No difference in outcomes between large- and small-pore meshes in a prospective, randomized, multicenter trial investigating open retromuscular meshplasty for incisional hernia repair.

STUDY DESIGN: A randomized, controlled, prospective multicenter clinical trial with a parallel group design was initiated in eight surgical centers to compare a large-pore polypropylene mesh (Ultrapro®) to a small-pore polypropylene mesh (Premilene®) within a standardized retromuscular meshplasty for incisional hernia repair. METHODS: Between 2004 and 2006, patients with a fascial defect with a minimum diameter of 4 cm after vertical midline laparotomy were recruited for the trial. Patients underwent retromuscular meshplasty with either a large-pore or a small-pore mesh to identify the superiority of the large-pore mesh. Follow-up visits were scheduled at 5 and 21 days and 4, 12, and 24 months after surgery. A clinical examination, a modified short form 36 (SF-36®), a daily activity questionnaire, and an ultrasound investigation of the abdominal wall were completed at every follow-up visit. The primary outcome criterion was foreign body sensation at the 12-month visit, and the secondary endpoint criteria were the occurrence of hematoma, seroma, and chronic pain within 24 months postoperatively. RESULTS: In 8 centers, 181 patients were included in the study. Neither foreign body sensation within the first year after surgery (27.5% Ultrapro®, 32.2% Premilene®) nor the time until the first occurrence of foreign body sensation within the first year was significantly different between the groups. Regarding the secondary endpoints, no significant differences could be observed. At the 2-year follow-up, recurrences occurred in 5 Ultrapro® patients (5.5%) and 4 Premilene® patients (4.4%). CONCLUSION: Despite considerable differences in theoretical and experimental works, we have not been able to identify differences in surgical or patient-reported outcomes between the use of large- and small-pore meshes for retromuscular incisional hernia repair. TRIAL REGISTRATION: Clinical Trials NCT04961346 (16.06.2021) retrospectively registered.

[Diagnostic algorithm "FAI and sports hernia" : Results of the consensus meeting for groin pain in athletes]. / Diagnostik-Algorithmus Grenzbereich "FAI und Sportlerleiste" : Ergebnisse des Konsensustreffens Leistenschmerz beim Sportler.

As a result of the complexity and diversity of diseases in the region of the groin, differentiation of the various soft-tissue and bone pathologies remains a challenge for differential diagnosis in routine clinical practice. In the case of athletes with pain localized in the area of the groin, femoroacetabular impingement (FAI) and athlete's groin must be considered as important causes of the groin pain, whereby the common occurrence of double pathologies further complicates diagnosis. Despite the importance of groin pain and its differential diagnoses in everyday clinical practice, there has been a lack of recognized recommendations for diagnostic procedure to date. To this end, a consensus meeting was held in February 2017, in which a group composed equally of groin and hip surgeons took part. With the formulation of recommendations and the establishment of a practicable diagnostic path, colleagues that are involved in treating such patients should be sensitized to this issue and the quality of the diagnosis of groin pain improved in routine clinical practice.

Classification of Rectus Diastasis-A Proposal by the German Hernia Society (DHG) and the International Endohernia Society (IEHS).

Introduction: Recently, the promising results of new procedures for the treatment of rectus diastasis with concomitant hernias using extraperitoneal mesh placement and anatomical restoration of the linea alba were published. To date, there is no recognized classification of rectus diastasis (RD) with concomitant hernias. This is urgently needed for comparative assessment of new surgical techniques. A working group of the German Hernia Society (DHG) and the International Endohernia Society (IEHS) set itself the task of devising such a classification. Materials and Methods: A systematic search of the available literature was performed up to October 2018 using Medline, PubMed, Scopus, Embase, Springer Link, and the Cochrane Library. A meeting of the working group was held in May 2018 in Hamburg. For the present analysis 30 publications were identified as relevant. Results: In addition to the usual patient- and technique-related influencing factors on the outcome of hernia surgery, a typical means of rectus diastasis classification and diagnosis should be devised. Here the length of the rectus diastasis should be classified in terms of the respective subxiphoidal, epigastric, umbilical, infraumbilical, and suprapubic sectors affected as well as by the width in centimeters, whereby W1 < 3 cm, W2 = 3- ≤ 5 cm, and W3 > 5 cm. Furthermore, gender, the concomitant hernias, previous abdominal surgery, number of pregnancies and multiple births, spontaneous birth or caesarian section, skin condition, diagnostic procedures and preoperative pain rate and localization of pain should be recorded. Conclusion: Such a unique classification is needed for assessment of the treatment results in patients with RD.

Polypropylene mesh implantation for hernia repair causes myeloid cell-driven persistent inflammation.

Polypropylene meshes that are commonly used for inguinal hernia repair may trigger granulomatous foreign body reactions. Here, we show that asymptomatic patients display mesh-associated inflammatory granulomas long after surgery, which are dominated by monocyte-derived macrophages expressing high levels of inflammatory activation markers. In mice, mesh implantation by the onlay technique induced rapid and strong myeloid cell accumulation, without substantial attenuation for up to 90 days. Myeloid cells segregated into distinct macrophage subsets with separate spatial distribution, activation profiles, and functional properties, showing a stable inflammatory phenotype in the tissue surrounding the biomaterial and a mixed, wound-healing phenotype in the surrounding stromal tissue. Protein mass spectrometry confirmed the inflammatory nature of the foreign body reaction, as characterized by cytokines, complement activation, and matrix-modulating factors. Moreover, immunoglobulin deposition increased over time around the implant, arguing for humoral immune responses in association with the cell-driven inflammation. Intravital multiphoton microscopy revealed a high motility and continuous recruitment of myeloid cells, which is partly dependent on the chemokine receptor CCR2. CCR2-dependent macrophages are particular drivers of fibroblast proliferation. Thus, our work functionally characterizes myeloid cell-dependent inflammation following mesh implantation, thereby providing insights into the dynamics and mechanisms of foreign body reactions to implanted biomaterials.

Time-dependent changes of magnetic resonance imaging-visible mesh implants in patients.

OBJECTIVES: Shrinkage and deformation of mesh implants used for hernia treatment can be the cause of long-term complications. The purpose of this study was to quantify noninvasively time-dependent mesh shrinkage, migration, and configuration changes in patients who were surgically treated for inguinal hernia using magnetic resonance imaging (MRI)-visible mesh implants. MATERIALS AND METHODS: In an agarose phantom, meshes in different shrinkage and folding conditions were used to validate the quantification process. Seven patients who were surgically (3 bilaterally) treated for inguinal hernia using iron-loaded mesh implants were prospectively examined using MRI. Gradient echo sequences in sagittal and transverse orientations were performed on day 1 after surgery and at day 90. The mesh-induced signal voids were semiautomatically segmented and a polygonal surface model was generated. A comparison of area and centroid position was performed between the 2 calculated surfaces (day 1 vs day 90). RESULTS: The phantom study revealed a maximum deviation of 3.6% between the MRI-based quantification and the actual mesh size. All 10 implants were successfully reconstructed. The mean (SD) observed mesh shrinkage 90 days after surgery was 20.9% (7.1%). The mean (SD) centroid movement was 1.17 (0.47) cm. Topographic analysis revealed mean (SD) local configuration changes of 0.23 (0.03) cm. CONCLUSIONS: In this study, significant mesh shrinkage (20.9%) but marginal changes in local mesh configuration occurred within 90 days after mesh implantation. Centroid shift of the mesh implant can be traced back to different patient positioning and abdominal distension. The developed algorithm facilitates noninvasive assessment of key figures regarding MRI-visible meshes. Consequently, it might help to improve mesh technology as well as surgical skills.

First in-human magnetic resonance visualization of surgical mesh implants for inguinal hernia treatment.

OBJECTIVES: Until today, there have been no conventional imaging methods available to visualize surgical mesh implants and related complications. In a new approach, we incorporated iron particles into polymer-based implants and visualized them by magnetic resonance imaging (MRI).After clinical approval of such implants, the purposes of this study were to evaluate the MRI conspicuity of such iron-loaded mesh implants in patients treated for inguinal hernias and to assess the immediate postsurgical mesh configuration. MATERIALS AND METHODS: Approved by the ethics committee, in this prospective cohort study, 13 patients (3 patients with bilateral hernia treatment) were surgically treated for inguinal hernia receiving iron-loaded mesh implants between March and October 2012. The implants were applied via laparoscopic technique (transabdominal preperitoneal technique; n = 8, 3 patients with bilateral hernia treatment) or via open surgical procedure (Lichtenstein surgery; n = 5). Magnetic resonance imaging was performed 1 day after the surgery at a 1.5-T scanner (Achieva; Philips, Best, The Netherlands) with a 16-channel receiver coil using 3 different gradient echo sequences (first gradient echo sequence, second gradient echo sequence, and third gradient echo sequence [GRE1-3]) and 1 T2-weighted turbo spin-echo sequence (T2wTSE). Three radiologists independently evaluated mesh conspicuity and diagnostic value with respect to different structures using a semiquantitative scoring system (1, insufficient; 2, sufficient; 3, good; 4, optimal). Mesh deformation and coverage of the hernia were visually assessed and rated using a 5-point semiquantitative scoring system. Statistical analysis was performed using mixed models and linear contrast. RESULTS: All 16 implants were successfully visualized by MRI. On gradient echo sequences, the mesh is clearly delineated as a thick hypointense line. On T2wTSE, the mesh was depicted as a faint hypointense line, which was difficult to identify. The first gradient echo sequence was rated best for visual conspicuity (mean [SD], 3.8 [0.4]). T2-weighted turbo spin-echo sequence was preferred for evaluation of the surrounding anatomy (mean [SD], 3.7 [0.3]). For the combined assessment of both mesh and anatomy, GRE3 was rated best (mean [SD], 2.9 [0.7]). Local air slightly reduced mesh delineation (lowest mean [SD] rating, 2.9 [0.7] for GRE3). Overall, in both implantation techniques, the meshes exhibited mild to moderate deformations (mean [SD], 3.3 [0.4], 3.1 [0.3], and 2.8 [0.3] on average with open technique, 2.7 [0.3], 2.7 [0.2], and 2.3 [0.3] with laparoscopic technique). Coverage of the hernia was achieved in 15 of the 16 implants. CONCLUSIONS: Combining iron-loaded implants and MRI, we achieved mesh visualization for the first time in patients. For MRI protocol, we propose a combination of different gradient echo sequences and T2-weighted turbo spin-echo sequences: first gradient echo sequence for mesh configuration, T2wTSE for anatomy assessment, and GRE3 for evaluation of hernia coverage and mesh localization. Using our approach, MRI could become a noninvasive alternative to open surgical exploration if mesh-related complications were suspected.

Intrathoracic versus cervical anastomosis after resection of esophageal cancer: a matched pair analysis of 72 patients in a single center study.

BACKGROUND: The aim of this study was to analyze the early postoperative outcome of esophageal cancer treated by subtotal esophageal resection, gastric interposition and either intrathoracic or cervical anastomosis in a single center study. METHODS: 72 patients who received either a cervical or intrathoracic anastomosis after esophageal resection for esophageal cancer were matched by age and tumor stage. Collected data from these patients were analyzed retrospectively regarding morbidity and mortality rates. RESULTS: Anastomotic leakage rate was significantly lower in the intrathoracic anastomosis group than in the cervical anastomosis group (4 of 36 patients (11%) vs. 11 of 36 patients (31%); p = 0.040). The hospital stay was significantly shorter in the intrathoracic anastomosis group compared to the cervical anastomosis group (14 (range 10-110) vs. 26 days (range 12 - 105); p = 0.012). Wound infection and temporary paresis of the recurrent laryngeal nerve occurred significantly more often in the cervical anastomosis group compared to the intrathoracic anastomosis group (28% vs. 0%; p = 0.002 and 11% vs. 0%; p = 0.046). The overall In-hospital mortality rate was 6% (4 of 72 patients) without any differences between the study groups. CONCLUSIONS: The present data support the assumption that the transthoracic approach with an intrathoracic anastomosis compared to a cervical esophagogastrostomy is the safer and more beneficial procedure in patients with carcinoma of the lower and middle third of the esophagus due to a significant reduction of anastomotic leakage, wound infection, paresis of the recurrent laryngeal nerve and shorter hospital stay.

Influence of two different levels of intra-abdominal hypertension on bacterial translocation in a porcine model.

BACKGROUND: The purpose of the present study was to quantify bacterial translocation to mesenteric lymph nodes due to different levels of intra-abdominal hypertension (IAH; 15 vs. 30 mmHg) lasting for 24 h in a porcine model. METHODS: We examined 18 anesthetized and intubated pigs (52.3 ± 4.7 kg) which were randomly allocated to three experimental groups (each n = 6) and studied over a period of 24 h. After preparation and establishing a steady state, the intra-abdominal pressure (IAP) was increased stepwise to 30 mmHg in six animals using a carbon dioxide (CO2) insufflator (IAP-30 group). In the second group, IAP was increased to 15 mmHg (IAP-15 group), while IAP remained unchanged in another six pigs (control group). Using a pulse contour cardiac output (PiCCO®) monitoring system, hemodynamic parameters as well as blood gases were recorded periodically. Moreover, peripheral and portal vein blood samples were taken for microbiological examinations. Lymph nodes from the ileocecal junction were sampled during an intra-vital laparotomy at the end of the observational period. After sacrificing the animals, bowel tissue samples and corresponding mesenteric lymph nodes (MLN) were extracted for histopathological and microbiological analyses. RESULTS: Cardiac output decreased in all groups. In IAP-30 animals, volumetric preload indices significantly decreased, while those of IAP-15 pigs did not differ from those of controls. Under IAH, the mean arterial pressure (MAP) in the IAP-30 group declined, while MAP in the IAP-15 group was significantly elevated (controls unchanged). PO2 and PCO2 remained unchanged. The grade of ischemic damage of the intestines (histopathologically quantified using the Park score) increased significantly with different IAH levels. Accordingly, the amount of translocated bacteria in intestinal wall specimens as well as in MLN significantly increased with the level of IAH. Lymph node cultures confirmed the relation between bacterial translocation (BT) and IAP. The most often cultivated species were Escherichia coli, Staphylococcus, Clostridium, Pasteurella, and Streptococcus. Bacteremia was detected only occasionally in all three groups (not significantly different) showing gut-derived bacteria such as Proteus, Klebsiella, and E. coli spp. CONCLUSION: In this porcine model, a higher level of ischemic damage and more BT were observed in animals subjected to an IAP of 30 mmHg when compared to animals subjected to an IAP of 15 mmHg or controls.

Gentamicin supplemented polyvinylidenfluoride mesh materials enhance tissue integration due to a transcriptionally reduced MMP-2 protein expression.

BACKGROUND: A beneficial effect of gentamicin supplemented mesh material on tissue integration is known. To further elucidate the interaction of collagen and MMP-2 in chronic foreign body reaction and to determine the significance of the MMP-2-specific regulatory element (RE-1) that is known to mediate 80% of the MMP-2 promoter activity, the spatial and temporal transcriptional regulation of the MMP-2 gene was analyzed at the cellular level. METHODS: A PVDF mesh material was surface modified by plasma-induced graft polymerization of acrylic acid (PVDF+PAAc). Three different gentamicin concentrations were bound to the provided active sites of the grafted mesh surfaces (2, 5 and 8 µg/mg). 75 male transgenic MMP-2/LacZ mice harbouring the LacZ reporter gene under control of MMP-2 regulatory sequence -1241/+423, excluding the RE-1 were randomized to five groups. Bilateral of the abdominal midline one of the five different meshes was implanted subcutaneously in each animal. MMP-2 gene transcription (anti-ß-galactosidase staining) and MMP-2 protein expression (anti-MMP-2 staining) were analyzed semiquantitatively by immunohistochemistry 7, 21 and 90 days after mesh implantation. The collagen type I/III ratio was analyzed by cross polarization microscopy to determine the quality of mesh integration. RESULTS: The perifilamentary ß-galactosidase expression as well as the collagen type I/III ratio increased up to the 90th day for all mesh modifications, whereas no significant changes could be observed for MMP-2 protein expression between days 21 and 90. Both the 5 and 8 µg/mg gentamicin group showed significantly reduced levels of ß-galactosidase expression and MMP-2 positive stained cells when compared to the PVDF group on day 7, 21 and 90 respectively (5 µg/mg: p < 0.05 each; 8 µg/mg: p < 0.05 each). Though the type I/III collagen ratio increased over time for all mesh modifications significant differences to the PVDF mesh were only detected for the 8 µg/mg group at all 3 time points (p < 0.05 each). CONCLUSIONS: Our current data indicate that lack of RE-1 is correlated with increased mesh induced MMP-2-gene expression for coated as well as for non-coated mesh materials. Gentamicin coating reduced MMP-2 transcription and protein expression. For the 8 µg/mg group this effect is associated with an increased type I/III collagen ratio. These findings suggest that gentamicin is beneficial for tissue integration after mesh implantation, which possibly is mediated via RE-1.

Biocompatibility of prosthetic meshes in abdominal surgery.

Surgical meshes today represent a group of implants mainly used for hernia repair. Modern hernia surgery is no longer imaginable without the application of these special biomaterials leading to millions of implantations each year worldwide. Because clinical trials are insufficient to evaluate the distinct effects of modified mesh materials in regard to tissue biocompatibility and functionality, a basic understanding of the physicochemical properties of mesh materials, as well as the underlying cause for hernia formation, is essential for a rational selection of the most appropriate device. The most important properties of meshes were found to be the type of filament, tensile strength, and experimental data, which indicate that particularly the mesh's porosity is of outstanding importance.

Abdominal compartment syndrome in childhood: diagnostics, therapy and survival rate.

PURPOSE: The abdominal compartment syndrome (ACS) in childhood is a rare but dire disease if diagnosed delayed and treated improperly. The mortality amounts up to 60% (Beck et al. in Pediatr Crit Care Med 2:51-56, 2001). ACS is defined by a sustained rise of the intraabdominal pressure (IAP) together with newly developed organ dysfunction. The present study reports on 28 children with ACS to evaluate its potential role in the diagnosis, treatment and outcome of ACS. METHODS: Retrospectively, medical reports and outcome of 28 children were evaluated who underwent surgical treatment for ACS. The diagnosis of ACS was established by clinical signs, intravesical pressure-measurements and concurrent organ dysfunction. RESULTS: Primary ACS was found in 25 children (89.3%) predominantly resulting from polytrauma and peritonitis. Three children presented secondary ACS with sepsis (2 cases) and combustion (1 case) being the underlying causative diseases. Therapy of choice was the decompression of the abdominal cavity with implantation of an absorbable Vicryl(®) mesh. In 18 cases the abdominal cavity could be closed later, while in the other ten cases granulation of the mesh was allowed. The overall survival rate was 78.6% (22 of 28 children). The cause of death in the remaining six cases (21.4%) was sepsis with multiorgan failure. CONCLUSION: Our results suggest that early establishment of the specific diagnosis of ACS followed by swift therapy with reduction of intraabdominal hypertension is essential in order to further reduce the high mortality rate associated with this condition.

New polymer for intra-abdominal meshes--PVDF copolymer.

PURPOSE: Full tissue integration without adhesion formation is still a challenge for intra-abdominal mesh materials. Purpose of this study was to investigate the adhesive potential and fibrocollagenous ingrowth of a polymer blend of polyvinylidene fluoride and hexafluorpropylene (co-PVDF), an established suture material in vascular surgery, when placed as a mesh in the intra-abdominal position. The results were compared with a matching polypropylene (PP) mesh. METHODS: In an established rabbit model, mesh implantation was performed by laparoscopy in the intraperitoneal onlay mesh technique. After 7, 21, and 90 days the degree of adhesion formation, foreign body reaction, bridging, and shrinkage of mesh area were investigated. RESULTS: In the early phase after 7 and 21 days we found significantly more adhesions for PP, but no differences after 90 days. Analysis of tissue reaction showed a significantly lower fibrotic reaction for co-PVDF. The degree of shrinkage revealed no significant difference. CONCLUSION: Large-pore PP and co-PVDF-meshes showed comparable good results in the intra-abdominal position, with a reduced inflammatory tissue reaction for co-PVDF. Large pore meshes should be considered an alternative for the development of intraperitoneal onlay meshes.

New technical approach for the repair of an abdominal wall defect after a transverse rectus abdominis myocutaneous flap: a case report.

INTRODUCTION: Breast reconstruction with autologous tissue transfer is now a standard operation, but abnormalities of the abdominal wall contour represent a complication which has led surgeons to invent techniques to minimize the morbidity of the donor site. CASE PRESENTATION: We report the case of a woman who had bilateral transverse rectus abdominis myocutaneous flap (TRAM-flap) breast reconstruction. The surgery led to the patient developing an enormous abdominal bulge that caused her disability in terms of abdominal wall and bowel function, pain and contour. In the absence of rectus muscle, the large defect was repaired using a combination of the abdominal wall component separation technique of Ramirez et al and additional mesh augmentation with a lightweight, large-pore polypropylene mesh (Ultrapro). CONCLUSION: The procedure of Ramirez et al is helpful in achieving a tension-free closure of large defects in the anterior abdominal wall. The additional mesh augmentation allows reinforcement of the thinned lateral abdominal wall.

Improved collagen type I/III ratio at the interface of gentamicin-supplemented polyvinylidenfluoride mesh materials.

BACKGROUND: Formation of recurrent inguinal and incisional hernia shows an underlying defect in the wound-healing process with an insufficient quality of scar formation. Even after mesh repair an altered collagen formation and insufficient mesh integration has been found as main reason for recurrences. Therefore, the development of bioactive mesh materials to achieve a local modification of the scar formation to improve patients outcome is advisable. MATERIALS AND METHODS: A polyvinylidenfluoride mesh material (PVDF) was constructed and surface modified by plasma-induced graft polymerization of acrylic acid (PVDF + PAAc). Surface supplementation was sought by binding of gentamicin to the provided active sites of the grafted mesh surfaces (PVDF+PAAc+Gentamicin). In vivo modulation of collagen formation was evaluated in a standardized animal model where an abdominal wall replacement was performed in 45 Sprague-Dawley rats. Seven, 21, and 90 days after mesh implantation, collagen/protein ratio and the collagen type I/III ratio as well as the expression of type I alpha 1 collagen mRNA (SYBR Green real-time RT-PCR) were analyzed at the perifilamentary region. Additionally, expression of matrix metalloproteinases (MMP-8/-13) has been investigated immunohistochemically. RESULTS: Implantation of the PVDF + PAAc + Gentamicin mesh induced a significantly decreased expression of MMP-8 and MMP-13 at the interface 21 and 90 days after implantation compared to the other groups. Whereas no significant effect was observed comparing the overall collagen/protein ratio, the quality of collagen formation expressed by the collagen type I/III ratio showed significantly higher ratios around the PVDF + PAAc + Gentamicin mesh 21 and 90 days after implantation. Correspondingly, an up to 5.3-fold expression of type I alpha 1 collagen mRNA was found. CONCLUSION: The present data confirm that a surface modification of PVDF mesh samples using plasma-induced graft polymerization of acrylic acid and supplementation of gentamicin is able to improve scar quality and mesh integration.

Incisional hernia: challenge of re-operations after mesh repair.

BACKGROUND AND AIMS: The widespread use of meshes for the repair of incisional hernia is currently followed by an increasing number of re-operations. The incidence of incisional hernia recurrence after mesh repair varies between 3 and 32%. The problem of mesh failure and options for another surgical intervention seem rather unattended. METHODS: We present our experience of 77 re-operations after previous mesh repair that were performed between 1995 and 2004 out of a total of 1,070 operations for incisional hernia. The retrospective analysis focused on recurrence in relation to location, material of the previous mesh repair and the surgical procedure to resolve the problem. RESULTS: The locations of the preceding meshes were epifascial as onlays (n=23), retromuscular as sublays (n=46), within the defect as inlays (n=6) or intraperitoneally (n=2). The direction of the incision was vertical medial (n=41) or horizontal crossing the linea semilunaris (n=36). Recurrences after median incisional hernia mesh repair mainly occurred at the cranial border of the mesh subxiphoidal. Except for two patients, all recurrences manifested at the margin of the enclosed mesh. CONCLUSIONS: Re-operation after previous mesh repair is a surgical challenge. The type of revision procedure has to consider the position and material of the previous mesh. In our clinic recurrences, heavyweight polypropylene meshes were mostly treated with mesh exchange and lightweight polypropylene meshes could be treated by extension with a second mesh. In contrast to suture techniques, deficient mesh repairs are more evidently related to technical problems.

Incisional hernia: open techniques.

Even with the routine use of mesh, repairing an incisional hernia is a challenge. Increasing evidence of impaired wound healing in these patients supports routine use of an open prefascial, retromuscular mesh repair. Basic pathophysiologic principles dictate that for a successful long-term outcome and prevention of recurrence a wide overlap underneath healthy tissue is required. The extent of this overlap should be 5 cm in all directions: surrounding the wound closure, subxiphoidal underneath the ribs, below the arcuate line, and retropubic.