- Email: [email protected]
Comparison of Different Extraction Sites Used During Laparoscopic Radical Nephrectomy
Comparison of Different Extraction Sites Used During Laparoscopic Radical Nephrectomy Vincent G. Bird,* Jason K. Au, Yekutiel Sandman, Rosely De Los Santos, Rajnikanth Ayyathurai and John M. Shields From the Department of Urology, Miller School of Medicine, University of Miami, Miami, Florida
Purpose: Laparoscopic radical nephrectomy is commonly performed for renal tumors that are not amenable to nephron sparing treatment. A number of techniques for intact specimen extraction are used. The development of incisional hernias from the extraction site is a known but infrequent delayed complication. We analyzed different extraction sites and risk factors for such hernias. Materials and Methods: We retrospectively analyzed a cohort of patients undergoing laparoscopic radical nephrectomy with intact specimen extraction through 3 sites. Patients and operation specific parameters were included with particular attention to factors predisposing patients to incisional hernia, including chronic obstructive pulmonary disease, diabetes mellitus, chronic steroid use and a high body mass index. Results: A total of 181 nephrectomies were performed in 175 patients and 175 kidneys (96.7%) had malignancy. Mean tumor size was 4.9 cm. Mean followup was 28.8 months. Extraction was done from a lower quadrant site in 55 patients (31.4%), from the umbilical site in 58 (33.2%) and from a paramedian site in 62 (35.4%). Patients with paramedian and lower quadrant extraction sites were older (p ⫽ 0.016), and had a higher body mass index (p ⫽ 0.001) and greater specimen weight (p ⫽ 0.003). In 4 patients an incisional hernia developed. An incisional hernia was significantly associated with the paramedian extraction site (p ⫽ 0.015). Conclusions: Incisional hernias may occur as a delayed complication of laparoscopic radical nephrectomy. This complication most commonly develops at the extraction site. In patients with a high body mass index using a paramedian extraction site is a significant risk factor for incisional hernia formation.
Abbreviations and Acronyms ASA ⫽ American Society of Anesthesiologists BMI ⫽ body mass index COPD ⫽ chronic obstructive pulmonary disease DM ⫽ diabetes mellitus IH ⫽ incisional hernia LRN ⫽ laparoscopic radical nephrectomy RCC ⫽ renal cell carcinoma Submitted for publication August 20, 2008. Study received internal review board approval. * Correspondence: Division of Endourology and Laparoscopy, Department of Urology, Miller School of Medicine, University of Miami, Dominion Tower, Suite 509 (M814), Miami, Florida 33136 (telephone: 305-243-7261; FAX: 305-2433381; e-mail: [email protected]).
Key Words: kidney, laparoscopy, nephrectomy, hernia, complications LAPAROSCOPIC radical nephrectomy is commonly performed when the size and location of a renal tumor necessitate complete organ removal. Specimen extraction methods include morcellation or removal of the specimen intact by port site expansion, connecting port site incisions, incising old abdominal scars or the creation of a new incision, such as a Pfannenstiel incision.1– 4 Common trocar sites used to
extract the specimen include a midline umbilical trocar site and a lower quadrant trocar site, which are created by extending the associated trocar site.4 In patients with a relatively larger body habitus the trocar site used for camera placement, which is at a paramedian location, may also be used as an extraction site. Paramedian incision is defined as an incision 2 finger breadths lateral to the rectus
0022-5347/09/1814-1565/0 THE JOURNAL OF UROLOGY® Copyright © 2009 by AMERICAN UROLOGICAL ASSOCIATION
Vol. 181, 1565-1570, April 2009 Printed in U.S.A. DOI:10.1016/j.juro.2008.11.113
www.jurology.com
1565
1566
LAPAROSCOPIC RADICAL NEPHRECTOMY EXTRACTION SITES
muscle. Various reports exist of the degree of pain associated with each of these extraction sites, although there are sparse data on long-term morbidity associated with these extraction sites, particularly with regard to the risk of hernia.5,6 We assessed the long-term morbidity of each of these extraction sites in conjunction with individual patient parameters to determine whether these extraction sites are associated with a higher risk of IH formation.
MATERIALS AND METHODS Internal review board approval was obtained for this study. From October 2002 to April 2007 we identified all patients who underwent LRN as performed by a single surgeon for localized renal mass/masses in our prospective database, which contains information on a number of variables related to patients and the surgical procedure. Patients who underwent hand assisted LRN were excluded from study, as were those with metastatic disease. Evaluable parameters were patient age, gender, BMI, side of surgery, specimen size, medical comorbidities and chronic steroid use for illnesses, including COPD, connective tissue disorders and as part of immunosuppression for renal transplantation, as well as previous abdominal surgery, length of stay, and any immediate and long-term complications related to the extraction site. Patients in this study underwent conventional LRN using a 4-trocar technique. If needed, a fifth trocar site lateral and just below the right costal margin was used for liver retraction. At the completion of the procedure in all patients the kidney/kidneys were placed in a LapSac® prepared with a Terumo® glidewire to aid with intracorporeal specimen entrapment, as described by Sundaram et al.7 After the specimen was placed in the entrapment sac a port site was expeditiously opened within 4 to 5 minutes for specimen extraction. One of 3 extraction sites was created. Site selection was not randomized, but rather chosen by the surgeon. When the trocar site used for camera insertion was located in the midline adjacent to the umbilicus, this site was extended in vertical fashion to be used as an extraction site (part A of figure). During the early course of this series in patients with a high BMI the trocar site for the camera port, which was placed via a paramedian incision, was extended in vertical fashion for extraction (part B of figure). These first 2 described sites were closed in a single layer with interrupted figure-of-8, 0 braided nonabsorbable polyester suture. In the latter part of this series in patients with a high BMI, although the trocar site for the camera port continued to be placed via a paramedian incision, the inferior trocar site located in the lower abdominal quadrant inferior and lateral to the umbilicus was extended in oblique, muscle splitting fashion for specimen extraction (part C of figure). This incision was closed in 2 layers with figure-of-8 interrupted 0 braided nonabsorbable polyester sutures. All patients undergoing LRN for a localized renal tumor enter a clinical pathway at hospital admission. They receive a patient controlled analgesia pump with morphine postoperatively. On postoperative day 1 pain management is converted to an oral narcotic analgesic. All
Extraction site incision. A, umbilical. B, paramedian. C, lower quadrant.
patients are discharged home with 1 prescription for a narcotic analgesic without refills. All patients were seen by the same operating surgeon for followup at 2 weeks, 6 weeks, 3 months, 6 months and 12 months, and at appropriate intervals thereafter related to cancer stage and other urological issues. Patients were evaluated by history/physical examination, serum analysis and radiological imaging as indicated. All wound sites were inspected for evidence of hernia and infection. Statistical analysis was performed using SPSS®, version 15.0. Differences in extraction site groups were analyzed with chi-square analysis. Analysis of any further significant association of other tested parameters was assessed by logistic regression or ANOVA. Statistical significance was considered at p ⬍0.05.
RESULTS A total of 181 nephrectomies were performed in 175 patients and 175 of 181 kidneys (96.7%) contained malignancy. All patients underwent successful
LAPAROSCOPIC RADICAL NEPHRECTOMY EXTRACTION SITES
1567
Table 1. Demographics of patients who underwent LRN Incision
No. pts* Mean ⫾ SD age No. men/women No. LRN (%): Rt/lt Bilat Mean ⫾ SD BMI (kg/m2) Mean ⫾ SD ASA score Mean ⫾ SD estimated blood loss (ml) Mean ⫾ SD specimen wt (kg) Mean ⫾ SD length of stay (hrs) Mean ⫾ SD followup (mos)
Entire Cohort
Umbilicus
Paramedian
Lower Quadrant
p Value
173 62 ⫾ 11.91 112/61
59 59 ⫾ 12.19 36/22
62 63 ⫾ 10.00 42/19
52 65 ⫾ 13.00 32/20
0.016 0.654
79 (45.6)/88 (50.9) 6 (3.5) 29 ⫾ 6.76 3 ⫾ 0.56 159 ⫾ 221.53 476 ⫾ 256.35 79.2 ⫾ 46.81 35 ⫾ 15.08
26/29 4 26 ⫾ 3.76 2 ⫾ 0.61 169 ⫾ 270.15 358 ⫾ 224.36 73.7 ⫾ 46.52 45 ⫾ 12.97
29/33 0 33 ⫾ 7.93 3 ⫾ 0.49 177 ⫾ 199 535 ⫾ 258.71 74.1 ⫾ 35.4 40 ⫾ 9.76
24/26 2 29 ⫾ 5.19 3 ⫾ 0.55 127 ⫾ 187.72 511 ⫾ 254.58 91.4 ⫾ 56.68 18 ⫾ 4.57
0.380 ⬍0.001 0.013 0.464 0.003 ⬍0.080 ⬍0.001
* Two patients who required conversion to hand assisted laparoscopy were removed from further analysis.
transperitoneal LRN. Four male and 2 female patients underwent bilateral nephrectomy with umbilical extraction at the time of the procedure. Final pathological results revealed bilateral RCC in 4 of these patients, right RCC and a left complex cyst in 1, and left RCC and a right benign tumor in 1. In 2 cases toward the end of the procedure a LapDisc® hand port was placed in the midline. One was used to manipulate the 16.5 cm tumor laden kidney for safe ligation of the renal artery and vein in sequential fashion. The other was used in a patient refusing blood products who sustained a laceration of the adrenal vein with subsequent bleeding. Neither patient had wound infection or hernia, although they were removed from any further analysis. Two patients underwent concomitant laparoscopic cholecystectomy with the gallbladder specimens removed via the kidney specimen extraction site. One patient in this study had a history of a ventral hernia. All patients underwent intact specimen removal with a LapSac. All patients were seen postoperatively at the clinic of our private urology office or at the urology clinic at our tertiary hospital. Table 1 shows the general demographic features of the patient cohort and breakdown by the 3 incision groups. Patients
with paramedian and lower quadrant extraction sites had a higher mean age, mean specimen weight and ASA score. BMI was highest in the group with a paramedian extraction site, followed by the group with a lower quadrant extraction site. Patients with a lower quadrant extraction site had shorter followup. Table 2 shows the breakdown of commonly recognized risk factors associated with IH and the presence of IH. IH occurred only in the group in which a paramedian extraction site was used (table 2). Table 3 lists demographics in patients in whom an IH developed. Analysis of other risk factors associated with IH by logistic regression did not show any significant association (table 4). None of the patients with an extraction site hernia experienced postoperative wound infection or ileus. There were no recognized cases of hernia at any of the 5 or 12 mm trocar sites. All 4 patients with an IH were sent to general surgery for consultation. One individual underwent laparoscopic repair with mesh with a good result. The other 3 patients elected conservative management. Three patients in whom the inferior extraction site was used complained of transient numbness and paresthesia in the incision region. These symptoms
Table 2. Incision sites and known IH risk factors
Overall BMI greater than 30 kg/m2 COPD DM Chronic steroid use Previous abdominal surgery Postop wound infection Renal failure Age greater than 50 Postop respiratory complications/chronic obstructive pulmonary disease exacerbation IH
No. Umbilicus (%)
No. Paramedian (%)
No. Lower Quadrant (%)
p Value
59 8 (13.6) 10 (16.9) 15 (25.4) 15 (25.4) 22 (37.3) 0 7 (11.9) 42 (71.2) 5 (8.5)
62 36 (58.1) 9 (14.5) 14 (22.3) 15 (24.2) 30 (48.4) 3 (4.8) 3 (4.8) 58 (93.5) 3 (4.8)
52 15 (28.8) 5 (9.6) 8 (15.4) 9 (17.3) 21 (40.4) 1 (1.9) 1 (1.9) 44 (84.6) 2 (3.8)
⬍0.001 0.528 0.419 0.551 0.443 0.204 0.084 0.012 0.537
0
4 (6.5)
0
0.026
1568
LAPAROSCOPIC RADICAL NEPHRECTOMY EXTRACTION SITES
Table 3. Demographics in patients with and without IH
Age Sex Wt (kg) Ht (cm) BMI (kg/m2) Skin incision length (cm) Specimen wt (gm) ASA score Significant medical history
Prior abdominal surgery Postop IH detection time (mos)
Pt 1
Pt 2
Pt 3
Pt 4
75 M 94.1 157.5 40.5 7.2 848 3 Hypertension, DM, pacemaker, gastroesophageal reflux disease, benign prostatic hyperplasia None 3
65 M 79.1 179.0 24.7 4.7 450 3 Coronary artery disease, postmyocardial infarction, hypertension, COPD
83 M 85.5 180.3 26.3 6.2 544 3 Hypertension, coronary artery disease, dyslipidemia, coagulopathy
49 F 100.9 158.7 40.4 4.8 474 2 Hypertension, gastroesophageal reflux disease, anemia
None
None
Hysterectomy
3
later resolved in all 3 cases. Two patients, including 1 with a lower quadrant extraction site and 1 with a paramedian extraction site, had wound infection requiring débridement and drainage. These 2 patients had a history of DM and obesity.
DISCUSSION Abdominal wall IHs have been studied extensively. These hernias result from a loss of connective tissue integrity after scar formation in a load bearing muscle tendon and/or fascial layer.8 After various types of abdominal incisions and closure techniques the incidence of these hernias is 5% to 15%.9,10 IH development may be multifactorial, related to operative/technical elements and to patient specific predisposing factors. Many risk factors for abdominal wall hernia formation have been identified, including previous abdominal surgery, obesity, renal insufficiency, renal failure, postoperative respiratory tract infection, diabetes, age older than 50 years, metastatic disease and impaired nutrition.2,10 Although many surgical series of various types exist documenting surgical outcomes, many of them do not specifically address or infer that IH formation was specifically part of the investigation. However, a review revealed a small number of studies that focused on or specifically pertained to the incidence of IH after operation. Table 5 lists the most relevant studies. These studies notably differ in the type of procedures performed, extraction sites used and paTable 4. IH and comorbidities Factors
p Value
Hypertension COPD DM Chronic steroid use BMI
0.218 0.998 0.182 0.998 0.494
9
12
tient specific parameters considered in the study population. Elashry et al described 5 patients with extraction site IH of 29 who underwent 5-port LRN with whole specimen extraction.2 Postoperatively respiratory complications developed in 3 of 5 patients with IH. The investigators concluded that they would avoid using the lower flank extraction site and they favored a midline or subcostal extraction site. They also stated that direct morcellation of nonurothelial cell renal tumors and extraction via a 10 to 12 mm port site are feasible. Troxel and Das investigated IH formation in 50 patients who underwent hand assisted laparoscopic surgery for renal cell tumor.10 To extract the specimen the hand port site was used at 1 of 2 sites, that is a midline incision or a muscle splitting, right lower quadrant incision. They reported 3 cases (6%) of IH formation postoperatively, which occurred at the midline where the hand port had been placed. They noted that obesity (mean weight 137 kg) coexisted in all patients with IH formation but they did not report BMI in their patient population. They suggested that early return to activity may predispose to hernia formation. Troxel and Das stated that wound strength is only at 40% to 80% of its maximum tensile strength between 3 and 6 weeks postoperatively, and premature resumption of normal activity may increase the risk of IH. It remains unclear whether particular aspects of hand assisted laparoscopy contribute to abdominal wall hernia formation. Using a hand through an incision for prolonged periods may result in tissue compression, trauma and transient ischemia. Tisdale et al compared extraction sites formed by expanded port site incisions vs Pfannenstiel incisions in a heterogeneous population that underwent LRN, laparoscopic nephroureterectomy and laparoscopic donor nephrectomy.4 The closure method for the extraction site was not mentioned. An IH developed in 3 patients (2.9%) who underwent LRN and
0 4 (7.4) 0 1 Layer 1 Layer 2 Layers No IH No IH No IH 25 No IH No IH 50 No IH No IH 34.9 No IH 25.5 33.2 29.0 58 63 64 Unreported Unreported Unreported
Unreported Unreported Unreported Unreported 29.0 57 Unreported Unreported
Unreported Infraumbilical/ Pfannenstiel Umbilicus Paramedian Lower quadrant
166
46
173
Singh et al13
Camargo et al12
Present series
4/5-Port LRN
Expanded port site, muscle cutting (16), infraumbilical/Pfannenstiel, muscle splitting (30) Vertical muscle splitting (59), vertical (62), oblique muscle splitting (52)
9 4 Unreported
Expanded port site, transverse (73), PFN (31) Vertical, linea alba division (74), vertical muscle splitting (92) 104 Tisdale et al4
LRN, trocar technique not specified Laparoscopic colorectal surgery, trocar technique not specified 4-Port LRN
50 Troxel and Das10
Hand assisted LRN, 2–3 ports
1 (2.2) 0
1 Polyglactin, method unreported 1 Polyglactin, method unreported Unreported
3 (2.9) 0 13 (7.8) 0 Unreported Unreported Unreported Unreported
38.2 Unreported 30.8 Unreported 30.2 28.0 27.3 29.8 62 56 64
Unreported 0 Unreported Unreported Unreported Unreported Unreported Unreported
Unreported Unreported Unreported Unreported
0 3 (6.0)
5 (16.8)
2-Zero chromic continuous 1 Polydioxanone continuous 3 Layer 0 0 34.9 26.0 63 10.4
Lower abdominal quadrant Rt lower abdominal quadrant Midline (supraumbilical/ infraumbilical) Paramedian Suprapubic Midline Paramedian Expanded port site, transverse muscle splitting (29) Transverse muscle splitting (9), vertical muscle splitting (41) 5-Port LRN 29 Elashry et al2
% COPD Mean BMI (kg/m2) Mean Age Mean Incision Length (cm) Extraction Site Incision (No. pts) Surgery References
No. Extraction Site Pts
Table 5. Reported incidence of IH secondary to transabdominal laparoscopic surgery with extraction site creation
Mean BMI (kg/m2)
IH
% DM
Closure
Peritoneum Suture
No. IH Pts (%)
LAPAROSCOPIC RADICAL NEPHRECTOMY EXTRACTION SITES
1569
whose specimen extraction sites were created by extending the trocar site at the lower quadrant. No patients with Pfannenstiel incisions had an IH. The investigators noted that the BMI in patients with an IH was greater than 35 kg/m2. Camargo et al studied the effects of kidney morcellation in 107 patients vs whole specimen extraction in 46 patients.12 Whole specimens were extracted through 1 of 3 incisions, including a muscle cutting incision connecting 2 port sites, a remote site through a muscle splitting infraumbilical incision or a Pfannenstiel incision. They noted only 2 complications relating to incision, including IH with concomitant small bowel obstruction and wound dehiscence in 1 case each. Singh et al examined midline and off midline extraction sites, and IH formation after laparoscopic colorectal surgery in 166 patients.13 They observed a higher IH rate for midline extraction sites compared to off midline extraction sites (7.8% vs 0%). They suggested that surgeon experience may have had a role since the midline incision was used more often in their early experience. In our study the incision site was the most significant factor with regard to IH formation postoperatively. We chose specific trocar and extraction sites since they appeared most conducive to performing the surgical procedure in the patient in question. In some studies the paramedian incision has been shown to result in a decreased incidence of IH compared to midline incisions. However, these studies may differ in patient specific factors, including age, body habitus, comorbidities, the pathological condition being treated and operative parameters, such as patient positioning. Furthermore, many studies do not consistently account for significant risk factors for IH formation (table 5). In our study patients who received midline incisions did not have an IH. However, midline extraction sites were reserved only for patients with a lower BMI and smaller body habitus. IHs only formed in patients who had received a paramedian incision for specimen extraction. Using this site was based on the patient body habitus. Before this observation the paramedian site was selected specifically for patients with a high BMI. These patients generally have attenuated body wall tissue and a relative lack of abdominal wall fascia for wound closure. Due to the observation of IH formation at the paramedian extraction site, the extraction site was changed to the inferior trocar site, where multiple abdominal wall muscular fascial layers are present. Care must be taken not to damage the lateral cutaneous nerves when using this site. Nonetheless, using the latter extraction site in patients with a high BMI has not resulted in IH formation. In our study 2 of 4 patients with an IH had a BMI of greater than 30 kg/m2 (table 3). A BMI of greater
1570
LAPAROSCOPIC RADICAL NEPHRECTOMY EXTRACTION SITES
than 30 kg/m2 is classified as obese according to WHO.14 These findings are consistent with those in previous studies and they reaffirm the risk of IH with increasing BMI.9 –11 A possible explanation for this rests on the anatomy of the abdominal wall muscles and the associated mechanics of the distribution of tension or the lack thereof in obese patients.15,16 Many obese patients in our series have numerous other comorbidities that may also contribute to delayed wound healing. Increasing age, chronic steroid use and DM have been shown to increase the risk of IH.9,11,13 Coughing secondary to COPD or a respiratory tract infection has also been shown to be a significant factor involved in IH formation.2,11 We accounted for all of these risk factors but we did not find any statistically significant association. The abdominal wall closure technique as well as the type of suture material used is also important. Results of randomized clinical trials and meta-analyses suggest that using continuous running nonabsorbable suture or slowly absorbed suture such as polydioxanone sulfate are the preferred suture materials for clo-
sure.15,17 Although Van’t Riet et al highlighted in their meta-analysis that no difference was detected in the IH incidence between slowly absorbable and nonabsorbable sutures, nonabsorbable sutures are associated with an increased incidence of prolonged wound pain and suture sinus formation.17 However, there are inherent limitations to our study. This is a retrospective study and as such it did not minimize confounding and bias, as may be done when performing a prospective, randomized study. This may account for the significant results found regarding ASA, BMI and specimen weight. Other factors, such as pain medication requirements, could also be of particular importance when analyzing the impact of the type of specimen extraction site.
CONCLUSIONS The site of intact specimen extraction in LRN cases should be considered because there is a risk of IH formation associated with using a paramedian extraction site in patients with a high BMI. Using the lower trocar site proved to be an effective alternative since the use of this site was not associated with IH development.
REFERENCES 1. Varkarakis I, Rha K, Hernandez F, Kavoussi LR and Jarrett TW: Laparoscopic specimen extraction: morcellation. BJU Int, suppl., 2005; 95: 27. 2. Elashry OM, Giusti G, Nadler RB, McDougall EM and Clayman RV: Incisional hernia after laparoscopic nephrectomy with intact specimen removal: caveat emptor. J Urol 1997; 158: 363. 3. Hemal AK, Kumar A, Gupta NP and Kumar R: Oncologic outcome of 132 cases of laparoscopic radical nephrectomy with intact specimen removal for T1-2N0M0 renal cell carcinoma. World J Urol 2007; 25: 619. 4. Tisdale BE, Kapoor A, Hussain A, Piercey K and Whelan JP: Intact specimen extraction in laparoscopic nephrectomy procedures: Pfannenstiel versus expanded port site incisions. Urology 2007; 69: 241. 5. Dunn MD, Portis AJ, Shalhav AL, Elbahnasy AM, Heidorn C and McDougall EM: Laparoscopic versus open radical nephrectomy: a 9-year experience. J Urol 2000; 164: 1153.
6. Ono Y, Kinukawa T, Hattori R, Yamada S, Nishiyama N, Mizutani K and Ohshima S: Laparoscopic radical nephrectomy for renal cell carcinoma: a five-year experience. Urology 1999; 53: 280. 7. Sundaram CP, Ono Y, Landman J, Rehman J and Clayman RV: Hydrophilic guide wire technique to facilitate organ entrapment using a laparoscopic sack during laparoscopy. J Urol 2002; 165: 1376. 8. Franz MG: The biology of hernia. Surg Clin North Am 2008; 88: 1. 9. Irvin TT, Stoddard CJ, Greavey MG and Duthie HL: Abdominal wound healing: a prospective clinical study. BMJ 1977; 2: 351. 10. Troxel SA and Das D: Incisional hernia following hand-assisted laparoscopic surgery for renal cell cancer. JSLS 2005; 9: 196. 11. Bucknall TE, Cox PJ and Ellis H: Burst abdomen and incisional hernia: a prospective study of 1129 major laparotomies. BMJ (Clin Res Ed) 1982; 284: 931. 12. Camargo AH, Rubenstein JN, Ershoff BD, Meng MV, Kane CJ and Stoller ML: The effect of kidney mor-
cellation on operative time, incision complications, and postoperative analgesia after laparoscopic nephrectomy. Int Braz J Urol 2006; 32: 273. 13. Singh R, Omiccioli A, Hegge S and McKinley C: Does the extraction site location in laparoscopic colorectal surgery have an impact on incisional hernia rates? Surg Endosc 2008; 22: 2596. 14. Kuczmarski R, Carroll M, Flegal KM and Toriano RP: Varying body mass index cutoff points to describe overweight prevalence among U. S. adults: NHANES III (1988 to 1994). Obes Res 1997; 5: 542. 15. O’Dwyer PJ and Courtney CA: Factors involved in abdominal wall closure and subsequent incisional hernia. Surgeon 2003; 1: 17. 16. Cox PJ, Ausobsky JR, Ellis H and Pollock AV: Towards no incisional hernias: lateral paramedian versus midline incisions. J R Soc Med 1986; 79: 711. 17. Van’t Riet M, Steyerberg EW, Nellensteyn J, Bonjer HJ and Jeekel J: Meta-analysis of techniques for closure of midline abdominal incisions. Br J Surg 2002; 89: 1350.
Purpose: Laparoscopic radical nephrectomy is commonly performed for renal tumors that are not amenable to nephron sparing treatment. A number of techniques for intact specimen extraction are used. The development of incisional hernias from the extraction site is a known but infrequent delayed complication. We analyzed different extraction sites and risk factors for such hernias. Materials and Methods: We retrospectively analyzed a cohort of patients undergoing laparoscopic radical nephrectomy with intact specimen extraction through 3 sites. Patients and operation specific parameters were included with particular attention to factors predisposing patients to incisional hernia, including chronic obstructive pulmonary disease, diabetes mellitus, chronic steroid use and a high body mass index. Results: A total of 181 nephrectomies were performed in 175 patients and 175 kidneys (96.7%) had malignancy. Mean tumor size was 4.9 cm. Mean followup was 28.8 months. Extraction was done from a lower quadrant site in 55 patients (31.4%), from the umbilical site in 58 (33.2%) and from a paramedian site in 62 (35.4%). Patients with paramedian and lower quadrant extraction sites were older (p ⫽ 0.016), and had a higher body mass index (p ⫽ 0.001) and greater specimen weight (p ⫽ 0.003). In 4 patients an incisional hernia developed. An incisional hernia was significantly associated with the paramedian extraction site (p ⫽ 0.015). Conclusions: Incisional hernias may occur as a delayed complication of laparoscopic radical nephrectomy. This complication most commonly develops at the extraction site. In patients with a high body mass index using a paramedian extraction site is a significant risk factor for incisional hernia formation.
Abbreviations and Acronyms ASA ⫽ American Society of Anesthesiologists BMI ⫽ body mass index COPD ⫽ chronic obstructive pulmonary disease DM ⫽ diabetes mellitus IH ⫽ incisional hernia LRN ⫽ laparoscopic radical nephrectomy RCC ⫽ renal cell carcinoma Submitted for publication August 20, 2008. Study received internal review board approval. * Correspondence: Division of Endourology and Laparoscopy, Department of Urology, Miller School of Medicine, University of Miami, Dominion Tower, Suite 509 (M814), Miami, Florida 33136 (telephone: 305-243-7261; FAX: 305-2433381; e-mail: [email protected]).
Key Words: kidney, laparoscopy, nephrectomy, hernia, complications LAPAROSCOPIC radical nephrectomy is commonly performed when the size and location of a renal tumor necessitate complete organ removal. Specimen extraction methods include morcellation or removal of the specimen intact by port site expansion, connecting port site incisions, incising old abdominal scars or the creation of a new incision, such as a Pfannenstiel incision.1– 4 Common trocar sites used to
extract the specimen include a midline umbilical trocar site and a lower quadrant trocar site, which are created by extending the associated trocar site.4 In patients with a relatively larger body habitus the trocar site used for camera placement, which is at a paramedian location, may also be used as an extraction site. Paramedian incision is defined as an incision 2 finger breadths lateral to the rectus
0022-5347/09/1814-1565/0 THE JOURNAL OF UROLOGY® Copyright © 2009 by AMERICAN UROLOGICAL ASSOCIATION
Vol. 181, 1565-1570, April 2009 Printed in U.S.A. DOI:10.1016/j.juro.2008.11.113
www.jurology.com
1565
1566
LAPAROSCOPIC RADICAL NEPHRECTOMY EXTRACTION SITES
muscle. Various reports exist of the degree of pain associated with each of these extraction sites, although there are sparse data on long-term morbidity associated with these extraction sites, particularly with regard to the risk of hernia.5,6 We assessed the long-term morbidity of each of these extraction sites in conjunction with individual patient parameters to determine whether these extraction sites are associated with a higher risk of IH formation.
MATERIALS AND METHODS Internal review board approval was obtained for this study. From October 2002 to April 2007 we identified all patients who underwent LRN as performed by a single surgeon for localized renal mass/masses in our prospective database, which contains information on a number of variables related to patients and the surgical procedure. Patients who underwent hand assisted LRN were excluded from study, as were those with metastatic disease. Evaluable parameters were patient age, gender, BMI, side of surgery, specimen size, medical comorbidities and chronic steroid use for illnesses, including COPD, connective tissue disorders and as part of immunosuppression for renal transplantation, as well as previous abdominal surgery, length of stay, and any immediate and long-term complications related to the extraction site. Patients in this study underwent conventional LRN using a 4-trocar technique. If needed, a fifth trocar site lateral and just below the right costal margin was used for liver retraction. At the completion of the procedure in all patients the kidney/kidneys were placed in a LapSac® prepared with a Terumo® glidewire to aid with intracorporeal specimen entrapment, as described by Sundaram et al.7 After the specimen was placed in the entrapment sac a port site was expeditiously opened within 4 to 5 minutes for specimen extraction. One of 3 extraction sites was created. Site selection was not randomized, but rather chosen by the surgeon. When the trocar site used for camera insertion was located in the midline adjacent to the umbilicus, this site was extended in vertical fashion to be used as an extraction site (part A of figure). During the early course of this series in patients with a high BMI the trocar site for the camera port, which was placed via a paramedian incision, was extended in vertical fashion for extraction (part B of figure). These first 2 described sites were closed in a single layer with interrupted figure-of-8, 0 braided nonabsorbable polyester suture. In the latter part of this series in patients with a high BMI, although the trocar site for the camera port continued to be placed via a paramedian incision, the inferior trocar site located in the lower abdominal quadrant inferior and lateral to the umbilicus was extended in oblique, muscle splitting fashion for specimen extraction (part C of figure). This incision was closed in 2 layers with figure-of-8 interrupted 0 braided nonabsorbable polyester sutures. All patients undergoing LRN for a localized renal tumor enter a clinical pathway at hospital admission. They receive a patient controlled analgesia pump with morphine postoperatively. On postoperative day 1 pain management is converted to an oral narcotic analgesic. All
Extraction site incision. A, umbilical. B, paramedian. C, lower quadrant.
patients are discharged home with 1 prescription for a narcotic analgesic without refills. All patients were seen by the same operating surgeon for followup at 2 weeks, 6 weeks, 3 months, 6 months and 12 months, and at appropriate intervals thereafter related to cancer stage and other urological issues. Patients were evaluated by history/physical examination, serum analysis and radiological imaging as indicated. All wound sites were inspected for evidence of hernia and infection. Statistical analysis was performed using SPSS®, version 15.0. Differences in extraction site groups were analyzed with chi-square analysis. Analysis of any further significant association of other tested parameters was assessed by logistic regression or ANOVA. Statistical significance was considered at p ⬍0.05.
RESULTS A total of 181 nephrectomies were performed in 175 patients and 175 of 181 kidneys (96.7%) contained malignancy. All patients underwent successful
LAPAROSCOPIC RADICAL NEPHRECTOMY EXTRACTION SITES
1567
Table 1. Demographics of patients who underwent LRN Incision
No. pts* Mean ⫾ SD age No. men/women No. LRN (%): Rt/lt Bilat Mean ⫾ SD BMI (kg/m2) Mean ⫾ SD ASA score Mean ⫾ SD estimated blood loss (ml) Mean ⫾ SD specimen wt (kg) Mean ⫾ SD length of stay (hrs) Mean ⫾ SD followup (mos)
Entire Cohort
Umbilicus
Paramedian
Lower Quadrant
p Value
173 62 ⫾ 11.91 112/61
59 59 ⫾ 12.19 36/22
62 63 ⫾ 10.00 42/19
52 65 ⫾ 13.00 32/20
0.016 0.654
79 (45.6)/88 (50.9) 6 (3.5) 29 ⫾ 6.76 3 ⫾ 0.56 159 ⫾ 221.53 476 ⫾ 256.35 79.2 ⫾ 46.81 35 ⫾ 15.08
26/29 4 26 ⫾ 3.76 2 ⫾ 0.61 169 ⫾ 270.15 358 ⫾ 224.36 73.7 ⫾ 46.52 45 ⫾ 12.97
29/33 0 33 ⫾ 7.93 3 ⫾ 0.49 177 ⫾ 199 535 ⫾ 258.71 74.1 ⫾ 35.4 40 ⫾ 9.76
24/26 2 29 ⫾ 5.19 3 ⫾ 0.55 127 ⫾ 187.72 511 ⫾ 254.58 91.4 ⫾ 56.68 18 ⫾ 4.57
0.380 ⬍0.001 0.013 0.464 0.003 ⬍0.080 ⬍0.001
* Two patients who required conversion to hand assisted laparoscopy were removed from further analysis.
transperitoneal LRN. Four male and 2 female patients underwent bilateral nephrectomy with umbilical extraction at the time of the procedure. Final pathological results revealed bilateral RCC in 4 of these patients, right RCC and a left complex cyst in 1, and left RCC and a right benign tumor in 1. In 2 cases toward the end of the procedure a LapDisc® hand port was placed in the midline. One was used to manipulate the 16.5 cm tumor laden kidney for safe ligation of the renal artery and vein in sequential fashion. The other was used in a patient refusing blood products who sustained a laceration of the adrenal vein with subsequent bleeding. Neither patient had wound infection or hernia, although they were removed from any further analysis. Two patients underwent concomitant laparoscopic cholecystectomy with the gallbladder specimens removed via the kidney specimen extraction site. One patient in this study had a history of a ventral hernia. All patients underwent intact specimen removal with a LapSac. All patients were seen postoperatively at the clinic of our private urology office or at the urology clinic at our tertiary hospital. Table 1 shows the general demographic features of the patient cohort and breakdown by the 3 incision groups. Patients
with paramedian and lower quadrant extraction sites had a higher mean age, mean specimen weight and ASA score. BMI was highest in the group with a paramedian extraction site, followed by the group with a lower quadrant extraction site. Patients with a lower quadrant extraction site had shorter followup. Table 2 shows the breakdown of commonly recognized risk factors associated with IH and the presence of IH. IH occurred only in the group in which a paramedian extraction site was used (table 2). Table 3 lists demographics in patients in whom an IH developed. Analysis of other risk factors associated with IH by logistic regression did not show any significant association (table 4). None of the patients with an extraction site hernia experienced postoperative wound infection or ileus. There were no recognized cases of hernia at any of the 5 or 12 mm trocar sites. All 4 patients with an IH were sent to general surgery for consultation. One individual underwent laparoscopic repair with mesh with a good result. The other 3 patients elected conservative management. Three patients in whom the inferior extraction site was used complained of transient numbness and paresthesia in the incision region. These symptoms
Table 2. Incision sites and known IH risk factors
Overall BMI greater than 30 kg/m2 COPD DM Chronic steroid use Previous abdominal surgery Postop wound infection Renal failure Age greater than 50 Postop respiratory complications/chronic obstructive pulmonary disease exacerbation IH
No. Umbilicus (%)
No. Paramedian (%)
No. Lower Quadrant (%)
p Value
59 8 (13.6) 10 (16.9) 15 (25.4) 15 (25.4) 22 (37.3) 0 7 (11.9) 42 (71.2) 5 (8.5)
62 36 (58.1) 9 (14.5) 14 (22.3) 15 (24.2) 30 (48.4) 3 (4.8) 3 (4.8) 58 (93.5) 3 (4.8)
52 15 (28.8) 5 (9.6) 8 (15.4) 9 (17.3) 21 (40.4) 1 (1.9) 1 (1.9) 44 (84.6) 2 (3.8)
⬍0.001 0.528 0.419 0.551 0.443 0.204 0.084 0.012 0.537
0
4 (6.5)
0
0.026
1568
LAPAROSCOPIC RADICAL NEPHRECTOMY EXTRACTION SITES
Table 3. Demographics in patients with and without IH
Age Sex Wt (kg) Ht (cm) BMI (kg/m2) Skin incision length (cm) Specimen wt (gm) ASA score Significant medical history
Prior abdominal surgery Postop IH detection time (mos)
Pt 1
Pt 2
Pt 3
Pt 4
75 M 94.1 157.5 40.5 7.2 848 3 Hypertension, DM, pacemaker, gastroesophageal reflux disease, benign prostatic hyperplasia None 3
65 M 79.1 179.0 24.7 4.7 450 3 Coronary artery disease, postmyocardial infarction, hypertension, COPD
83 M 85.5 180.3 26.3 6.2 544 3 Hypertension, coronary artery disease, dyslipidemia, coagulopathy
49 F 100.9 158.7 40.4 4.8 474 2 Hypertension, gastroesophageal reflux disease, anemia
None
None
Hysterectomy
3
later resolved in all 3 cases. Two patients, including 1 with a lower quadrant extraction site and 1 with a paramedian extraction site, had wound infection requiring débridement and drainage. These 2 patients had a history of DM and obesity.
DISCUSSION Abdominal wall IHs have been studied extensively. These hernias result from a loss of connective tissue integrity after scar formation in a load bearing muscle tendon and/or fascial layer.8 After various types of abdominal incisions and closure techniques the incidence of these hernias is 5% to 15%.9,10 IH development may be multifactorial, related to operative/technical elements and to patient specific predisposing factors. Many risk factors for abdominal wall hernia formation have been identified, including previous abdominal surgery, obesity, renal insufficiency, renal failure, postoperative respiratory tract infection, diabetes, age older than 50 years, metastatic disease and impaired nutrition.2,10 Although many surgical series of various types exist documenting surgical outcomes, many of them do not specifically address or infer that IH formation was specifically part of the investigation. However, a review revealed a small number of studies that focused on or specifically pertained to the incidence of IH after operation. Table 5 lists the most relevant studies. These studies notably differ in the type of procedures performed, extraction sites used and paTable 4. IH and comorbidities Factors
p Value
Hypertension COPD DM Chronic steroid use BMI
0.218 0.998 0.182 0.998 0.494
9
12
tient specific parameters considered in the study population. Elashry et al described 5 patients with extraction site IH of 29 who underwent 5-port LRN with whole specimen extraction.2 Postoperatively respiratory complications developed in 3 of 5 patients with IH. The investigators concluded that they would avoid using the lower flank extraction site and they favored a midline or subcostal extraction site. They also stated that direct morcellation of nonurothelial cell renal tumors and extraction via a 10 to 12 mm port site are feasible. Troxel and Das investigated IH formation in 50 patients who underwent hand assisted laparoscopic surgery for renal cell tumor.10 To extract the specimen the hand port site was used at 1 of 2 sites, that is a midline incision or a muscle splitting, right lower quadrant incision. They reported 3 cases (6%) of IH formation postoperatively, which occurred at the midline where the hand port had been placed. They noted that obesity (mean weight 137 kg) coexisted in all patients with IH formation but they did not report BMI in their patient population. They suggested that early return to activity may predispose to hernia formation. Troxel and Das stated that wound strength is only at 40% to 80% of its maximum tensile strength between 3 and 6 weeks postoperatively, and premature resumption of normal activity may increase the risk of IH. It remains unclear whether particular aspects of hand assisted laparoscopy contribute to abdominal wall hernia formation. Using a hand through an incision for prolonged periods may result in tissue compression, trauma and transient ischemia. Tisdale et al compared extraction sites formed by expanded port site incisions vs Pfannenstiel incisions in a heterogeneous population that underwent LRN, laparoscopic nephroureterectomy and laparoscopic donor nephrectomy.4 The closure method for the extraction site was not mentioned. An IH developed in 3 patients (2.9%) who underwent LRN and
0 4 (7.4) 0 1 Layer 1 Layer 2 Layers No IH No IH No IH 25 No IH No IH 50 No IH No IH 34.9 No IH 25.5 33.2 29.0 58 63 64 Unreported Unreported Unreported
Unreported Unreported Unreported Unreported 29.0 57 Unreported Unreported
Unreported Infraumbilical/ Pfannenstiel Umbilicus Paramedian Lower quadrant
166
46
173
Singh et al13
Camargo et al12
Present series
4/5-Port LRN
Expanded port site, muscle cutting (16), infraumbilical/Pfannenstiel, muscle splitting (30) Vertical muscle splitting (59), vertical (62), oblique muscle splitting (52)
9 4 Unreported
Expanded port site, transverse (73), PFN (31) Vertical, linea alba division (74), vertical muscle splitting (92) 104 Tisdale et al4
LRN, trocar technique not specified Laparoscopic colorectal surgery, trocar technique not specified 4-Port LRN
50 Troxel and Das10
Hand assisted LRN, 2–3 ports
1 (2.2) 0
1 Polyglactin, method unreported 1 Polyglactin, method unreported Unreported
3 (2.9) 0 13 (7.8) 0 Unreported Unreported Unreported Unreported
38.2 Unreported 30.8 Unreported 30.2 28.0 27.3 29.8 62 56 64
Unreported 0 Unreported Unreported Unreported Unreported Unreported Unreported
Unreported Unreported Unreported Unreported
0 3 (6.0)
5 (16.8)
2-Zero chromic continuous 1 Polydioxanone continuous 3 Layer 0 0 34.9 26.0 63 10.4
Lower abdominal quadrant Rt lower abdominal quadrant Midline (supraumbilical/ infraumbilical) Paramedian Suprapubic Midline Paramedian Expanded port site, transverse muscle splitting (29) Transverse muscle splitting (9), vertical muscle splitting (41) 5-Port LRN 29 Elashry et al2
% COPD Mean BMI (kg/m2) Mean Age Mean Incision Length (cm) Extraction Site Incision (No. pts) Surgery References
No. Extraction Site Pts
Table 5. Reported incidence of IH secondary to transabdominal laparoscopic surgery with extraction site creation
Mean BMI (kg/m2)
IH
% DM
Closure
Peritoneum Suture
No. IH Pts (%)
LAPAROSCOPIC RADICAL NEPHRECTOMY EXTRACTION SITES
1569
whose specimen extraction sites were created by extending the trocar site at the lower quadrant. No patients with Pfannenstiel incisions had an IH. The investigators noted that the BMI in patients with an IH was greater than 35 kg/m2. Camargo et al studied the effects of kidney morcellation in 107 patients vs whole specimen extraction in 46 patients.12 Whole specimens were extracted through 1 of 3 incisions, including a muscle cutting incision connecting 2 port sites, a remote site through a muscle splitting infraumbilical incision or a Pfannenstiel incision. They noted only 2 complications relating to incision, including IH with concomitant small bowel obstruction and wound dehiscence in 1 case each. Singh et al examined midline and off midline extraction sites, and IH formation after laparoscopic colorectal surgery in 166 patients.13 They observed a higher IH rate for midline extraction sites compared to off midline extraction sites (7.8% vs 0%). They suggested that surgeon experience may have had a role since the midline incision was used more often in their early experience. In our study the incision site was the most significant factor with regard to IH formation postoperatively. We chose specific trocar and extraction sites since they appeared most conducive to performing the surgical procedure in the patient in question. In some studies the paramedian incision has been shown to result in a decreased incidence of IH compared to midline incisions. However, these studies may differ in patient specific factors, including age, body habitus, comorbidities, the pathological condition being treated and operative parameters, such as patient positioning. Furthermore, many studies do not consistently account for significant risk factors for IH formation (table 5). In our study patients who received midline incisions did not have an IH. However, midline extraction sites were reserved only for patients with a lower BMI and smaller body habitus. IHs only formed in patients who had received a paramedian incision for specimen extraction. Using this site was based on the patient body habitus. Before this observation the paramedian site was selected specifically for patients with a high BMI. These patients generally have attenuated body wall tissue and a relative lack of abdominal wall fascia for wound closure. Due to the observation of IH formation at the paramedian extraction site, the extraction site was changed to the inferior trocar site, where multiple abdominal wall muscular fascial layers are present. Care must be taken not to damage the lateral cutaneous nerves when using this site. Nonetheless, using the latter extraction site in patients with a high BMI has not resulted in IH formation. In our study 2 of 4 patients with an IH had a BMI of greater than 30 kg/m2 (table 3). A BMI of greater
1570
LAPAROSCOPIC RADICAL NEPHRECTOMY EXTRACTION SITES
than 30 kg/m2 is classified as obese according to WHO.14 These findings are consistent with those in previous studies and they reaffirm the risk of IH with increasing BMI.9 –11 A possible explanation for this rests on the anatomy of the abdominal wall muscles and the associated mechanics of the distribution of tension or the lack thereof in obese patients.15,16 Many obese patients in our series have numerous other comorbidities that may also contribute to delayed wound healing. Increasing age, chronic steroid use and DM have been shown to increase the risk of IH.9,11,13 Coughing secondary to COPD or a respiratory tract infection has also been shown to be a significant factor involved in IH formation.2,11 We accounted for all of these risk factors but we did not find any statistically significant association. The abdominal wall closure technique as well as the type of suture material used is also important. Results of randomized clinical trials and meta-analyses suggest that using continuous running nonabsorbable suture or slowly absorbed suture such as polydioxanone sulfate are the preferred suture materials for clo-
sure.15,17 Although Van’t Riet et al highlighted in their meta-analysis that no difference was detected in the IH incidence between slowly absorbable and nonabsorbable sutures, nonabsorbable sutures are associated with an increased incidence of prolonged wound pain and suture sinus formation.17 However, there are inherent limitations to our study. This is a retrospective study and as such it did not minimize confounding and bias, as may be done when performing a prospective, randomized study. This may account for the significant results found regarding ASA, BMI and specimen weight. Other factors, such as pain medication requirements, could also be of particular importance when analyzing the impact of the type of specimen extraction site.
CONCLUSIONS The site of intact specimen extraction in LRN cases should be considered because there is a risk of IH formation associated with using a paramedian extraction site in patients with a high BMI. Using the lower trocar site proved to be an effective alternative since the use of this site was not associated with IH development.
REFERENCES 1. Varkarakis I, Rha K, Hernandez F, Kavoussi LR and Jarrett TW: Laparoscopic specimen extraction: morcellation. BJU Int, suppl., 2005; 95: 27. 2. Elashry OM, Giusti G, Nadler RB, McDougall EM and Clayman RV: Incisional hernia after laparoscopic nephrectomy with intact specimen removal: caveat emptor. J Urol 1997; 158: 363. 3. Hemal AK, Kumar A, Gupta NP and Kumar R: Oncologic outcome of 132 cases of laparoscopic radical nephrectomy with intact specimen removal for T1-2N0M0 renal cell carcinoma. World J Urol 2007; 25: 619. 4. Tisdale BE, Kapoor A, Hussain A, Piercey K and Whelan JP: Intact specimen extraction in laparoscopic nephrectomy procedures: Pfannenstiel versus expanded port site incisions. Urology 2007; 69: 241. 5. Dunn MD, Portis AJ, Shalhav AL, Elbahnasy AM, Heidorn C and McDougall EM: Laparoscopic versus open radical nephrectomy: a 9-year experience. J Urol 2000; 164: 1153.
6. Ono Y, Kinukawa T, Hattori R, Yamada S, Nishiyama N, Mizutani K and Ohshima S: Laparoscopic radical nephrectomy for renal cell carcinoma: a five-year experience. Urology 1999; 53: 280. 7. Sundaram CP, Ono Y, Landman J, Rehman J and Clayman RV: Hydrophilic guide wire technique to facilitate organ entrapment using a laparoscopic sack during laparoscopy. J Urol 2002; 165: 1376. 8. Franz MG: The biology of hernia. Surg Clin North Am 2008; 88: 1. 9. Irvin TT, Stoddard CJ, Greavey MG and Duthie HL: Abdominal wound healing: a prospective clinical study. BMJ 1977; 2: 351. 10. Troxel SA and Das D: Incisional hernia following hand-assisted laparoscopic surgery for renal cell cancer. JSLS 2005; 9: 196. 11. Bucknall TE, Cox PJ and Ellis H: Burst abdomen and incisional hernia: a prospective study of 1129 major laparotomies. BMJ (Clin Res Ed) 1982; 284: 931. 12. Camargo AH, Rubenstein JN, Ershoff BD, Meng MV, Kane CJ and Stoller ML: The effect of kidney mor-
cellation on operative time, incision complications, and postoperative analgesia after laparoscopic nephrectomy. Int Braz J Urol 2006; 32: 273. 13. Singh R, Omiccioli A, Hegge S and McKinley C: Does the extraction site location in laparoscopic colorectal surgery have an impact on incisional hernia rates? Surg Endosc 2008; 22: 2596. 14. Kuczmarski R, Carroll M, Flegal KM and Toriano RP: Varying body mass index cutoff points to describe overweight prevalence among U. S. adults: NHANES III (1988 to 1994). Obes Res 1997; 5: 542. 15. O’Dwyer PJ and Courtney CA: Factors involved in abdominal wall closure and subsequent incisional hernia. Surgeon 2003; 1: 17. 16. Cox PJ, Ausobsky JR, Ellis H and Pollock AV: Towards no incisional hernias: lateral paramedian versus midline incisions. J R Soc Med 1986; 79: 711. 17. Van’t Riet M, Steyerberg EW, Nellensteyn J, Bonjer HJ and Jeekel J: Meta-analysis of techniques for closure of midline abdominal incisions. Br J Surg 2002; 89: 1350.