- Email: [email protected]
A beneficial effect of purse-string skin closure after ileostomy takedown: A retrospective cohort study
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54
IJSU1336_proof ■ 5 May 2014 ■ 1/6
International Journal of Surgery xxx (2014) 1e6
Contents lists available at ScienceDirect
International Journal of Surgery journal homepage: www.journal-surgery.net
Original research
A beneficial effect of purse-string skin closure after ileostomy takedown: A retrospective cohort study Q2
Yong Joon Suh a, Ji Won Park b, *, Yong Suk Kim b, Sung Chan Park b, Jae Hwan Oh b a
Department of Surgery, Seoul National University Hospital and College of Medicine, Seoul, Republic of Korea Center for Colorectal Cancer, Research Institute and Hospital, National Cancer Center, 323 Ilsan-ro Ilsandong-gu, Goyang-si, Gyeonggi-do 410-769, Republic of Korea b
a r t i c l e i n f o
a b s t r a c t
Article history: Received 11 March 2014 Received in revised form 28 March 2014 Accepted 25 April 2014 Available online xxx
Objectives: The aim of this study was to investigate the clinical benefits of purse-string skin closure (PS) in daily practice compared with conventional linear skin closure (CL) after ileostomy takedown in Korean population. Methods: These retrospectively collected data were based on 157 consecutive patients who underwent ileostomy takedown between November 2010 and September 2011. Before March, 2011, CL was performed in 79 patients. Thereafter, PS was performed in 78 patients. The medical records including pain score recorded daily were reviewed and the postoperative outcomes, including SSI, were analyzed. Results: PS group had a significantly lower overall complication rate than the CL group (8.97% vs. 25.32%, p ¼ 0.010). Among complications, PS group had a significantly lower SSI rate than the CL group (0% vs. 11.39%, p ¼ 0.003). After adjusted for other risk factors (smoking, body mass index, anastomosis method), PS method was associated significantly and independently with a lower SSI rate than CL method (adjusted odds ratio: 26.63, 95% confidence interval: 3.02e267.70, p ¼ 0.001). And the two groups did not differ in terms of postoperative pain (p ¼ 0.323) or pain pattern (p ¼ 0.548). Conclusion: In daily practice, PS had a beneficial effect on SSI in patients who underwent ileostomy takedown in the Korean population. Ó 2014 Surgical Associates Ltd. Published by Elsevier Ltd. All rights reserved.
Keywords: Ileostomy Abdominal wound closure techniques Surgical wound infection
1. Introduction Ileostomy is frequently performed for preventing complications associated with leakage due to poor gastrointestinal anastomosis [1]. It is then followed by ileostomy takedown. Surgeons must perform this procedure with great care because it is associated with a high risk of wound infection due to the presence of microorganisms on the skin around the ileostomy site and possible contamination with the intestinal content during the open-end manipulation of the bowel [2]. Surgical site infection (SSI) can prolong the hospital stay and increase cost of treatment [3]. To avoid this, some surgeons use secondary closure or delayed primary closure [4,5]. Recently, modified secondary closure such as purse-string skin closure (PS) or gunsight skin closure was introduced [6e9]. Several studies showed that PS can reduce SSI compared with conventional linear skin closure (CL) in western countries. However, the clinical benefit of PS in daily routine
Q1
* Corresponding author. Current address: Department of Surgery, Seoul National University Hospital and College of Medicine, Seoul, Republic of Korea. E-mail address: [email protected] (J.W. Park).
practice is poorly understood in Korea. The aim of this study was to investigate the clinical benefits of PS in daily practice compared with CL after ileostomy takedown in the Korean population. 2. Methods Between November 1, 2010 and September 31, 2011, 157 consecutive patients, who underwent ileostomy takedown at National Cancer Center in Korea, were reviewed and adequate sample size was achieved for analysis. All had undergone ileostomy after restorative proctectomy for rectal neoplasms. CL was used for ileostomy skin closure until March, 2011. PS was used in all subsequent patients. The present study was approved by the Institutional Review Board (NCCNCS-11-545) of National Cancer Center (Goyang, Korea). In this observational study, all the collected data were reviewed. The following variables were recorded: demographic data (gender, age, body mass index (BMI), diagnosis, history of previous abdominal surgery, chemotherapy, radiation, alcohol intake, and smoking), American Society of Anesthesiologists (ASA) score, underlying disease, reasons for ileostomy formation, anastomosis
http://dx.doi.org/10.1016/j.ijsu.2014.04.008 1743-9191/Ó 2014 Surgical Associates Ltd. Published by Elsevier Ltd. All rights reserved.
Please cite this article in press as: Y.J. Suh, et al., A beneficial effect of purse-string skin closure after ileostomy takedown: A retrospective cohort study, International Journal of Surgery (2014), http://dx.doi.org/10.1016/j.ijsu.2014.04.008
55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65
IJSU1336_proof ■ 5 May 2014 ■ 2/6
2
Y.J. Suh et al. / International Journal of Surgery xxx (2014) 1e6
method, ileostomy skin closure method, time from ileostomy formation to reversal, transfusion history, length of hospital stay, duration of antibiotic administration, and readmission within 30 days after discharge. And perioperative numerical rating scale (NRS) pain score during admission period was also obtained serial assessment of pain has been conducted by medical personnel in all inpatients at National Cancer Center every day during hospitalization period. In all cases, the elective operation was performed under general anesthesia without preoperative bowel preparation. A circular incision was made in the area surrounding the stoma and the ileum was taken out after being detached from the abdominal wall. The ileum was then subjected to side-to-side (functional endto-end) double stapling with a TLC 75 (Ethicon, Somerville, NJ, USA). End-to-end hand-sewing was only performed if bowel extraction was difficult due to adhesions. Thereafter, the anastomosed bowel was pushed back into its original position. The peritoneum was then subjected to continuous suture with 1e 0 vicryl and the fascia was closed with interrupted 1e0 vicryl. In some patients at high risk of developing seroma, 3.2-mm-diameter JacksonePratt closed suction drain (BarovacÒ 100 ml, Sewoon Medical Co., Seoul, Korea) was inserted into the subcutaneous space under negative pressure. When CL was performed, the skin was closed tight with 2e0 nylon and a skin stapler. When PS was performed, the same methods used to close the peritoneum and fascia described above were used (Fig. 1a). Thereafter, however, Scarfa’s fascia of the subcutaneous layer was approximated by purse-string suturing using 2e0 monosyn (Ethicon), after which the dermal layer was subjected to purse-string suturing using 3e 0 monosyn (Ethicon) (Fig. 1b). The gap was then packed by rolling two-by-two gauze and the upper end of the gauze was pulled out of the wound for drainage. Equally in both groups, cefotetan was administered for antibiotic prophylaxis during the 24-h perioperative period and analgesics recommended by an anesthesiologist at the center were used until 3 days after surgery. Dressing was conducted every other day in both groups from postoperative second day. The presence of SSI was determined by the surgeons on the basis of Center of Disease Control (CDC) guidelines [10]. After discharge, the condition of the wound was monitored by the surgeons during follow-up at an outpatient department (Fig. 1c). The monitoring continued every week until wounds healed up. Even in the meantime, patients or their guardians were instructed to contact the center if there was any sign of wound problem. If there were no problems in wounds, patients were followed up 1, 3, and 6 months after surgery. SSI was defined as a superficial incisional infection, a deep incisional infection, or an organ/space infection that occurred within 30 postoperative days. Healed wound was defined as a wound which required no additional dressing. G*Power 3.1.7 (Institute of Experimental Psychology, Heinrich Heine University, Düsseldorf, Germany) was used to calculate the sample size for a definitive study with a power of 80% and a-level of 0.05. The CL and PS groups were compared by Chi-square test or Fisher’s exact test for categorical data or Student’s t test for continuous data (or the Wilcoxon rank sum test if the data were not normal). The two groups were compared in terms of perioperative NRS pain score by using non-parametric repeated measure ANOVA. Significant relationships (p < 0.1) between the development of SSI and other correction factors were tested for independence by multivariate logistic regression. Adjusted odds ratios were estimated by using the penalized maximum likelihood estimation method in SAS 9.1 (SAS Institute Inc., Cary, NC, USA). The significance of each factor was confirmed by penalized likelihood ratio testing. In two-tailed tests, p < 0.05 was deemed to indicate statistical significance.
Fig. 1. Representative photos of purse-string skin closure. (a) The wound after closing the peritoneum and then the fascia. (b) Purse-string skin closure after two layers were completed. (c) A scab formed on the purse-string-closed wound 3 weeks after surgery.
3. Results 3.1. Characteristics Between November, 2010 and March 2011, CL was performed in 79 patients. Thereafter, PS was performed in 78 patients. There was
Please cite this article in press as: Y.J. Suh, et al., A beneficial effect of purse-string skin closure after ileostomy takedown: A retrospective cohort study, International Journal of Surgery (2014), http://dx.doi.org/10.1016/j.ijsu.2014.04.008
66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65
IJSU1336_proof ■ 5 May 2014 ■ 3/6
Y.J. Suh et al. / International Journal of Surgery xxx (2014) 1e6
no follow-up loss for the current study. The two groups did not differ significantly in terms of baseline characteristics (Table 1). The frequencies of any comorbidity were not different between two groups: 46.8% (37/79) of the PS group and 50% (39/78) of the CL group had one or more comorbidities (p ¼ 0.750). 3.2. Outcomes All patients except for two in each group underwent stapled bowel anastomosis. Operative time, the rate of transfusion, length of hospital stay, and the rate of readmission were not significantly different between two groups (Table 2). Six and five patients in the CL and PS groups were readmitted within 30 days after discharge, respectively. Of the six readmitted patients in the CL group, two had constipation, three had small bowel obstruction, and one had a wound problem that required reoperation. Of the five readmitted patients in the PS group, one had diarrhea, two had constipation, one had small bowel obstruction, and one was admitted for supportive care. The PS group had a significantly lower overall complication rate than the CL group (p ¼ 0.010). However, there was no difference in complications other than SSI. The PS group had a significantly lower SSI rate (0% vs. 11.39%, p ¼ 0.003). The patients who developed SSI spent, on average, 1 week longer in hospital than the patients who did not develop SSI (18.00 7.12 days vs. 11.25 6.15 days, p ¼ 0.002). SSI extended the time for healing by up to 19.74 days in the CL group (p ¼ 0.021). Both groups exhibited a sharp increase in pain on the day of surgery that then dropped over time. As time passed, the NRS pain scores of both groups dropped significantly (p < 0.001; Fig. 2). However, the two
Table 1 Demographic characteristics. Characteristic
CL (n ¼ 79)
Age, years 60.13 10.53 Gender, male 49 (62.03) 2 Body mass index, kg/m 22.59 3.06 ASA score 1.68 0.49 1 26 (32.91) 2 52 (65.82) 3 1 (1.27) Alcohol Yes 50 (63.29) No 29 (36.71) Smoking Yes 38 (48.10) No 41 (51.90) Chemotherapy Yes 62 (78.48) No 17 (21.52) Radiation Yes 50 (63.29) No 29 (36.71) Past abdominal operation history Yes 22 (27.85) No 57 (72.15) Time from ileostomy creation 189.58 102.21 to closure, days Co-morbidity 37 (46.84) Diabetes mellitus 9 (11.39) Hypertension 23 (29.11) Chronic liver disease 6 (7.59) Tuberculosis 6 (7.59) a Others 15 (18.99) The reasons for ileostomy formation Low anastomosis 76 (96.20) Rectovaginal fistula 1 (1.27) Anastomotic leakage 2 (2.53)
PS (n ¼ 78)
p-value
59.53 10.74 50 (64.10) 22.5 2.83 1.58 0.57 36 (46.15) 39 (50.00) 3 (3.85)
0.724 0.869 0.791 0.155 0.104 0.053 0.367 0.516
45 (57.69) 33 (42.31) 0.874 39 (50.00) 39 (50.00) 0.708 59 (75.64) 19 (24.36) 0.870 48 (61.54) 30 (38.46) 0.461 17 (21.79) 61 (78.21) 215.40 149.13 0.251 39 13 24 3 3 12
(50.00) (16.67) (30.77) (3.85) (3.85) (15.38)
0.750 0.367 0.863 0.495 0.495 0.673
74 (94.87) 1 (1.28) 3 (3.85)
0.719 1.000 0.681
CL, conventional linear; PS, purse-string. a Others included asthma, arrhythmia, stroke, heart failure, COPD, and etc.
3
Table 2 Surgical characteristics, postoperative outcomes, and complication rates. Characteristic
CL (n ¼ 79)
PS (n ¼ 78)
p-value
Hand-sewn anastomosis Operative time, minutes Transfusion Length of hospital stay, days Readmission within 30 days Complication SSI Small bowel obstruction Leakage Incisional hernia Sepsis
2 (2.53) 76.7 48.2 2 (2.53) 11.08 4.14 6 (7.59) 20 (25.32) 9 (11.39)a 9 (11.39)b 1 (1.27)c 1 (1.27) 0 (0)
2 (2.56) 77.37 52.93 4 (5.13) 12.21 8.03 5 (6.41) 7 (8.97) 0 (0) 7 (8.97) 0 (0) 0 (0) 0 (0)
1.000 0.920 0.443 0.663 1.000 0.010 0.003 0.617 1.000 1.000 1.000
CL, conventional linear; PS, purse-string; SSI, surgical site infection. a 7 were classified as superficial incisional SSI, 1 as deep incisional SSI, and 1 as organ/space SSI. b 3 also experienced SSI. c 1 also experienced SSI.
groups did not differ significantly in terms of NRS pain scores on any of the days (p ¼ 0.323) or in terms of the pain pattern (p ¼ 0.548). 3.3. Factors associated with the development of SSI The overall rate of SSI was 5.7% (9/157). Univariate analyses revealed that anastomosis method and skin closure method were significantly associated with SSI (Table 3). SSI was more likely to occur after hand-sewn anastomosis than after stapled anastomosis (p ¼ 0.017). It was also significantly more likely to occur after CL than after PS (p ¼ 0.003). It tended to occur more frequently in obese patients (BMI 25) and in smokers, but these trends did not achieve statistical significance (both p ¼ 0.094). These four factors (BMI, smoking history, anastomosis method, and ileostomy skin closure method) were then examined by multivariate logistic regression analysis to determine whether they associated independently with the development of SSI. There were significantly higher odds that patients undergoing ileostomy takedown would develop SSI if CL method was used (adjusted odds ratio 26.63, 95% CI 3.02e267.70; p ¼ 0.001) and hand-sewn anastomosis was performed (adjusted odds ratio 31.16, 95% CI 1.41e607.64; p ¼ 0.029) (Table 4). While a high BMI tended to be associated with a high SSI rate, this trend only achieved marginal statistical significance (p ¼ 0.050). Smoking did not associate significantly with SSI. After stepwise selection, the ileostomy skin closure and anastomosis method variable were selected and the significance of the associations of SSI with these variables in the final model was assessed. This analysis showed significant associations between lower SSI rates and two factors: PS method (p ¼ 0.001) and stapled bowel anastomosis (p ¼ 0.003). 4. Discussion In the present study, the PS group had a significantly lower SSI rate than the CL group. PS method was introduced by Anjan Banerjee [7]. In retrospective studies, the SSI rate associated with PS method ranges from 0% to 6.7% [8,17e19]. Those studies favored PS in terms of SSI. Of these studies, one study could not show significant difference due to the small number of subject patients although PS showed comparable outcomes [18]. They reported that the incidences of wound infection were 16.7% (5/30) and 5.6% (1/ 18) in CL and PS groups, respectively. In one randomized controlled study, an interim analysis showed sufficient reductions of SSI in PS group compared with CL group (6% vs. 38.7%) [13,14]. Enrollment
Please cite this article in press as: Y.J. Suh, et al., A beneficial effect of purse-string skin closure after ileostomy takedown: A retrospective cohort study, International Journal of Surgery (2014), http://dx.doi.org/10.1016/j.ijsu.2014.04.008
66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65
IJSU1336_proof ■ 5 May 2014 ■ 4/6
4
Y.J. Suh et al. / International Journal of Surgery xxx (2014) 1e6
Fig. 2. Numeric pain rating scale before and after conventional linear skin closure and purse-string skin closure. CL, conventional linear; PS, purse-string; NRS, numerical rating scale; PREOP, preoperative; POD, postoperative day.
Table 3 Univariate analysis of eligible factors related to surgical site infection. Characteristic
Age, years 2
BMI, kg/m Gender Alcohol Smoking
Chemotherapy Radiation Co-morbidity Transfusion Anastomosis method Skin closure method
Surgical site infection
60 <60 25 <25 Male Female Yes No Yes No Yes No Yes No Yes No Yes No Hand-sewn Stapled CL PS
No (n ¼ 148)
Yes (n ¼ 9)
79 69 29 119 91 57 88 60 70 78 115 33 94 54 72 76 6 142 2 146 70 78
3 6 4 5 8 1 7 2 7 2 6 3 4 5 4 5 0 9 2 7 9 0
(53.38) (46.62) (19.59) (80.41) (61.49) (38.51) (59.46) (40.54) (47.30) (52.70) (77.70) (22.30) (63.51) (36.49) (48.65) (51.35) (4.05) (95.95) (1.35) (98.65) (47.30) (52.70)
(33.33) (66.67) (44.44) (55.56) (88.89) (11.11) (77.78) (22.22) (77.78) (22.22) (66.67) (33.33) (44.44) (55.56) (44.44) (55.56) (0) (100) (22.22) (77.78) (100) (0)
p-value
0.312 0.094 0.156 0.484 0.094 0.429 0.298 1.000 1.000 0.017 0.003
BMI, body mass index; CL, conventional linear; PS, purse-string.
was discontinued by reason of the result. The recent randomized controlled trials also showed that PS group resulted in lowering of SSI compared with CL group (0% vs. 24%e36.6%) [15,16]. A drain hole in the middle left by PS as basically a method of secondary
Table 4 Multivariate analysis of eligible factors related to surgical site infection. Characteristic
Adjusted OR
95% C.I.
p-value
Smoking (Yes vs. No) BMI (25 vs. <25 kg/m2) Skin closure method (CL vs. PS) Anastomosis method (Hand-sewn vs. stapled)
4.11 5.43 26.63 31.16
0.83e28.86 1.00e32.02 3.02e267.70 1.41e607.64
0.086 0.050 0.001 0.029
OR, odds ratio; C.I., confidence interval; BMI, body mass index; CL, conventional linear; PS, purse-string.
wound healing, may contribute to a lower SSI rate after ileostomy takedown by draining wound discharges. The SSI rate after ileostomy takedown has been reported variously in a few studies. Herwig Porkorny et al. [11] reported that 9% of patients who underwent stoma closure developed SSI. The study of Harold et al. [12] revealed a similar rate while GarciaBotello et al. [5] reported SSI rates up to 18.3%. These studies indicate that along with small bowel obstruction, SSI is a major complication of ileostomy reversal [1,11]. In the present study, the overall SSI rate was relatively low (5.7%). This result may come from the introduction of PS. And this study indicated that PS gave postoperative benefits after ileostomy reversal in daily routine practice. Despite this advantage, PS gave concerns about patients’ discomfort, cosmetic aspect, postoperative pain, wound care, and etc. However, patients in PS group reported greater satisfaction of cosmetic outcome in one systemic review [20]. Similarly, Daniel Camacho-Mauries et al. [16] showed superior patient satisfaction levels and better cosmetic outcomes in their study. Milanchi et al. [19] documented no significant difference although PS group showed higher patient satisfaction score in subjective factors: cosmesis, expectation before surgery, postoperative pain, wound care, and activity. In general accord with their studies, two groups did not differ significantly in terms of NRS pain scores in the present study. The two groups in the present study were homogeneous, which allowed the SSI rates of unselected patients who underwent elective ileostomy reversal to be compared directly. Moreover, unlike in the other studies described above including recent randomized controlled trials, the patients underwent stitching with a double layer of absorbable sutures that did not have to be removed later [15,16]. The SSI rate in secondary closure or delayed primary closure varies differently from 0 to 15% [2,4,6,12,21e24]. Ileostomy takedown is more likely to be classified as clean-contaminated category [2]. The wound to be suspected of contamination would rather be left open logically. The burden of interval closure cannot be negligible in delayed primary closure. In this point of view, secondary closure appears to be an adjunct to the techniques designed to decrease the SSI rate although wound healing is expected to be delayed. PS allows partial wound edge apposition [25]. This apposition may accelerate healing compared to secondary closure.
Please cite this article in press as: Y.J. Suh, et al., A beneficial effect of purse-string skin closure after ileostomy takedown: A retrospective cohort study, International Journal of Surgery (2014), http://dx.doi.org/10.1016/j.ijsu.2014.04.008
66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65
IJSU1336_proof ■ 5 May 2014 ■ 5/6
Y.J. Suh et al. / International Journal of Surgery xxx (2014) 1e6
Analysis of the complications in the CL and PS groups in the present study revealed that the PS group developed significantly fewer complications overall (p ¼ 0.004). According to the Claviene Dindo Classification, seven patients in CL group experienced more than grade II while one in PS group did [26,27]. Considering that SSI accounts for a large percentage of complication after ileostomy takedown, PS may decrease SSI-related morbidity in ileostomy takedown. While most patients underwent routine side-to-side double stapling for anastomosis, the hand-sewn anastomosis method was used in four patients (two from each group). Two of these four patients developed SSI and the multivariate analysis showed that the hand-sewn anastomosis method associated independently with SSI (adjusted odds ratio 31.16, 95% CI 1.41e607.64; p ¼ 0.029). This result is inconsistent with a recent meta-analysis [28]. It is because patients with hand-sewn anastomosis were mostly difficult cases for ileostomy takedown. It should be also emphasized that there were only four cases of hand-sewn anastomosis, and thus the association between hand-sewn anastomosis and SSI should be viewed with caution. In the present study, patients with BMI 25 had a higher SSI rate than patients with BMI <25 and multivariate analysis excluding all other major factors showed that BMI associated marginally significantly with SSI (adjusted odds ratio 5.43, 95% CI 1e32.02; p ¼ 0.050). This may reflect the fact that there were only four patients with BMI 25, which in turn may reflect the low BMI of the Korean population compared to Western populations. A larger sample size may show a significant and independent association between BMI and SSI. If this is the case, it may be advisable to perform PS after ileostomy takedown in patients with BMI 25. Notably, the large-scale retrospective study conducted by Riou et al. [29] revealed that obesity had an adverse effect on wound healing. The hospital stay of both groups in the present study was a few days longer than in the other studies, because Korean healthcare system depends much on hospitals and not enough on communitybased medical practices, and therefore much of their hospital cost is covered by the national health insurance. When 16 patients of CL group who had JP wound drain bags were compared with 63 patients of CL group who did not, no significant difference in SSI rate was found (p ¼ 0.443) as N. Mirbagheri et al. [25] showed in their study. Because JP wound drain bag was removed just before discharge, the JP insertion group was under the same condition as the non-insertion group in terms of the SSI development during the period between discharge and outpatient visit. Accordingly, no significant difference was probably shown. In addition, selection bias of the surgeons might be involved because patients who were thought to have a high risk of developing SSI selectively underwent JP insertion. Therefore, the protective effect of JP insertion against SSI could not be determined in the present study. The present study suffers from the usual limitations of observational studies. In particular, the CL and PS groups were not compared at the same time. This means that patient randomization could not be performed, although it should be noted that the two groups largely experienced identical conditions apart from the method of ileostomy skin closure. Another limitation of this study was the outcome evaluation which was not performed by blinded observers. 5. Conclusion PS had a beneficial effect on SSI in patients who underwent ileostomy takedown in daily practice. PS can be considered as standard of care for wound closure after ileostomy takedown in daily routine practice.
5
Ethical approval This study was approved by the institutional review board of National Cancer Center of Korea (NCCNCS-11-545). Funding None. Conflict of interest The authors declare no conflict of interest. Author contribution Study conception and design: YJ Suh, JW Park. Acquisition of data: YJ Suh, YS Kim, SC Park, JW Park, JH Oh. Analysis and interpretation of data: YJ Suh, JW Park. Drafting of the manuscript: YJ Suh, JW Park. Critical revision: YJ Suh, YS Kim, SC Park, JW Park, JH Oh. Acknowledgments All statistical analyses were supported by the Medical Research Collaborating Center, Seoul National University and Seoul National University Hospital, Seoul, Korea. References [1] O. Kaidar-Person, B. Person, S.D. Wexner, Complications of construction and closure of temporary loop ileostomy, J. Am. Coll. Surg. 201 (2005) 759e773. [2] D.J. Hackam, O.D. Rotstein, Stoma closure and wound infection: an evaluation of risk factors, Can. J. Surg. 38 (1995) 144e148. [3] A. Barbul, When will the war end? Wound Repair Regen. 21 (2013) 179. [4] P.T. Phang, J.M. Hain, J.J. Perez-Ramirez, R.D. Madoff, B.T. Gemlo, Techniques and complications of ileostomy takedown, Am. J. Surg. 177 (1999) 463e466. [5] S.A. Garcia-Botello, J. Garcia-Armengol, E. Garcia-Granero, et al., A prospective audit of the complications of loop ileostomy construction and takedown, Dig. Surg. 21 (2004) 440e446. [6] E.D. Verrier, K.J. Bossart, F.W. Heer, Reduction of infection rates in abdominal incisions by delayed wound closure techniques, Am. J. Surg. 138 (1979) 22e 28. [7] A. Banerjee, Pursestring skin closure after stoma reversal, Dis. Colon Rectum 40 (1997) 993e994. [8] C.D. Sutton, N. Williams, L.J. Marshall, G. Lloyd, W.M. Thomas, A technique for wound closure that minimizes sepsis after stoma closure, ANZ J. Surg. 72 (2002) 766e767. [9] J.T. Lim, S.M. Shedda, I.P. Hayes, “Gunsight” skin incision and closure technique for stoma reversal, Dis. Colon Rectum 53 (2010) 1569e1575. [10] A.J. Mangram, T.C. Horan, M.L. Pearson, L.C. Silver, W.R. Jarvis, Guideline for Prevention of Surgical Site Infection, Centers for Disease Control and Prevention (CDC) Hospital Infection Control Practices Advisory Committee, Am. J. Infect. Control 1999 (27) (1999) 97e132. [11] H. Pokorny, H. Herkner, R. Jakesz, F. Herbst, Mortality and complications after stoma closure, Arch. Surg. 140 (2005) 956e960. [12] D.M. Harold, E.K. Johnson, J.A. Rizzo, S.R. Steele, Primary closure of stoma site wounds after ostomy takedown, Am. J. Surg. 199 (2010) 621e624. [13] K. Reid, P. Pockney, T. Pollitt, B. Draganic, S.R. Smith, Randomized clinical trial of short-term outcomes following purse-string versus conventional closure of ileostomy wounds, Br. J. Surg. 97 (2010) 1511e1517. [14] A. Moreton, S. Shankar, S. Jones, Randomized clinical trial of short-term outcomes following purse-string versus conventional closure of ileostomy wounds (Br. J. Surg. 2010; 97: 1511e1517), Br. J. Surg. 98 (2011) 458. [15] N. Dusch, D. Goranova, F. Herrle, M. Niedergethmann, P. Kienle, Randomized controlled trial: comparison of two surgical techniques for closing the wound following ileostomy closure: purse string vs direct suture, Colorectal Dis. 15 (2013) 1033e1040. [16] D. Camacho-Mauries, J.L. Rodriguez-Diaz, N. Salgado-Nesme, Q.H. Qonzalez, O. Vergara-Fernandez, Randomized clinical trial of intestinal ostomy takedown comparing pursestring wound closure vs conventional closure to eliminate the risk of wound infection, Dis. Colon Rectum 56 (2013) 205e211. [17] C.D. Klink, M. Wünschmann, M. Binnebösel, et al., Influence of skin closure technique on surgical site infection after loop ileostomy reversal: retrospective cohort study, Int. J. Surg. 11 (2013) 1123e1125.
Please cite this article in press as: Y.J. Suh, et al., A beneficial effect of purse-string skin closure after ileostomy takedown: A retrospective cohort study, International Journal of Surgery (2014), http://dx.doi.org/10.1016/j.ijsu.2014.04.008
66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130
1 2 3 4 5 6 7 8 9 10 11 12 13 14
IJSU1336_proof ■ 5 May 2014 ■ 6/6
6
Y.J. Suh et al. / International Journal of Surgery xxx (2014) 1e6
[18] J.R. Lee, Y.W. Kim, J.J. Sung, et al., Conventional linear versus purse-string skin closure after loop ileostomy reversal: comparison of wound infection rates and operative outcomes, J. Korean Soc. Coloproctol. 27 (2011) 58e63. [19] S. Milanchi, Y. Nasseri, T. Kidner, P. Fleshner, Wound infection after ileostomy closure can be eliminated by circumferential subcuticular wound approximation, Dis. Colon Rectum 52 (2009) 469e474. [20] D.P. McCartan, J.P. Burke, S.R. Walsh, J.C. Coffey, Purse-string approximation is superior to primary skin closure following stoma reversal: a systematic review and meta-analysis, Tech. Coloproctol. 17 (2013) 345e351. [21] S.D. Wexner, D.A. Taranow, O.B. Johansen, et al., Loop ileostomy is a safe option for fecal diversion, Dis. Colon Rectum 36 (1993) 349e354. [22] K.S. Wong, F.H. Remzi, E. Gorgun, et al., Loop ileostomy closure after restorative proctocolectomy: outcome in 1,504 patients, Dis. Colon Rectum 48 (2005) 243e250. [23] G. Lahat, H. Tulchinsky, G. Goldman, J.M. Klauzner, M. Rabau, Wound infection after ileostomy closure: a prospective randomized study comparing primary vs. delayed primary closure techniques, Tech. Coloproctol. 9 (2005) 206e208.
[24] N. Vermulst, J. Vermeulen, E.J. Hazebroek, P.P.L.O. Coene, E. van der Harst, Primary closure of the skin after stoma closure. Management of wound infections is easy without (long-term) complications, Dig. Surg. 23 (2006) 255e 258. [25] N. Mirbagheri, J. Dark, S. Skinner, Factors predicting stomal wound closure infection rates, Tech. Coloproctol. 17 (2013) 215e220. [26] D. Dindo, N. Demartines, P.A. Clavien, Classification of surgical complications: a new proposal with evaluation in a cohort of 6336 patients and results of a survey, Ann. Surg. 240 (2004) 205e213. [27] P.A. Clavien, J. Barkun, M.L. de Oliveira, et al., The Clavien-Dindo classification of surgical complications: five-year experience, Ann. Surg. 250 (2009) 187e 196. [28] J. Gong, Z. Guo, Y. Li, et al., Stapled versus hand suture closure of loop ileostomy: a meta-analysis, Colorectal Dis. 10 (2013) e561ee568. [29] J.P. Riou, J.R. Cohen, H. Johnson Jr., Factors influencing wound dehiscence, Am. J. Surg. 163 (1992) 324e330.
Please cite this article in press as: Y.J. Suh, et al., A beneficial effect of purse-string skin closure after ileostomy takedown: A retrospective cohort study, International Journal of Surgery (2014), http://dx.doi.org/10.1016/j.ijsu.2014.04.008
15 16 17 18 19 20 21 22 23 24 25 26 27 28
IJSU1336_proof ■ 5 May 2014 ■ 1/6
International Journal of Surgery xxx (2014) 1e6
Contents lists available at ScienceDirect
International Journal of Surgery journal homepage: www.journal-surgery.net
Original research
A beneficial effect of purse-string skin closure after ileostomy takedown: A retrospective cohort study Q2
Yong Joon Suh a, Ji Won Park b, *, Yong Suk Kim b, Sung Chan Park b, Jae Hwan Oh b a
Department of Surgery, Seoul National University Hospital and College of Medicine, Seoul, Republic of Korea Center for Colorectal Cancer, Research Institute and Hospital, National Cancer Center, 323 Ilsan-ro Ilsandong-gu, Goyang-si, Gyeonggi-do 410-769, Republic of Korea b
a r t i c l e i n f o
a b s t r a c t
Article history: Received 11 March 2014 Received in revised form 28 March 2014 Accepted 25 April 2014 Available online xxx
Objectives: The aim of this study was to investigate the clinical benefits of purse-string skin closure (PS) in daily practice compared with conventional linear skin closure (CL) after ileostomy takedown in Korean population. Methods: These retrospectively collected data were based on 157 consecutive patients who underwent ileostomy takedown between November 2010 and September 2011. Before March, 2011, CL was performed in 79 patients. Thereafter, PS was performed in 78 patients. The medical records including pain score recorded daily were reviewed and the postoperative outcomes, including SSI, were analyzed. Results: PS group had a significantly lower overall complication rate than the CL group (8.97% vs. 25.32%, p ¼ 0.010). Among complications, PS group had a significantly lower SSI rate than the CL group (0% vs. 11.39%, p ¼ 0.003). After adjusted for other risk factors (smoking, body mass index, anastomosis method), PS method was associated significantly and independently with a lower SSI rate than CL method (adjusted odds ratio: 26.63, 95% confidence interval: 3.02e267.70, p ¼ 0.001). And the two groups did not differ in terms of postoperative pain (p ¼ 0.323) or pain pattern (p ¼ 0.548). Conclusion: In daily practice, PS had a beneficial effect on SSI in patients who underwent ileostomy takedown in the Korean population. Ó 2014 Surgical Associates Ltd. Published by Elsevier Ltd. All rights reserved.
Keywords: Ileostomy Abdominal wound closure techniques Surgical wound infection
1. Introduction Ileostomy is frequently performed for preventing complications associated with leakage due to poor gastrointestinal anastomosis [1]. It is then followed by ileostomy takedown. Surgeons must perform this procedure with great care because it is associated with a high risk of wound infection due to the presence of microorganisms on the skin around the ileostomy site and possible contamination with the intestinal content during the open-end manipulation of the bowel [2]. Surgical site infection (SSI) can prolong the hospital stay and increase cost of treatment [3]. To avoid this, some surgeons use secondary closure or delayed primary closure [4,5]. Recently, modified secondary closure such as purse-string skin closure (PS) or gunsight skin closure was introduced [6e9]. Several studies showed that PS can reduce SSI compared with conventional linear skin closure (CL) in western countries. However, the clinical benefit of PS in daily routine
Q1
* Corresponding author. Current address: Department of Surgery, Seoul National University Hospital and College of Medicine, Seoul, Republic of Korea. E-mail address: [email protected] (J.W. Park).
practice is poorly understood in Korea. The aim of this study was to investigate the clinical benefits of PS in daily practice compared with CL after ileostomy takedown in the Korean population. 2. Methods Between November 1, 2010 and September 31, 2011, 157 consecutive patients, who underwent ileostomy takedown at National Cancer Center in Korea, were reviewed and adequate sample size was achieved for analysis. All had undergone ileostomy after restorative proctectomy for rectal neoplasms. CL was used for ileostomy skin closure until March, 2011. PS was used in all subsequent patients. The present study was approved by the Institutional Review Board (NCCNCS-11-545) of National Cancer Center (Goyang, Korea). In this observational study, all the collected data were reviewed. The following variables were recorded: demographic data (gender, age, body mass index (BMI), diagnosis, history of previous abdominal surgery, chemotherapy, radiation, alcohol intake, and smoking), American Society of Anesthesiologists (ASA) score, underlying disease, reasons for ileostomy formation, anastomosis
http://dx.doi.org/10.1016/j.ijsu.2014.04.008 1743-9191/Ó 2014 Surgical Associates Ltd. Published by Elsevier Ltd. All rights reserved.
Please cite this article in press as: Y.J. Suh, et al., A beneficial effect of purse-string skin closure after ileostomy takedown: A retrospective cohort study, International Journal of Surgery (2014), http://dx.doi.org/10.1016/j.ijsu.2014.04.008
55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65
IJSU1336_proof ■ 5 May 2014 ■ 2/6
2
Y.J. Suh et al. / International Journal of Surgery xxx (2014) 1e6
method, ileostomy skin closure method, time from ileostomy formation to reversal, transfusion history, length of hospital stay, duration of antibiotic administration, and readmission within 30 days after discharge. And perioperative numerical rating scale (NRS) pain score during admission period was also obtained serial assessment of pain has been conducted by medical personnel in all inpatients at National Cancer Center every day during hospitalization period. In all cases, the elective operation was performed under general anesthesia without preoperative bowel preparation. A circular incision was made in the area surrounding the stoma and the ileum was taken out after being detached from the abdominal wall. The ileum was then subjected to side-to-side (functional endto-end) double stapling with a TLC 75 (Ethicon, Somerville, NJ, USA). End-to-end hand-sewing was only performed if bowel extraction was difficult due to adhesions. Thereafter, the anastomosed bowel was pushed back into its original position. The peritoneum was then subjected to continuous suture with 1e 0 vicryl and the fascia was closed with interrupted 1e0 vicryl. In some patients at high risk of developing seroma, 3.2-mm-diameter JacksonePratt closed suction drain (BarovacÒ 100 ml, Sewoon Medical Co., Seoul, Korea) was inserted into the subcutaneous space under negative pressure. When CL was performed, the skin was closed tight with 2e0 nylon and a skin stapler. When PS was performed, the same methods used to close the peritoneum and fascia described above were used (Fig. 1a). Thereafter, however, Scarfa’s fascia of the subcutaneous layer was approximated by purse-string suturing using 2e0 monosyn (Ethicon), after which the dermal layer was subjected to purse-string suturing using 3e 0 monosyn (Ethicon) (Fig. 1b). The gap was then packed by rolling two-by-two gauze and the upper end of the gauze was pulled out of the wound for drainage. Equally in both groups, cefotetan was administered for antibiotic prophylaxis during the 24-h perioperative period and analgesics recommended by an anesthesiologist at the center were used until 3 days after surgery. Dressing was conducted every other day in both groups from postoperative second day. The presence of SSI was determined by the surgeons on the basis of Center of Disease Control (CDC) guidelines [10]. After discharge, the condition of the wound was monitored by the surgeons during follow-up at an outpatient department (Fig. 1c). The monitoring continued every week until wounds healed up. Even in the meantime, patients or their guardians were instructed to contact the center if there was any sign of wound problem. If there were no problems in wounds, patients were followed up 1, 3, and 6 months after surgery. SSI was defined as a superficial incisional infection, a deep incisional infection, or an organ/space infection that occurred within 30 postoperative days. Healed wound was defined as a wound which required no additional dressing. G*Power 3.1.7 (Institute of Experimental Psychology, Heinrich Heine University, Düsseldorf, Germany) was used to calculate the sample size for a definitive study with a power of 80% and a-level of 0.05. The CL and PS groups were compared by Chi-square test or Fisher’s exact test for categorical data or Student’s t test for continuous data (or the Wilcoxon rank sum test if the data were not normal). The two groups were compared in terms of perioperative NRS pain score by using non-parametric repeated measure ANOVA. Significant relationships (p < 0.1) between the development of SSI and other correction factors were tested for independence by multivariate logistic regression. Adjusted odds ratios were estimated by using the penalized maximum likelihood estimation method in SAS 9.1 (SAS Institute Inc., Cary, NC, USA). The significance of each factor was confirmed by penalized likelihood ratio testing. In two-tailed tests, p < 0.05 was deemed to indicate statistical significance.
Fig. 1. Representative photos of purse-string skin closure. (a) The wound after closing the peritoneum and then the fascia. (b) Purse-string skin closure after two layers were completed. (c) A scab formed on the purse-string-closed wound 3 weeks after surgery.
3. Results 3.1. Characteristics Between November, 2010 and March 2011, CL was performed in 79 patients. Thereafter, PS was performed in 78 patients. There was
Please cite this article in press as: Y.J. Suh, et al., A beneficial effect of purse-string skin closure after ileostomy takedown: A retrospective cohort study, International Journal of Surgery (2014), http://dx.doi.org/10.1016/j.ijsu.2014.04.008
66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65
IJSU1336_proof ■ 5 May 2014 ■ 3/6
Y.J. Suh et al. / International Journal of Surgery xxx (2014) 1e6
no follow-up loss for the current study. The two groups did not differ significantly in terms of baseline characteristics (Table 1). The frequencies of any comorbidity were not different between two groups: 46.8% (37/79) of the PS group and 50% (39/78) of the CL group had one or more comorbidities (p ¼ 0.750). 3.2. Outcomes All patients except for two in each group underwent stapled bowel anastomosis. Operative time, the rate of transfusion, length of hospital stay, and the rate of readmission were not significantly different between two groups (Table 2). Six and five patients in the CL and PS groups were readmitted within 30 days after discharge, respectively. Of the six readmitted patients in the CL group, two had constipation, three had small bowel obstruction, and one had a wound problem that required reoperation. Of the five readmitted patients in the PS group, one had diarrhea, two had constipation, one had small bowel obstruction, and one was admitted for supportive care. The PS group had a significantly lower overall complication rate than the CL group (p ¼ 0.010). However, there was no difference in complications other than SSI. The PS group had a significantly lower SSI rate (0% vs. 11.39%, p ¼ 0.003). The patients who developed SSI spent, on average, 1 week longer in hospital than the patients who did not develop SSI (18.00 7.12 days vs. 11.25 6.15 days, p ¼ 0.002). SSI extended the time for healing by up to 19.74 days in the CL group (p ¼ 0.021). Both groups exhibited a sharp increase in pain on the day of surgery that then dropped over time. As time passed, the NRS pain scores of both groups dropped significantly (p < 0.001; Fig. 2). However, the two
Table 1 Demographic characteristics. Characteristic
CL (n ¼ 79)
Age, years 60.13 10.53 Gender, male 49 (62.03) 2 Body mass index, kg/m 22.59 3.06 ASA score 1.68 0.49 1 26 (32.91) 2 52 (65.82) 3 1 (1.27) Alcohol Yes 50 (63.29) No 29 (36.71) Smoking Yes 38 (48.10) No 41 (51.90) Chemotherapy Yes 62 (78.48) No 17 (21.52) Radiation Yes 50 (63.29) No 29 (36.71) Past abdominal operation history Yes 22 (27.85) No 57 (72.15) Time from ileostomy creation 189.58 102.21 to closure, days Co-morbidity 37 (46.84) Diabetes mellitus 9 (11.39) Hypertension 23 (29.11) Chronic liver disease 6 (7.59) Tuberculosis 6 (7.59) a Others 15 (18.99) The reasons for ileostomy formation Low anastomosis 76 (96.20) Rectovaginal fistula 1 (1.27) Anastomotic leakage 2 (2.53)
PS (n ¼ 78)
p-value
59.53 10.74 50 (64.10) 22.5 2.83 1.58 0.57 36 (46.15) 39 (50.00) 3 (3.85)
0.724 0.869 0.791 0.155 0.104 0.053 0.367 0.516
45 (57.69) 33 (42.31) 0.874 39 (50.00) 39 (50.00) 0.708 59 (75.64) 19 (24.36) 0.870 48 (61.54) 30 (38.46) 0.461 17 (21.79) 61 (78.21) 215.40 149.13 0.251 39 13 24 3 3 12
(50.00) (16.67) (30.77) (3.85) (3.85) (15.38)
0.750 0.367 0.863 0.495 0.495 0.673
74 (94.87) 1 (1.28) 3 (3.85)
0.719 1.000 0.681
CL, conventional linear; PS, purse-string. a Others included asthma, arrhythmia, stroke, heart failure, COPD, and etc.
3
Table 2 Surgical characteristics, postoperative outcomes, and complication rates. Characteristic
CL (n ¼ 79)
PS (n ¼ 78)
p-value
Hand-sewn anastomosis Operative time, minutes Transfusion Length of hospital stay, days Readmission within 30 days Complication SSI Small bowel obstruction Leakage Incisional hernia Sepsis
2 (2.53) 76.7 48.2 2 (2.53) 11.08 4.14 6 (7.59) 20 (25.32) 9 (11.39)a 9 (11.39)b 1 (1.27)c 1 (1.27) 0 (0)
2 (2.56) 77.37 52.93 4 (5.13) 12.21 8.03 5 (6.41) 7 (8.97) 0 (0) 7 (8.97) 0 (0) 0 (0) 0 (0)
1.000 0.920 0.443 0.663 1.000 0.010 0.003 0.617 1.000 1.000 1.000
CL, conventional linear; PS, purse-string; SSI, surgical site infection. a 7 were classified as superficial incisional SSI, 1 as deep incisional SSI, and 1 as organ/space SSI. b 3 also experienced SSI. c 1 also experienced SSI.
groups did not differ significantly in terms of NRS pain scores on any of the days (p ¼ 0.323) or in terms of the pain pattern (p ¼ 0.548). 3.3. Factors associated with the development of SSI The overall rate of SSI was 5.7% (9/157). Univariate analyses revealed that anastomosis method and skin closure method were significantly associated with SSI (Table 3). SSI was more likely to occur after hand-sewn anastomosis than after stapled anastomosis (p ¼ 0.017). It was also significantly more likely to occur after CL than after PS (p ¼ 0.003). It tended to occur more frequently in obese patients (BMI 25) and in smokers, but these trends did not achieve statistical significance (both p ¼ 0.094). These four factors (BMI, smoking history, anastomosis method, and ileostomy skin closure method) were then examined by multivariate logistic regression analysis to determine whether they associated independently with the development of SSI. There were significantly higher odds that patients undergoing ileostomy takedown would develop SSI if CL method was used (adjusted odds ratio 26.63, 95% CI 3.02e267.70; p ¼ 0.001) and hand-sewn anastomosis was performed (adjusted odds ratio 31.16, 95% CI 1.41e607.64; p ¼ 0.029) (Table 4). While a high BMI tended to be associated with a high SSI rate, this trend only achieved marginal statistical significance (p ¼ 0.050). Smoking did not associate significantly with SSI. After stepwise selection, the ileostomy skin closure and anastomosis method variable were selected and the significance of the associations of SSI with these variables in the final model was assessed. This analysis showed significant associations between lower SSI rates and two factors: PS method (p ¼ 0.001) and stapled bowel anastomosis (p ¼ 0.003). 4. Discussion In the present study, the PS group had a significantly lower SSI rate than the CL group. PS method was introduced by Anjan Banerjee [7]. In retrospective studies, the SSI rate associated with PS method ranges from 0% to 6.7% [8,17e19]. Those studies favored PS in terms of SSI. Of these studies, one study could not show significant difference due to the small number of subject patients although PS showed comparable outcomes [18]. They reported that the incidences of wound infection were 16.7% (5/30) and 5.6% (1/ 18) in CL and PS groups, respectively. In one randomized controlled study, an interim analysis showed sufficient reductions of SSI in PS group compared with CL group (6% vs. 38.7%) [13,14]. Enrollment
Please cite this article in press as: Y.J. Suh, et al., A beneficial effect of purse-string skin closure after ileostomy takedown: A retrospective cohort study, International Journal of Surgery (2014), http://dx.doi.org/10.1016/j.ijsu.2014.04.008
66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65
IJSU1336_proof ■ 5 May 2014 ■ 4/6
4
Y.J. Suh et al. / International Journal of Surgery xxx (2014) 1e6
Fig. 2. Numeric pain rating scale before and after conventional linear skin closure and purse-string skin closure. CL, conventional linear; PS, purse-string; NRS, numerical rating scale; PREOP, preoperative; POD, postoperative day.
Table 3 Univariate analysis of eligible factors related to surgical site infection. Characteristic
Age, years 2
BMI, kg/m Gender Alcohol Smoking
Chemotherapy Radiation Co-morbidity Transfusion Anastomosis method Skin closure method
Surgical site infection
60 <60 25 <25 Male Female Yes No Yes No Yes No Yes No Yes No Yes No Hand-sewn Stapled CL PS
No (n ¼ 148)
Yes (n ¼ 9)
79 69 29 119 91 57 88 60 70 78 115 33 94 54 72 76 6 142 2 146 70 78
3 6 4 5 8 1 7 2 7 2 6 3 4 5 4 5 0 9 2 7 9 0
(53.38) (46.62) (19.59) (80.41) (61.49) (38.51) (59.46) (40.54) (47.30) (52.70) (77.70) (22.30) (63.51) (36.49) (48.65) (51.35) (4.05) (95.95) (1.35) (98.65) (47.30) (52.70)
(33.33) (66.67) (44.44) (55.56) (88.89) (11.11) (77.78) (22.22) (77.78) (22.22) (66.67) (33.33) (44.44) (55.56) (44.44) (55.56) (0) (100) (22.22) (77.78) (100) (0)
p-value
0.312 0.094 0.156 0.484 0.094 0.429 0.298 1.000 1.000 0.017 0.003
BMI, body mass index; CL, conventional linear; PS, purse-string.
was discontinued by reason of the result. The recent randomized controlled trials also showed that PS group resulted in lowering of SSI compared with CL group (0% vs. 24%e36.6%) [15,16]. A drain hole in the middle left by PS as basically a method of secondary
Table 4 Multivariate analysis of eligible factors related to surgical site infection. Characteristic
Adjusted OR
95% C.I.
p-value
Smoking (Yes vs. No) BMI (25 vs. <25 kg/m2) Skin closure method (CL vs. PS) Anastomosis method (Hand-sewn vs. stapled)
4.11 5.43 26.63 31.16
0.83e28.86 1.00e32.02 3.02e267.70 1.41e607.64
0.086 0.050 0.001 0.029
OR, odds ratio; C.I., confidence interval; BMI, body mass index; CL, conventional linear; PS, purse-string.
wound healing, may contribute to a lower SSI rate after ileostomy takedown by draining wound discharges. The SSI rate after ileostomy takedown has been reported variously in a few studies. Herwig Porkorny et al. [11] reported that 9% of patients who underwent stoma closure developed SSI. The study of Harold et al. [12] revealed a similar rate while GarciaBotello et al. [5] reported SSI rates up to 18.3%. These studies indicate that along with small bowel obstruction, SSI is a major complication of ileostomy reversal [1,11]. In the present study, the overall SSI rate was relatively low (5.7%). This result may come from the introduction of PS. And this study indicated that PS gave postoperative benefits after ileostomy reversal in daily routine practice. Despite this advantage, PS gave concerns about patients’ discomfort, cosmetic aspect, postoperative pain, wound care, and etc. However, patients in PS group reported greater satisfaction of cosmetic outcome in one systemic review [20]. Similarly, Daniel Camacho-Mauries et al. [16] showed superior patient satisfaction levels and better cosmetic outcomes in their study. Milanchi et al. [19] documented no significant difference although PS group showed higher patient satisfaction score in subjective factors: cosmesis, expectation before surgery, postoperative pain, wound care, and activity. In general accord with their studies, two groups did not differ significantly in terms of NRS pain scores in the present study. The two groups in the present study were homogeneous, which allowed the SSI rates of unselected patients who underwent elective ileostomy reversal to be compared directly. Moreover, unlike in the other studies described above including recent randomized controlled trials, the patients underwent stitching with a double layer of absorbable sutures that did not have to be removed later [15,16]. The SSI rate in secondary closure or delayed primary closure varies differently from 0 to 15% [2,4,6,12,21e24]. Ileostomy takedown is more likely to be classified as clean-contaminated category [2]. The wound to be suspected of contamination would rather be left open logically. The burden of interval closure cannot be negligible in delayed primary closure. In this point of view, secondary closure appears to be an adjunct to the techniques designed to decrease the SSI rate although wound healing is expected to be delayed. PS allows partial wound edge apposition [25]. This apposition may accelerate healing compared to secondary closure.
Please cite this article in press as: Y.J. Suh, et al., A beneficial effect of purse-string skin closure after ileostomy takedown: A retrospective cohort study, International Journal of Surgery (2014), http://dx.doi.org/10.1016/j.ijsu.2014.04.008
66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65
IJSU1336_proof ■ 5 May 2014 ■ 5/6
Y.J. Suh et al. / International Journal of Surgery xxx (2014) 1e6
Analysis of the complications in the CL and PS groups in the present study revealed that the PS group developed significantly fewer complications overall (p ¼ 0.004). According to the Claviene Dindo Classification, seven patients in CL group experienced more than grade II while one in PS group did [26,27]. Considering that SSI accounts for a large percentage of complication after ileostomy takedown, PS may decrease SSI-related morbidity in ileostomy takedown. While most patients underwent routine side-to-side double stapling for anastomosis, the hand-sewn anastomosis method was used in four patients (two from each group). Two of these four patients developed SSI and the multivariate analysis showed that the hand-sewn anastomosis method associated independently with SSI (adjusted odds ratio 31.16, 95% CI 1.41e607.64; p ¼ 0.029). This result is inconsistent with a recent meta-analysis [28]. It is because patients with hand-sewn anastomosis were mostly difficult cases for ileostomy takedown. It should be also emphasized that there were only four cases of hand-sewn anastomosis, and thus the association between hand-sewn anastomosis and SSI should be viewed with caution. In the present study, patients with BMI 25 had a higher SSI rate than patients with BMI <25 and multivariate analysis excluding all other major factors showed that BMI associated marginally significantly with SSI (adjusted odds ratio 5.43, 95% CI 1e32.02; p ¼ 0.050). This may reflect the fact that there were only four patients with BMI 25, which in turn may reflect the low BMI of the Korean population compared to Western populations. A larger sample size may show a significant and independent association between BMI and SSI. If this is the case, it may be advisable to perform PS after ileostomy takedown in patients with BMI 25. Notably, the large-scale retrospective study conducted by Riou et al. [29] revealed that obesity had an adverse effect on wound healing. The hospital stay of both groups in the present study was a few days longer than in the other studies, because Korean healthcare system depends much on hospitals and not enough on communitybased medical practices, and therefore much of their hospital cost is covered by the national health insurance. When 16 patients of CL group who had JP wound drain bags were compared with 63 patients of CL group who did not, no significant difference in SSI rate was found (p ¼ 0.443) as N. Mirbagheri et al. [25] showed in their study. Because JP wound drain bag was removed just before discharge, the JP insertion group was under the same condition as the non-insertion group in terms of the SSI development during the period between discharge and outpatient visit. Accordingly, no significant difference was probably shown. In addition, selection bias of the surgeons might be involved because patients who were thought to have a high risk of developing SSI selectively underwent JP insertion. Therefore, the protective effect of JP insertion against SSI could not be determined in the present study. The present study suffers from the usual limitations of observational studies. In particular, the CL and PS groups were not compared at the same time. This means that patient randomization could not be performed, although it should be noted that the two groups largely experienced identical conditions apart from the method of ileostomy skin closure. Another limitation of this study was the outcome evaluation which was not performed by blinded observers. 5. Conclusion PS had a beneficial effect on SSI in patients who underwent ileostomy takedown in daily practice. PS can be considered as standard of care for wound closure after ileostomy takedown in daily routine practice.
5
Ethical approval This study was approved by the institutional review board of National Cancer Center of Korea (NCCNCS-11-545). Funding None. Conflict of interest The authors declare no conflict of interest. Author contribution Study conception and design: YJ Suh, JW Park. Acquisition of data: YJ Suh, YS Kim, SC Park, JW Park, JH Oh. Analysis and interpretation of data: YJ Suh, JW Park. Drafting of the manuscript: YJ Suh, JW Park. Critical revision: YJ Suh, YS Kim, SC Park, JW Park, JH Oh. Acknowledgments All statistical analyses were supported by the Medical Research Collaborating Center, Seoul National University and Seoul National University Hospital, Seoul, Korea. References [1] O. Kaidar-Person, B. Person, S.D. Wexner, Complications of construction and closure of temporary loop ileostomy, J. Am. Coll. Surg. 201 (2005) 759e773. [2] D.J. Hackam, O.D. Rotstein, Stoma closure and wound infection: an evaluation of risk factors, Can. J. Surg. 38 (1995) 144e148. [3] A. Barbul, When will the war end? Wound Repair Regen. 21 (2013) 179. [4] P.T. Phang, J.M. Hain, J.J. Perez-Ramirez, R.D. Madoff, B.T. Gemlo, Techniques and complications of ileostomy takedown, Am. J. Surg. 177 (1999) 463e466. [5] S.A. Garcia-Botello, J. Garcia-Armengol, E. Garcia-Granero, et al., A prospective audit of the complications of loop ileostomy construction and takedown, Dig. Surg. 21 (2004) 440e446. [6] E.D. Verrier, K.J. Bossart, F.W. Heer, Reduction of infection rates in abdominal incisions by delayed wound closure techniques, Am. J. Surg. 138 (1979) 22e 28. [7] A. Banerjee, Pursestring skin closure after stoma reversal, Dis. Colon Rectum 40 (1997) 993e994. [8] C.D. Sutton, N. Williams, L.J. Marshall, G. Lloyd, W.M. Thomas, A technique for wound closure that minimizes sepsis after stoma closure, ANZ J. Surg. 72 (2002) 766e767. [9] J.T. Lim, S.M. Shedda, I.P. Hayes, “Gunsight” skin incision and closure technique for stoma reversal, Dis. Colon Rectum 53 (2010) 1569e1575. [10] A.J. Mangram, T.C. Horan, M.L. Pearson, L.C. Silver, W.R. Jarvis, Guideline for Prevention of Surgical Site Infection, Centers for Disease Control and Prevention (CDC) Hospital Infection Control Practices Advisory Committee, Am. J. Infect. Control 1999 (27) (1999) 97e132. [11] H. Pokorny, H. Herkner, R. Jakesz, F. Herbst, Mortality and complications after stoma closure, Arch. Surg. 140 (2005) 956e960. [12] D.M. Harold, E.K. Johnson, J.A. Rizzo, S.R. Steele, Primary closure of stoma site wounds after ostomy takedown, Am. J. Surg. 199 (2010) 621e624. [13] K. Reid, P. Pockney, T. Pollitt, B. Draganic, S.R. Smith, Randomized clinical trial of short-term outcomes following purse-string versus conventional closure of ileostomy wounds, Br. J. Surg. 97 (2010) 1511e1517. [14] A. Moreton, S. Shankar, S. Jones, Randomized clinical trial of short-term outcomes following purse-string versus conventional closure of ileostomy wounds (Br. J. Surg. 2010; 97: 1511e1517), Br. J. Surg. 98 (2011) 458. [15] N. Dusch, D. Goranova, F. Herrle, M. Niedergethmann, P. Kienle, Randomized controlled trial: comparison of two surgical techniques for closing the wound following ileostomy closure: purse string vs direct suture, Colorectal Dis. 15 (2013) 1033e1040. [16] D. Camacho-Mauries, J.L. Rodriguez-Diaz, N. Salgado-Nesme, Q.H. Qonzalez, O. Vergara-Fernandez, Randomized clinical trial of intestinal ostomy takedown comparing pursestring wound closure vs conventional closure to eliminate the risk of wound infection, Dis. Colon Rectum 56 (2013) 205e211. [17] C.D. Klink, M. Wünschmann, M. Binnebösel, et al., Influence of skin closure technique on surgical site infection after loop ileostomy reversal: retrospective cohort study, Int. J. Surg. 11 (2013) 1123e1125.
Please cite this article in press as: Y.J. Suh, et al., A beneficial effect of purse-string skin closure after ileostomy takedown: A retrospective cohort study, International Journal of Surgery (2014), http://dx.doi.org/10.1016/j.ijsu.2014.04.008
66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130
1 2 3 4 5 6 7 8 9 10 11 12 13 14
IJSU1336_proof ■ 5 May 2014 ■ 6/6
6
Y.J. Suh et al. / International Journal of Surgery xxx (2014) 1e6
[18] J.R. Lee, Y.W. Kim, J.J. Sung, et al., Conventional linear versus purse-string skin closure after loop ileostomy reversal: comparison of wound infection rates and operative outcomes, J. Korean Soc. Coloproctol. 27 (2011) 58e63. [19] S. Milanchi, Y. Nasseri, T. Kidner, P. Fleshner, Wound infection after ileostomy closure can be eliminated by circumferential subcuticular wound approximation, Dis. Colon Rectum 52 (2009) 469e474. [20] D.P. McCartan, J.P. Burke, S.R. Walsh, J.C. Coffey, Purse-string approximation is superior to primary skin closure following stoma reversal: a systematic review and meta-analysis, Tech. Coloproctol. 17 (2013) 345e351. [21] S.D. Wexner, D.A. Taranow, O.B. Johansen, et al., Loop ileostomy is a safe option for fecal diversion, Dis. Colon Rectum 36 (1993) 349e354. [22] K.S. Wong, F.H. Remzi, E. Gorgun, et al., Loop ileostomy closure after restorative proctocolectomy: outcome in 1,504 patients, Dis. Colon Rectum 48 (2005) 243e250. [23] G. Lahat, H. Tulchinsky, G. Goldman, J.M. Klauzner, M. Rabau, Wound infection after ileostomy closure: a prospective randomized study comparing primary vs. delayed primary closure techniques, Tech. Coloproctol. 9 (2005) 206e208.
[24] N. Vermulst, J. Vermeulen, E.J. Hazebroek, P.P.L.O. Coene, E. van der Harst, Primary closure of the skin after stoma closure. Management of wound infections is easy without (long-term) complications, Dig. Surg. 23 (2006) 255e 258. [25] N. Mirbagheri, J. Dark, S. Skinner, Factors predicting stomal wound closure infection rates, Tech. Coloproctol. 17 (2013) 215e220. [26] D. Dindo, N. Demartines, P.A. Clavien, Classification of surgical complications: a new proposal with evaluation in a cohort of 6336 patients and results of a survey, Ann. Surg. 240 (2004) 205e213. [27] P.A. Clavien, J. Barkun, M.L. de Oliveira, et al., The Clavien-Dindo classification of surgical complications: five-year experience, Ann. Surg. 250 (2009) 187e 196. [28] J. Gong, Z. Guo, Y. Li, et al., Stapled versus hand suture closure of loop ileostomy: a meta-analysis, Colorectal Dis. 10 (2013) e561ee568. [29] J.P. Riou, J.R. Cohen, H. Johnson Jr., Factors influencing wound dehiscence, Am. J. Surg. 163 (1992) 324e330.
Please cite this article in press as: Y.J. Suh, et al., A beneficial effect of purse-string skin closure after ileostomy takedown: A retrospective cohort study, International Journal of Surgery (2014), http://dx.doi.org/10.1016/j.ijsu.2014.04.008
15 16 17 18 19 20 21 22 23 24 25 26 27 28