Uterine closure after cesarean delivery: surgical technique, biological rationale, and clinical implications

Healing of a hysterotomy unfolds in 3 overlapping phases: inflammation, repair, and remodeling. The inflammatory phase begins immediately and typically lasts 3 to 5 days, characterized by hemostasis, leukocyte infiltration, and activation of cytokine networks. 72 – 74 The reparative phase follows from around day 4 through 2 to 3 weeks, with fibroblast proliferation, angiogenesis, and deposition of extracellular matrix. Remodeling begins within 3 to 4 weeks and may continue for months to a year, marked by collagen reorganization and gradual recovery of tensile strength. Endometrial reepithelialization plays a crucial role in maintaining the integrity of uterine function and can be disrupted by a hysterotomy and its repair ( Box 1 ). Unlike cutaneous wounds, however, uterine repair occurs in the unique context of involution and profound hormonal changes. The rapid fall in estrogen and progesterone after delivery, combined with myometrial contraction and a catabolic state, creates a biologic environment unlike any other site of surgical healing. These biological phases have direct clinical relevance. For example, short interpregnancy intervals increase the likelihood of uterine rupture and abnormal placentation, most likely because conception occurs before scar remodeling is complete. 75 – 77 The prolonged nature of the remodeling phase highlights why adequate spacing between deliveries is important, and why the technique of hysterotomy closure has implications not only for immediate repair but also for outcomes in subsequent pregnancies.

Key questions about the optimal method of hysterotomy closure remain unanswered and warrant renewed experimental and clinical investigation. Well-designed studies are needed to determine how variations in suture technique, tissue handling, and myometrial thickness influence the quality of healing and long-term reproductive outcomes. The 3-layer method proposed in this article applies primarily to cesarean deliveries performed before labor, or in early labor, yet the best approach for cesarean conducted in the second stage—when the lower uterine segment is extremely thin and may involve cervical tissue—has not been established. Historically, efforts to shorten operative time have strongly influenced surgical technique. Yet, the long-term implications of a poorly healed uterine scar—such as infertility, pain, abnormal bleeding, and life-threatening complications in future pregnancies—argue for a renewed emphasis on quality rather than speed. In an age when most cesarean deliveries are performed under regional anesthesia, the priority should be meticulous restoration of uterine anatomy, over a rapid closure. It goes without saying that if patients were asked to choose between a fast operation and one that best protects their future reproductive health, most would undoubtedly prefer the latter. Institutions and healthcare systems should ensure that maternity care is staffed and structured to support careful, deliberate surgical practice that safeguards maternal well-being beyond the immediate postoperative period. We believe strongly that future research should extend beyond short-term surgical metrics such as operative duration or blood loss and focus on outcomes that matter most to patients: restoration of uterine structure, reproductive performance, and long-term gynecologic health. The integration of imaging, biomechanical, and molecular studies will be essential to elucidate the biological pathways that underlie successful healing. Cesarean delivery has lasting consequences for mothers and their families. Because of its global frequency and lifelong impact, research aimed at optimizing uterine closure and improving maternal outcomes should be recognized as an immediate public health priority.

The goal of uterine closure after cesarean delivery is to restore the organ’s structural integrity while preserving its function for future pregnancies. Halsted’s principles of safe surgery—gentle handling, meticulous hemostasis, preservation of blood supply, precise apposition of tissues, elimination of dead space, and tension-free closure—remain the cornerstone of durable repair. 154 The uterus, although biologically distinct and structurally much thicker, follows similar logic: avoiding incorporation of the endometrium, reapproximating myometrium to myometrium, and protecting the serosa leads to a stronger, more functional scar. To provide clinicians with a practical framework, we propose the REPAIR mnemonic, which translates these enduring surgical principles into the specific context of cesarean hysterotomy closure ( Box 4 ). The closure of a low transverse hysterotomy can be conceptualized as 3 distinct steps, each corresponding to a specific anatomical layer: the endometrium, the myometrium, and the serosa ( Figure 7 ). The first step reapproximates the endomyometrial junction while excluding the decidual surface. The second step restores uterine wall integrity by reapproximating the myometrium and minimizing dead space. The third step reapproximates the remaining superficial myometrium and serosa, ensuring coverage of exposed tissue, optimizing hemostasis, and reducing the need for cauterization. This 3-step technique is particularly suitable for cesareans performed before or in early labor, when the lower uterine segment is relatively thick, and the different layers can be clearly distinguished. These are also the cases most at risk for scar defects and long-term complications. In situations where the myometrium is thin, as in advanced labor, the first 2 steps can often be combined. The optimal method of closure for second-stage cesarean delivery remains uncertain, as the incision frequently extends into cervical tissue. 50 , 155 , 156 The goal of the first layer is to approximate the endomyometrial junction while avoiding incorporation of the endometrium. This prevents introducing bacteria into the wound, tearing fragile endometrium, and displacement of decidual cells into the myometrium, which may impair healing or promote adenomyosis and endometriosis. Inclusion of both myometrium and decidua increases the risk of scar defects and long-term complications. 86 , 88 , 130 , 131 , 157 Because approximating decidua alone is technically difficult, sutures should be placed in the myometrium immediately above the junction (endomyometrial junction). Sutures are generally placed parallel to the incision ( Figure 8A ); in friable tissue, a slight oblique angle may provide a more secure closure ( Figure 8B ). The aim is to eliminate dead space while avoiding endometrial displacement into the myometrium. Figure 9 shows an example of parallel approximation. Large myometrial bites should be avoided, as they increase ischemia and tissue strangulation. The second layer reapproximates myometrium to myometrium and is not intended to bury the first layer. Unless active bleeding is present, it is unnecessary to take large bites of myometrium. As in the first step, sutures placed parallel to the incision and near the myometrium—serosa junction ensure apposition with minimal tension or strangulation ( Figure 10 ). This technique also keeps sutures beneath the uterine surface, reducing peritoneal exposure and adhesion formation. Figure 11 shows an example of myometrial approximation using a subserosal entry. The third layer aims to restore surface anatomy and optimize healing ( Figure 12 ). This layer ensures that superficial myometrium left exposed after the second step is covered, reducing the risk of adhesions. When the myometrium is thick, the second layer should be placed deeply to avoid dead space, while the third layer closes the remaining myometrium and serosa. This layer also contributes to hemostasis, as small vessels are often present on the surface ( Figure 13 ). Rather than relying on cautery, which may impair healing, gentle approximation with a fine absorbable suture can cover exposed myometrium and sutures, reduce bleeding, and minimize thermal injury. When the myometrium is thin (<5 mm), as often occurs after prolonged labor, a single first layer may be sufficient to reapproximate the entire thickness of the myometrium, with the suture positioned between the endometrium and the serosa. The next layer, in this context the equivalent of a ‘third layer’, can then be used to approximate the serosa with a fine absorbable suture, thereby reinforcing the closure in tissue that is often more fragile ( Figure 14 ). The 3-step technique does not appear to increase the need for hemostatic sutures. When required, they can be used without compromising anatomic reapproximation and tissue handling principles. Although excessive hemostatic suturing could theoretically impair healing by distorting tissue or entrapping decidual cells, limited use is unlikely to affect overall apposition.

In advanced labor, the lower uterine segment becomes markedly thinned, and the incision may extend into cervical tissue. Under these circumstances, it is often difficult to distinguish and separately reapproximate the endometrium and myometrium according to standard surgical principles. In a randomized trial of 122 participants, Vikhareva et al reported that placing the incision approximately 2 cm above the vesicouterine fold, rather than below it, may reduce the risk of subsequent scar defects in women undergoing cesarean during advanced labor. 155 Nonetheless, additional studies are needed to define the optimal surgical technique in this specific context, as uterine and cervical tissues remodel substantially in the hours and days following delivery, often returning to a configuration very different from that observed intraoperatively. Based on available evidence, a slightly higher hysterotomy is advisable, as this increases the likelihood of entering the lower uterine segment rather than the cervix. For closure, the markedly thinned myometrium usually requires modification of standard techniques. We recommend combining the first 2 parallel layers into a single layer placed parallel to the incision, thereby approximating the thin residual myometrial tissue between the endometrium and serosa. A second layer should then be performed with a rapidly resorbable synthetic suture, reapproximating the serosa while incorporating a small amount of superficial myometrium to reinforce the closure and restore the uterine surface.

Two prospective comparative studies—one quasi-randomized and one randomized—have evaluated closure techniques similar to this 3-step method, in which the first layer is unlocked (or interrupted) and excludes the endometrium and the second layer reapproximates the residual myometrium. These designs differ slightly from the parallel second layer described in our proposed approach. Each trial included 2 control groups: a single-layer closure incorporating both myometrium and endometrium and a 2-layer closure in which the first layer included the endometrium and the second simply covered it. 158 , 159 In the quasi-randomized trial of 137 participants, Hayakawa et al 159 found significant differences in scar defect rates: 6% (2/36) for the 2-layer closure excluding the endometrium, 34% (17/50) for interrupted single-layer closure including the endometrium, and 16% (8/51) for 2-layer closure with the first layer including the endometrium ( P <.01). In the randomized trial by Roberge et al 158 involving 81 participants, the corresponding rates of severe scar defect were 4% (1/26), 20% (5/25), and 5% (1/22), respectively ( P =.10), while the residual myometrial thickness was significantly greater with the 2-layer closure excluding the endometrium (6.1±2.2 mm) compared with the other 2 methods (3.8±1.6 mm and 4.8±1.3 mm, P <.001). The operative time between the 3 groups was similar ( P =0.85). More recently, Özler et al 160 compared double-layer closure with an unlocked first layer avoiding the endometrium (n=28) to single-layer unlocked closure also avoiding the endometrium (n=26). At 6 months, the residual myometrial thickness was significantly greater in the double-layer group (5.1±0.4 mm vs 4.1±0.4 mm; P <.001). Interpretation of these latter results is limited, however, as most participants had at least one previous cesarean delivery. Overall, these studies consistently demonstrate that a double-layer closure with the first layer excluding the endometrium is associated with thicker residual myometrium and lower rates of uterine scar defect. The potential benefit of adding a third (serosal) layer remains less certain. While some observational data suggest that peritoneal closure may reduce postoperative adhesions, 161 randomized trials have not confirmed a durable advantage. 162 At present, this additional step is based more on surgical judgment and established principles of tissue handling than on definitive evidence ( Box 5 ). 95 , 96 , 163

Systematic investigation of uterine scar healing began in the early twentieth century, motivated by the occurrence of uterine rupture after cesarean delivery and Cragin’s dictum, “once a cesarean, always a cesarean”. 78 At the Harvard Surgical Laboratory, Mason and Williams (1910) combined animal experiments and clinical observations to demonstrate that with careful suturing, the uterus could withstand subsequent pregnancy and labor, and that rupture nearly always occurred along the scar rather than through intact myometrium. 79 Histologic descriptions from Spalding (1917) and Losee (1917) provided early evidence that cesarean scars consisted mainly of fibrous tissue with limited muscle regeneration and that healing was impaired when infection or placental implantation involved the incision site. 80 , 81 Further progress came from Washington University, St. Louis, where Schwarz and Paddock (1925) studied both human and experimental material. They found that healed incisions were composed largely of fibrous tissue with minimal myocyte regeneration, establishing that the scar was biologically distinct from normal myometrium. 82 In 1938, Schwarz et al confirmed these findings using mechanical testing: scarred uteri ruptured at lower intraluminal pressures than unscarred controls, correlating biomechanical weakness with disorganized fibrotic architecture. 83 The focus then turned to minimizing inflammation from suture material and improving tissue approximation. In 1942, Potter and Elton proposed an interrupted closure technique that excluded the endometrium and brought only the external myometrium into accurate apposition to reduce ischemia and foreign-body reaction in vertical incisions. 84 In 1954, Potter and Johnston applied similar principles to low transverse hysterotomy, using interrupted silk sutures through the outer third of the myometrium and fine plain catgut for the peritoneal flap. 85 These studies advanced the concept that optimal repair requires the use of the smallest amount of nonreactive suture necessary to ensure accurate apposition and hemostasis while avoiding excessive tension that could lead to ischemia. An important contribution came from the work of Leslie O.S. Poidevin (1961), whose doctoral thesis at the University of Adelaide combined animal experiments and clinical hysterography. 11 , 86 , 88 He showed that healing was strongest when closure achieved muscle-to-muscle apposition and excluded endometrium, whereas inclusion of endometrium led to weak, collagen-rich scars and pouch-like defects—later recognized as niches or isthmoceles.His research established the biologic rationale for endometrium-excluding techniques (details are described in the next section). Csúcs et al (1970) in Hungary proposed a single-layer closure limited to the superficial myometrium, allowing the endometrium and deeper layers to heal spontaneously. 89 , 90 In rabbits, this method provided healing comparable to 2-layer closure, and clinical observations suggested its feasibility in humans. This study is one of several that have been conducted using animal models to investigate surgical techniques for uterine closure ( Box 2 ). In the 1970s, locked single-layer closure including the endometrium gained popularity because of technical simplicity and shorter operative time (details of the debate are described in the next section). However, accumulating evidence now links this method to higher rates of scar defects and increased risk of uterine rupture in subsequent pregnancies. 91 – 94 Three main mechanisms have been proposed for the increased risk of defective scars with single-layer techniques: 1) incomplete apposition of the inner myometrial layer ( Figure 1 ); 2) malalignment or strangulation of myometrium leading to ischemia ( Figure 2 ); and 3) implantation of endometrial cells into the suture line, resulting in ectopic endometrial tissue within the scar ( Figure 3 ) - that issue is not resolved with a second layer ( Figure 4 ). 3 , 11 , 95 , 96 In summary, over the past century, experimental, histologic, and clinical investigations have revealed that uterine healing is a complex process influenced by anatomic site, infection, labor, hormonal state, and surgical technique. Recognition of these factors has reshaped the understanding of why some scars remain strong and asymptomatic, while others predispose to rupture, abnormal bleeding, infertility, or placental invasion disorders in future pregnancies. 97 Box 3 summarizes recent studies on uterine wound healing and emerging therapeutic approaches. For much of the twentieth century, the uterine incision at cesarean delivery was routinely closed in 2 or 3 layers, following Kerr’s 1926 description of the low transverse hysterotomy and the guidance of successive editions of Williams Obstetrics . 98 – 107 In 1976, Pritchard and MacDonald observed that “the uterine incision may be closed with either one or the more traditional 2 layers of continuous chromic suture,” noting that a thin lower uterine segment could often be satisfactorily approximated with one layer. 108 This statement marked a turning point that prompted modern evaluation of the optimal closure method. The first randomized trial, by Hauth et al published in 1992, compared a single locked continuous layer with a 2-layer closure. 109 The single-layer technique reduced operative time without increasing hemorrhage, infection, or extra hemostatic sutures. In the follow-up of 145 subsequent pregnancies (70 with single-layer and 75 with double-layer closure), no uterine ruptures were observed. 110 Subsequent studies produced conflicting results. Bujold et al 93 , 94 reported that single-layer closure was associated with a higher risk of rupture during trial of labor after cesarean, whereas Roberge et al clarified that the excess risk was limited to locked single-layer closures that included endometrium, while unlocked single-layer closures were not associated with rupture. 91 In a meta-analysis (2014), Roberge et al found that double-layer closure resulted in a thicker residual myometrium (+2.6 mm; 95% confidence interval [CI], 2.2—3.1) and that an unlocked first layer produced a thicker scar than a locked one (+2.5 mm; 95% CI, 1.8—3.2). 92 Reviews that did not distinguish these technical variations found no differences in clinical outcomes. 111 , 112 Blumenfeld et al later showed an increased risk of bladder adhesions after single-layer closure (odds ratio, 7.0; 95% CI, 1.7—28.1). 113 Among major randomized trials, the Caesarean section surgical techniques: a randomised factorial trial (CAESAR) 114 and the Caesarean section surgical techniques (CORONIS) trial 115 – 118 demonstrated no difference in short-term or long-term maternal outcomes between single-layer and double-layer closure, although specific suture techniques were not standardized. In contrast, the Cesarean-scar Thickness and Closure Technique trial by Bamberg et al 119 , 120 found greater residual myometrial thickness with a double-layer closure (7.9 mm; 95% CI, 6.4—10.2) than with either locked (6.7 mm; 95% CI, 5.1—8.8) or unlocked (6.6 mm; 95% CI, 5.0—8.8) single-layer methods ( P =.04) among women who underwent primary cesarean. In the Niche In CEsarean Scar Trial (NICEST) trial, 121 double-layer closure produced fewer severe scar niches and thicker residual myometrium at 6 and 12 months, while in the 2Close study [a multicenter randomized controlled trial that compared single- versus double-layer closure of the uterus in the prevention of gynaecological symptoms in relation to niche development], 122 – 125 no differences in niche frequency, gynecologic symptoms, or fertility were observed. However, when the endometrium was excluded within the single-layer closure group, the prevalence of niche formation was significantly reduced (59% [150/253] vs 72% [471/656]; relative risk [RR], 0.83; 95% CI, 0.74—0.93; P =.001). 124 In summary, current evidence suggests that while a second layer may improve residual myometrial thickness, the critical factor for optimal healing is exclusion of the endometrium/decidua. The number of layers alone does not ensure an intact scar unless myometrial approximation is achieved and endometrial inclusion is avoided. 122 , 126 , 127 From the nineteenth century onward, surgeons understood that incorporating the uterine lining into a hysterotomy closure was detrimental. In his classical report in 1891, Howard A. Kelly described the essential “do’s and don’ts” of cesarean technique and explicitly described exclusion of the endometrium. 128 Early obstetric surgeons were trained in general surgery, and their approach to cesarean repair reflected the principles of visceral closure established in operations on the bowel and stomach. In such procedures, the mucosal layer was excluded from the muscular suture line to prevent infection, leakage, or fistula formation. By analogy, the endometrium uterine mucosal lining was considered a delicate and potentially contaminated surface that should not be incorporated into the sutures. This principle guided early cesarean techniques, ensuring apposition of myometrium to myometrium while avoiding decidual inclusion. This concept persisted until the latter half of the twentieth century, when the introduction of single-layer continuous closure led some surgeons to include the endometrium in the suture line for simplicity and speed. However, careful experimental and clinical work by Leslie O.S. Poidevin at the University of Adelaide (1961) demonstrated that inclusion of the endometrium impaired healing. 11 In animal experiments using cats and rabbits, he compared closure techniques in opposite uterine horns: when the suture line included endometrium, defective healing occurred in 78% (14/18), whereas exclusion of the endometrium resulted in no defects (0/18). 88 Poidevin extended these findings to humans in an observational study of 202 women. 129 On hysterosonography, scar defects were observed far more frequently when the endometrium had been included in the closure—73% (93/127)—than when it had been excluded—8% (6/75)—and all severe defects (100%, 14/14) occurred in the endometrium-included group. Histologic examination of hysterectomy specimens further confirmed these results, showing endometrial glands deep within the fibrotic scar in 89% (8/9) of women whose closures had incorporated the endometrium. 11 Subsequent investigators reaffirmed Poidevin’s conclusions. Antoine et al summarized the accumulating experimental and clinical evidence that inclusion of the endometrium predisposes to scar defects and abnormal placentation, explicitly crediting Poidevin’s pioneering contributions. 127 , 130 – 133 Most recently, Lino et al (2025) synthesized data from 4 randomized trials including 392 women and confirmed that endometrium-excluding techniques halved the risk of scar defect (RR, 0.53; 95% CI, 0.34—0.82; I 2 =0%). 134 Together, these findings reaffirm a principle first drawn from general surgical experience and now supported by modern evidence: exclusion of the endometrium is essential for durable uterine healing regardless of whether closure is performed in 1 or 2 layers. Two main strategies have been used to close the hysterotomy: continuous and interrupted suturing. Poidevin 11 observed that when the endometrium was excluded, interrupted sutures resulted in no scar defects (0%, 0/30), whereas continuous sutures produced defects in 13% (6/45). However, when the endometrium was included, defect rates were high regardless of technique—73% (91/124) with continuous and 67% (2/3) with interrupted sutures. These data suggested that the inclusion of endometrium, rather than the choice of suture technique, was the key factor determinant of poor healing. A randomized trial by Tsuji et al 135 confirmed these experimental findings: severe defects were observed in only 2% (2/89) of women after interrupted 2-layer closure excluding the endometrium, compared with 22% (19/88) after continuous unlocked 2-layer closure including the endometrium ( P <.001). Similarly, Sumigama et al 136 reported a higher risk of placenta accreta spectrum with continuous sutures (odds ratio, 6.0; 95% CI, 1.4—25.2), although this retrospective study was limited by design. Taken together, these studies indicate that interrupted suturing that excludes the endometrium is associated with the most favorable histologic and clinical outcomes. In routine obstetric practice, however, continuous suturing remains far more common, largely because it is faster, simpler, uses less suture material, and easier to perform. Locking is a frequent modification of continuous suturing in which each needle pass loops and “locks” the thread, preventing slippage and allowing tighter tension control along the suture line. Although this technique provides mechanical security, its independent effect on uterine healing remains difficult to isolate because locking is often combined with single-layer closure and inclusion of the endometrium. From a physiological standpoint, locking may concentrate stress and compromise microvascular perfusion at the myometrial edges, whereas nonlocking sutures distribute tension more evenly, reducing focal ischemia and promoting tissue viability. Evidence from other visceral and mucosal closures supports this principle: locking sutures may offer short-term security but at the cost of increased tissue compression and inflammatory response. Tarafdari et al 137 compared locked vs unlocked first layers—both including endometrium—and found that scar defects developed in all women at 6 months (30/30 locked, 26/26 unlocked). Although mean residual myometrial thickness did not differ significantly (4.12±0.48 mm vs 4.44±1.07 mm, P =.14), scar defect depth was greater in the unlocked group (3.77±1.11 mm vs 3.16±1.10 mm, P =.04). A larger trial involving 435 women also found no difference between locked and unlocked single-layer closures, both including endometrium, but demonstrated that double-layer closure with an unlocked first layer produced thicker residual myometrium. 119 , 120 In an observational study of 388 women, Hudic et al 138 observed similarly high rates of complete uterine rupture after both locked and unlocked single-layer closures (2.1% [1/47] vs 2.4% [2/85], P =.93). Overall, the available data suggest that locking itself is not independently harmful, but it provides no measurable benefit and may impair microcirculation when used under high tension. Consequently, unlocked continuous sutures are generally preferred for hysterotomy closure, particularly when the first layer excludes the endometrium and achieves accurate myometrial apposition. Suture materials used for hysterotomy closure have evolved markedly over the past century. In early days, surgeons relied on silk and plain catgut, both of which produced marked tissue inflammation and were soon replaced by chromic catgut, a treated variant intended to slow absorption and reduce reactivity. However, even chromic catgut elicits a significant inflammatory response and has been associated with poorer-quality scars and delayed healing. 139 This recognition led to the adoption of synthetic absorbable polymers, such as polyglactin (Vicryl) and poliglecaprone (Monocryl), which provide more predictable absorption and minimal tissue reaction. In an observational study of 1613 women, Vachon-Marceau et al 140 found no difference in third-trimester lower uterine segment thickness between catgut and synthetic multifilament sutures, and the CORONIS trial 116 , 117 similarly found no significant difference in uterine rupture risk at subsequent delivery (0.2%; n=3/1660) catgut vs (0.06%; n=1/1647) synthetic (RR, 3.1; 95% CI, 0.3—29.3). However, Hosseini et al 141 observed higher rates of scar defects with catgut (18.2%, n=20/110) than with Vicryl (9.3%; n=13/140) and greater residual myometrial thickness with Vicryl (5.0±2.2 mm vs 3.7±1.5 mm; P <.01). Recent randomized trials suggest that monofilament sutures may yield slightly thicker uterine scars than multifilament sutures (+1.1 mm residual myometrium across 3 trials, 499 women) but require approximately 3 additional minutes of operative time. 142 Saccone et al, 143 in a trial of 300 women, found no significant differences in scar thickness or symptoms between monofilament and multifilament materials. Two meta-analyses suggested that the use of barbed sutures reduced uterus closure time by approximately 2 minutes but had no impact on intra-operative complications. 142 , 144 However, a recent trial observed less scar niche using a double-layer closure with barbed sutures excluding the endometrium (29.1%; n=32/110) compared to conventional double-layer sutures (68.2%; n=75/110; P <.001). 145 Because braided multifilament sutures can harbor bacteria, triclosan-coated variants have been developed to reduce surgical site infection. Large meta-analyses in general surgery—Jalalzadeh et al 146 and Depuydt et al 147 —reported approximately a 25% reduction in wound infection rates with antimicrobial-coated sutures. Although these have not been specifically studied in cesarean hysterotomy closure, they may benefit patients at high risk of infection (eg those with chorioamnionitis, prolonged rupture of membranes, or confirmed intra-amniotic infection). Overall, current evidence supports synthetic absorbable sutures as safe and effective, with monofilament materials showing modest potential advantages for long-term scar integrity. For the serosal layer, a rapidly resorbable monofilament appears a reasonable and biologically sound choice. The extent and depth of tissue captured with each suture pass may influence long-term scar quality as much as the number or type of layers used. Taking full-thickness, widely spaced sutures (large bites) can lead to tissue strangulation and ischemia, whereas very superficial or widely gapped passes may leave residual cavities or dead space. In uterine closure, precise, closely spaced, partial-thickness sutures (ie small bites) promote accurate reapproximation of the myometrial edges while preserving perfusion. Evidence from general surgery reinforces this principle: in the small bites versus large bites for closure of abdominal midline incisions (STITCH) trial of abdominal fascial closure, fine, closely spaced suture placement (approximately 5 mm×5 mm) significantly reduced incisional hernia rates compared with wider placement (10 mm×10 mm) (13%; n=35/268 vs 21%; 57/277 at 1 year, P <.05). 148 Although this study addressed fascial rather than uterine repair, the underlying biomechanics are similar—narrow, equidistant suture placement distributes tension evenly along the incision, reducing localized ischemia and supporting stronger healing. Figure 5 illustrates how precise, short-interval suturing facilitates uniform apposition and minimizes reliance on thread tension for wound stability. The orientation of sutures relative to the incision line may influence wound mechanics and healing. Experimental evidence from general surgery supports this concept. In a rat laparotomy model, Rappaport et al (1990) compared sutures placed parallel and perpendicular to the incision and found that parallel placement produced greater bursting strength, less inflammation, and less tissue necrosis, likely because tension was distributed along the wound axis rather than concentrated at discrete points. 149 Similarly, Lear et al (2020) showed in excisional skin wounds that tension distribution—more than suture material or type—determines whether sutures “cheese-wire” through tissue; parallel orientation provided more uniform stress and resistance to tearing. 150 Although the uterine wall differs in composition and mechanical loading, analogous principles may apply. Aligning sutures parallel to the hysterotomy could reduce focal ischemia and stress across myometrial fibers, enhancing healing quality. Supporting this hypothesis, Alper et al 151 observed in 497 women undergoing primary cesarean before labor that parallel-layered closure was associated with fewer scar defects (7.9% vs 16.3%, P =.009) and fewer long-term gynecologic symptoms (1.5% vs 7.8%, P =.004) compared with perpendicular single-locked closure. The purse-string technique described below also employs a parallel suture orientation, further underscoring the potential relevance of suture direction for optimizing uterine repair. Traditional uterine repair involves direct approximation of the upper and lower edges of the hysterotomy. An alternative approach, described by Turan et al 152 in Turkey, proposed a purse-string double-layer closure designed to reduce residual scar size. This technique employs 2 layers, excludes the endometrium, and places sutures parallel to the incision. Rather than securing knots at the ends, the suture is run circumferentially and tightened centrally, producing the characteristic purse-string appearance ( Figure 6 ). A meta-analysis of 8 randomized trials (751 women) suggested that the purse-string technique is associated with lower uterine scar defect than conventional single-layer or double-layer closures that included the endometrium (RR, 0.45; 95% CI, 0.36—0.58; I 2 =0%). 153 While these findings are encouraging, concerns remain regarding possible circumferential ischemia and tissue strangulation, and long-term safety data are still limited. Current evidence supports a double-layer closure using synthetic absorbable suture, unlocked, and placed parallel to the incision, with the first layer excluding the endometrium, as the configuration most likely to promote optimal uterine healing. These principles appear especially beneficial for primary cesarean deliveries and for procedures performed before the onset of labor.

We propose that successful uterine closure after cesarean delivery depends on adherence to fundamental surgical principles: exclusion of the endometrium in the first layer, restoration of myometrial thickness with a tension-free second layer, and serosal coverage to minimize adhesions. The videos ( Supplemental Material ) illustrate the 3-layer closure technique, which embodies these principles and provides a practical framework for routine surgical practice ( Figure 15 ). The REPAIR mnemonic offers a concise reminder of the key elements underlying optimal closure, aligning surgical technique with the biological requirements of uterine healing.

Closure of the hysterotomy during cesarean delivery is often considered a routine step. Yet, it is a surgical and biologically consequential act that influences not only immediate operative outcomes but also long-term gynecologic health, fertility, and risks in future pregnancies ( Table ). Two factors are fundamental to successful healing: the site of the incision and the method of closure. The anatomic location of the hysterotomy determines the nature of the tissue incised—whether it involves the uterine body, the lower uterine segment, or the cervix—each characterized by distinct structural and vascular properties. The chosen closure technique then dictates how these tissues are reapproximated: whether the endometrium is excluded, whether vascular perfusion is preserved, and whether the serosal surface is restored. Together, these elements determine the structural integrity of the uterine wall, the quality of the cesarean scar, and its far-reaching clinical implications. This review integrates experimental observations, histologic and imaging studies, and surgical experience to examine the biology of uterine scar formation after cesarean delivery. We trace the evolution of closure techniques, identify factors that contribute to optimal or defective healing, and discuss the implications for future reproductive outcomes. Our objective is to provide clinicians with a biologically informed and surgically practical framework for uterine closure that minimizes short-term complications and reduces the burden of long-term sequelae. A substantial body of research has examined the biology of uterine healing, yet this knowledge is rarely integrated into everyday clinical practice. To bridge this gap, this review includes a series of concise summary boxes that highlight key biologic and surgical concepts, offering a quick reference to foundational studies and practical insights that inform optimal closure techniques.