Gynecologic surgeons often use a suprapubic incision for obstetric and pelvic surgery. Since the initial description by J. Pfannenstiel,1 other transverse incisions have been reported. The main difference is transection of the rectus muscle as performed by Maylard,2 which was not in the Pfannenstiel1 technique. The Maylard incision is used by some surgeons as an alternative to midline laparotomy when good exposure of the pelvis is needed.3,4 Ayers and Morley5 concluded in a randomized study of pregnant women that the Maylard incision should be considered when maximal surgical exposure is necessary. However, guidelines for choosing the type of incision for cesarean delivery have not been well defined, and the choice depends largely on the surgeon's preference.
We investigated whether the transverse muscle-cutting Maylard incision affects postoperative complications, pain, quality of life, or functional abdominal muscle recovery compared with the Pfannenstiel incision for cesarean delivery.
PATIENTS AND METHODS
This randomized double-blind study comparing two techniques of laparotomy was conducted at Montpellier University Hospital between April 1998 and April 1999. Patients eligible for inclusion were those who had elective or emergency cesarean delivery. They had to be more than 18 years old and at more than 37 weeks' gestation. Indications for cesarean delivery included fetal-pelvic disproportion, nonvertex presentation, fetal distress, and dystocia. Exclusion criteria were a scarred abdominal wall, previous cesarean delivery, abdominal wall weakness (hernia), multiple gestation, grand multiparity (more than four deliveries), diabetes mellitus, myopathy, corticosteroid therapy during pregnancy, abnormality of hemostasis or anticoagulant treatment, and delivery using general anesthesia. When the neonate was at risk for transfer to the neonatal unit (for fetal distress, fetal hypotrophy, chorioamnionitis, or fetal malformation), the mother was not asked to participate in the study.
Women who satisfied the inclusion criteria were invited to participate, and they signed a consent form approved by the institutional review board. They were assigned to Pfannenstiel incision or Maylard incision using a random number table with group allocation predetermined and placed in consecutively numbered, sealed envelopes. Envelopes were supplied by an independent institution (Clinical Research Center, School of Medicine, University of Montpellier). The patient was masked to her assignment throughout the study period and the envelope remained sealed until the suprapubic skin incision was completed.
The operative procedures were similar and followed the same technical steps. All operations were performed by a resident with an attending physician assisting. Three surgeons were involved in the study. The skin incision was made in the suprapubic site just above the symphysis pubis. Subcutaneous fat and superficial fascia were incised transversely down to the aponeurosis. The aponeurosis was then incised at its midpoint with a scalpel and extended laterally with scissors. The lateral extent of the incision corresponded to that of the skin incision. After this step, the two laparotomy procedures were different.
When the Pfannenstiel incision was performed, the fascial aponeurosis was freed from the underlying rectus abdominis muscles in both cranial and caudal directions. After the caudal-cut aponeurosis was elevated under tension, the rectus muscles were separated in the midline and the peritoneum was opened in an identical manner using the vertical midline incision. When the Maylard incision was performed, the rectus muscles were undermined from the midline and completely transected just upon the pyramidales muscles using electrocautery. The peritoneum was then cut transversely. The inferior epigastric pedicle was not clamped unless it was bleeding.
Thereafter, the cesarean followed our standard procedure—transverse lower uterine segment incision and delivery of the fetus and placenta. Suturing of the uterine incision was performed without exteriorization of the uterus using Vicryl size 1 monolayer continuous absorbable sutures (Ethicon SAS, Issy-les-moulineaux, France). Vicryl 2–0 continuous absorbable sutures (Ethicon) were then used to close the serosa. The abdominal wall was closed in two layers. For the parietal peritoneum we used Vicryl 2–0 continuous absorbable sutures, and for the fascial aponeurosis we used two Vicryl size 2 continuous absorbable sutures (Ethicon). In the transverse suprapubic incision the muscles were not sutured. Subfascial drainage was not used, and the skin was closed with a subcuticular suture using a 3–0 absorbable Monocril monofilament (Ethicon). The wound dressing was the same and had the same length in all cases (Mepore, 9 cm × 20 cm; Mölnlycke Health Care AB, Linselles, France).
During surgery, after the umbilical cord was divided, each woman received intraveneous antibiotic prophylaxis with amoxicillin plus clavulanic acid (2 g) or erythromycin (1 g) in case of allergy to penicillin. After the cesarean, each woman received intrathecal analgesia with the same protocol (5 mL of saline serum containing 2.5 mg of morphine plus 50–75 μg of clonidine).
All patients were cared for on the same surgical unit. All patients, persons involved in postoperative care (ie, nurses, students, and doctors), and outcome assessors were masked to the incision assignment. The postoperative care was standardized and included intravenous analgesia using acetaminophen (2 g every 8 hours) and ketoprofene (100 mg every 8 hours) for 24 hours, and early feeding before bowel movements (liquids begun 6 hours after surgery and solid food at day 1). When intravenous analgesia was discontinued, the patients were transferred to oral analgesia consisting of acetaminophen (500 mg) or acetaminophen plus dextropropoxyphene (400 mg acetaminophen plus 30 mg dextropropoxyphene) at each intake). The choice of analgesic was left to the patient. The Foley catheter was removed after 24 hours, and patients were encouraged to ambulate. Ocytocics were administered systematically intravenously the first day at a dose of 20 IU. Standard criteria for discharge were used for all study patients.
Outcome measures included intraoperative and postoperative morbidity, postoperative pain, health-related quality of life, and abdominal wall muscle assessment. The demographic characteristics taken into account were well defined morbidity risk factors, such as obesity, duration of ruptured membranes before cesarean, and duration of labor before cesarean.6
The intraoperative variables were scar length and operative time (from skin incision to the end of skin closure). These values were recorded by the second author in a masked manner from direct postoperative measurement of the scar length and the anesthesiologist's observation of operative duration. The use of retractors during the operation was the only variable recorded by the gynecologist involved in the operation.
Postoperative morbidity was evaluated from the records of any postoperative complications, including fever (temperature of more than 38C on two occasions 4 hours apart) and lower urinary tract infection (equal to or more than 105 bacteria per mL of urine). A uritest was used for bacteriologic assessment when reading of leukocyturia and nitrites were positive. Other specific complications, such as wound healing problems (hematomas, infections), phlebitis, and need for postoperative blood transfusion, were also recorded.
Postoperative pain was assessed by using a 100-mm visual analog scale and by analyzing the need for postoperative medications.7 The visual analog scale was presented twice a day, at 9 AM (at rest) (± 1 hour) and after ambulating at 3 PM (± 1 hour) from the first day (day 1) to the fifth day after surgery (day 5). Each time, the patient was reminded of her previous score and was then asked to focus on present abdominal pain. The number of analgesic tablets was recorded from day 1 to day 5. All the relevant preoperative, operative, and postoperative data were collected and checked by the second author.
The related quality of life was evaluated with the Nottingham Health Profile.8 After the cesarean each woman completed a retrospective questionnaire about preoperative status. The patients then completed two questionnaires mailed 1 month and 3 months after surgery to assess late postoperative pain. The first part of this questionnaire has 38 statements divided over six dimensions (mobility, pain, sleep, energy, social isolation, and emotional reaction). Within each dimension, statements are weighted for severity. Positive answers are given the appropriate weight, and the score in any section becomes a measure of the number and severity of perceived problems for that dimension.8 The maximum score on any dimension is 100 and the minimum is 0. For study purposes, we analyzed only the pain and physical mobility dimensions.
Evaluation of abdominal wall function was done 3 months postoperatively by a physical therapist masked to the incision assignment. A subjective evaluation of the strength of the abdominal wall muscles was first obtained with a standard grading system for muscles.9,10 For Janda's test, the patient was on her back with her legs bent. She was asked to perform a slow sit-up while maintaining the spine in flexion. The therapist observed the curve of her spine and assessed supraumbilical muscle strength by direct palpation (range 0–5).11 The Ito test was used to assess the strength of the subumbilical zone. The patient was in the same position but was asked to slowly raise her feet as the therapist evaluated subumbilical muscle strength and the maximal duration (seconds) of muscle work by direct palpation.12
Overall performance of the abdominal wall was measured objectively using a dynamometer (Biodex Medical Systems, Inc, New York, NY) during a series of exercises in the sitting position.13 The dynamometer measures muscle activity at a constant velocity over the course of a full-amplitude of movement in flexion or extension. The velocity is preprogrammed and can range from 0 degrees per second to 450 degrees per second. The initial work of Smith et al12 established the norms for flexor and extensor muscles, and several measures can be used for different angular velocities. The mean work-power seems to be the most reliable index, with reproducibility ranging from 0.93 to 0.99 and a high correlation (ie, r = 0.9) with other factors of measure. The maximal peak torque and maximal work-power of the trunk flexor and extensor muscles were measured at angular velocities of 60 degrees per second and 120 degrees per second. The results are reported in Newton/m and watts, respectively, for torque and power. Each variable was measured five times, with the greatest range used for data analysis.
Baseline demographic and postoperative variables were compared with analysis of variance for continuous data and with χ2 analysis or, when appropriate, the Fisher exact test for categoric data. The nonparametric Kruskal-Wallis test was used to compare the range of variables. Analysis of variance with a fixed factor (group) and repeated measures (time) was used to compare outcome variables (postoperative pain, analgesic intake, and quality of life), with automatic correction of the lack of sphericity. When the group-by-time interactions were significant, time-specific comparisons were performed using the nonparametric Friedman test plus repeated measures. P < .05 was considered statistically significant. The data analyst was masked to group assignment.
During the study period, 450 cesarean deliveries were performed in our unit. One hundred twenty women (28%) were eligible for recruitment and underwent randomly assigned operations. Postoperative questionnaires and outcome variables were available for 97 (81%) of those women. Fifty-four women had the Pfannenstiel incision and 43 had the Maylard incision. Postoperative isokinetic assessment was performed on 30 (55.5%) patients in the Pfannenstiel incision group and 24 (55.8%) in Maylard incision group. There were no statistically significant differences in pretreatment maternal characteristics between patients who responded and those who failed to respond to the therapist's convocation at 3 months. The characteristics included age (P = .46), weight before pregnancy (P = .62) and weight at surgery (P = .88), body-mass index (P = .27), parity (P = .69), time from 5-cm dilatation to cesarean (P = .73), and duration of ruptured membranes (P = .63). The groups were also similar demographically (Table 1). There were no differences in median (range) parity between the Pfannenstiel and Maylard groups (1 [1–3] versus 1 [1–5], P = .65).
There were no differences between Pfannenstiel and Maylard groups in intraoperative characteristics such as (mean ± standard deviation [SD]) scar length (11.7 ± 1.4 cm versus 11.9 ± 1.2 cm, respectively) and operative time (38.6 ± 11.1 versus 38.5 ± 10.9 minutes, respectively), and retractors were used with the same frequency (38 versus 28, respectively).
The data on postoperative complications were available for all patients but five, for whom febrile morbidity was not known. The uritest was not done for eight patients, resulting in a lack of precise data on asymptomatic urinary infection for these patients. The mean (SD) number of hospital days was 6.3 (1.3) in the Pfannenstiel group and 6.7 (2.2) in the Maylard incision group. The incidences of postoperative complications were similar in the two groups regarding febrile morbidity (one case in each group, P = .85), urinary tract infection (one case in each group, P = .85), and wound healing problems (three cases in each group, P = .82). Blood transfusion was performed in one woman in the Pfannenstiel group. No phlebitis developed in either group.
Using the visual analog scale scores for postoperative pain, the median (range) scores varied from 3.5 (0–9) at day 1 (9 AM) to 1 (0–3) at day 5 (3 PM) in the Pfannenstiel group and 3 (0–6) at day 1 (9 AM) to 1 (0–4.5) at day 5 (3 PM) for the Maylard incision group. Group-by-time interactions measured by analysis of variance with repeated measures were not significant between groups (P = .55). The number of analgesic tablets used is reported in Table 2. No significant group-by-time interactions were obtained concerning acetaminophen use (P = .88) or acetaminophen plus dextropropoxyphene intake (P = .86).
When health-related quality of life was assessed using the same statistical methods, no significant group-by-time interactions were found in the physical mobility (P = .57) and pain (P = .20) dimensions of the Nottingham Health Profile questionnaire (data not shown).
The 3-month physical test results were not different between Pfannenstiel and Maylard incision patients. Janda's test did not show a difference (3.3 ± 0.9 versus 3.4 ± 1.2) between the two groups. The average and range obtained by the Ito test did not differ between groups (50 [7–215] seconds versus 55 [10–160] seconds, respectively). Using isokinetic measurement, no differences were found for the maximal peak torque of the trunk extensor and flexor muscles at the two angular velocities tested (60 and 120 degrees per second). The data on maximal work-power are presented in Table 3. No differences were found for any of the abdominal muscles evaluated.
We focused our study on the following three important areas: homogeneity of the patient population and surgical procedure, long-term outcome, and rigorous tools of evaluation.
Two previously published studies5,13 compared the Pfannenstiel incision and incision with rectus muscle section. Ayers and Morley5 compared the Maylard incision with the Pfannenstiel incision. The mean length of the Maylard incision was 18 cm, which was significantly longer than the Pfannenstiel incision (14 cm). They found no differences in perioperative morbidity in the two groups. Pain and quality of life were not assessed, and objective evaluation tools were not used. Berthet et al13 randomly assigned 119 women to a Pfannenstiel incision group or a transverse muscle-cutting incision group and compared the 3-month clinical abdominal wall strength. The method of abdominal wall evaluation was not described, and the authors were unable to demonstrate any difference between groups.
Data on postoperative pain are difficult to interpret because pain thresholds are subject to individual variation. Studies have proven that pain can be influenced by factors such as the patient's sex, the caregiver's attitude, the frequency and quality of relational coordination with the patients, and the type of wound dressing.14–17 For these reasons, we analyzed pain from analgesic requirements, scores on the visual analog scale, and related quality of life responses. Our study showed no significant differences with rectus transection. Two studies compared postoperative pain and analgesic use. One study reported results of Joel-Cohen laparotomy versus Pfannenstiel incision18 and another on Kustner laparotomy versus Pfannenstiel incision.19 Although the rectus muscle was not cut in either study, no difference in early postoperative pain was demonstrated. Finally, there are no studies on the quality of life after laparotomy.
The effects of muscle harvesting on abdominal wall strength have been assessed by reconstructive surgeons who performed rectus abdominis muscle flap breast reconstruction. Zauner-Dungl et al20 reported a case-control study of patients who had rectus abdominis muscle flap reconstruction and noted that the patients had impaired isokinetic strength compared with controls. Conversely, Suominen et al,21 Kind et al,22 and Edsander-Nord et al23 reported preoperative and postoperative evaluation of patients who had rectus abdominis muscle flaps and concluded that the use of either the entire rectus muscle or only a part did not influence postoperative abdominal strength measured objectively by isokinetic techniques. Although transverse suprapubic incision compares closely with rectus abdominis muscle flaps, it is likely that the latter surgery causes more abdominal trauma and sequelae because of the concomitant aponeurosis resection. However, all our patients were pregnant and we cannot generalize our findings to nonpregnant women.
Although our study failed to show any differences in the variables tested, statistical significance might have been reached if more women had been enrolled. The lowest P value (.07) obtained in our study was in the comparison of isokinetic parameters (maximum work-power for extensor muscles at gular velocity of 120 degrees per second). Using this difference, we determined that a sample of 55 women undergoing isokinetic evaluation would have revealed a difference, if it existed, with a power of 90% at the .05 significance level. The study would have taken 2 years. A comparison between preoperative and postoperative values in the same patient would also have been of great interest. Unfortunately, cesarean delivery is often unplanned and thus preoperative measurements are not possible.
In summary, our study found that transecting the rectus muscle was no more deleterious than the Pfannenstiel incision.
1. Pfannenstiel J. On the advantages of a transverse cut of the fascia above the symphysis for gynecological laparotomies, and advice on surgical methods and indications. Samml Klin Votr Gynäkol 1897;68:98.
2. Maylard AE. Direction of abdominal incision. Brit Med J 1907;2:895–901.
3. Helmkamp BF, Krebs HB. The Maylard incision in gynecologic surgery. Am J Obstet Gynecol 1990;163:1554–7.
4. Orr JW, Orr PJ, Bolen DD, Holimon JL. Radical hysterectomy: Does the type of incision matter? Am J Obstet Gynecol 1995;173:399–405.
5. Ayers JW, Morley GW. Surgical incision for cesarean section. Obstet Gynecol 1987;70:706–8.
6. Nielsen TF, Hokegard KH. Postoperative cesarean section morbidity: A prospective study. Am J Obstet Gynecol 1983;146:911–5.
7. DeLoach LJ, Higgins MS, Caplan AB, Stiff JL. The visual analog scale in the immediate postoperative period: Intra-subject variability and correlation with a numeric scale. Anesth Analg 1998;86:102–6.
8. Hunt SM, McEwen J, McKenna SP. Measuring health status: A new tool for clinicians and epidemiologists. J R Coll Gen Pract 1985;35:185–8.
9. Daniels L, Worthingham C. Muscles testing. Techniques of manual examination. Philadelphia: WB Saunders; 1986.
10. Lacôte M, Chevalier AM, Miranda A, Bleton JP. Clinical evaluation of muscle function. London, UK: Churchill Livingstone; 1990.
11. Kumar S, Dufresne RM, Van Schoor T. Human strength profile in flexion and extension. Spine 1995;20:160–8.
12. Smith SS, Mayer TG, Gatchel RJ, Becker TJ. Quantification of lumbar function. Measures in sagittal and axial planes in normal subjects. Spine 1985;10:757–64.
13. Berthet J, Peresse JF, Rosier P, Racinet C. Comparative study of Pfannenstiel's incision and transverse abdominal incision in gynecologic and obstetric surgery. Presse Med 1989;18:1431–3.
14. Morin C, Lund JP, Villarroel T, Clokie CM, Feine JS. Differences between the sexes in post-surgical pain. Pain 2000;85:79–85.
15. Miaskowski C, Crews J, Ready LB, Paul SM, Ginsberg B. Anesthesia-based pain services improve the quality of postoperative pain management. Pain 1999;80:23–9.
16. Gittell JH, Fairfield KM, Bierbaum B, Head W, Jackson R, Kelly M, et al. Impact of relational coordination on quality of care, postoperative pain and functioning, and length of stay: A nine-hospital study of surgical patients. Med Care 2000;38:807–19.
17. Briggs M. Surgical wound pain: A trial of two treatments. J Wound Care 1996;5:456–60.
18. Stark M, Chavkin Y, Kupfersztain C, Guedj P, Finkel AR. Evaluation of combinations of procedures in cesarean section. Int J Gynaecol Obstet 1995;48:273–6.
19. Vercellini P, Cortesi I, Oldani S, et al. Comparison of postoperative complications after Kustner and Pfannenstiel transverse suprapubic incisions. Arch Gynecol Obstet 1996;258:201–6.
20. Zauner-Dungl A, Resch KL, Herczeg E, Piza-Katzer H. Quantification of functional deficits associated with rectus abdominis muscle flaps. Plast Reconstr Surg 1995;96:1623–8.
21. Suominen S, Asko-Seljavaara S, von Smitten K, et al. Sequelae in the abdominal wall after pedicled or free TRAM flap surgery. Ann Plast Surg 1996;36:629–36.
22. Kind GM, Rademaker AW, Mustoe TA. Abdominal-wall recovery following TRAM flap: A functional outcome study. Plast Reconstr Surg 1997;99:417–28.
23. Edsander-Nord A, Jurell G, Wickman M. Donor-site morbidity after pedicled or free TRAM flap surgery: A prospective and objective study. Plast Reconstr Surg 1998; 102:1508–16.