{"paper_id":"7d83d8ff-990f-4e36-abfa-ca0a000a1ebc","body_text":"Various types of incisions used in abdominal operations are an important source of post-operative morbidities such as pain, surgical site infections, scarring, tumour implantation and incisional hernia [ 1 ]. One of the objectives of laparoscopic- or other minimally invasive surgical approaches is to minimise incision-related complications and to improve post-operative outcomes. Laparoscopic colorectal surgery, as compared with open surgery, has been reported to improve short-term and long-term outcomes in patients suffering from various colorectal disorders. Laparoscopic colorectal resection (LCR) has therefore become a preferred technique for treating both benign and malignant conditions of the colon and rectum [ 2–7 ]. After a successful dissection in laparoscopic colorectal surgery, the enlargement of a trocar incision, resulting in ‘minilaparotomy', is invariably necessary for two major reasons; firstly for intestinal anastomosis, to maintain the continuity of the gastro-intestinal tract, and secondly for the purpose of retrieval of the specimen. The extension of a port-site incision causes more tissue trauma than one would expect from a smaller port wound and thus potentially reduces the aforementioned advantages of LCR [ 8 ]. This poses a special challenge to operating surgeons, due to the size of the specimen and the desire to keep the retrieval incision as small as possible to retain the benefits of laparoscopic surgery. In addition, the potential problems of dissemination of tumour cells, implantation of tumour cells in the wound; metastasis and wound contamination must be kept in mind during the process of specimen retrieval [ 9 ]. A number of solutions have been reported, for the purpose of avoiding minilaparotomy altogether or placing another incision away from the port incisions to retrieve the specimen. These include transverse incision in the left iliac fossa, transverse incision in the right iliac fossa, McBurney’s incision, extension of the umbilical port incision in midline, and stoma site incision. Additionally, specimen retrieval through natural orifices—such as through the anus or vagina—has also been reported as a relatively preferable solution.\nThe objective of this article is to review the various specimen retrieval techniques reported in the medical literature during LCR and generate a summary conclusion based on the level of evidence available.\n\nAll the published articles on specimen retrieval techniques during laparoscopic colorectal resection were identified through searches of the Medline, Embase, CINAHL, Cochrane library and Pubmed databases. The search terms “colorectal surgery”, “laparoscopic”, “minimal invasive surgery”, “natural orifice retrieval”, “trocar incision”, “midline incision”, “periumbilical incision”, “Pfannensteil incision” and “transverse incision” were used alone and in various combinations. Relevant articles referenced in these publications were also downloaded from databases. The ‘related article' function was used to widen the search results. All abstracts, case reports, case series and published single-centre or multi-centre studies were retrieved and searched comprehensively.\nFor inclusion in the literature review, a study had to meet the following criteria: (i) randomized, controlled trial, case controlled trial, cohort studies, all types of comparative studies, case series and case reports, (ii) laparoscopic colorectal resections for both benign and malignant conditions, (iii) evaluation of surgical site infection rate, and (iv) trials in patients undergoing any kind of surgery.\nUsing a predefined data format, two independent reviewers (MSS and MIB) extracted data from each study, which resulted in high and satisfactory interobserver agreement. Information collected included the name(s) of the author(s), title of the study, journal in which the study was published, country and year of the study, treatment regimen, length of the therapy, method by which specimens were retrieved, testing sample size (with sex differentiation if applicable) and the number of patients receiving each regimen. Within each arm in case of comparative study, the reviewers noted the number of patients who responded to- and the number of patients who failed to respond to treatment, the patient compliance rate in each group, the number of patients reporting complications and the number of patients with absence of complications. After completing the data extraction, the two independent reviewers discussed the results and, if discrepancies were present, a consensus was reached.\nWhere applicable, the RevMan 5.2 software package [ 10 ,  11 ], provided by the Cochrane Collaboration, was used for the statistical analysis to achieve a combined outcome. The odds ratio (OR) with a 95% confidence interval (CI) was calculated for binary data. The random- and fixed-effects models (where applicable) were used to calculate the combined outcomes of both binary and continuous variables [ 12 ,  13 ]. If the standard deviation was not available, then it was calculated according to the Cochrane Collaboration's guidelines [ 10 ]. This process involved assumptions that both groups had the same variance—which may not have been true—and variance was either estimated from the range or from the  P -value. The estimate of the difference between the two techniques was pooled, depending upon the effect weights in results determined by each trial estimate variance. A forest plot was used for the graphical display of the results. The square around the estimate stood for the accuracy of the estimation (sample size), and the horizontal line represented the 95% CI.\n\nThere were 43 studies reporting various approaches used for the purpose of specimen retrieval in 2388 patients undergoing LCR [ 14–56 ]. These approaches can be categorized as transvaginal, transanal, transverse incision in the right or left iliac fossa, periumbilical midline incision, Pfannensteil incision and approach through the stoma site. The literature search strategy and methodology is given in  Figure 1 .\n Figure 1. PRISMA flow diagram.\nPRISMA flow diagram.\nThirteen studies [ 14–26 ] on 143 patients reported on the use of transvaginal approach to retrieve the specimen after LCR. These included five case reports [ 16–18 ,  23 ,  26 ] and eight case series [ 14 ,  15 ,  19–22 ,  24 ,  25 ]. There was no reported incision site herniation or tumour recurrence among these 143 patients. Overall, there were eight patients (5.5%) with various complications including surgical site infection (SSI) ( Table 2 ).\n Table 1. Characteristics and variables of articles reporting transvaginal specimen retrieval in patients undergoing laparoscopic colorectal surgery Trials Year Country Study type Patient number Age (years) Surgery details Follow up (months) Incisional hernia Complications/ incision site infection Recurrence Use of wound protector Awad  et al.  [ 14 ] 2011 USA Case series 14 62 (range 50–80) Right hemicolectomy for both benign and malignant conditions 17.8 (range 8–32) 0 0 0 Hubert bag Boni  et al.  [ 15 ] 2007 Italy Case series 11 45 ± 12 Rectal resection for benign conditions 4 ± 2 0 0 0 Standard vaginal extractor with endobag Dozois et [ 16 ] 2008 USA Case report 1 53 Hysterectomy 1 0 0 0 No Salpingo-ophorectomy Total colectomy Franklin  et al.  [ 17 ] 2008 USA Case report 1 88 Right hemicolectomy for caecal carcinoma 1 week 0 0 0 Specimen bag García–Flórez  et al.  [ 18 ] 2010 Spain Case report 1 86 Anterior resection with en bloc salpingo-ophorectomy for sigmoid carcinoma 3 0 0 0 Plastic retractor Ghezzi  et al.  [ 19 ] 2007 Italy Case series 33 33.4 (range 25–43) Rectosigmoid resection for endometriosis 13 (range 3–27) 0 1 case of pelvic seroma 0 Retrieval bag McKenzie  et al.  [ 20 ] 2010 USA Case series 4 74 (range 68–81) Right hemicolectomy for both benign and malignant conditions 3 1 1 case of internal hernia not related to incision 0 Specimen bag Palanivelu  et al.  [ 21 ] 2008 India Case series 7 49.5 (range 34–65) Proctocolectomy for colorectal cancer and familial adenomatosis coli 12 0 1 case of pouchitis 0 Endobag 1 case of anastomotic leak Park  et al.  [ 22 ] 2010 South Korea Case series 14 66 (range 44–74) Colectomy for colorectal carcinoma 34 0 0 1 case of distant metastasis Plastic bag Sanchez  et al.  [ 23 ] 2009 USA Case report 1 63 Sigmoid colectomy for rectal prolapse 3 0 0 0 No Tarantino  et al.  [ 24 ] 2011 Switzerland Case series 34 64 (range 35–88) Anterior resections for diverticular disease 6 0 1 case of wound dehiscence 0 Ring wound protector 1 case of colpitis 1 case of anastomotic leak 2 cases of wound haematoma Torres  et al.  [ 25 ] 2012 Argentina Case series 21 50 (range 32–69) Colectomy for both benign and malignant conditions 34 0 0 0 Alexis wound retractor Wilson  et al.  [ 26 ] 2007 UK Case report 1 84 Right hemicolectomy for hepatic flexure carcinoma 1 0 0 0 Hubert bag \n Table 2. Characteristics and variables of articles reporting transanal specimen retrieval in patients undergoing laparoscopic colorectal surgery Trials Year Country Study type Patient number Age (years) Surgery details Follow up (months) Incisional hernia Complications/incision site infection Recurrence Use of wound protector Akamatsu  et al.  [ 27 ] 2009 Japan Case series 16 n/a Anterior resection for sigmoid carcinoma 2–15 0 0 0 No Awad  et al.  [ 28 ] 2012 USA Case report 1 27 Colonic resections for benign disorders 1 0 0 0 No Cheung  et al.  [ 29 ] 2009 China Case series 10 66 (range 55–81) Left colonic resections for carcinoma 1 0 0 0 TEO device Co  et al.  [ 30 ] 2010 China Case report 1 80 Left colonic resection for carcinoma 1 0 0 0 TEO device Franklin  et al.  [ 31 ] 2012 USA Case series 179 66.9 ± 14.4 Anterior resection for rectal cancer 24 0 3 cases of anastomotic leakage 9 Plastic bag 3 cases of anal stenoses Fuchs  et al.  [ 32 ] 2012 Germany Case series 14 61 (range 28–86) Colonic resection for benign conditions 6 0 0 0 TEO device Hara  et al.  [ 33 ] 2011 Japan Case series 8 71 (range 48–75) Anterior resection for rectal carcinoma 1 0 0 0 No Knol  et al.  [ 34 ] 2009 Belgium Case report 1 20 Rectal resection for benign condition 1 0 0 0 Novymed proctoscope Lacy  et al.  [ 35 ] 2012 Spain Case report 1 36 Colectomy for ulcerative colitis 1 0 0 0 Endo Catch II Leroy  et al.  [ 36 ] 2011 France Case series 16 61.2 Anterior resection for diverticular disease 1 0 0 0 No Makris  et al.  [ 37 ] 2012 USA Case report 1 n/a n/a n/a n/a n/a n/a No Nishimura  et al.  [ 38 ] 2011 Japan Case series 18 46–84 Anterior resection for colorectal cancer 5–20 0 1 case of anastomotic leakage 0 Alexis wound retractor 1 case of umbilical port infection Ooi  et al.  [ 39 ] 2009 Singapore Case report 1 51 Anterior resection for rectal cancer 1 0 0 0 No Saad  et al.  [ 40 ] 2011 Germany Case series 15 61 (range 46–76) Anterior resection for both benign and neoplastic conditions 1 0 0 0 TEO device Saad  et al.  [ 41 ] 2010 Germany Case series 8 n/a Anterior resection for both benign and neoplastic conditions 1 0 0 0 McCartney tube Wolthius  et al.  [ 42 ] 2011 Belgium Case series 21 41 (34–66) Anterior resection for both benign and neoplastic conditions 3.6 0 1 case of anastomotic leakage 0 Specimen retrieval pouch n/a = not available, TEO = transanal endoscopic operation.\nCharacteristics and variables of articles reporting transvaginal specimen retrieval in patients undergoing laparoscopic colorectal surgery\nCharacteristics and variables of articles reporting transanal specimen retrieval in patients undergoing laparoscopic colorectal surgery\nn/a = not available, TEO = transanal endoscopic operation.\nA transanal approach to retrieve the specimen after LCR was reported in sixteen case series [ 27–42 ] recruiting 311 patients. Nine patients (2.9%) developed various complications, set out in  Table 2 . The risk of developing post-operative complications was higher following a transvaginal approach than with a transanal approach (OR 0.50; 95% CI 0.19–0.33).\nJones  et al.  [ 43 ] published data on 500 patients undergoing LCR for diverticular disease, where the specimen was retrieved through a transverse incision in the left iliac fossa ( Table 3 ). Risk of incisional hernia was 0.4%; that for SSI was 1.2% and risk of reported anastomotic leak was 1.4%.\n Table 3. Characteristics of studies reporting various other approaches of specimen retrieval in patients undergoing laparoscopic colorectal surgery Trials Year Country Study type Approach of specimen retrieval Patient number Age (years) Follow up (months) Surgery details Incisional hernia Complications/ incision site infection Recurrence Use of wound protector Jones  et al.  [ 43 ] 2008 Australia Case series Transverse incision in the left iliac fossa 500 58 n/a All types of colonic resection for diverticular disease 2 7 cases of anastomotic leakage 0 No 6 cases of wound infections Casciola  et al.  [ 44 ] 2008 Italy Case series Periumbilical midline incision 352 n/a 6 69 splenectomies 1 3 n/a Endobag 138 right hemicolectomies 115 gastric resections López-Köstner  et al.  [ 45 ] 2008 Chile Case series Periumbilical midline incision 106 54 27 Sigmoid colectomy for diverticular disease 3 3 1 No Wilhelm  et al.  [ 46 ] 2006 Germany Case series Pfannensteil incision 100 58 19 Sigmoid colectomy for diverticular disease 1 11 1 diverticulitis n/a Sahakitrungruang  et al.  [ 47 ] 2008 Thailand Case series Pfannensteil incision 7 na 1 Proctocolectomy for ulcerative colitis 0 0 0 n/a n/a = not available.\nCharacteristics of studies reporting various other approaches of specimen retrieval in patients undergoing laparoscopic colorectal surgery\nn/a = not available.\nTwo studies [ 44 ,  45 ] reported data on 458 patients undergoing colorectal and upper gastro-intestinal surgical resections where the specimen was retrieved through a periumbilical (extended port side wound) midline incision ( Table 3 ). There were four cases of incisional hernia (0.87%), six cases of SSI (1.3%) and one case of distant recurrence. The risk of developing incisional hernia was greater in cases of periumbilical midline incision than with left iliac fossa transverse incision (OR 2.19; 95% CI 0.40–12.3).\nTwo articles reported on the use of Pfannensteil incision for specimen retrieval in 100 patients undergoing laparoscopic sigmoid colectomy for diverticular disease [ 46 ], and in seven patients undergoing laparoscopic panproctocolectomy for ulcerative colitis [ 47 ] ( Table 3 ). There was only one case of incisional hernia (0.93%) and 11 cases of SSI in this case series [ 46 ].\nThis approach was published in two comparative studies [ 48 ,  49 ], in which 68 patients underwent LCR for benign colorectal conditions ( Table 4 ). Statistically, the duration of operation [Standardized Mean Difference (SMD) 1.81; 95% CI -1.99–5.62;  Z =  0.93;  P =  0.35] and risk of incisional hernia (OR 6.46; 95% CI 0.24–174.08;  Z =  1.11;  P =  0.27) were similar (SMD 1.81; 95% CI -1.99–5.62;  Z =  0.93;  P =  0.35) in both approaches. However, the transanal approach was associated with higher risk of SSI (OR 17.40; 95% CI 1.50–202.47;  Z =  2.28;  P =  0.02) ( Figure 1 ).\n Table 4. Characteristics and variables of studies reporting comparisions between various approaches of specimen retrieval in patients undergoing laparoscopic colorectal surgery Trials Year Country Study type Approach of specimen retrieval Patient number Age (years) Surgery details Operation time (minutes) Follow-up (Months) Hernia Infection/ complications Recurrence Christoforidis  et al.  [ 48 ] 2012 Switzerland Case control Transanal 10 47 (range 26–62) Left colonic resections for benign disease 200 ± 60 n/a 1 0 0 Periubmilical midline incision 20 56 (range 38–81) 205.5 ± 49 0 0 0 Eshuis  et al.  [ 49 ] 2010 Netherlands Case control Transanal 8 31 (range 19–61) Ileocolic resections for inflammatory bowel disease 208 ± 45.1 3 0 3 0 Periubmilical midline incision 30 115 ± 15.1 0 1 0 Lee  et al.  [ 50 ] 2012 Canada Case control Periumbilical midline incision 68 63.0 All types of colorectal resections for malignant lesions of the colorectum n/a 37 20 n/a n/a Pfannensteil 24 65.8 0 De Souza  et al.  [ 51 ] 2010 USA Case control Periumbilical midline incision 231 62.68 All types of colorectal resections for both benigh and malignant lesions of the colorectum n/a 17.5 56 n/a n/a Pfannensteil 139 61.32 0 Lim  et al.  [ 52 ] 2012 Korea Case control Periumbilical midline incision 92 63 Left colonic resections for colorectal cancer 164.5 ± 8.6 20 2 12 n/a LIF transverse 55 66 167.4 ± 8.6 0 7 Lee  et al.  [ 50 ] 2012 Canada Case control Periumbilical midline incision 68 63.0 All types of colorectal resections for malignant lesions of the colorectum n/a 37 20 n/a n/a LIF/RIF transverse 7 60.8 1 Wolthuis  et al.  [ 53 ] 2011 Belgium Case control Periumbilical midline incision 21 35 (range 30–38) Rectal resection for benign conditions 90 ± 5 10.3 0 1 0 LIF transverse 21 34 (range 32–35) 105 ± 6.4 18.7 0 5 0 Gardenbroek  et al.  [ 54 ] 2012 Netherlands Case series Stoma site 3 21.5 Subtotal colectomy for inflammatory bowel disease 219 7 n/a 0 n/a Transanal 7 0 Choi  et al.  [ 55 ] 2009 Korea Case series Transanal 11 53.6 ± 12.8 Robot-assisted laparosocopic anterior resection for rectal and sigmoid carcinoma 260.8 ± 62.9 1 0 1 leak 0 Transvaginal 2 260.8 ± 62.9 0 1 bleed 0 Costantino  et al.  [ 56 ] 2012 France Case control Transanal 29 60.1 Left-sided colorectal resections for diverticular disease 122 ± 25.1 1 0 2 0 Pfannensteil 23 59.5 105 ± 25.1 0 1 0 LIF = left iliac fossa, n/a = not available, RIF = right iliac fossa.\nCharacteristics and variables of studies reporting comparisions between various approaches of specimen retrieval in patients undergoing laparoscopic colorectal surgery\nLIF = left iliac fossa, n/a = not available, RIF = right iliac fossa.\nTwo studies reported the comparison between Pfannensteil  vs.  periumbilical midline incision approach to retrieve the specimen in 462 patients undergoing LCR for both benign and malignant conditions [ 50 ,  51 ] ( Table 4 ). The risk of developing incisional hernia was significantly higher (OR 53.72; 95% CI 7.48–386.04;  Z =  3.96;  P =  0.0001) following periumbilical midline incision compared to Pfannensteil incision ( Figure 2 ).\n Figure 2. Forest plot for surgical site infection following the use of transanal  vs.  periumbilical midline incision for specimen retrieval in patients undergoing laparoscopic colorectal resections. Odds ratios are shown with 95% confidence intervals.\nForest plot for surgical site infection following the use of transanal  vs.  periumbilical midline incision for specimen retrieval in patients undergoing laparoscopic colorectal resections. Odds ratios are shown with 95% confidence intervals.\nTwo studies published data comparing the use of periumbilical midline incision against transverse incision in the right or left iliac fossa in 222 patients ( Table 4 ). The risk of developing incisional hernia was significantly higher (OR 0.37; 95% CI 0.06–2.20;  Z =  1.09;  P =  0.028) following periumbilical midline incision than after transverse incision but there was no difference in the risk of SSI between the two approaches to specimen retrieval.\n Figure 3. Forest plot for incisional hernia following the use of periumbilical midline  vs.  Pfannensteil incision for specimen retrieval in patients undergoing laparoscopic colorectal resections. Odds ratios are shown with 95% confidence intervals.\nForest plot for incisional hernia following the use of periumbilical midline  vs.  Pfannensteil incision for specimen retrieval in patients undergoing laparoscopic colorectal resections. Odds ratios are shown with 95% confidence intervals.\nThere was higher risk of developing SSI in the transanal approach than in the transverse incision approach [ 53 ] for specimen retrieval ( Table 4 ). One study [ 54 ] reported specimen retrieval through the stoma site in comparison with a transanal approach, and reported no difference in SSI. One study on 13 patients reported a comparison between transvaginal and transanal approaches [ 55 ]; there were no cases of hernia or recurrence in this study. Statistically, the complication rate and duration of operation were similar in both techniques ( P =  1.0). The transanal approach was compared against Pfannensteil in a study of 52 patients undergoing left-sided laparoscopic colonic resection for diverticular disease [ 56 ]. The transanal approach was associated with slightly higher risk of SSI but operative time and incidence of incisional hernia were similar.\n\nColorectal surgeons employ numerous approaches to retrieve specimens following LCR. The most common of these are periumbilical midline incision (1260 reported case in the literature), transverse incision (583 reported cases in the literature) in the right or left iliac fossa depending upon the side of colonic resection and Pfannensteil incision (293 reported cases in the literature). Periumbilical midline incision is associated with the highest risk of developing incisional hernia. There is no difference between these three common approaches to specimen retrieval, in terms of SSI. Transanal and transvaginal approaches are associated with higher risk of SSI. This conclusion is based on the combined findings of published case series, case reports and comparative studies. It may therefore be considered biased, less reliable and weaker. Randomized, controlled trials with longer follow-up are required to achieve reliable evidence before recommending the routine use of any approach for specimen retrieval in patients undergoing LCR.\nConflict of interest:  none declared.","source_license":"CC-BY-4.0","license_restricted":false}