G. M. Richter

Helical hydro-CT for diagnosis and staging of gastric carcinoma.

PURPOSEnThe purpose of this work was to define the accuracy of helical hydro-CT (HHCT) in the diagnosis and staging of gastric carcinoma.nnnMETHODnOne hundred twelve patients with gastric carcinoma were preoperatively imaged by HHCT. Gastric distension was achieved by ingestion of up to 1,500 ml of water. Bolus tracking was performed, and peristalsis was minimized by intravenously administered spasmolytics. Contrast material was then injected, and helical scanning was performed at the time of peak enhancement of the liver. CT images were analyzed for tumor infiltration of the gastric wall, and TNM staging criteria were applied according to the International Union Against Cancer (UICC) classification. The results were correlated with histopathologic findings.nnnRESULTSnOne hundred two of 115 (89%) gastric carcinomas were correctly diagnosed by HHCT. Small malignant ulcers (< or =2 cm) that corresponded to early gastric carcinoma were not visible on CT scans. T and N staging accuracies were 51% each; abdominal M staging was correct in 79% of all cases. The positive and negative predictive values of HHCT to foresee curative resection of gastric carcinoma were 75 and 84%, respectively.nnnCONCLUSIONnMural thickening as well as marked contrast enhancement of the gastric wall are firmly related to gastric carcinoma. The accuracy of HHCT is acceptable for M staging but inadequate for local staging of gastric carcinoma. Nonetheless, HHCT is a useful guide for choosing between tumor resection and nonoperative treatment of patients. We therefore recommend HHCT as the method of choice for preoperative imaging of gastric carcinoma.

A New Invasion Score for Determining the Resectability of Pancreatic Carcinomas with Contrast-Enhanced Multidetector Computed Tomography

Objective: It was the aim of this study to evaluate a new infiltration score to determine the resectability of pancreatic carcinomas in preoperative planning. Materials and Methods: Eighty patients with suspected pancreatic tumor were examined prospectively using 16-row spiral CT. The scans were evaluated for the presence of pancreatic carcinoma, peripancreatic tumor extension and vascular invasion using a standardized questionnaire. Invasion of the surgically relevant vessels was evaluated using a new invasion score. The operative and histological findings and the clinical follow-up served as the gold standard. Results: Forty patients had a pancreatic carcinoma, 5 had metastasis of a different primary tumor, and in 35 patients, there was no malignant pancreatic disease. The sensitivity for tumor detection was 100%, with a specificity of 88% for differentiating between malignant and benign pancreatic tumors. Invasion of the surrounding vessels was evaluated correctly using the invasion score, with a sensitivity of 89% and a specificity of 99%. In evaluation of resectability, a sensitivity of 94% and a specificity of 89% were achieved. Conclusion: Using 16-row spiral CT, the invasion score is a valid tool for correctly assessing invasion in relevant vessels in cases of pancreatic carcinoma and for determining resectability.

Assessment of hepatic perfusion in pigs by pharmacokinetic analysis of dynamic MR images

The purpose of this study was to evaluate a new method based on magnetic resonance imaging for the characterization of hepatic perfusion. In nine pigs dynamic MRI was performed before and after partial occlusion of the portal vein. The pharmacokinetic analysis of the contrast enhancement resulted in a set of parameters (amplitude, A; perfusion rate, kp; elimination rate, kel; lag time, tlag) of which kp was expected to correlate with hepatic perfusion. Reference measurements were done with ultrasound flowmeters and with a thermal diffusion probe (TDP). MR perfusion rate kp significantly dropped under partial portal vein occlusion from an average of 11.3 to 4.9 min−1 (P < 0.001), while the difference in amplitude A was not significant. The correlation between kp and the TDP measurement was r = 0.89 (P < 0.001). Pharmacokinetic analysis of MRI contrast enhancement provides a non‐invasive assessment of hepatic perfusion.J. Magn. Reson. Imaging 1999;9:568–572.

Experimental Evaluation of Early and Long-Term Effects of Microparticle Embolization in Two Different Mini-Pig Models. Part I: Kidney

PurposeUsing a pig model: (1) to evaluate the vascular distribution pattern, including the homogeneity and completeness of the intra-arterial microsphere distribution, of 40–120-μm trisacryl-gelatin microspheres (Embospheres) in acute whole-kidney embolization; (2) to evaluate the durability and biocompatibility of 40–120-μm trisacryl-gelatin microspheres (Embospheres) in chronic partial kidney embolization.MethodsTwenty-two animals were divided into four groups: group 1 (nxa0=xa04) underwent total arterial renal occlusion with immediate euthanasia. Groups 2–4 had chronic superselective and partial renal embolization with increasing follow-up times: group 2 (nxa0=xa02), 1 week; group 3 (nxa0=xa07), 4 weeks; and group 4 (nxa0=xa09), 14 weeks. Key endpoints in group 1 were homogeneity and completeness of acute embolizations. In groups 2–4 the key endpoints were durability of embolization and particle-related inflammation in chronic partial embolizations as assessed by quantitative angiography or histomorphometry. A numerical angiographic occlusion score (0.0 to 4.0, where 3.0 is optimal) was developed to assess and quantify the angiographic durability of superselective embolizations (groups 2–4).ResultsIn group 1, a relatively homogeneous distribution of the particles from segmental arteries to the precapillary level was shown by histomorphometry. Some particles reached the glomerular vas afferens (10xa0μm diameter). In groups 2–4, a mild recanalization appeared during follow-up. The immediate average postembolization occlusion score of 3.18xa0±xa00.73 was reduced to 1.44xa0±xa00.73 (statistically significant). Microscopy revealed subtotal necrosis but no foreign body granuloma formation. The intra-arterial appearance of giant cells closely attaching to the surface of the embolic spheres inside the vessel lumen was noted. Vessel walls showed major ischemic reactions.ConclusionMicrospheres 40–120xa0μm in diameter might achieve total occlusion of the arterial kidney vasculature when injected centrally as a result of their fairly homogeneous distribution. Segmental renal infarction occurs after chronic partial embolization despite recanalizations during follow-up. Only mild specific intra-arterial foreign body reactions were found.

In Vitro Comparison of Self-Expanding Versus Balloon-Expandable Stents in a Human Ex Vivo Model

The objective was to compare the radial strength and expansile precision of self-expanding stents and balloon-expandable stents in a human cadaver bifurcation model. Seven different self-expanding (LUMINEXX, JOSTENT SelfX, JOSTENT SelfX hrf, Sinus-Repo, Sinus SuperFlex, Easy Wallstent, SMART) and four different balloon-expandable stent models (Palmaz, Sinus Stent, SAXX Medium, JOSTENT peripheral), each type 10 stents (total n = 110 stents) were implanted into the common iliac arteries of human cadaver corpses. The maximum stent diameter was 10 mm for all models. After stent implantation, the specimens were filled with silicone caoutchouc. After 24 h, the vascular walls including the stents were removed from the hardened casts. Diameters were taken and the weight of the cast cylinders was measured in air and in purified water to calculate the volume of the bodies (according to Archimedes Law) as a relative but precise degree for the radial strength of the implanted stents. The cylindrical casts of the self-expanding stents showed lower mean diameters (8.2 ± 1.0 mm) and mean volumes (0.60 ± 0.14 ml/cm) than in the balloon-expandable stent group (10.1 ± 0.3 mm and 0.71 ± 0.04 ml/cm, respectively; p < 0.01). The nominal maximum diameter of 10 mm was not achieved in any of the self-expanding stents, but this was achieved in more than 70% (29/40) of the balloon-expandable stent specimens (p < 0.05). The variation between achieved volumes was significantly larger in self-expanding (range: 0.23–0.78 ml/cm) than in balloon-expandable stents (range: 0.66–0.81 ml/cm; p < 0.05). Self-expanding stents presented considerably lower radial expansion force and lower degree of precision than balloon-expandable stents.

Morphologic and clinical results of the transjugular intrahepatic portosystemic stent-shunt (TIPSS)

The concept of transjugular intrahepatic portosystemic stent-assisted shunt (TIPSS) using the Palmaz iliac stent has been successfully accomplished in 18 of 24 patients representing a technical success rate of 75%. Fourteen were male, 4 female with a mean age of 60 years (range 34–84 years). According to classification of Child’s and Turcotte, 6 were in stage A, 6 in stage B, and 6 in stage C. Five patients were treated on an emergency basis because of massive active bleeding. In 10 patients the portosystemic tract was created between the middle hepatic vein and the right main stem of the portal vein in 8, and the left main stem in 2 patients. In 8 patients, the shunt was established between the right hepatic vein and the right main branch of the portal vein. The portosystemic gradient in 18 patients was 29.9±6 mm Hg and dropped to an average of 16.9±4 mm Hg after shunt establishment. Within the early postprocedural period of 30 days, 1 patient died of direct complications of the procedure. Because of catheter dislocation, embolization of the percutaneous transhepatic approach to the portal vein after successful shunt “creation” could not be done and was followed by intraabdominal exsanguination. One patient died of an ARDS after TIPSS. A third developed pulmonary infection. In 13 patients, because of hematomas at the puncture site of the transhepatic approach, only the transjugular approach was elected for establishing TIPSS. The mean portosystemic gradient in 18 patients prior to TIPSS was 29±6 mm Hg (range 19–41 mm Hg), dropped to an average of 16.9±4 mm Hg (range 7–21 mm Hg), and showed no significant change 6 months after TIPSS with a pressure of 16±1.8 mm Hg. The 1-year survival rate was 75% (8/12); the 2-year rate was 50% (3/6).

Acute Pancreatitis after Embolization of Liver Tumors: Frequency and Associated Risk Factors

Introduction: Acute pancreatitis (AP) is a rare complication after liver embolization (LE) of primary and secondary liver tumors (approximately 1.7%), but it has a significant morbidity and mortality potential if associated with other complications. It usually develops early within 24 h after the LE procedure. Study Purpose: To calculate the frequency of AP after LE in our institution and to analyze the factors involved in this procedure (anatomical features, embolization materials, cytostatic drugs, technical factors). Materials and Methods: 118 LE (bland embolization and transarterial chemoembolization) were performed in our institution. The study group included 59 patients who met the following inclusion criteria: one or more LE events, with complete pre- and post-interventional laboratory studies including: serum Ca2+, creatinine, blood urea nitrogen, glucose, lactate dehydrogenase, aminotransferases, alkaline phosphatase, amylase, lipase, C-reactive protein, hematocrit and leukocytes. The diagnosis of AP was established according to the criteria of the Atlanta system of classification. For the statistical analysis the association between two response variables (e.g. AP after embolization and risk factor during the embolization, AP after embolization and volume of embolic material) was evaluated using Pearson’s χ2 test and Fisher’s exact test. Results: The calculated frequency of AP after LE in our series was 15.2%. Amylase and lipase were elevated up to 8.7 and 20.1 times, respectively, 24 h after LE. We observed a statistically significantly lower incidence of AP in those patients who received 2 ml or less of embospheres compared with those with an embolization volume of >2 ml (Pearson’s χ2 = 4.5000, Pr = 0.034, Fisher’s exact test = 0.040). Although carboplatin was administered to 7 of 9 of the patients who developed AP after the embolization procedure, there was no statistical significance (Fisher’s exact test = 0.197) for carboplatin as an AP risk factor when compared with all the patients who received this drug (n = 107). Conclusion: Although AP after LE seems to have a multifactorial etiology, both the toxicity of the antineoplastic drugs (carboplatin-related toxicity) as well as direct ischemic mechanisms (non-target embolization, reflux mechanisms) may be the most important causes of the inflammatory pancreatic reaction after LE. We suggest that systematic measurement of serum pancreatic enzymes should be performed in cases of abdominal pain following selective LE and transarterial chemoembolization in order to confirm acute pancreatitis after embolization, which can clinically mimic a postembolization syndrome.

TNM staging of gastrointestinal tumors by hydrosonography: results of a histopathologically controlled study in 60 patients

Abstract.Background: The study is a prospective evaluation of preoperative TNM staging of gastrointestinal tumors by hydrosonography (HUS).nnMethods: Sixty patients with suspected gastric or colorectal cancer underwent HUS for TNM staging. All patients were operated on and the tumors completely removed when possible. HUS findings were correlated with histopathologic staging.nnResults: HUS correctly localized tumors in 75% of patients. T stage accuracy was low for gastric cancers (41%). N staging of gastric cancers was accurate in 68% of all cases and was highly specific (100%). Staging was more accurate for colorectal tumors (70%), especially with respect to infiltration of other structures (sensitivity 100%, specificity 95%). N staging, however, was not reliable, mostly owing to impaired examination conditions.nnConclusion: HUS easily misses tumors of the gastric cardia and distal part of the rectum. T staging of colorectal tumors with HUS is highly accurate, reaching 92% if the tumor is localized. T1 cancers of the stomach tend to be overstaged, and serosal infiltration by gastric cancers is often misjudged. With the exception of cardial gastric and distal rectal cancers, HUS comes close to endosonography for staging gastrointestinal tumors. HUS does not require intraluminal access.nnn

Experimental study of the effectiveness of capillary embolization using contrast-enhanced Ethibloc.

In experimental and clinical use, Ethibloc in combination with 40% glucose preinjection has proven to be of major advantage in tumor embolization. However, its low radiographic contrast is a limiting factor in monitoring its vascular distribution and venous propagation. Various contrast media were tested in order to enhance this contrast in laboratory and animal experiments. Normal rat kidneys (N = 96) and renal tumors, induced by Dimethylnitrosamine (N = 66) were tested as in previous studies. Ethibloc-N was produced by substituting Lipiodol for poppy seed oil which is an ingredient of the original Ethibloc. This proved to be the only embolization medium that combined the excellent properties of the original Ethibloc with increased contrast. All other embolization media tested resulted in new complications such as under-or overembolization and pulmonary embolism.

Die Aortendissektion : Wann operieren, wann endoluminal therapieren?

ZusammenfassungZiel. Darstellung der Heidelberger Ergebnisse zur operativen und interventionellen Therapie der akuten Aortendissektion auf der Basis einer Analyse der Jahre 1999–2000.nn Material und Methode. 93 Patienten mit akuter Aortendissektion wurden im Untersuchungszeitraum durch Herzchirurgie, Gefäßchirurgie und Interventionelle Radiologie mit dem Ziel einer optimierten Stufentherapie behandelt. Entsprechend der klinischen und morphologischen Situation erfolgte die Therapie klassisch offen (Typ A) mit Herz-Lungen-Maschine und seit der Verfügbarkeit großlumiger Stentgrafts endoluminal (Typ B) oder dann konservativ, wenn weder klinisch noch morphologisch eine invasive Therapie erforderlich erschien.nn Ergebnisse. 36 Patienten hatten eine A- und 57 eine B-Dissektion. Fast alle A-Patienten wurden operiert (32), 3 der nichtoperierten verstarben kurz nach Aufnahme. 20 der Patienten mit B-Dissektion wurden therapiert, davon 12 endoluminal. Die Mortalität der A-Patienten lag insgesamt bei 35%, die der B-Patienten bei 15%, wobei allerdings die Mortalität in beiden Gruppen bei den unbehandelten am höchsten war. Die Mortalität der endoluminal behandelten Patienten war 0%. Der wichtigste Mortalitätsfaktor bei allen A-Patienten war die viszerale Ischämie, die trotz erfolgreicher Korrektur der Aorta ascendens durch symptomatischen ”True lumen collaps”, ggf. verschärft durch den Einsatz der Herz-Lungen-Maschine, in der unmittelbar postoperativen Periode entscheidend wurde. Paraplegische Kompliationen traten bei keinem Patient auf. Bei insgesamt 4 Patienten konnten erstmals Kombinationsverfahren aus offener Chirurgie und endoluminaler Korrektur der fortbestehenden (symptomatischen) B-Dissektion erfolgreich eingesetzt werden.nn Diskussion. Die Typ-A-Dissektion kann mit hoher technischer Erfolgsrate behandelt werden. Das Fortbestehen ungünstiger hämodynamischer Verhältnisse in der distalen Aorta begünstigt jedoch Organmalperfusionen. Die erfolgreiche Kombination mit endoluminalen Verfahren legt nahe, ein solches Konzept einer breiteren klinischen Erprobung zu unterziehen. Akute und klinisch symptomatische B-Dissektionen können mit hoher technischer Erfolgsrate endoluminal behandelt werden. Ungelöste Fragen betreffen allerdings noch das endgültige Prothesendesign.AbstractGoal. To demonstrate the Heidelberg results of the previous 2 years in patients referred for acute aortic dissection.nn Material and Methods. 93 patients referred for acute aortic dissection were treated by cardiac surgery, vascular surgery and interventional radiology according to a novel therapeutic algorithm including stent-grafts and combined open and interventional procedures and conservative medical therapy when no malperfusion syndrome was present or patients were considered prohibitive for even minor surgical procedures. Stent-graft placements were done assisted by short term cardiac arrest to facilitate correct device deployment.nn Results. 36 patients presented with type A and the other 57 with type B dissection. 32 of the A patients were operated and 20 of the B patients, respectively. 12 patients with B dissection were treated with stent-grafts. 3 required additional interventional therapy for organ malperfusion. The mortality was 0% in these 12 patients The overall mortality rate in the A group was close to 40% mainly as a result of postoperative organ malperfusion while it was 15% in the B group. In both groups mortality was highest in the respective untreated patient subgroup (3/4 and 8/37, respectively). The main mortality factor was visceral (mesenteric or liver) ischemia. Paraplegic complications occured in neither group. In 4 patients a combined approach applying cardiac surgery of the ascending aorta and endluminal stent-graft placement for the residual B dissection was successfully performed. In one patient this was done simultaneously.nn Discussion. Acute aortic dissection of type A with or without valve involvement, coronary artery ischemia can be treated with high technical success rates. However, remaining distal aortic dissection associated with true lumen collapse and organ malperfusion is the main causative factor for clinical failures. Successful combination of open proximal aortic surgery with endoluminal treatment of residual B dissection encourages further use of this novel approach. Acute B type dissection appears to be effectively and safely treated by endoluminal approach in selected cases. Unsolved questions of this less invasive therapeutic approach focus mainly on the design of the proximal anchoring part of the devices.