Randomized study of closure of the peritoneum at cesarean delivery.

This study was conducted to test the hypothesis that nonclosure of the visceral and parietal peritoneum during low transverse cervical cesarean delivery is not associated with increased intraoperative or immediate postoperative complications. One hundred thirteen patients scheduled for low transverse cervical cesarean were randomized to either closure of both the visceral and parietal peritoneum with absorbable suture (N = 59) or no peritoneal closure (N = 54). Patients were cared for in the usual postoperative manner without reference to treatment group. There were no demographic differences between the groups and no differences in method(s) of anesthesia, operative indication(s), or use of peripartum epidural narcotics. The incidence of fever, endometritis, or wound infection was similar between groups. There were no differences in the number of patients requiring parenteral narcotic analgesia or in the number of doses per patient. The number of oral analgesic doses was significantly greater with closure than without (P = .014). The frequency with which postoperative ileus was diagnosed in each group was similar, and there was no difference regarding the day on which patients were advanced to liquid or select diets. Bowel stimulants were administered more frequently to the closure than to non-closure patients (P = .03). The average operating time was shorter for the open group than for the closure group (P less than .005). We conclude that non-closure of the visceral and parietal peritoneum at low transverse cervical cesarean delivery appears to have no adverse effect on immediate postoperative recovery, may decrease postoperative narcotic requirements, allows less complicated return of bowel function, and provides a simplified and shorter surgical procedure.

A Randomized Study of Closure of the Peritoneum at Cesarean Delivery DALE B. HULL, MD, AND MICHAEL W. VARNER, MD This study was conducted to test the hypothesis that non­closure of the visceral and parietal peritoneum during low transverse cervical cesarean delivery is not associated with increased intraoperative or immediate postoperative compli­cations. One hundred thirteen patients scheduled for low transverse cervical cesarean were randomized to either clo­sure of both the visceral and parietal peritoneum with absorbable suture (N = 59) or no peritoneal closure (N = 54). Patients were cared for in the usual postoperative manner without reference to treatment group. There were no demo­graphic differences between the groups and no differences in method(s) of anesthesia, operative indication(s), or use of peripartum epidural narcotics. The incidence of fever, en­dometritis, or wound infection was similar between groups. There were no differences in the number of patients requir­ing parenteral narcotic analgesia or in the number of doses per patient. The number of oral analgesic doses was signif­icantly greater with closure than without (P = .014). The frequency with which postoperative ileus was diagnosed in each group was similar, and there was no difference regard­ing the day on which patients were advanced to liquid or select diets. Bowel stimulants were administered more fre­quently to the closure than to non-closure patients (P = .03). The average operating time was shorter for the open group than for the closure group (P < .005). We conclude that non-closure of the visceral and parietal peritoneum at low transverse cervical cesarean delivery appears to have no adverse effect on immediate postoperative recovery, may decrease postoperative narcotic requirements, allows less complicated return of bowel function, and provides a sim­plified and shorter surgical procedure. (Obstet Gynecol 77:818, 1991) Cesarean delivery is the most common intraperitoneal surgical procedure performed in the United States.1 The description by Kerr2 of the transperitoneal low transverse cervical incision includes active closure of the visceral peritoneum across the uterine incision. In addition to the visceral peritoneum, the parietal peri­toneum has traditionally also been closed in a separate From the Department of Obstetrics and Gynecology, University of Utah College of Medicine, Salt Lake City, Utah. layer from the remainder of the abdominal-wall clo­sure. 3 This closure of the peritoneum has persisted in obstetrics despite published reports, primarily in the general surgery literature, indicating that reapproxi­mation is not only unnecessary for wound healing and wound strength but may actually delay healing and promote adhesion formation.4-10 We conducted this study to test the hypothesis that non-closure of the visceral and parietal peritoneum at low transverse cervical cesarean is not associated with increased intra­operative or immediate postoperative complications. Materials and Methods Between June and September 1988, 117 women who were to undergo cesarean delivery were randomized to one of two categories. Group assignment was based on the last digit of the patient's medical record. The control group consisted of women undergoing low transverse cervical cesarean with active closure of both the visceral and parietal peritoneum using a delayed absorbable suture and a continuous running tech­nique. The non-closure (study) group consisted of patients undergoing the same procedure but without reapproximation of either peritoneal membrane. Four subjects were disqualified because they received verti­cal hysterotomies on the basis of intraoperative find­ings. Fifty-nine women were randomized to the con­trol group and 54 to the study group. In both groups, the fascia was closed as a separate layer using delayed absorbable suture and the skin incision was reapproximated using staples. The oper­ations were performed by a variety of resident and faculty physicians in the Department of Obstetrics and Gynecology of the University of Utah. The patients were managed in the usual postopera­tive manner. The nursing staff was not aware of the patient's treatment group. All medications were ad­ministered on an as-needed basis. Postoperative man­agement decisions were made without reference to 818 0029-7844/91/$3.50 Obstetrics & Gynecology Table 1. Patient Demographics, Anesthesia Type, and Epidural Narcotic Rate Control group (N = 59) Study group (N = 54) Significance Age 26.8 ± 6.1 25.4 ± 5.2 NS Range 15-38 ' 15-37 Parity 2.4 ± 1.3 2.6 ± 2.2 NS Range l-£ 1-14 Gestational age 36.7 ± 3.5 36.7 ± 4.2 NS Range 26-41 26-42 Primary cesarean 38 (64%) 35 (64.8%) NS Anesthesia Epidural 45 (76.2%) 42 (77.8%) NS General 11 (18.6%) 9 (16.7%) NS Spinal 3 (5.1%) 3 (5.6%) NS Epidural narcotics 25 (42.3%) 24 (44.4%) NS NS = not significant. treatment group. Some subjects in each group were treated with a single dose of epidural morphine sulfate (3-5 mg) perioperatively by the anesthesiologist as­signed to the case. This decision was independent of treatment group. Following the study period, we reviewed the hospi­tal records for patient demographics, duration of sur­gery, and postoperative complications, including fever (temperature of 38C or greater at least once during the postoperative hospitalization), endometritis, antibiotic usage, ileus or partial ileus, and wound problems. Diet advancement and length of hospital stay were also evaluated. Bowel stimulants were used in any woman who complained of postoperative gas pains or consti­pation without evidence of bowel obstruction. Postop­erative narcotic usage was quantified by the number of administered doses of parenteral meperidine or mor­phine and oral oxycodone or codeine. This study was conducted under a protocol ap­proved by the Human Subjects Study Institutional Review Board of the University of Utah School of Medicine. Statistical analysis was performed using Student t test for continuous variables and x1 analysis for discrete variables. Statistical significance was de­fined as P < .05. Results Table 1 presents the patient demographic and anesthe­sia data; we found no significant differences between the groups. Potential sources of antepartum and peri- partum bias appear in Table 2; again, there were no significant differences between the groups. Table 3 outlines the febrile and infectious morbidity as recorded in the hospital records. There were no significant differences between the groups. Five of the VOL. 77, NO. 6, JUNE 1991 Table 2. Antepartum and Peripartum Factors Control group (N = 59) Study group (N = 54) Significance Tubal ligation 10 (16.9%) 5 (9.3%) NS Severe preeclampsia 3 (5.1%) 5 (9.3%) NS Intravenous magnesium 7 (11.9%) 6 (11.1%) NS sulfate Chorioamnionitis 1 (1.7%) 5 (9.3%) NS NS = not significant. six study subjects with endometritis were discharged on the fourth postoperative day. No patients in either group had pelvic abscess, septic pelvic thrombophle­bitis, or peritonitis and none required reoperation or other invasive procedures. There was no significant difference between the groups in the number of doses of postoperative paren­teral narcotics per subject (control group 4.6 ± 2.7, study group 5.9 ± 4.1). The controls did require significantly more oral narcotics (control group 11.1 ± 6.3, study group 8.6 ± 4.9; P = .014). This significant difference persisted when women undergoing concur­rent sterilization procedures were excluded (control [N = 49] 10.5 ± 6.4, study [N = 49] 8.0 ± 5.4; P = .034). Fourteen controls and 15 study subjects required no parenteral narcotic analgesia and all had previously received intraoperative epidural narcotics. The clinical diagnosis of ileus or partial ileus was similar in both groups (control group three of 59 or 5.1%, study group two of 54 or 3.7%). There was no difference between groups in regard to the day on which the patients were advanced to liquid or select diets. However, there was a significant difference in the use of bowel stimulants; 17 controls (28.8%) re­quired suppositories or enemas, in contrast to eight in the study group (14.8%) (P = .03). The mean operative time in the control group was 57.9 ± 13.9 minutes (range 30-113), versus 50.0 ± 13.5 minutes (range 26-100) in the study group (P < .005). Table 3. Postoperative Febrile and Infectious Morbidity and Antibiotic Usage Control group (N = 59) Study group (N = 54) Significance Fever 8 (13.5%) 9 (16.6%) NS Endometritis 3 (5.1%) 6(11.1%) NS Wound infection 5 (8.5%) 3 (5.6%) NS Therapeutic antibiotic administration 6 (10.2%) 9 (16.6%) NS NS = not significant. Hull and Varner Peritoneal Closure a t Cesarean 819 Anesthesia and operating room charges for individual subjects were not recorded. Although not statistically significant, controls had slightly longer postoperative hospitalizations (control group 4.25 ± 0.98 days, range 3-7; study group 4.02 ± 0.79 days, range 3-7). However, a significantly greater proportion of study subjects were discharged by the fourth hospital day (study 46 of 54 or 85.2%, control 43 of 59 or 72.8%; P = .05). Discussion When left undisturbed, peritoneal defects demonstrate mesothelial integrity (reperitonealization) by 48 hours and complete indistinguishable healing (ie, no scarring) by 5 days.4"'6 Reapproximation of peritoneal edges or repair of defects via grafts, even with suture material considered to be minimally reactive, results in increased tissue ischemia and necrosis and foreign-body tissue reactions, and may lead to increased adhesion formation at the sites of reperitonealization.6^7,9,11 Only recently have these principles been recognized and discussed in the obstetric and gynecologic literature.5,12-14 In this sample of abdominal deliveries, there were no significant increases in febrile morbidity, endo­metritis, antibiotic usage, length of hospital stay, or return of bowel function when the peritoneum was left unsutured. On the contrary, the non-closure subjects had less oral pain medication requirements, less trou­blesome bowel function, and no extra hospital days. Although the controls required more bowel stimu­lants, this may merely reflect the increased oral nar­cotic requirements rather than any intrinsic postoper­ative bowel dysfunction. Although the number of subjects experiencing en­dometritis in the study group was larger, the difference did not achieve statistical significance. All but one of the six women were discharged on the fourth postop­erative day, and the remaining patient was released on the fifth postoperative day. In addition, all five study subjects with chorioamnionitis were also discharged by the fourth postoperative day, suggesting that non­closure of the peritoneum did not prolong or prevent the treatment of these infections. The decrease in operative time associated with non­closure of the peritoneum was also associated with less anesthesia time and less time that the wound was exposed to environmental contaminants. Although not specifically addressed by this study, potential eco­nomic benefits include decreased anesthesia and oper­ating room costs, personnel time and expense, and suture costs. Adhesions are caused by ischemia, inflammation, and infection rather than by open surfaces. Omental and bowel adhesions to the hysterotomy site are rare in the low transverse cervical procedure as compared with classical cesarean. Adhesions at peritoneal clo­sure sites following gynecologic surgery have been associated with subsequent small-bowel obstruction.12 A recent study14 confirmed that non-closure of the parietal peritoneum after gynecologic surgery did not increase adhesion formation found at second-look lap­aroscopy. Certainly if the bladder flap is less adherent to the lower uterine segment and fewer adhesions are formed, bladder discomfort may be decreased and subsequent pelvic surgery may be simplified. We believe that our data support the following conclusions regarding non-closure of the visceral and parietal peritoneum at the time of cesarean: 1) It appears to have no detrimental effect in the immediate postoperative recovery period; 2) it may decrease post­operative narcotic requirements; 3) it is associated with less complicated return of bowel function; and 4) it provides a simplified surgical technique requiring less operative time. References 1. Placek PJ, Taffel SM. Recent patterns in cesarean delivery in the United States. Obstet Gynecol Clin North Am 1987;15:607-27. 2. Kerr JMM. The technique of cesarean section with special refer­ence to the lower uterine segment incision. Am J Obstet Gynecol 1926;12:729-34. 3. Cunningham FG, MacDonald PC, Gant NL. Williams obstetrics. 18th ed. Norwalk, Connecticut: Appleton and Lange, 1989:451. 4. Buckman RF, Jr, Buckman PD, Hufnagel HV, Gervin AS. A physiologic basis for the adhesion-free healing of deperitoneal- ized surfaces. J Surg Res 1976;21:67-76. 5. Elkins TE, Stovall TG, Warren J, Ling FW, Meyer NL. A histo­logical evaluation of peritoneal injury and repair: Implications for adhesion formation. Obstet Gynecol 1987;70:225-8. 6. Ellis H. The aetiology of post operative abdominal adhesions, an experimental study. Br J Surg 1962;50:10-6. 7. Ellis H, Heddle R. Does the peritoneum need to be closed at laparotomy? Br ] Surg 1977;64:733-7. 8. Karipineni RC, Wilk PJ, Danese CA. The role of the peritoneum in the healing of abdominal incisions. Surg Gynecol Obstet 1976;142:729-30. 9. McFadden PM, Peacock EE Jr. Preperitoneal abdominal wound repair: Incidence of dehiscence. Am J Surg 1983;145:213-4. 10. Poole GV Jr. Mechanical factors in abdominal wound closure: The prevention of fascial dehiscence. Surgery 1985;97:631-40. 11. Williams DC. The peritoneum. A plea for a change in attitude toward this membrane. Br J Surg 1953;42:401-5. 12. Ratcliff JB, Kapemick P, Brooks GG, Dunnihoo RR. Small bowel obstruction and previous gynecologic surgery. South Med J 1983;76:1349-50. 13. Gabert HA, Bey M. History and development of cesarean oper­ation. Obstet Gynecol Clin North Am 1988;15:591-605. 14. Tulandi T, Hum HS, Gelfand MM. Closure of laparotomy inci­sions with or without peritoneal suturing and second-look lapa­roscopy. Am J Obstet Gynecol 1988;158:536-7. 820 Hull and Varner Peritoneal Closure at Cesarean Obstetrics & Gynecology Address reprint requests to: Michael W. Varner, MD Department of Obstetrics and Gynecology University of Utah School of Medicine 50 North Medical Drive Salt Lake City, UT 84132 Difficult clinical problems confronting the OBSTETRICIAN August 8-10, 1991 The American College of Obstetricians and Gynecologists is sponsoring a course in clinical problems in obstetrics at the Westin Hotel at Copley Place, Boston, Massachusetts. This course has been approved for 16 cognates, Formal Learning, by The American College of Obstetricians and Gynecologists. For further informa­tion, contact the Registrar, The American College of Obstetricians and Gynecolo­gists, 409 12th Street SW, Washington, DC 20024-2188; [phone]. Received August 9, 1990. Received in revised form January 11, 1991. Accepted January 25, 1991. Copyright © 1991 by The American College of Obstetricians and Gynecologists. VOL. 77, NO. 6, JUNE 1991 Hull and Varner Peritoneal Closure at Cesarean 821