4].
More complex closure techniques like interrupted 'X' sutures, interrupted Smead-Jones sutures with or without a non-absorbable suture material for closure of linea alba combined with mass closure involving all the layers; also with a non-absorbable suture material; have been tried in the past [25][26][27]. Currently, a trial is going on using the 'Hughes Repair', which combines a standard mass closure with a series of horizontal and two vertical mattress sutures within a single suture [28]. This, theoretically, distributes the load along the incision length as well as across it. But this involves the use of two sutures, has the risk of bowel trapping and risk of suture entanglement, takes longer time, longer suture, and adds to the cost.
Wound infection (SSI) is the most important single factorraising the risk of development of IH by 2-10 times [11,15,29]. In an attempt to minimize SSI, investigators have tried using braided suture materials with antibacterial activity, but it did not result in decreasing the incidence of IH [30]. A long stitch length is associated with an increase in the rate of wound infection and IH [31]. Another significant factor is wound dehiscence, which is almost invariably followed by IH; regardless of suture material or technique of repair [14,29]. It has been emphasized that it is not the type of suture, but the technique of suturing which determines the outcome as there was no significant difference in IH rate between absorbable and non-absorbable sutures [32][33][34]. However, Polypropylene is found to be associated with more persistent wound pain and more sinus formation as compared to Vicryl [14,22,35].
Concerns about unacceptably high IH rates have led to some 'out of box' thinking about secure abdominal wall closure. These include use of prophylactic retention sutures, use of rectus sheath relaxation incisions, as well as studies on histology and arterial perfusion of abdominal wall, novel plasma proteins as biomarkers for the development of incisional hernias and suture tension dynamics [36][37][38][39][40]. In the last few years investigators' focus has been on 'small tissue bites of 5mm every 5mm' and use of prophylactic mesh placement to minimize IH. From small comparative studies to large multicenter Randomized Control Trials and systemic reviews -all have shown that both these techniques reduce the incidence of IH [31,[41][42][43][44][45][46][47]. Recent success of small stitch trial has questioned the traditional guideline of 4-4.9:1 relationship between suture and wound length, as it requires a much longer suture and does not add to the risk of SSI and IH as was previously thought [33,48].
The authors' reluctance to use a foreign body, i.e., a mesh in emergency situations in patients with potentially contaminated cases (perforation peritonitis, intestinal obstruction) led to hypothesizing that a simple 'herring bone' stitch repair can provide secure abdominal wall closure and minimize the incidence of IH in high risk emergency cases. Alexis Carrel, widely considered the father of vascular surgery for developing the technique of vascular anastomosis, was awarded Nobel Prize in Medicine in 1912 for his pioneering work which paved the path for organ transplantation. His interest in embroidery, piqued in childhood by watching his mother, led him to learn use of tiny needles and thread which later led to the development of his technique of vascular anastomosis [49]. Taking an inspirational leaf out of Carrel's book, authors had earlier applied the art of Tatting (lace making) to develop a simple, inexpensive, easy to use and safe technique of extracorporeal knotting [50]. In the present study, the inspiration came from the art of knitting, embroidery, and crochet and led to use of Herring bone stitch for secure abdominal wall closure.
The literal meaning of 'Herring bone' is a pattern consisting of columns of short parallel lines with all lines in onecolumn, sloping one way and all thelines inthenext column sloping the other way; is so named as it resembles the bones extending from the spine of a herring fish. Herring bone stitch has many advantages. This technique relies on each successive suture to reduce the tension on the stitch for the next suture to be passed. Herring bone suturing technique has the advantage of both continuous and cross suture method. The suture does not cross the midline on the peritoneal aspect thus avoiding the trauma to the moving gut loops by wiring effect. Instead the suture crosses the midline over the anterior rectus sheath in a crisscross manner making a network of suture giving the additional advantage of darning effect.
This leads to new fibrous tissue being laid down in the interstices of darn which strengthens the repair. Additionally, vector physics has shown that the herring bone suture is less likely to cut through the tissue during lateral expansion than the simple running suture (Figure 2). This is due to the fact that any tension on the suture line is preferentially distributed parallel to the wound boundary and can better accommodate any increase in the intra-abdominal pressure in the post-operative period. There is additional advantage of burying the knot inside the rectus sheath as it theoretically leads to decreased incidence of stitch granulomas. It is technically easy, reproducible, safe and can be performed quickly.
Our results in emergency setting in relatively high risk (>80% patients had perforation peritonitis, had anaemia: 30% in study group and 27% in control group, hypoproteinaemia: 23% in study group and 25% in control group, raised bilirubin: 44% in study group and 40% in control group. (Table 1) patient cohorts go in favour of our hypothesis. Despite using polypropylene, knot granuloma was seen in only 1 patient in study group and in 2 patients in control group. Other complications (SSI, and superficial wound dehiscence) were comparable among the two groups. The incidence of fascial dehiscence (7.14% as compared to 15.74%; p = 0.045) and incisional hernia was significantly less (2.68% as compared to 12.04%; p = 0.009) in study group (Table 2). Fascial dehiscence and IH have similar causative factors; but fascial dehiscence occurring within 2 weeks of emergency surgery reflects more about the patient related factors (anemia, hypoproteinemia, intraabdominal sepsis, post-operative abdominal distention, chest infection etc.) while IH occurring much later is more reflective of abdominal closure technique [12,25].
The present study shows superiority of 'herring bone suture' over conventional closure of rectus sheath in emergency midline laparotomy. However, there are a few limitations of our study: Small numbers of patients have been studied in the two arms and the follow-up period is only 1 year. It is known that the incidence of IH increases with the duration of follow up; only about a third of IH develop in the first 6 months after the operation, 54.4% after 12 months, 74.8% after 2 years and 88.9% after 5 years [7]. Hence validation of this hypothesis in larger number of patients, followed up for longer period will strengthen the evidence.