Hypoperfusion of segment 4 in right in situ split-liver transplantation.

To expand the donor pool, split-liver transplantation has been implemented in recent years. In the classic technique, the arterial axis with the artery for segment 4 (S4) coming from the left hepatic artery (HA) is included with the right graft. To give a surgical advantage to pediatric recipients in our center, the left HA, the common HA, and the celiac trunk are generally retained with the left liver. Thus the artery for S4 is sacrificed. We compared the outcomes of S4 in 290 whole grafts (WG; group A) with 28 right in situ split-liver grafts (SSLG; group B), which were transplanted over the past 10 years (January 1999-December 2009). The rates of major biliary and of hemorrhagic complications were similar. In most of cases (16/24, 66%) S4, on computerized tomographic scan appeared to show signs of hypoperfusion, sometimes with a peripheral aspect of hyperperfusion in the arterial phase. S1 showed signs of hypoperfusion in only 2 cases. A biliary collection near the resection line present in 8 cases was treated in 6 of them with percutaneous drainage and in 2 with laparotomy. These complications did not influence graft or patient survival. Graft survivals at 1, 5, and 10 years for WG and SSLG were not different among the groups: 85%, 74%, and 66% vs 89%, 79%, and 63%, respectively (P = .8). Although our technique cannot be considered to be anatomically correct, the ischemia of S4 did not influence the outcome. The rate of retransplantations for hepatic artery thrombosis was 17.9% in RSSG and 3.4% in WG (P = .001), which was probably due at least in part to the insertion of interposition grafts.

Lisinopril-induced acute pancreatitis.

Drugs are an uncommon but well-recognised cause of acute pancreatitis and new agents of drug-induced pancreatitis continue to be reported. We describe only the 10th reported case of lisinopril-induced pancreatitis in a young female patient.

Expression of lactoferrin in human stomach.

The expression of the haeme-binding protein, lactoferrin, was studied in human gastric tissues displaying normal, benign hyperplastic or malignant histology. A single 2.5-kb mRNA was detected in only 14% (2/14) of normal resections. This was similar to the finding that 85% of tumours were also negative, with 4/27 positive. In contrast, samples with superficial or atrophic gastritis had a high frequency of expression, with 5/7 and 9/14 positive respectively. The higher incidence of lactoferrin mRNA in antral samples was a reflection of the greater proportion of these compared with body resections of patients with gastritis. No expression was seen in any of 5 gastric carcinoma cell lines. High levels were observed in the cardia, in contrast to complete absence in the oesophagus. Immunocytochemistry showed localization of lactoferrin in cells of both antral and body glands. Chief cells, but not adjacent parietal cells, were strongly stained. In tissues exhibiting superficial or atrophic gastritis we observed a greater degree and intensity of staining as compared with samples with normal histology. We also observed some staining of tumour cells, though this was very patchy. Lactoferrin may have a role in mucosal iron transport in both normal and highly proliferating tissue, but does not appear to be significantly associated with malignant lesions.

Increased production of growth factors and epidermal growth factor receptor by gastric carcinoma.

Four growth factors and one growth factor receptor have been studied in carcinomas from 22 gastrectomy specimens and compared to non-malignant tissue from the same specimen. cDNA probes for transforming growth factors alpha and beta, platelet-derived growth factor A and B, insulin-like growth factor II and epidermal growth factor receptor were used to assay messenger RNA transcripts for the growth factors by dot hybridization. Increased levels of all the transcripts were found in carcinomas compared to benign tissue (P less than 0.05). No correlation was found between any of the growth factors studied and tumour stage or patient survival. Increased growth factor production by gastric cancers may be important in the pathogenesis of these tumours and further work is required to establish their role.

Expression of growth factor and epidermal growth factor receptor encoded transcripts in human gastric tissues.

Expression of mRNA-encoding transforming growth factors alpha and beta (TGF alpha and beta), epidermal growth factor receptor (EGFR), and platelet-derived growth factors (PDGF) A and B chains was examined in 63 human gastric biopsies. Despite considerable individual variation, transcript levels were generally higher in 16 paired gastric tumors compared with surrounding epithelium. Marked increases were observed for the TGFs and c-sis, whereas EGFR mRNA was poorly expressed; there was no correlation with pathological staging of the cancers. In the nonneoplastic tissues, 14 had normal histology and 27 displayed superficial (SG) or atrophic gastritis (AG). Transcript levels greater than or equal to + were similar between these categories for all the growth factors, but were about 50% higher for EGFR in the tissues with gastritis. Concurrent expression of TGF alpha and EGFR (greater than or equal to + level) was more frequent in the paired tumors (38%) than in adjacent nonmalignant tissue (6%) and was seen in only one of 14 (7%) normal samples, in three of 19 (16%) of those with AG, and none of eight of those displaying SG. High levels of TGF beta and PDGFA mRNA were expressed in gastric ulcers, with little or no TGF alpha and EGFR transcripts; in contrast both TGFs and EGFR message were found in normal oesophagus. Stomach tissues are thus capable of synthesizing a variety of growth factors. These may be associated with nonneoplastic hyperplasia and/or malignant proliferation. Coexpression of TGF alpha/EGFR supports the possibility of an autocrine loop sustaining tumor growth which is different from the mechanisms responsible for normal cellular proliferation.

Anastomotic leakage: an avoidable complication of Lewis-Tanner oesophagectomy.

Data were collected prospectively from 159 patients undergoing Lewis-Tanner oesophagectomy for carcinoma of the lower two-thirds of the oesophagus and analysed with regard to anastomotic leakage. The 30-day mortality rate was 3.8 per cent and none of these deaths was due to anastomotic leakage. In hospital, mortality was 7.5 per cent. Six patients (3.8 per cent) had evidence of a leak. In four patients this was from the suture line, but in the remaining two it was due to patches of gangrene in the fundus of the transposed stomach and was separate from the anastomosis. Two patients died with an unhealed leak. Anastomotic leaks were minor in three instances and healed with conservative treatment but the fourth required exploration. Gangrene of the oesophageal substitute resulted in gross leakage with mediastinitis and required exploration in both cases. An apparent predisposing cause was evident in five of the six cases with leakage and some of these might have been avoidable. Transthoracic oesophagectomy can be performed with acceptably low mortality and leak rates and may therefore be considered as the treatment of choice for most patients with oesophageal cancer.

A study on the mechanism of the postoperative fall in serum iron concentration.

The purpose of this study was to investigate the mechanism of the postoperative fall in serum iron concentration. Simultaneous measurements of iron uptake from, and release to, the plasma were made using two iron radioisotopes. Eight rabbits were submitted to laparotomy and eight were used as controls. Six hours after laparotomy, when the plasma iron concentration was decreasing, the clearance half-time from the plasma was decreased from 82.0 +/- 3.0 min in controls to 25.0 +/- 4.0 min (P less than 0.001) but the amount of iron uptake was increased from 0.289 +/- 0.013 to 0.574 +/- 0.092 mumole/liter blood/day (P less than 0.001). Twenty-four hours after laparotomy, when the plasma iron concentration was increasing, the half-time remained shortened in the operative group but the amount of iron uptake was decreased from 0.39 +/- 0.28 to 0.25 +/- 0.1 mumole/liter blood/day although this difference is not significantly different. At both times studied there was no difference in the amount of iron released to the plasma. These results suggest that the fall in iron concentration after surgery is due to increased uptake from the plasma.