Transverse uterine incision non-closure versus closure: an experimental study in dogs

European Journal of Obstetrics & Gynecology and Reproductive Biology 88 (2000) 95–99

www.elsevier.com / locate / ejogrb

Original Article

Transverse uterine incision non-closure versus closure: an experimental study in dogs ¨ ¨ a , *, C¸etin Kotan b , Serdar Ugras ˇ ¸ c , Muhammet Alan d , Talip Gul ¨ e Abdulaziz Gul a

¨ ¨ ¨ Yil University, Van, Turkey Department of Obstetrics and Gynecology, Faculty of Medicine, Yuzuncu b ¨ ¨ ¨ Yil University, Van, Turkey Department of General Surgery, Faculty of Medicine, Yuzuncu c ¨ ¨ ¨ Yil University, Van, Turkey Department of Pathology, Faculty of Medicine, Yuzuncu d ¨ ¨ ¨ Yil University, Van, Turkey Department of Obstetrics, Faculty of Veterinary, Yuzuncu e Department of Obstetrics and Gynecology, Faculty of Medicine, Dicle University, Diyarbakir, Turkey Received 27 January 1999; received in revised form 3 March 1999; accepted 28 April 1999

Abstract This study was conducted to test the hypothesis that non-closure of all layers of the uterus during low transverse cesarean section is not associated with increased intra-operative or immediate and late postoperative complication. Eleven pregnant dogs underwent cesarean section for the evaluation of non-closure and closure of all layers of the uterus on immediate or early and late postoperative complication and the effect of suture in tissue. Statistical analysis was performed using Student’s t-test for continuous variables and analysis for qualitative variables. Significance was defined as P , 0.05. The ranges of wound infection, other morbidity, and mortality were similar between the groups. The average operating time was significantly less for the non-closure group (71.0067.11 min) than for the closure group (92.0066.12 min; P , 0.005). Adhesion was significantly less (P , 0.001) for the non-closure group than for the closure group. The ranges of myometrial necrosis (5 / 5: 100% versus 0 / 5: 00%; P , 0.001) and fibrosis (2 / 5: 40% versus 0 / 5: 00%; P , 0.01) were significantly higher for the closure group than for the open group. It was found that non-closure of all layers of the uterus at low transverse cesarean incision had no adverse effect on immediate and late postoperative complication in dogs. Our data show that non-closure of all layers of the uterus at low transverse cesarean incision results in significantly less muscular necrosis and fibrosis than in the closure group. We suggest that non-closure and / or at least non-vigorous locking but very simple closure of all layers of the uterus at low transverse cesarean incision may be preferential in appropriate cases.  2000 Elsevier Science Ireland Ltd. All rights reserved. Keywords: Cesarean incision; Non-closure; Closure; Dogs

1. Introduction Cesarean section is defined as delivery of the fetus through an incision in the abdominal wall (laparotomy) and the uterine wall (hysterotomy) [1]. Cesarean section is the most widespread surgical procedure in the United States [2]. Many reports have addressed an increasing incidence and indication for cesarean delivery [2,3]. However, there is little information relating to the optimum operative technique for this method of delivery [3]. *Corresponding author. Tel.: 190-432-216-4706; fax: 190-432-2167519.

In 1926, Kerr [4] described the transverse lower uterine segment incision and recommended a two-layer closure. In the 15th edition of Williams Obstetrics, Pritchard and MacDonald [3,5] first noted that ‘‘the uterine incision may be closed with either one, or the more traditional, two layers of continuous chromic catgut suture’’. They recommended that the first layer be running and locked and that ‘‘when the lower segment is thin, satisfactory approximation of the cut (we also assume torn) edge can usually be obtained with one layer of suture. If approximation is not satisfactory after a single-layer continuous closure then another layer may be placed’’. Hauth et al. [3] recommended a one-layer closure when its use is anatomically

0301-2115 / 00 / $ – see front matter  2000 Elsevier Science Ireland Ltd. All rights reserved. PII: S0301-2115( 99 )00118-9

96

¨ et al. / European Journal of Obstetrics & Gynecology and Reproductive Biology 88 (2000) 95 – 99 A. Gul

feasible. Theoretically, a one-layer closure damages the tissue less, introduces less foreign material, requires less operative time, and achieves homostasis more effectively. Adhesion formation is related to suture reactivity, tissue de-vascularization, ischemia, and infection [6–9], and it has been proposed that routine closure of the peritoneum should be re-examined [6,8]. The amount of adhesion formation has been correlated with the presence and quantity of suture material [10]. Fibrinolytic and microscopic cellular studies in animals have demonstrated that the broad peritoneal reparative process is different from that of the edge-to-edge skin scatrization [11,12]. After 48–72 h, the entire surface is re-mesothelialized without re-approximation with suture material [13,14]. More recent studies have used less reactive suture materials and different techniques to diminish the inflammatory response associated with tissue healing. However, this inflammatory response is still known to be an important factor in the process of postoperative adhesion formation [6,13]. It is known that the average gestational age of dogs is 65 (63–70) days, their uterus has two horns and one cervix and their placenta is endotheliochorial type [15]. Due to the above hypothesis and classical information, this study was scheduled to compare all layers non-closure and closure of a lower segment transverse incision of the uterus. The aim of this experimental study was to investigate and determine whether non-closure of all layers of a uterine incision heal or not and, if so, whether or not there is any advantage of closure.

2. Material and methods This prospective experimental study received institutional approval. Eleven pregnant dogs were randomly divided into two groups (first group, non-closure; second group, closure). The non-closure group included six dogs and the closure group five dogs, with the average gestational age of each group being the same (non-closure group, 5564.70 days; closure group, 5266.20 days). Each dog’s left abdomen was shaved and prepared with povidone iodine. The drapes, instruments and operating staff were all sterile. Each dog was anesthetized with xylidino hydrochloride 23.32 mg / ml (1 ml / 10 kg) and ketamin hydroclorur 50 mg / ml (2 ml / 10 kg) (Rompun  / Bayer, Turkey and Ketalar  / Eczacibasi, Turkey respectively) administered intramuscularly. Cesarean delivery was accomplished by a left transverse abdominal incision. To deliver the whelp, a 7–8 cm transverse incision was carried out keeping 2–3 cm from the cervix. In the first group the uterine incision line was left open. Because there was no significant bleeding from the incision line, no process was made for hemostasis. However, in the second group, the uterine incision line was sutured by No. 1 chromic cat-gut suture using the Schimiden technique (Fig. 1). Placentas were extracted in both groups

Fig. 1. All layers non-closure and closure of the uterine incision line. Top, closure; bottom, non-closure.

due to endotheliochorial-type placentas. Soon after delivery, in both groups, one dose of oxytocin 5 IU (Synpitan Forte  / Deva, Turkey) was administered intramuscularly (IM). For parenteral prophylactic antibiotic a combination of 232 mg streptomycin and 226.000 IU penicillin G (Vetimisin  / Vetas, Turkey) was administered to the dogs. In both groups, all layers of the abdominal wall except for the skin were sutured en bloc with No. 2 polyglactin (Vicryl  CTI / Ethicon, Nordersdet) using the continuous suture technique. Skin was sutured with No. 3-0 silk interrupted sutures. In the early postoperative period, there was one mortality in the non-closure group. Autopsy revealed no hemorrhage or peritonitis (Fig. 2) and mortality was attributed to cardio-respiratory arrest due to high dose anesthesia. After re-laparotomising the dogs at the end of 5 months postpartum, their abdominal cavities were opened and explored. Finally adhesions were scored blindly as shown in Table 1. Tissues taken from the incision line of each uterus were fixed in 10% neutral buffered formalin prior to routine processing and were embedded in a paraffin block. Sections (4 mm thick) were cut and stained with hematoxylin– eosin. A pathologist, who knew nothing about the study, evaluated all sections, and reported especially on the muscular necrosis, endometriosis, and fibrosis. Statistical analysis was performed using Student’s t-test for continuous variables and analysis for qualitative variables. Significance was defined as P , 0.05.

3. Results The mean gestational age was 5564.70 days for the non-closure group and 5266.20 days for the closure group

¨ et al. / European Journal of Obstetrics & Gynecology and Reproductive Biology 88 (2000) 95 – 99 A. Gul

97

Fig. 2. Abdominal exploration (large arrow) of the dog that died 6 h after the operation. There was no hemorrhage (small arrow).

Table 1 Adhesion scores Score

Criterion

0 1 2 3

No adhesion Mild Moderate Severe

4

More severe

No macroscopic adhesions observed One filmy macroscopic fibrin adhesion Multiple macroscopic filmy fibrous adhesions One dense macroscopic fibrous adhesion with or without additional macroscopic filmy adhesion Multiple dense macroscopic fibrous adhesions with or without additional filmy adhesion

(P 5 0.28, Student’s t-test). The mean operative time was 71.0067.11 min for the non-closure group and 92.0066.12 min for the closure group. Therefore, it was

significantly shorter (P , 0.009) for the non-closure group than for the closure group (Table 2). Intra-abdominal adhesion was significantly less (P ,

Table 2 Gestational age and operating time for the two groups Non-closure group (N 5 5)

Closure group (N 5 5)

t

P

Mean

6SD

Mean

6SD

Gestational age (days)

55.00

4.70

52.00

6.20

0.86

0.42

Gestational age (days)

55 49 52 60 59

6.12

25.01

0.009

Row data

Operation time (min)

71.00

Operation time (min)

62 81 68 74 70

Row data

58 53 58 46 45 7.11

92.00 99 83 89 95 94

98

¨ et al. / European Journal of Obstetrics & Gynecology and Reproductive Biology 88 (2000) 95 – 99 A. Gul

Table 3 Adhesion ratio and scores between the non-closure and closure groups

Adhesion No. Adhesion score

Closure group (N 5 5) No. (%)

Non-closure group (N 5 5) No. (%)

P

4 (80) 12

0 (00) 0

,0.001 ,0.001

0.001) for the non-closure group than for the closure group (Table 3). Myometrial necrosis (5 / 5: 100% versus 0 / 5: 00%; P , 0.001) and fibrosis (2 / 5: 40% versus 0 / 5: 00%; P , 0.01) were significantly higher for the closure group than for the non-closure group (Fig. 3). While muscular necrosis and fibrosis were not observed in any of the non-closure group, muscular necrosis was observed in all cases, and fibrosis was observed in two of the closure group (Table 4).

4. Discussion

Fig. 3. (A) Endometrium (E1), myometrium (M), and serous layer (S) were normally observed in the uterus of dogs of the non-closure group (hematoxylin–eosin stain, original magnification 310). (B–D) Serous layer (S) myometrium (M), muscular necrosis (N), endometriosis focus (E2) between muscular fibers, and fibrosis (F) were observed in the uterus of dogs of the closure group. (hematoxylin–eosin stain, original magnification 325).

In 1882, Sanger [16] recommended that all uterine incisions be sutured immediately after delivery. This procedure immediately reduced the maternal mortality from this previously most often fatal operation [3]. Actually, how the closure of the myometrium reduces mortality is not easy to explain. If there is intensive bleeding, hemostasis is to be performed. We suggest that all layers of the uterus be left open in appropriate and safe hemostasis cases. Currently, most obstetricians continue to perform the lower uterine segment transverse incision described by Kerr [4] in 1926. A transverse incision should not be used if the lower uterine segment has not been sufficiently thinned and widened so as to make the juncture of the active myometrium and the thinned, relatively inactive lower uterine segment easily discernible. It is recommended that closure of a low transverse uterine incision be accomplished with one or two layers of a 36-inch length of 0 chromic cat-gut suture [17,18]. Pritchard and MacDonald [5] suggested that a one layer closure can be easily accomplished. We recommend that non-closure of the myometrium be used if the lower uterine segment is sufficiently thinned and widened and if hemostasis is safe. Our experimental study has demonstrated the benefits of this non-closure technique, including less operating time, myometrial necrosis and fibrosis. We suggest that muscular non-closure be applied in certain cases, considering spontaneous healing of the peritoneal defect, and closure of the wound edges in the incision line in the course of uterine involution. Suture of low caliber, as is commonly used in conservative or reconstructive surgery, should be less frequently associated with adhesion induction by reducing the development of tension and, therefore, is-

Table 4 Comparison of histopathological findings of the incision line in the non-closure and closure groups

Muscular necrosis Fibrosis Endometriosis

Closure group (N 5 5) No. (%)

Non-closure group (N 5 5) No. (%)

P

5 (100) 2 (40) 1 (20)

0 (0) 0 (0) 0 (0)

,0.001 ,0.01 .0.05

¨ et al. / European Journal of Obstetrics & Gynecology and Reproductive Biology 88 (2000) 95 – 99 A. Gul

chemia along the lines of the incision [6,9]. There is no reason why the same should not be true for muscular suture. In our study we concluded that muscular necrosis was caused by ischemia. Closure was in no way superior to non-closure except for worse outcome. We suggest that leaving the myometrium open will decrease the risk of rupture during a subsequent pregnancy and result in a subsequent safe labor. It should be noted that, in our study, the cervix was closed and when the dog rose to its feet, the uterus incision line was facing down. Compared with women, in spite of all the disadvantages, the method applied proved to be successful for dogs. Therefore, this method can also be used for women.

5. Conclusion The uterus is not the same as the intestine, stomach or bladder. Its contents do not cause peritonitis. Postpartum bleeding is controlled physiologically by constriction of the interlacing myometrial fibers that surround the blood vessels supplying the placental implantation site, and closure of the wound edges in the incision line in the course of uterine involution [17,18]. Within 48–72 h of injury, either visceral or parietal, the entire surface of the peritoneum is re-mesothelialized without re-approximation with suture material [13,14]. The line of the uterine incision closes spontaneously with these mechanisms. In many appropriate cases suturing the myometrim is not needed. Suture material, which has many negative effects on tissues, is being used unnecessarily. Williams believed that the uterus heals by regeneration of the muscular fibers and not by scar tissue formation [1]. Our experimental study demonstrated that the uterus of all the non-closure group healed by regeneration of the muscular fibers. We recommend that lower transverse incision non-closure and / or at least non-vigorous locking but very simple closure be used in selected cases if anatomically feasible.

6. Condensation The aim of this experimental study was to investigate and determine whether or not non-closure of all layers of a uterine incision heal and, if so, whether or not there is any advantage of closure.

99

References [1] Cunningham FG, MacDonald PC, Gant NF. Cesarean section and cesarean hysterectomy. In: Cunningham FG, MacDonald PC, Gant NF, editors, Williams obstetrics, 19th ed., Norwalk, CT: Appleton & Lange, 1993, p. 591. [2] Taffel SM, Placek PJ, Moien M et al. 1989 U.S. cesarean section rate steadies: Vaginal Birth After Cesarean (VBAC) rate rises to nearly one in five. Birth 1991;18:73–7. [3] Hauth JC, Owen J, Davis RO. Transverse uterine incision closure: one versus two layers. Am J Obstet Gynecol 1992;167:1108–11. [4] Kerr JMM. The technic of cesarean section, with special reference to the lower uterine segment incision. Am J Obstet Gynecol 1926;12:729–33. [5] Pritchard JA, MacDonald PC. Cesarean section and cesarean hysterectomy. In: Pritchard JA, MacDonald PC, editors, Williams obstetrics, 15th ed., New York: Appleton-Century-Crofts, 1976, p. 903. [6] Holtz G. Adhesion induction by suture of varying tissue reactivity and caliber. Int J Fertil 1982;27:134–5. [7] Elkins TE, Stovall TG, Warren J et al. A histologic evaluation of peritoneal injury and repair: implication for adhesion formation. Obstet Gynecol 1987;70:225–8. [8] O’brien WF, Collins E, Knuppel RA et al. Peritoneal closure or non-closure at cesarean. Obstet Gynecol 1991;77:293–6. [9] Neff MR, Holtz GL, Betsill WL. Adhesion formation and histologic reaction with polydiaxnone and polyglactin suture. Am J Obstet Gynecol 1985;51:20–1. [10] Luciano AA, Hauser KS, Benda J. Evaluation of commonly used adjuvants in the prevention of postoperative adhesions. Am J Obstet Gynecol 1983;146:88–92. [11] Porter JM, McGregor FH, Mullen DC et al. Fibrinolytic activity of mesothelial surface. Surg Forum 1969;20:80–2. [12] Robbins GF, Brunchwig A, Foote FW. Deperitonealization: clinical and experimental observations. Ann Surg 1949;130:466–72. [13] Hubbard TB, Khan MZ, Carag VR et al. The pathology of peritoneal repair: its relation to the formation of adhesions. Ann Surg 1967;165:908–16. [14] Raftery AT. Effect of peritoneal trauma on peritoneal fibrinolytic activity and intraperitoneal adhesion formation. Eur Surg Res 1981;13:397–401. ¨ ¨ ¨ S¸. Placettation and grosses process. In: Alac¸am E, editor, [15] Kuplulu Theriogenologi nural, Ankara: Matbaacilik A.S¸, 1990, p. 97. [16] Eastman NJ. The role of frontier America in the development of cesarean section. Am J Obstet Gynecol 1932;24:919–29. [17] Cunningham FG, MacDonald PC, Gant NF. Cesarean section and cesarean hysterectomy. In: Cunningham FG, MacDonald PC, Gant NF, editors, William obstetrics, 18th ed., Norwalk, CT: Appleton & Lange, 1989, p. 451. [18] Hibbard LT. Cesarean section and other surgical procedures. In: Gabbe SG, Niebyl JR, Simpson JL, editors, Obstetrics: normal and problem pregnancies, New York: Churchill Livingstone, 1986, p. 522.