European Journal of Obstetrics & Gynecology and Reproductive Biology 172 (2014) 40–45
Contents lists available at ScienceDirect
European Journal of Obstetrics & Gynecology and Reproductive Biology journal homepage: www.elsevier.com/locate/ejogrb
Sharp compared with blunt fascial incision at cesarean delivery: a randomized controlled trial with each case as her own control Anna J.M. Aabakke a,*, Kristine J. Hare b, Lone Krebs a, Niels J. Secher c,d a
Department of Obstetrics and Gynecology, University of Copenhagen, Holbæk Hospital, Denmark Department of Obstetrics and Gynecology, University of Copenhagen, Hvidovre Hospital, Denmark c The Research Unit Women’s and Children’s Health, The Juliane Marie Center, Copenhagen University Hospital, Rigshospitalet, Denmark d Department of Obstetrics and Gynecology, Aarhus University Hospital, Denmark b
A R T I C L E I N F O
A B S T R A C T
Article history: Received 12 August 2013 Received in revised form 4 October 2013 Accepted 25 October 2013
Objective: To compare patient preference for either sharp incision with scissors or blunt manual cleavage of the fascia at cesarean delivery in a randomized controlled trial in which each woman was her own control. Study design: Women undergoing primary cesarean delivery (n = 34) were randomized to side distribution of sharp or blunt incision of the fascia (sharp right and blunt left or blunt right and sharp left) and followed three months postoperatively. The primary outcome was patient preference for the right or left side of the scar 3 months postoperatively and modeled by polytomous logistic regression. The secondary outcome was difference in pain between the two sides measured on a 0.0–10.0 numerical rating scale at 1, 3, and 7 days and 1 and 3 months postoperatively. Pain scores were analyzed with a Wilcoxon signed rank test. Results: 28 cases were analyzed and no significant difference was found in preference after three months. Nine women preferred the sharp (32%, 95% CI 16–52%) and 7 the blunt side (25%, 95% CI 11–45%) (P = 0.804). Pain scores did not differ significantly between the two sides at any time postoperatively either at rest or during mobilization. Conclusion: No significant difference was found in patient preference with regard to sharp or blunt incision of the fascia, nor was there a significant difference in postoperative pain scores. Clinical Trial Registration: ClinicalTrials.gov: www.clinicaltrials.org; NCT01297725. ß 2013 Elsevier Ireland Ltd. All rights reserved.
Keywords: Cesarean delivery Randomized controlled trial Fascia Pain Postoperative
1. Introduction Cesarean delivery rates in the United States, Latin America and Europe are presently around 30% [1–3] and are increasing worldwide [4,5]. Because cesarean delivery is a common procedure, the surgical techniques should be evidence-based [6]. Two types of abdominal entry methods – Pfannenstiel and Joel-Cohen – have been compared in several studies which have examined the postoperative results of a combination of techniques to open the abdominal wall layers [7–10]. To the authors’ knowledge, however, no studies to date have focused on the techniques used to incise the fascia. The fascia can be opened either sharply with scissors or bluntly by traction. Both techniques are presently practised
* Corresponding author at: Department of Obstetrics and Gynecology, University of Copenhagen, Holbæk Hospital, Smedelundsgade 60, 4300 Holbæk, Denmark. Tel.: +45 5948 4268. E-mail addresses: [email protected], [email protected] (Anna J.M. Aabakke). 0301-2115/$ – see front matter ß 2013 Elsevier Ireland Ltd. All rights reserved. http://dx.doi.org/10.1016/j.ejogrb.2013.10.029
according to the choice of the surgeon. It is unknown if one method causes more pain than the other. Studies of surgical techniques commonly have objective outcomes, but the subjective preference of the patient should not be neglected. The aim of this study was to compare blunt and sharp extension of the fascia at cesarean delivery in a study design in which each patient served as her own control. This design eliminates several confounders and strengthens the statistical analysis [11]. 2. Materials and methods This was a single-center, prospective, randomized, controlled, double-blind study with each case as her own control. Before patient enrollment, the trial was approved by the regional ethics committee (reg. no. H-2-2010-129) and the Danish data protection agency and was registered with ClinicalTrials.gov (reg. no. NCT01297725; 16/02/2011). We followed the CONSORT recommendations for reporting randomized, controlled, clinical trials with non-pharmacological treatment [12,13].
A.J.M. Aabakke et al. / European Journal of Obstetrics & Gynecology and Reproductive Biology 172 (2014) 40–45
41
2.1. Participants Eligible participants were women with no history of previous lower abdominal surgery undergoing scheduled primary cesarean delivery. Patients had to be able to speak and understand Danish and provide informed oral and written consent. Exclusion criteria were age below 18 years, pre-pregnancy diabetes mellitus, ongoing infection, daily use of immunosuppressives, alcohol and drug abuse, diseases with chronic pain (e.g. fibromyalgia), and BMI above 35. Participants were recruited from January to July 2011 at Hvidovre University Hospital, Copenhagen, Denmark, which has the largest obstetric unit in Denmark with more than 6000 births yearly, and followed for 3 months. Participants received no financial compensation. 2.2. Design Women were enrolled by one of three investigators, and all recruited women gave written informed consent to participate and were consecutively numbered. Patients were randomized to side distribution of sharp or blunt opening of the fascia (sharp right and blunt left or blunt right and sharp left) and included when surgery had been performed. Recruited women were replaced if procedures did not follow the protocol. Randomization was computergenerated at a 1:1 allocation ratio by a third party not otherwise involved in the trial. The allocation was concealed in 30 identical opaque, sequentially numbered sealed envelopes. The appropriate numbered envelope was opened by the surgeon shortly before initiation of the cesarean delivery, and the allocation was not spoken out loud in the operating theater. All surgeons were instructed in the surgical techniques by a video especially produced for the study (Supplemental Digital Content, Video 1). All women had spinal anesthesia with 10 mg bupivacain, 2.5 mg sufentanil and 0.1 mg morphine, and standard procedures for anesthesia were followed. The surgical procedure was as follows. A Joel-Cohen incision was placed 3 cm below the line joining the anterior superior iliac spines, and the subcutaneous tissue and the fascia incised in the midline only. On the ‘‘sharp’’ side the subcutaneous fat was dissected with a finger, and the fascial incision extended with the tip of a pair of scissors (Fig. 1a). On the ‘‘blunt’’ side, the fascia and subcutaneous tissue were dissected in one pull with the fingertips placed under the rectus muscle (Fig. 1b). The rest of the cesarean delivery procedure was performed using the same techniques bilaterally. A bladder flap was not made and the placenta was delivered spontaneously with gentle cord traction and uterine massage. The fascia was closed with a continuous polyglactin 910 suture (Vicryl Plus; Ethicon Inc., a company in Johnson & Johnson, Norderstedt Hamburg, Germany). If the subcutaneous fat layer was more than 2.5 cm thick, single polyglactin 910 sutures were placed. The skin was closed with subcuticular sutures (Vicryl Rapide 3-0; Ethicon Inc., Nordersted, Germany). All women received 1.5 g i.v. cefuroxime preoperatively. Postoperative analgesics consisted of a diclofenac 100 mg suppository immediately after surgery and oral paracetamol 1 g and oral diclofenac 75 mg twice daily until the 5th postoperative day. Rescue analgesic was oral ketobemidone 10 mg combined with 50 mg dimethylaminodiphenylbuten. Patients were mobilized on the day of surgery. Trial assessment took place on the 1st, 3rd and 7th postoperative day and again 1 and 3 months postoperatively by telephone interview carried out by one of two investigators. Participants, care-givers and outcome assessors were blinded to the allocation. The allocation list was stored in a locked room by a third party not clinically involved in the study. Data were recorded
Fig. 1. (a) Sharp opening of the fascia and (b) blunt opening of the fascia.
on clinical registration forms, and after completed follow-up entered into EpiData Entry version 3.1 (EpiData Association, Odense, Denmark). Data were cleaned and consecutively locked. A copy of the locked data was passed on in exchange for the allocation list before it was broken. 2.3. Outcomes The primary study outcome was the woman’s overall preferred side 3 months postoperatively (right/left/no difference). Secondary outcomes were (a) side with more pain (right/ left/no difference) and (b) difference in pain scores (absolute and relative) between the two sides 1, 3 and 7 days and 1 and 3 months postoperatively, and (c) the occurrence and side distribution of infection assessed after 1 and 3 months. Pain was measured on a numerical rating scale from 0.0 (no pain) to 10.0 (worst pain imaginable) on each side of the scar at rest and during mobilization and assessed by telephone interview. The absolute difference was calculated by subtraction of the scores. The relative difference in pain scores was calculated as: absolute difference/maximum pain score at the time 100%. Infection was defined as the occurrence of any infection reported by the patient and the treatment was registered. During assessment of the primary outcome patients were asked to list the reason for their preference (pain/cosmetics/change in sensitivity). 2.4. Statistics Calculation of sample size for the primary effect variable was based on the following reasoning: if the probability of having a preference is 0.062, then the probability of detecting at least one woman with a side preference out of 25 women is 0.80. Due to the
42
A.J.M. Aabakke et al. / European Journal of Obstetrics & Gynecology and Reproductive Biology 172 (2014) 40–45
Assessed for eligibility Primary caesarean delivery (n=118)
Excluded (n=84): Declined to participate (n=73) Not eligible (n=11)
Randomized (n=34)
Allocated to sharp opening right and blunt left (n=18) Received intervention (n=13) Excluded (n=5) o Surgery not according to randomization (n=3) o Sharp extension on blunt side (n=2)
Excluded (n=0):
Allocated to blunt opening right and sharp left (n=16) Received intervention (n=15) Excluded (n=1) o Vaginal delivery (n=1)
Excluded (n=0):
Lost to follow-up (n=0) Insufficient Registration (n=0)
Lost to follow-up (n=0) Insufficient Registration (n=0)
Analyzed (n=13)
Analyzed (n=15)
Fig. 2. Flow through the study.
risk of drop-outs, 30 women were included in the trial. No interim analyses were performed. The primary endpoint was stated as proportions with 95% exact confidence intervals and modeled by polytomous logistic regression with the odds ratio reported with 95% confidence interval calculated on a log-scale and back transformed. The secondary endpoints of absolute and relative differences in pain scores were analyzed with a Wilcoxon signed rank test for robustness, and the confidence intervals obtained by a t-test. A P value below 0.01 was considered statistically significant due to a post hoc Bonferroni’s correction equivalent to the number of time points investigated (n = 5). Continuous variables were assessed for normal distribution with the Kolmogorov–Smirnov test and presented as means with standard deviations when normally distributed or as medians with interquartile ranges when not. Categorical variables were reported as counts with percentage. Dichotomous data were analyzed with the chi-squared test, and the odds ratios presented with 95% confidence intervals. Data analyses were carried out using IBM SPSS Statistics 19 (SPSS Inc., Chicago, IL, USA). P values below 0.05 were considered statistically significant.
3. Results Thirty-four women undergoing primary cesarean delivery were randomized and 28 included in the analysis. Trial flow, randomization, and reasons for exclusion are illustrated in Fig. 2. Patient and surgery characteristics are shown in Table 1. Three months postoperatively, there was no significant difference in preference of technique, with 9 women preferring the sharp side (32.1%) and 7 the blunt side (25.0%), (OR 1.29 [0.48; 3.45], P = 0.618). Twelve women (42.9%) had no side preference. There was no significant difference with regard to which technique caused more pain at any time during follow-up (Tables 2 and 3). Nor was there a significant absolute or relative difference in pain scores at any time (Fig. 3). Infection was registered in two women (7.1%). One woman had infection on the side of the sharp technique while the other had bilateral infection. The incidence was too low to be statistically analyzed. No side-specific serious adverse events were registered. One woman developed Ogilvie’s syndrome, which was treated conservatively.
A.J.M. Aabakke et al. / European Journal of Obstetrics & Gynecology and Reproductive Biology 172 (2014) 40–45 Table 1 Baseline characteristics. Data are expressed as mean (SD), median (IQR) or n (%). BMI: body mass index. (n = 28)
Variable Patient characteristics Age (years) BMI (kg/m2)
32.1 (5.5) 22.1 (20.8–26.3)
Gravidity 1 2 3 4
17 6 4 1
Parity 0 1 2
19 (68%) 7 (25%) 2 (7%)
Pregnancy characteristics Gestational age at cesarean delivery (days) No. of fetuses Singleton pregnancy Twin pregnancy
(61%) (21%) (14%) (4%)
272 (269–274) 26 (93%) 2 (7%)
Surgery characteristics Indication Maternal request Placenta praevia Breech position Twin pregnancy Previous anal sphincter rupture Other
7 1 11 2 4 3
Surgery type Elective Acute
26 (93%) 2 (7%)
Subcutaneous sutures
(25%) (4%) (39%) (7%) (14%) (11%)
8 (29%)
4. Comment This randomized controlled trial found no significant difference in patient preference for the blunt or sharp technique to incise the fascia at cesarean delivery. Nor were there any significant differences in pain scores after either technique up to 3 months postoperatively.
43
The study design of this trial with each patient as her own control is rarely implemented but suitable for reducing the influence of possible confounders [11]. This study shows that the design is also suitable for trials of different surgical techniques. The experience of pain was the most frequently reported reason for technique preference (88%) (data not shown). Postoperative pain is caused by traumatized nerves [14,15]. The cutaneous innervation of the lower anterior abdominal wall in the area of the transverse incision of cesarean delivery is supplied by the iliohypogastric and subcostal nerves [15]. The left and right nerves innervate their respective sides, and thus it should be possible to distinguish between pain from trauma to the right or the left nerve. It may be, however, that pain from other steps of the cesarean delivery causing visceral pain overshadowed the pain from the fascial incision even though the study design controlled for this. Pain intensity is the most clinically relevant dimension of the pain experience and should be assessed by unidimensional scales based on self-reporting [16]. In this study, pain intensity was measured on a numerical rating scale from 0.0 to 10.0. Compared to the well-known visual analog scale, the numerical rating scale is the instrument of choice in an age-mixed population and has better compliance [16,17]. It is suitable for phone interviews, such as those used in this study. Differences in numerical rating scores have to be both statistically and clinically significant, and the test-retest reliability has to be considered [18]. The level of absolute differences in numerical rating scores of pain intensity depends on the baseline level of pain [19,20]. This relationship is reduced if relative differences in pain scores are used [19,20]. The minimum clinically significant difference in pain on the numerical rating scale, defined as a little/minimal change in pain intensity, is represented by an absolute difference of 1.3– 1.8 units and a relative change of 20.1–20.3% [20,21]. In this study only two of the absolute differences had confidence intervals including 1.3 cm (Fig. 3a and b), indicating there may be no clinically significant absolute differences in pain scores between the two techniques. None of the mean relative differences was above 10%, but several confidence intervals barely included the 20.1–20.3% limit (Fig. 3c and d). With means so close to zero, however, it is doubtful that this might have clinical significance.
Table 2 Technique causing more pain over time. Data are expressed as n (%) and ORsharp/blunt presented with a 95% confidence interval. The data were analyzed with polytomous logistic regression. P < 0.01 is interpreted as significant due to a Bonferroni correction of repeated measures. Time
1 3 7 1 3
day days days month months
Technique causing more pain Sharp
Blunt
No difference
OR (95% CI)
P
7 9 8 6 6
8 7 10 5 2
11 12 8 17 20
0.87 1.28 0.80 1.20 3.00
0.796 0.618 0.638 0.763 0.178
(26.9%) (32.1%) (30.8%) (21.4%) (21.4%)
(30.8%) (25.0%) (38.5%) (17.9%) (7.1%)
(42.3%) (42.9%) (30.8%) (60.7%) (71.4%)
(0.32; (0.48; (0.32; (0.37; (0.61;
2.41) 3.45) 2.03) 3.93) 14.86)
Table 3 Pain scores over time. Pain scores were measured on a Numerical Rating Scale from 0.0 (no pain) to 10.0 (worst pain imaginable) on each side of the scar at rest and during mobilization. Data are expressed as median (IQR) and analyzed with Wilcoxon signed rank test. A P value below 0.01 was considered statistical significant due to a post hoc Bonferroni’s correction equivalent to the number of time points investigated (n = 5). Time
1 3 7 1 3
day (n = 26) days (n = 28) days (n = 26) month (n = 28) months (n = 28)
At rest
During mobilization
Sharp
Blunt
P
Sharp
Blunt
P
2.0 2.0 1.3 0.0 0.0
2.0 2.0 1.0 0.0 0.0
0.168 0.622 0.570 0.276 0.705
3.8 3.3 2.0 0.8 0.0
4.5 3.0 2.8 1.0 0.0
0.409 0.640 0.284 0.428 0.149
(0.8–3.3) (1.0–4.0) (0.0–3.0) (0.0–0.4) (0.0–0.0)
(0.5–4.8) (1.0–5.5) (0.0–2.1) (0.0–1.0) (0.0–0.0)
(2.0–6.0) (2.0–5.0) (0.9–3.1) (0.0–1.4) (0.0–1.0)
(3.0–6.0) (1.5–5.0) (1.0–5.1) (0.0–2.0) (0.0–0.4)
44
A.J.M. Aabakke et al. / European Journal of Obstetrics & Gynecology and Reproductive Biology 172 (2014) 40–45
a
c
b
d
Fig. 3. Difference in pain scores. Absolute differences in pain scores during mobilization (a) and at rest (b), and relative differences in pain scores during mobilization (c) and at rest (d). Data are expressed as mean absolute and relative differences with 95% confidence intervals and significance levels at 1, 3, and 7 days and 1 and 3 months postoperatively. Positive differences express more pain on the sharp side and negative differences more pain on the blunt side. Relative difference = absolute difference/ maximum pain score at the time 100%.
4.1. Limitations The sample size of this study was rather small. Despite this, our results clearly show no trend toward a difference in patient preference for the blunt or sharp techniques and no trend toward clinically relevant absolute and relative pain differences between the two sides. We therefore doubt whether a larger sample size would have changed the outcome other than with regard to the width of the confidence intervals. Two-thirds of the patients assessed as eligible declined to participate (Fig. 1). The influence of this inclusion bias was reduced by the study design. We also found that the characteristics of the patients included and the operations performed (Table 1) corresponded to the average of the center. It may be, however, that the patients declining to participate were more anxious about pain. 4.2. Generalizability The study was performed at a university hospital with a large number of yearly births and a large percentage of trainees. The surgery was performed by a number of different surgeons (n = 14), which could have influenced the internal validity of the results. However, this is the situation at most centers, thereby increasing the external validity of the study. For practical reasons we mainly included elective cesarean deliveries, but the results should also be valid for acute cesarean deliveries, where the quickest technique is preferred. Patients with
a BMI above 35 and diabetes were excluded because of a higher risk of infection, which could confound the outcome, but the results are probably applicable to these groups of patients as well, since abdominal pain perception should not be affected by these conditions. A history of chronic pain was also a reason for exclusion since the perception of pain could be affected in this group. Whether our results are applicable to this group is unknown. Cesarean delivery is a frequent surgical procedure performed with increasing incidence in a young group of patients, so evidence behind the surgical techniques is essential. Several studies have compared groups of cesarean procedures without discerning the individual steps. We recommend future studies of the individual surgical steps comprising the cesarean delivery in order to optimize the techniques. A study design using each case as her own control can be applied in future trials to reduce the influence of confounders and the extent of the study. Acknowledgements The authors thank Associate Professor Lene Theil Skovgaard, Department of Biostatistics, University of Copenhagen and Statistician Steen Ladelund, University of Copenhagen, Hvidovre Hospital for great statistical advice, and the doctors at the Department of Obstetrics and Gynecology, University of Copenhagen, Hvidovre Hospital for cooperative assistance with the study.
A.J.M. Aabakke et al. / European Journal of Obstetrics & Gynecology and Reproductive Biology 172 (2014) 40–45
None of the above received financial compensation for their assistance. This trial was supported by grants from the Faculty of Health, University of Copenhagen and The Region Zealand Health Sciences Research Foundation. Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.ejogrb.2013.10.029. References [1] CDC National Vital Statistics Report. Births: preliminary data for 2009; 2011, http://www.cdc.gov/nchs/data/nvsr/nvsr59/nvsr59_03.pdf [accessed 22.06.11]. [2] Villar J, Valladares E, Wojdyla D, et al. Caesarean delivery rates and pregnancy outcomes: the 2005 WHO global survey on maternal and perinatal health in Latin America. Lancet 2006;367:1819–29. [3] Lumbiganon P, Laopaiboon M, Gulmezoglu AM, et al. Method of delivery and pregnancy outcomes in Asia: the WHO global survey on maternal and perinatal health 2007–2008. Lancet 2010;375:490–9. [4] Berghella V, Baxter JK, Chauhan SP. Evidence-based surgery for cesarean delivery. Am J Obstet Gynecol 2005;193:1607–17. [5] National Institutes of Health state-of-the-science conference statement: cesarean delivery on maternal request, March 27–29, 2006. Obstet Gynecol 2006;107:1386–97. [6] Jenkins TR. It’s time to challenge surgical dogma with evidence-based data. Am J Obstet Gynecol 2003;189:423–7. [7] Mathai M, Hofmeyr GJ. Abdominal surgical incisions for caesarean section. Cochrane Database Syst Rev 2007;(1):CD004453. [8] Stark M, Finkel AR. Comparison between the Joel-Cohen and Pfannenstiel incisions in cesarean section. Eur J Obstet Gynecol Reprod Biol 1994;53:121–2.
45
[9] Mathai M, Ambersheth S, George A. Comparison of two transverse abdominal incisions for cesarean delivery. Int J Gynaecol Obstet 2002;78:47–9. [10] Franchi M, Ghezzi F, Raio L. Joel-Cohen or Pfannenstiel incision at cesarean delivery: does it make a difference? Acta Obstet Gynecol Scand 2002;81:1040–6. [11] Aabakke AJM, Krebs L, Pipper CB, Secher NJ. Subcuticular suture compared with staples for skin closure after cesarean delivery: a randomized controlled trial. Obstet Gynecol 2013;122:878–84. [12] Moher D, Hopewell S, Schulz KF, et al. CONSORT 2010 explanation and elaboration: updated guidelines for reporting parallel group randomised trials. BMJ 2010;340:c869. [13] Boutron I, Moher D, Altman DG, Schulz KF, Ravaud P. Extending the CONSORT statement to randomized trials of nonpharmacologic treatment: explanation and elaboration. Ann Intern Med 2008;148:295–309. [14] Mandelkow H, Loeweneck H. The iliohypogastric and ilioinguinal nerves. Distribution in the abdominal wall, danger areas in surgical incisions in the inguinal and pubic regions and reflected visceral pain in their dermatomes. Surg Radiol Anat 1988;10:145–9. [15] Loos MJ, Scheltinga MR, Roumen RM. Surgical management of inguinal neuralgia after a low transverse Pfannenstiel incision. Ann Surg 2008;248:880–5. [16] Hjermstad MJ, Fayers PM, Haugen DF, et al. Studies comparing numerical rating scales, verbal rating scales, and visual analogue scales for assessment of pain intensity in adults: a systematic literature review. J Pain Symptom Manage 2011;41:1073–93. [17] Williamson A, Hoggart B. Pain: a review of three commonly used pain rating scales. J Clin Nurs 2005;14:798–804. [18] Todd KH. Clinical versus statistical significance in the assessment of pain relief. Ann Emerg Med 1996;27:439–41. [19] Farrar JT, Polomano RC, Berlin JA, Strom BL. A comparison of change in the 0– 10 numeric rating scale to a pain relief scale and global medication performance scale in a short-term clinical trial of breakthrough pain intensity. Anesthesiology 2010;112:1464–72. [20] Cepeda MS, Africano JM, Polo R, Alcala R, Carr DB. What decline in pain intensity is meaningful to patients with acute pain? Pain 2003;105:151–7. [21] Bijur PE, Latimer CT, Gallagher EJ. Validation of a verbally administered numerical rating scale of acute pain for use in the emergency department. Acad Emerg Med 2003;10:390–2.
Contents lists available at ScienceDirect
European Journal of Obstetrics & Gynecology and Reproductive Biology journal homepage: www.elsevier.com/locate/ejogrb
Sharp compared with blunt fascial incision at cesarean delivery: a randomized controlled trial with each case as her own control Anna J.M. Aabakke a,*, Kristine J. Hare b, Lone Krebs a, Niels J. Secher c,d a
Department of Obstetrics and Gynecology, University of Copenhagen, Holbæk Hospital, Denmark Department of Obstetrics and Gynecology, University of Copenhagen, Hvidovre Hospital, Denmark c The Research Unit Women’s and Children’s Health, The Juliane Marie Center, Copenhagen University Hospital, Rigshospitalet, Denmark d Department of Obstetrics and Gynecology, Aarhus University Hospital, Denmark b
A R T I C L E I N F O
A B S T R A C T
Article history: Received 12 August 2013 Received in revised form 4 October 2013 Accepted 25 October 2013
Objective: To compare patient preference for either sharp incision with scissors or blunt manual cleavage of the fascia at cesarean delivery in a randomized controlled trial in which each woman was her own control. Study design: Women undergoing primary cesarean delivery (n = 34) were randomized to side distribution of sharp or blunt incision of the fascia (sharp right and blunt left or blunt right and sharp left) and followed three months postoperatively. The primary outcome was patient preference for the right or left side of the scar 3 months postoperatively and modeled by polytomous logistic regression. The secondary outcome was difference in pain between the two sides measured on a 0.0–10.0 numerical rating scale at 1, 3, and 7 days and 1 and 3 months postoperatively. Pain scores were analyzed with a Wilcoxon signed rank test. Results: 28 cases were analyzed and no significant difference was found in preference after three months. Nine women preferred the sharp (32%, 95% CI 16–52%) and 7 the blunt side (25%, 95% CI 11–45%) (P = 0.804). Pain scores did not differ significantly between the two sides at any time postoperatively either at rest or during mobilization. Conclusion: No significant difference was found in patient preference with regard to sharp or blunt incision of the fascia, nor was there a significant difference in postoperative pain scores. Clinical Trial Registration: ClinicalTrials.gov: www.clinicaltrials.org; NCT01297725. ß 2013 Elsevier Ireland Ltd. All rights reserved.
Keywords: Cesarean delivery Randomized controlled trial Fascia Pain Postoperative
1. Introduction Cesarean delivery rates in the United States, Latin America and Europe are presently around 30% [1–3] and are increasing worldwide [4,5]. Because cesarean delivery is a common procedure, the surgical techniques should be evidence-based [6]. Two types of abdominal entry methods – Pfannenstiel and Joel-Cohen – have been compared in several studies which have examined the postoperative results of a combination of techniques to open the abdominal wall layers [7–10]. To the authors’ knowledge, however, no studies to date have focused on the techniques used to incise the fascia. The fascia can be opened either sharply with scissors or bluntly by traction. Both techniques are presently practised
* Corresponding author at: Department of Obstetrics and Gynecology, University of Copenhagen, Holbæk Hospital, Smedelundsgade 60, 4300 Holbæk, Denmark. Tel.: +45 5948 4268. E-mail addresses: [email protected], [email protected] (Anna J.M. Aabakke). 0301-2115/$ – see front matter ß 2013 Elsevier Ireland Ltd. All rights reserved. http://dx.doi.org/10.1016/j.ejogrb.2013.10.029
according to the choice of the surgeon. It is unknown if one method causes more pain than the other. Studies of surgical techniques commonly have objective outcomes, but the subjective preference of the patient should not be neglected. The aim of this study was to compare blunt and sharp extension of the fascia at cesarean delivery in a study design in which each patient served as her own control. This design eliminates several confounders and strengthens the statistical analysis [11]. 2. Materials and methods This was a single-center, prospective, randomized, controlled, double-blind study with each case as her own control. Before patient enrollment, the trial was approved by the regional ethics committee (reg. no. H-2-2010-129) and the Danish data protection agency and was registered with ClinicalTrials.gov (reg. no. NCT01297725; 16/02/2011). We followed the CONSORT recommendations for reporting randomized, controlled, clinical trials with non-pharmacological treatment [12,13].
A.J.M. Aabakke et al. / European Journal of Obstetrics & Gynecology and Reproductive Biology 172 (2014) 40–45
41
2.1. Participants Eligible participants were women with no history of previous lower abdominal surgery undergoing scheduled primary cesarean delivery. Patients had to be able to speak and understand Danish and provide informed oral and written consent. Exclusion criteria were age below 18 years, pre-pregnancy diabetes mellitus, ongoing infection, daily use of immunosuppressives, alcohol and drug abuse, diseases with chronic pain (e.g. fibromyalgia), and BMI above 35. Participants were recruited from January to July 2011 at Hvidovre University Hospital, Copenhagen, Denmark, which has the largest obstetric unit in Denmark with more than 6000 births yearly, and followed for 3 months. Participants received no financial compensation. 2.2. Design Women were enrolled by one of three investigators, and all recruited women gave written informed consent to participate and were consecutively numbered. Patients were randomized to side distribution of sharp or blunt opening of the fascia (sharp right and blunt left or blunt right and sharp left) and included when surgery had been performed. Recruited women were replaced if procedures did not follow the protocol. Randomization was computergenerated at a 1:1 allocation ratio by a third party not otherwise involved in the trial. The allocation was concealed in 30 identical opaque, sequentially numbered sealed envelopes. The appropriate numbered envelope was opened by the surgeon shortly before initiation of the cesarean delivery, and the allocation was not spoken out loud in the operating theater. All surgeons were instructed in the surgical techniques by a video especially produced for the study (Supplemental Digital Content, Video 1). All women had spinal anesthesia with 10 mg bupivacain, 2.5 mg sufentanil and 0.1 mg morphine, and standard procedures for anesthesia were followed. The surgical procedure was as follows. A Joel-Cohen incision was placed 3 cm below the line joining the anterior superior iliac spines, and the subcutaneous tissue and the fascia incised in the midline only. On the ‘‘sharp’’ side the subcutaneous fat was dissected with a finger, and the fascial incision extended with the tip of a pair of scissors (Fig. 1a). On the ‘‘blunt’’ side, the fascia and subcutaneous tissue were dissected in one pull with the fingertips placed under the rectus muscle (Fig. 1b). The rest of the cesarean delivery procedure was performed using the same techniques bilaterally. A bladder flap was not made and the placenta was delivered spontaneously with gentle cord traction and uterine massage. The fascia was closed with a continuous polyglactin 910 suture (Vicryl Plus; Ethicon Inc., a company in Johnson & Johnson, Norderstedt Hamburg, Germany). If the subcutaneous fat layer was more than 2.5 cm thick, single polyglactin 910 sutures were placed. The skin was closed with subcuticular sutures (Vicryl Rapide 3-0; Ethicon Inc., Nordersted, Germany). All women received 1.5 g i.v. cefuroxime preoperatively. Postoperative analgesics consisted of a diclofenac 100 mg suppository immediately after surgery and oral paracetamol 1 g and oral diclofenac 75 mg twice daily until the 5th postoperative day. Rescue analgesic was oral ketobemidone 10 mg combined with 50 mg dimethylaminodiphenylbuten. Patients were mobilized on the day of surgery. Trial assessment took place on the 1st, 3rd and 7th postoperative day and again 1 and 3 months postoperatively by telephone interview carried out by one of two investigators. Participants, care-givers and outcome assessors were blinded to the allocation. The allocation list was stored in a locked room by a third party not clinically involved in the study. Data were recorded
Fig. 1. (a) Sharp opening of the fascia and (b) blunt opening of the fascia.
on clinical registration forms, and after completed follow-up entered into EpiData Entry version 3.1 (EpiData Association, Odense, Denmark). Data were cleaned and consecutively locked. A copy of the locked data was passed on in exchange for the allocation list before it was broken. 2.3. Outcomes The primary study outcome was the woman’s overall preferred side 3 months postoperatively (right/left/no difference). Secondary outcomes were (a) side with more pain (right/ left/no difference) and (b) difference in pain scores (absolute and relative) between the two sides 1, 3 and 7 days and 1 and 3 months postoperatively, and (c) the occurrence and side distribution of infection assessed after 1 and 3 months. Pain was measured on a numerical rating scale from 0.0 (no pain) to 10.0 (worst pain imaginable) on each side of the scar at rest and during mobilization and assessed by telephone interview. The absolute difference was calculated by subtraction of the scores. The relative difference in pain scores was calculated as: absolute difference/maximum pain score at the time 100%. Infection was defined as the occurrence of any infection reported by the patient and the treatment was registered. During assessment of the primary outcome patients were asked to list the reason for their preference (pain/cosmetics/change in sensitivity). 2.4. Statistics Calculation of sample size for the primary effect variable was based on the following reasoning: if the probability of having a preference is 0.062, then the probability of detecting at least one woman with a side preference out of 25 women is 0.80. Due to the
42
A.J.M. Aabakke et al. / European Journal of Obstetrics & Gynecology and Reproductive Biology 172 (2014) 40–45
Assessed for eligibility Primary caesarean delivery (n=118)
Excluded (n=84): Declined to participate (n=73) Not eligible (n=11)
Randomized (n=34)
Allocated to sharp opening right and blunt left (n=18) Received intervention (n=13) Excluded (n=5) o Surgery not according to randomization (n=3) o Sharp extension on blunt side (n=2)
Excluded (n=0):
Allocated to blunt opening right and sharp left (n=16) Received intervention (n=15) Excluded (n=1) o Vaginal delivery (n=1)
Excluded (n=0):
Lost to follow-up (n=0) Insufficient Registration (n=0)
Lost to follow-up (n=0) Insufficient Registration (n=0)
Analyzed (n=13)
Analyzed (n=15)
Fig. 2. Flow through the study.
risk of drop-outs, 30 women were included in the trial. No interim analyses were performed. The primary endpoint was stated as proportions with 95% exact confidence intervals and modeled by polytomous logistic regression with the odds ratio reported with 95% confidence interval calculated on a log-scale and back transformed. The secondary endpoints of absolute and relative differences in pain scores were analyzed with a Wilcoxon signed rank test for robustness, and the confidence intervals obtained by a t-test. A P value below 0.01 was considered statistically significant due to a post hoc Bonferroni’s correction equivalent to the number of time points investigated (n = 5). Continuous variables were assessed for normal distribution with the Kolmogorov–Smirnov test and presented as means with standard deviations when normally distributed or as medians with interquartile ranges when not. Categorical variables were reported as counts with percentage. Dichotomous data were analyzed with the chi-squared test, and the odds ratios presented with 95% confidence intervals. Data analyses were carried out using IBM SPSS Statistics 19 (SPSS Inc., Chicago, IL, USA). P values below 0.05 were considered statistically significant.
3. Results Thirty-four women undergoing primary cesarean delivery were randomized and 28 included in the analysis. Trial flow, randomization, and reasons for exclusion are illustrated in Fig. 2. Patient and surgery characteristics are shown in Table 1. Three months postoperatively, there was no significant difference in preference of technique, with 9 women preferring the sharp side (32.1%) and 7 the blunt side (25.0%), (OR 1.29 [0.48; 3.45], P = 0.618). Twelve women (42.9%) had no side preference. There was no significant difference with regard to which technique caused more pain at any time during follow-up (Tables 2 and 3). Nor was there a significant absolute or relative difference in pain scores at any time (Fig. 3). Infection was registered in two women (7.1%). One woman had infection on the side of the sharp technique while the other had bilateral infection. The incidence was too low to be statistically analyzed. No side-specific serious adverse events were registered. One woman developed Ogilvie’s syndrome, which was treated conservatively.
A.J.M. Aabakke et al. / European Journal of Obstetrics & Gynecology and Reproductive Biology 172 (2014) 40–45 Table 1 Baseline characteristics. Data are expressed as mean (SD), median (IQR) or n (%). BMI: body mass index. (n = 28)
Variable Patient characteristics Age (years) BMI (kg/m2)
32.1 (5.5) 22.1 (20.8–26.3)
Gravidity 1 2 3 4
17 6 4 1
Parity 0 1 2
19 (68%) 7 (25%) 2 (7%)
Pregnancy characteristics Gestational age at cesarean delivery (days) No. of fetuses Singleton pregnancy Twin pregnancy
(61%) (21%) (14%) (4%)
272 (269–274) 26 (93%) 2 (7%)
Surgery characteristics Indication Maternal request Placenta praevia Breech position Twin pregnancy Previous anal sphincter rupture Other
7 1 11 2 4 3
Surgery type Elective Acute
26 (93%) 2 (7%)
Subcutaneous sutures
(25%) (4%) (39%) (7%) (14%) (11%)
8 (29%)
4. Comment This randomized controlled trial found no significant difference in patient preference for the blunt or sharp technique to incise the fascia at cesarean delivery. Nor were there any significant differences in pain scores after either technique up to 3 months postoperatively.
43
The study design of this trial with each patient as her own control is rarely implemented but suitable for reducing the influence of possible confounders [11]. This study shows that the design is also suitable for trials of different surgical techniques. The experience of pain was the most frequently reported reason for technique preference (88%) (data not shown). Postoperative pain is caused by traumatized nerves [14,15]. The cutaneous innervation of the lower anterior abdominal wall in the area of the transverse incision of cesarean delivery is supplied by the iliohypogastric and subcostal nerves [15]. The left and right nerves innervate their respective sides, and thus it should be possible to distinguish between pain from trauma to the right or the left nerve. It may be, however, that pain from other steps of the cesarean delivery causing visceral pain overshadowed the pain from the fascial incision even though the study design controlled for this. Pain intensity is the most clinically relevant dimension of the pain experience and should be assessed by unidimensional scales based on self-reporting [16]. In this study, pain intensity was measured on a numerical rating scale from 0.0 to 10.0. Compared to the well-known visual analog scale, the numerical rating scale is the instrument of choice in an age-mixed population and has better compliance [16,17]. It is suitable for phone interviews, such as those used in this study. Differences in numerical rating scores have to be both statistically and clinically significant, and the test-retest reliability has to be considered [18]. The level of absolute differences in numerical rating scores of pain intensity depends on the baseline level of pain [19,20]. This relationship is reduced if relative differences in pain scores are used [19,20]. The minimum clinically significant difference in pain on the numerical rating scale, defined as a little/minimal change in pain intensity, is represented by an absolute difference of 1.3– 1.8 units and a relative change of 20.1–20.3% [20,21]. In this study only two of the absolute differences had confidence intervals including 1.3 cm (Fig. 3a and b), indicating there may be no clinically significant absolute differences in pain scores between the two techniques. None of the mean relative differences was above 10%, but several confidence intervals barely included the 20.1–20.3% limit (Fig. 3c and d). With means so close to zero, however, it is doubtful that this might have clinical significance.
Table 2 Technique causing more pain over time. Data are expressed as n (%) and ORsharp/blunt presented with a 95% confidence interval. The data were analyzed with polytomous logistic regression. P < 0.01 is interpreted as significant due to a Bonferroni correction of repeated measures. Time
1 3 7 1 3
day days days month months
Technique causing more pain Sharp
Blunt
No difference
OR (95% CI)
P
7 9 8 6 6
8 7 10 5 2
11 12 8 17 20
0.87 1.28 0.80 1.20 3.00
0.796 0.618 0.638 0.763 0.178
(26.9%) (32.1%) (30.8%) (21.4%) (21.4%)
(30.8%) (25.0%) (38.5%) (17.9%) (7.1%)
(42.3%) (42.9%) (30.8%) (60.7%) (71.4%)
(0.32; (0.48; (0.32; (0.37; (0.61;
2.41) 3.45) 2.03) 3.93) 14.86)
Table 3 Pain scores over time. Pain scores were measured on a Numerical Rating Scale from 0.0 (no pain) to 10.0 (worst pain imaginable) on each side of the scar at rest and during mobilization. Data are expressed as median (IQR) and analyzed with Wilcoxon signed rank test. A P value below 0.01 was considered statistical significant due to a post hoc Bonferroni’s correction equivalent to the number of time points investigated (n = 5). Time
1 3 7 1 3
day (n = 26) days (n = 28) days (n = 26) month (n = 28) months (n = 28)
At rest
During mobilization
Sharp
Blunt
P
Sharp
Blunt
P
2.0 2.0 1.3 0.0 0.0
2.0 2.0 1.0 0.0 0.0
0.168 0.622 0.570 0.276 0.705
3.8 3.3 2.0 0.8 0.0
4.5 3.0 2.8 1.0 0.0
0.409 0.640 0.284 0.428 0.149
(0.8–3.3) (1.0–4.0) (0.0–3.0) (0.0–0.4) (0.0–0.0)
(0.5–4.8) (1.0–5.5) (0.0–2.1) (0.0–1.0) (0.0–0.0)
(2.0–6.0) (2.0–5.0) (0.9–3.1) (0.0–1.4) (0.0–1.0)
(3.0–6.0) (1.5–5.0) (1.0–5.1) (0.0–2.0) (0.0–0.4)
44
A.J.M. Aabakke et al. / European Journal of Obstetrics & Gynecology and Reproductive Biology 172 (2014) 40–45
a
c
b
d
Fig. 3. Difference in pain scores. Absolute differences in pain scores during mobilization (a) and at rest (b), and relative differences in pain scores during mobilization (c) and at rest (d). Data are expressed as mean absolute and relative differences with 95% confidence intervals and significance levels at 1, 3, and 7 days and 1 and 3 months postoperatively. Positive differences express more pain on the sharp side and negative differences more pain on the blunt side. Relative difference = absolute difference/ maximum pain score at the time 100%.
4.1. Limitations The sample size of this study was rather small. Despite this, our results clearly show no trend toward a difference in patient preference for the blunt or sharp techniques and no trend toward clinically relevant absolute and relative pain differences between the two sides. We therefore doubt whether a larger sample size would have changed the outcome other than with regard to the width of the confidence intervals. Two-thirds of the patients assessed as eligible declined to participate (Fig. 1). The influence of this inclusion bias was reduced by the study design. We also found that the characteristics of the patients included and the operations performed (Table 1) corresponded to the average of the center. It may be, however, that the patients declining to participate were more anxious about pain. 4.2. Generalizability The study was performed at a university hospital with a large number of yearly births and a large percentage of trainees. The surgery was performed by a number of different surgeons (n = 14), which could have influenced the internal validity of the results. However, this is the situation at most centers, thereby increasing the external validity of the study. For practical reasons we mainly included elective cesarean deliveries, but the results should also be valid for acute cesarean deliveries, where the quickest technique is preferred. Patients with
a BMI above 35 and diabetes were excluded because of a higher risk of infection, which could confound the outcome, but the results are probably applicable to these groups of patients as well, since abdominal pain perception should not be affected by these conditions. A history of chronic pain was also a reason for exclusion since the perception of pain could be affected in this group. Whether our results are applicable to this group is unknown. Cesarean delivery is a frequent surgical procedure performed with increasing incidence in a young group of patients, so evidence behind the surgical techniques is essential. Several studies have compared groups of cesarean procedures without discerning the individual steps. We recommend future studies of the individual surgical steps comprising the cesarean delivery in order to optimize the techniques. A study design using each case as her own control can be applied in future trials to reduce the influence of confounders and the extent of the study. Acknowledgements The authors thank Associate Professor Lene Theil Skovgaard, Department of Biostatistics, University of Copenhagen and Statistician Steen Ladelund, University of Copenhagen, Hvidovre Hospital for great statistical advice, and the doctors at the Department of Obstetrics and Gynecology, University of Copenhagen, Hvidovre Hospital for cooperative assistance with the study.
A.J.M. Aabakke et al. / European Journal of Obstetrics & Gynecology and Reproductive Biology 172 (2014) 40–45
None of the above received financial compensation for their assistance. This trial was supported by grants from the Faculty of Health, University of Copenhagen and The Region Zealand Health Sciences Research Foundation. Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.ejogrb.2013.10.029. References [1] CDC National Vital Statistics Report. Births: preliminary data for 2009; 2011, http://www.cdc.gov/nchs/data/nvsr/nvsr59/nvsr59_03.pdf [accessed 22.06.11]. [2] Villar J, Valladares E, Wojdyla D, et al. Caesarean delivery rates and pregnancy outcomes: the 2005 WHO global survey on maternal and perinatal health in Latin America. Lancet 2006;367:1819–29. [3] Lumbiganon P, Laopaiboon M, Gulmezoglu AM, et al. Method of delivery and pregnancy outcomes in Asia: the WHO global survey on maternal and perinatal health 2007–2008. Lancet 2010;375:490–9. [4] Berghella V, Baxter JK, Chauhan SP. Evidence-based surgery for cesarean delivery. Am J Obstet Gynecol 2005;193:1607–17. [5] National Institutes of Health state-of-the-science conference statement: cesarean delivery on maternal request, March 27–29, 2006. Obstet Gynecol 2006;107:1386–97. [6] Jenkins TR. It’s time to challenge surgical dogma with evidence-based data. Am J Obstet Gynecol 2003;189:423–7. [7] Mathai M, Hofmeyr GJ. Abdominal surgical incisions for caesarean section. Cochrane Database Syst Rev 2007;(1):CD004453. [8] Stark M, Finkel AR. Comparison between the Joel-Cohen and Pfannenstiel incisions in cesarean section. Eur J Obstet Gynecol Reprod Biol 1994;53:121–2.
45
[9] Mathai M, Ambersheth S, George A. Comparison of two transverse abdominal incisions for cesarean delivery. Int J Gynaecol Obstet 2002;78:47–9. [10] Franchi M, Ghezzi F, Raio L. Joel-Cohen or Pfannenstiel incision at cesarean delivery: does it make a difference? Acta Obstet Gynecol Scand 2002;81:1040–6. [11] Aabakke AJM, Krebs L, Pipper CB, Secher NJ. Subcuticular suture compared with staples for skin closure after cesarean delivery: a randomized controlled trial. Obstet Gynecol 2013;122:878–84. [12] Moher D, Hopewell S, Schulz KF, et al. CONSORT 2010 explanation and elaboration: updated guidelines for reporting parallel group randomised trials. BMJ 2010;340:c869. [13] Boutron I, Moher D, Altman DG, Schulz KF, Ravaud P. Extending the CONSORT statement to randomized trials of nonpharmacologic treatment: explanation and elaboration. Ann Intern Med 2008;148:295–309. [14] Mandelkow H, Loeweneck H. The iliohypogastric and ilioinguinal nerves. Distribution in the abdominal wall, danger areas in surgical incisions in the inguinal and pubic regions and reflected visceral pain in their dermatomes. Surg Radiol Anat 1988;10:145–9. [15] Loos MJ, Scheltinga MR, Roumen RM. Surgical management of inguinal neuralgia after a low transverse Pfannenstiel incision. Ann Surg 2008;248:880–5. [16] Hjermstad MJ, Fayers PM, Haugen DF, et al. Studies comparing numerical rating scales, verbal rating scales, and visual analogue scales for assessment of pain intensity in adults: a systematic literature review. J Pain Symptom Manage 2011;41:1073–93. [17] Williamson A, Hoggart B. Pain: a review of three commonly used pain rating scales. J Clin Nurs 2005;14:798–804. [18] Todd KH. Clinical versus statistical significance in the assessment of pain relief. Ann Emerg Med 1996;27:439–41. [19] Farrar JT, Polomano RC, Berlin JA, Strom BL. A comparison of change in the 0– 10 numeric rating scale to a pain relief scale and global medication performance scale in a short-term clinical trial of breakthrough pain intensity. Anesthesiology 2010;112:1464–72. [20] Cepeda MS, Africano JM, Polo R, Alcala R, Carr DB. What decline in pain intensity is meaningful to patients with acute pain? Pain 2003;105:151–7. [21] Bijur PE, Latimer CT, Gallagher EJ. Validation of a verbally administered numerical rating scale of acute pain for use in the emergency department. Acad Emerg Med 2003;10:390–2.
