This article has Open Peer Review reports available.
Correspondence: Laparoscopic repair of abdominal wall hernia - “How I do it” - synopsis of a seemingly straightforward technique
© Berney. 2015
Received: 15 January 2015
Accepted: 28 July 2015
Published: 19 August 2015
Abdominal wall hernia repairs are commonly performed worldwide in general surgery. There is still no agreed consensus on the optimal surgical approach. Since the turn of the twenty-first century, minimally invasive techniques have gained in popularity as they combine the advantages of limited abdominal wall dissection, reduced post-operative pain and risk of complications, and shorter hospital stay. Although the added cost incurred by using sophisticated laparoscopic instruments may be quite substantial, it is precisely counterbalanced by an improved morbidity rate, faster discharge home and time to return to work. Laparoscopic abdominal wall hernia repair is often challenging, as it requires good anatomical knowledge, eye-hand coordination and diversified laparoscopic skills. The objective of this article is not to present another set of personal data and to compare it with already published results on this matter, but simply to offer comprehensive guidelines on the practical aspects of this relatively new technique. Some of these steps have already been discussed but most of the time in a scattered way in the surgical literature, while others are the fruit of a personal expertise grasped over the years.
Mesh repair of primary ventral abdominal wall and incisional hernias is a common procedure performed worldwide by general surgeons. The use of mesh-based techniques has revolutionized our practice as it has significantly dropped the unacceptably elevated long-term recurrence rates, previously reported to be as high as 60 % with a simple primary open surgical repair without mesh [1, 2]. Except from general acceptance that use of prosthetic material for good hernia repair is essential, it is extremely difficult to find agreed international consensus on what would be the most optimal surgical approach for treatment of a particular abdominal wall defect. This is in part due to the ongoing development and release of new prosthetic materials, improvement of existing technologies and description of innovative surgical techniques.
Since the turn of the twenty-first century minimally invasive laparoscopic approach has gained in popularity and may become the preferred method for hernia defects smaller than 10 cm in width . The advantages of this technique over traditional open repair approach are reduced post-operative pain and wound complication rates, quicker recovery and return to normal physical activities, the potential to identify and repair any associated subclinical (occult) hernia defect concomitantly , including Swiss-cheese weaknesses, and improved cosmesis.
Unlike the laparoscopic technique for repair of inguinal hernias that requires a slow and often laborious learning curve to gaining complete familiarity with the anatomy of the groin [5, 6], the use of such approach to ventral or incisional hernia repairs is relatively simpler and more straightforward to acquire as the operator is not restricted by a confined operative space and the overall surgical view is generally better. Having said that it is still very important to become familiar with the anatomical landmarks, to possess good eye-hand coordination and suitable skills in minimally invasive surgical techniques in order to minimize the potential risk of developing major complications such as small bowel injury, or increase the hernia recurrence rate.
The concept of laparoscopic approach for repair of ventral and incisional hernias (LRVIH) is inspired from the principles of the open retrorectus tension-free mesh repair (sublay) technique championed by Rives-Stoppa, where the prosthesis is held in place between the muscle and its posterior sheath by transfixing sutures and intra-abdominal pressure [7, 8]. However, due to its simplicity and the fact that it requires much less extensive lateral dissection laparoscopic repair differs significantly in the sense that the mesh is preferentially secured directly in the abdominal cavity as the so-called IntraPeritoneal Onlay Mesh (IPOM) technique, first described by LeBlanc and Booth in 1993 . In other words with the IPOM approach a considerable size prosthetic mesh can be easily inserted and secured to the posterior abdominal wall offering a wide overlap of the defect and greater mesh to hernia ratio, thus improving even further long-term clinical outcome (low recurrence rate) . In comparison both the open onlay or sublay techniques would necessitate either raising large subcutaneous flaps or developing considerable space behind the rectus muscles bilaterally to deploy a similar large-sized mesh, thus resulting in significantly higher morbidity rate.
Apart from its relatively higher cost, other disadvantages of the laparoscopic technique compared to a standard open approach are the potential risk of visceral injury and not being able to restore the anatomy and physiology of a fully engineered abdominal wall, if the fascial defect has been left open before securing the mesh to its edges [11, 12].
Abdominal wall hernia disease is often a complex problem and the treatment options are numerous, especially with the ongoing development of new prosthetic materials or tacking devices. As such, a ‘reductionist’ scientific approach to study and conclude what would be the best treatment option according to ‘evidence-best medicine’ is unfeasible and beyond the scope of this article. The purpose of this ‘How I do it’ paper is therefore to offer an experienced surgical opinion on the practical aspects of LRVIH for those interested to learn and/or improve their technique for the benefit of their patients. Indeed, the key for a successful laparoscopic repair relies on the repetition of identical moves that follow a very detailed ‘step by step’ scheme. Some of the basic rules here have already been discussed in various formats in the surgical literature, while others are the fruit of a personal expertise grasped over the years. I will initially and mainly talk about the IPOM repair, which is the most widely used and simpler technique to learn. I will also discuss another approach, which is our preferred method for abdominal wall hernia repairs whenever suitable, namely the TransAbdominal PrePeritoneal (TAPP) technique that is more commonly used in inguinal hernia repairs, and briefly about the Totally Extraperitoneal (TEP) approach for Spigelian hernias. I do not intend to talk about laparoscopic parastomal hernia repair, which is a more complex procedure than standard LRVIH and should be only attempted by expert laparoscopic hernia surgeons. Approval to prepare this manuscript was given by our local Ethics Committee of the Department of Surgery at Bankstown-Lidcombe Hospital.
The indications for laparoscopic repair of abdominal wall hernias are essentially the same as for open surgery and absolute contraindications to this method, although quite rare would include patients with severe cardiovascular, pulmonary and liver impairment, intra-abdominal sepsis, children and pregnancy. Specific contraindications are the presence of a strangulated hernia, or patients in whom a safe access to the intra-abdominal cavity cannot be achieved due to dense adhesions. Relative contraindications include non-reducible (incarcerated), large size (>10 cm in diameter) and loss of domain hernias. This will of course depend on the surgeon’s skills.
Preoperative specific measures are essential for patients on antiplatelet medication and anticoagulant such as aspirin, clopidogrel or warfarin. If aspirin has been prescribed as secondary prevention treatment, it will be withheld for one week before surgery. Otherwise, as a primary prevention the aspirin won’t be stopped and the patient will be warned about the higher risk of post-operative bleeding. In case of dual antiplatelet regimen following previous coronary stenting, the clopidogrel will be discontinued preferentially ten days prior to the procedure and the aspirin maintained unless otherwise instructed by the cardiologist. Finally, for patients on warfarin the decision will be adapted depending on the indication for anticoagulation and in collaboration with their specialist cardiologist and/or hematologist. Whether anticoagulation can only be withheld for several hours or a few days, preference will still be for a laparoscopic IPOM approach as the level of surgical abdominal wall dissection will still be minimal as compared to a traditional open technique.
In theatre, the patient is shaved using a clipper in order to avoid skin micro traumatism, which can potentially increase the risk of infection . He/she is also asked to empty his/her bladder before surgery in order to avoid urine catheterization and potential risk of catheter associated UTI. As a preference an indwelling catheter (IDC) will be only used if the surgery is likely to last for more than two hours, if a male patient suffers from prostatism, or if the hernia defect is suprapubic. Single-shot preoperative low-molecular weight heparin (Enoxaparin 20 mg) and prophylactic first generation cephalosporin (Cephalothin 1gr) injections are preferably used. The procedure is performed under general anesthesia and muscular relaxation. Sequential calf compressors are used throughout the entire length of the procedure. The patient is placed in a supine position with both arms tucked in at his side. If deemed necessary, an orogastric tube is inserted to decompress the stomach and will be removed at the end of the procedure. The surgeon and camera holder stand on the side opposite to the TV monitor, which position will vary depending on the location of the abdominal wall defect and suspected presence of underlying adhesions. As most ventral/incisional hernias arise at the level of the midline through the linea alba, the surgeon and his assistant should preferentially stand on the left side of the patient. Indeed, the descending colon is narrower and more dorsally situated than the ascending colon; it is therefore preferable to position the working ports from the left side of the abdomen as it represents the most lateral insertion site, meaning the further away from the hernia defect.
Laparoscopic IntraPeritoneal Onlay Mesh (IPOM) technique
Conventional IPOM technique corresponds to a ‘bridging’ repair where the abdominal wall defect is left open (tension-free) and covered by the mesh. In this situation post-operative seroma formation will inevitably occur as well as residual abdominal bulging at the site of hernia repair, which is not cosmetically rewarding and can be regarded by the patients as a hernia recurrence. Another important disadvantage is that no attempt is made to restore the integrity of the linea alba, which is essential for achieving proper anatomical and physiological function of the abdominal wall. This is the reason why, whenever achievable without creating too much tension, I recommend primarily closing the defect before performing the mesh repair. This also maximizes the amount of tissue ingrowth into the mesh as the surface area between the prosthesis and the abdominal wall is significantly increased. Such variation is called ‘augmentation repair’, or IPOM-Plus . It is important to consider the shape of the abdominal wall defect prior to opting for a primary closure or not. Indeed, the more elliptical the opening the likeliest primary approximation of the hernia contours will be achievable with acceptable wound edge tension. Conversely the further the length-to-width ratio of the hernia defect approaches the value of 1, meaning the more circular it is in shape, the greatest the tension to the wound edge will apply if trying to primarily close it, and then a ‘bridging’ repair will be more appropriate. Laparoscopic bilateral component separation technique for increasing the chance of primary approximation of the rectus abdominis muscles has been recently proposed , but we will not expand on the subject as this is beyond the scope of this article.
If conventional IPOM technique is indicated, the hernia defect can be accurately measured either by placement of spinal needles through the abdominal wall, or using a small disposable ruler that can be easily placed intraperitoneally. The size of the defect can be also calculated directly on the skin, but bear in mind that due to the curvature of the abdominal wall external measurements are always overestimated. This difference increases with the size of the patient and is accentuated by the pneumoperitoneum, reason why from now on the intra-abdominal pressure is maintained at a maximum of 8 mmHg in order to reduce this magnification phenomenon.
As previously mentioned, the aim is to select a prosthesis that extends 3 to 5 cm beyond all edges of the defect, especially if the defect has not been primarily closed. There are several ‘anatomical’ circumstances where a smaller overlap is acceptable and sometimes necessary. For instance when fixing the mesh to the pubic tubercle, under the costal margin, or on the iliac crest. This will be discussed later.
Once the mesh size has been appropriately chosen and orientated its configuration is directly drawn with a sterile pen onto the abdominal Steri-Drape, surrounding the hernia defect. Of importance is that the mesh must be hydrated in a sterile saline solution before being inserted into the abdominal cavity. Dual method of mesh fixation with tackers and transfacial sutures is our preferred technique. On average six sutures of absorbable 2/0 polydioxanone (PDS® II, Ethicon Endo-Surgery, Inc., CA) are initially positioned along the periphery of the prosthesis before the mesh being rolled (collagen side on the inside) and delivered into the abdomen, via the 12 mmHg size port. The two ends of each of those sutures should have the same length in order to facilitate subsequent mesh fixation. One 2/0 PDS is long enough to place two sutures.
Laparoscopic TransAbdominal PrePeritoneal (TAPP) technique
Currently, laparoscopic IPOM is the most frequently used technique, as it is relatively straightforward and requires reasonably short operating time. A concern is that following hernia repair adhesions may still develop in a significant number of patients and irrespective of the materials used , even though Chelala et al.  have reported a merely 11 % incidence of serosal adhesions formation with the polyester-based mesh Parietex™ Composite (precursor of the PCOx Mesh), in 85 redo surgeries after LRVIH. Nevertheless, ideally the prosthetic material should be placed in the extraperitoneal or retrorectus space as per the Rives-Stoppa principles [7, 8], but the laparoscopic TAPP approach is technically more challenging and will require longer operating time. In those situations I would recommend placement of an IDC at the beginning of the procedure. In the few cases where the mesh could be entirely secured in the extraperitoneal space, then a simple uncoated mesh would be sufficient (and cheaper) as there would be no potential risk of adhesion formation with the underlying viscus, separated by the peritoneum. The TAPP technique is mainly used in the lower abdomen when the hernia extends towards the symphysis pubis following a low midline laparotomy or Pfannenstiel incision post gynaecological surgery. Indeed, in these circumstances it is imperative to enter the extraperitoneal space, push back the bladder and expose the major vascular and nerve structures in the pelvis, otherwise the mesh cannot be properly and safely secured inferiroly.
On some occasions it might be also necessary to extend our blunt dissection laterally towards the deep inguinal ring and posteriorly with exposure of part of the psoas muscle and the genito-femoral nerve that runs downwards on it. In those cases I recommend identification and reduction of any associated inguinal hernia, and in female patients to divide the round ligament (corresponds embryologically to the vas deferens). Once divided, the proximal end of the ligament may be secured on with an Endoloop of PDS if judged necessary as it may sometimes bleed. After completion of the dissection, the ‘triangle of doom’ and to a lesser degree the ‘triangle of pain’ will be exposed and it is very important to be familiar with those two anatomical danger zones before considering placement of the prosthetic mesh. The ‘triangle of doom’ is bordered medially by the vas deferens, laterally by the gonadal vessels and inferiorly by the reflected peritoneum, with its apex corresponding to the deep inguinal ring (Fig. 10d). It contains the external iliac vessels, the deep circumflex iliac vein as well as the genital branch of the genito-femoral nerve. The ‘triangle of pain’ lies laterally from this, delimited medially by the gonadal vessels, superiorly by the iliopubic tract and laterally by the reflected peritoneum. Within this triangle lies the femoral branch of the genito-femoral nerve, the lateral cutaneous femoral nerve of the thigh and more deeply the femoral nerve . No sutures or stapling material can be used in this region for mesh fixation.
When performing a laparoscopic TAPP repair I also prefer to primarily close the defect if feasible, select a PCOx mesh and fix the upper part of the prosthesis with sutures and absorbable Sorbafix™, similar to the IPOM-Plus technique. Caudally, the medial part of the PCOx mesh is anchored onto the pubic symphysis, but both infero-lateral aspects of the mesh have to be fixed differently as it is unsafe to use either stapling or suturing devices in this region. In this situation and similar to our previously published TEP technique for inguinal hernia , I secure this part of the prosthetic mesh with 4mls of fibrin glue (TISSEEL [Fibrin Sealant], Baxter, Deerfield, Il, USA) uniformly sprayed (Fig. 10e). When adequately positioned the mesh should stretch out from the retropubic space, lean against the posterior aspect of the superior pubic ramus inferiorly, and lying infero-laterally over the psoas muscle. This way, the mesh will also reinforce the Hesselbach’s triangle and cover all the potential hernia sites in this region; namely direct, indirect, femoral and obturator. Once the PCOx mesh has been completely secured the previously incised parietal peritoneum is closed, thus partially covering the distal end of the prosthesis as shown on Fig. 10f.
Although quite uncommon, incisional hernia of the subxiphoid region may also occur following a median sternotomy, insertion of a chest tube in the mediastinum or simply after a laparoscopic procedure (such as cholecystectomy) where a 10mm working port has been placed in the epigastrium. The difficulty of the repair is mainly due to the close proximity of the rib cage and xiphoid process, but also the diaphragm, pericardium and pleural cavity. In this case, it is preferable to have the patient in a lithotomy and reverse Trendelenburg position, and perform part of the dissection from between the legs with the TV monitor located cephalad towards the right shoulder. The retroxiphoid space must be entered to allow a subfascial position and adequate mesh overlap. To do so, and following adequate dissection of the ligamentum teres of the liver and falciform ligament (Fig. 3b, c) I prefer the use of a harmonic scalpel. Indeed this device is safe and offers a sharp tissue dissection if deemed necessary, with better haemostasis. Furthermore, the potential risk of small bowel injury is almost inexistent as the anterior surface of the liver generally prevents intestinal incarceration . Once the posterior lamina of the rectus sheath that inserts on the posterior side of the xiphoid process has been divided, subsequent development of the retrosternal space can be achieved by blunt dissection.
Laparoscopic IPOM-plus and Totally Extraperitoneal (TEP) techniques for Spigelian hernia
Quite often the defect is small and I therefore recommend primary closure with either Endoloops of PDS (Fig. 14c, d) or figure-of-eight of absorbable 1PDS as previously described (Fig. 4). Using a standard flat polyester or polypropylene mesh for hernia reinforcement would be sufficient as there is no risk of adhesion formation with underlying viscus, but I still prefer to use the round-shaped 9cm diameter PCOx mesh (occasionally the 12cm one) as it is soft, tends to spontaneously stick onto the abdominal cavity, pre-shaped and already containing two stay sutures (Fig. 14e, f). Furthermore, the mesh-to-defect ratio is such that the smaller round PCOx size is more than adequate. Finally, the mesh is preferentially secured with absorbable tacks, but in some occasions where the defect is very small fibrin glue alone may also be sufficient.
Whatever the technique used IPOM, IPOM-Plus, TAPP, or TEP, after satisfactory repair haemostasis is checked and preferentially no drain is used. The pneumoperitoneum (or extra-) is totally deflated under direct vision, local anesthetic is injected into the wounds and the skin closed with interrupted subcuticular 3/0 Caprosyn stitches. Steristrips and waterproof dressing are kept intact for five days. All patients are fitted with an abdominal binder just prior to being extubated and once fully awake they can resume a normal diet, and are able to walk freely if comfortable.
Intravenous antibiotic is generally given for the first 24 h and low molecular weight heparin ceased at the discretion of the operating surgeon. In some occasions, the patient might be discharged home (generally within two days) on extended VTE prophylaxis for few weeks. He is prescribed with simple analgesia and will need to remain on light duties with no straining or heavy lifting for a minimum of 3–4 weeks. Ideally I like to keep the binder tight 24/7 for two weeks and then daytime for another 4 weeks minimum. In our experience this significantly reduces the level of post-operative pain and also provides additional external support during the healing process. Follow up is usually organized at two weeks, six weeks and 3 months post-operatively. If deemed necessary, subsequent post-operative review is also scheduled until patients are symptom free and on some occasions a repeat abdominal CT-scan is organized at 6-month post-operative.
Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated.
- Burger JW, Luijendijk RW, Hop WC, Halm JA, Verdaasdonk EG, Jeekel J. Long-term follow-up of a randomized controlled trialof suture versus mesh repair of incisional hernia. Ann Surg. 2004;240:578–83.PubMedPubMed CentralGoogle Scholar
- Luijendijk RW, Hop WC, van den Tol MP, de Lange DC, Braaksma MM, Ijzermans JN, et al. A comparison of suture repair with mesh repair for incisional hernia. N Engl J Med. 2000;343:392–8.View ArticlePubMedGoogle Scholar
- Moreno-Egea A, Carrillo-Alcaraz A, Aguayo-Albasini JL. Is the outcome of laparoscopic incisional hernia repair affected by defect size? A prospective study. Am J Surg. 2012;203:87–94.View ArticlePubMedGoogle Scholar
- Sharma A, Mehrotra M, Khullar R, Soni V, Baijal M, Chowbey PK. Laparoscopic ventral/incisional hernia repair: a single centre experience of 1,242 patients over a period of 13 years. Hernia. 2011;15:131–9.View ArticlePubMedGoogle Scholar
- Edwards 2nd CC, Bailey RW. Laparoscopic hernia repair: the learning curve. Surg Laparosc Endosc Percutan Tech. 2000;10:149–53.View ArticlePubMedGoogle Scholar
- Lal P, Kaila RK, Chander J, Ramteke VK. Laparoscopic total extraperitoneal (TEP) inguinal hernia repair: overcoming the learning curve. Surg Endosc. 2004;18:642–5.View ArticlePubMedGoogle Scholar
- Rives J, Pire JC, Flament JB, Palot JP, Body C. Treatment of large eventrations. New therapeutic indications a propos of 322 cases. Chirurgie. 1985;111:215–25.PubMedGoogle Scholar
- Stoppa RE. The treatment of complicated groin and incisional hernias. World J Surg. 1989;13:545–54.View ArticlePubMedGoogle Scholar
- LeBlanc KA, Booth WV. Laparoscopic repair of incisional abdominal hernias using expanded polytetrafluoroethylene: preliminary findings. Surg Laparosc Endosc. 1993;3:39–41.PubMedGoogle Scholar
- Schumpelick V, Fitzgibbons RJ. Recurrent hernia. Prevention and treatment. Berlin Heidelberg: Springer-Verlag; 2007. p. 223–54.View ArticleGoogle Scholar
- Kingsnorth A. The management of incisional hernia. Ann R Coll Surg Engl. 2006;88:252–60.View ArticlePubMedPubMed CentralGoogle Scholar
- Palanivelu C, Jani KV, Senthilnathan P, Parthasarathi R, Madhankumar MV, Malladi VK. Laparoscopic sutured closure with mesh reinforcement of incisional hernias. Hernia. 2007;11:223–8.View ArticlePubMedGoogle Scholar
- Tanner J, Moncaster K, Woodings D. Preoperative hair removal: a systematic review. J Perioper Pract. 2007;17(118–121):124–32.Google Scholar
- Bedi AP, Bhatti T, Amin A, Zuberi J. Laparoscopic incisional and ventral hernia repair. J Minim Access Surg. 2007;3:83–90.View ArticlePubMedPubMed CentralGoogle Scholar
- Heniford BT, Park A, Ramshaw BJ, Voeller G. Laparoscopic repair of ventral hernias. Nine years’ experience with 850 consecutive hernias. Ann Surg. 2003;238:391–400.PubMedPubMed CentralGoogle Scholar
- Bittner R, Bingener-Casey J, Dietz U, Fabian M, Ferzli GS, Fortelny RH, et al. Guidelines for laparoscopic treatment of ventral and incisional abdominal wall hernias (International Endohernia Society[IEHS])-Part 1. Surg Endosc. 2014;28:2–29.View ArticlePubMedGoogle Scholar
- Moazzez A, Mason RJ, Katkhouda N. A new technique for minimally invasive abdominal wall reconstruction of complex incisional hernias: totally laparoscopic component separation and incisional hernia repair. Surg Technol Int. 2010;20:185–91.PubMedGoogle Scholar
- Orenstein SB, Dumeer JL, Monteagudo J, Poi MJ, Novitsky YW. Outcomes of laparoscopic ventral hernia repair with routine defect closure using “shoelacing” technique. Surg Endosc. 2011;25:1452–7.View ArticlePubMedGoogle Scholar
- Berney CR. The Endoloop technique for the primary closure of direct inguinal hernia defect during the endoscopic totally extraperitoneal approach. Hernia. 2012;16:301–5.View ArticlePubMedGoogle Scholar
- Doctor HG. Evaluation of various prosthetic materials and newer meshes for hernia repairs. J Minim Access Surg. 2006;2:110–6.View ArticlePubMedPubMed CentralGoogle Scholar
- Bittner R, Bingener-Casey J, Dietz U, Fabian M, Ferzli GS, Fortelny RH, et al. Guidelines for laparoscopic treatment of ventral and incisional abdominal wall hernias (International Endohernia Society [IEHS])-Part 2. Surg Endosc. 2014;28:353–79.View ArticlePubMedGoogle Scholar
- Morales-Conde S, Cadet H, Cano A, Bustos M, Martin J, Morales-Mendez S. Laparoscopic ventral hernia repair without sutures-double crown technique: our experience after 140 cases with a mean follow-up of 40 months. Int Surg. 2005;90(3 Suppl):S56–62.PubMedGoogle Scholar
- Bittner R, Bingener-Casey J, Dietz U, Fabian M, Ferzli G, Fortelny R, et al. Guidelines for laparoscopic treatment of ventral and incisional abdominal wall hernias (International Endohernia Society [IEHS])-Part III. Surg Endosc. 2014;28:380–404.View ArticlePubMedGoogle Scholar
- Chelala E, Debardemaeker Y, Elias B, Charara F, Dessily M, Allé JL. Eighty-five redo surgeries after 733 laparoscopic treatments for ventral and incisional hernia: adhesion and recurrence analysis. Hernia. 2010;14:123–9.View ArticlePubMedGoogle Scholar
- Putnis S, Berney CR. Totally extraperitoneal repair of inguinal hernia: techniques and pitfalls of a challenging procedure. Langenbeck’s Arch Surg. 2012;397:1343–51.View ArticleGoogle Scholar
- Conze J, Prescher A, Kisielinski K, Klinge U, Schumpelick V. Technical consideration for subxiphoidal incisional hernia repair. Hernia. 2005;9:84–7.View ArticlePubMedGoogle Scholar