Intrahepatic portal occlusion by microspheres: a new model of portal hypertension in the rat.

Available experimental models of portal hypertension are based either on cirrhosis or externally applied portal vein constricting devices. A new method is described of raising portal pressure, which uses intraportally injected microspheres to block intrahepatic portal radicles, which has the advantages of retaining normal liver architecture and providing a more clinically relevant intrahepatic obstruction to portal flow. Measured aliquots of microspheres (15, 25, 50, 90 microns) or equivalent volumes of saline were injected into a peripheral portal tributary (caecal vein) of 22 normal rats. The resultant changes in arterial, portal, and splenic pulp pressures were monitored. Sequential microsphere injections produced graduated rises in portal pressure up to a peak of 18.5-22.5 mm Hg (8.7-12.4 mm Hg increase from basal), which declined gradually to a steady state pressure of 13.3-15.1 mm Hg (4.0-5.0 mm Hg increase). There was no significant difference between pressure increases produced by microspheres of differing sizes. It is concluded that portal hypertension can be produced acutely by blocking portal radicles with microspheres. The maximum pressure achieved, however, is substantially less than that obtained by total portal vein occlusion (mean: 57.6 mm Hg). This suggests the existence of functional intrahepatic portal systemic shunts not previously described in the normal liver.

The effect of portal perfusion on intrasplenic hepatocytes.

A newly developed rat model of portal diversion, in which the subcutaneously transposed spleen is perfused by splanchnic effluent as a result of portal vein ligation, has been used to monitor the effect of such portal perfusion on intrasplenic hepatocyte implants. Three groups of animals (controls, n = 42; transposed only, n = 70; and portally diverted, n = 58) received 2 million syngeneic liver cells. The number of hepatocytes in each spleen was assessed 5 days to 9 months later by direct counting of splenic sections. The transposed spleen was capable of supporting hepatocyte grafts even over long periods, although the number of cells was reduced in comparison with controls. Diversion of portal flow across the transposed spleen significantly increased hepatocyte numbers in the first 6 weeks (median number of cells (with inner quartile range), 1380 (70-1300) versus 600 (347-4050); n = 75), but no differences were detected thereafter. It appears that the initial lag phase of hepatocyte grafts can be partially abrogated by portal perfusion, but the subsequent 'proliferative' phase is unaffected. These effects correlate well with established theories on the importance of portal flow to the intact liver.

General surgery with a special interest in vascular surgery: is the tail starting to wag the dog?

Most general surgeons are expected to develop an interest in a surgical specialty in addition to their general surgical workload. Over a 2-year period we have prospectively analysed the type of patient admitted to a general surgical unit with a special interest in vascular surgery. Half of the admissions were emergencies, over which we had no control. Of the elective admissions, there was a steady increase in the proportion of patients with a vascular diagnosis. This was due to increasing numbers of arterial reconstructions and interventional radiological procedures being undertaken. As cutbacks are made in the numbers of available beds and operating sessions, the increasing number of patients requiring admission for the management of arterial disease can only be achieved at the cost of patients awaiting routine general surgical procedures.

General surgery with a special interest in vascular surgery: an audit of relative workload.

In a district general hospital (DGH) almost all vascular surgery is provided by general surgeons with a vascular interest and training. There is a growing view, however, that vascular surgeons should be 'pure' to the exclusion of other surgery. In an attempt to define the relationship between general and vascular work in a DGH we have analysed, prospectively, out-patient, in-patient and theatre workload over a three-month period. Eight-hundred-and-forty-four patients (277 new, 567 follow-up) were seen in the clinics. Nine per cent of new vascular referrals and 33 per cent of new general referrals were booked for admission. There were 356 admissions (50 per cent 'emergencies') representing a wide spectrum of general surgery. Elective and emergency vascular cases stayed in hospital for twice and four times longer, respectively, than general patients. Vascular patients represented 26 per cent of the caseload but accounted for 46 per cent of the bed occupancy. Vascular operations made up only 21 per cent of the total theatre caseload (233 procedures) but consumed 34 per cent of theatre time. The vascular unit in a DGH deals with a substantial number and wide variety of general surgical cases. Vascular surgery, however, consumes disproportionately large amounts of out-patient and theatre time and hospital beds. This has implications both for the planning of vascular services and also for the resource allocation within the hospital.

The growth of liver cells in the pancreas after intra-splenic implantation: the effects of portal perfusion.

A newly developed rat model of portal diversion, in which the pancreas gland is perfused by splanchnic effluent, has been used to monitor the effects of such portal perfusion on the development of isolated hepatocytes within the pancreatic parenchyma. Two groups of animals (control and portally-perfused) received intra-splenic hepatocyte implants (2 x 10(6) cells). The development of liver cells in the pancreas was then assessed histologically 10 days-9 months later. Portal diversion did not alter the incidence of hepatocytes within the pancreas as a whole (70% of all animals). However hepatocyte-islet rosettes (the intimate association between liver cells and endocrine tissue) were found in a significantly higher percentage of portally-perfused pancreases (92%) in comparison to controls (43%) 6 and 9 months after implantation. It appears that liver cell translocation from the spleen to the pancreas is dependent largely on factors associated with the implantation procedure itself. The growth of liver cells specifically around the islets of Langerhans, however, seems to be positively influenced by portal factors. These studies provide further evidence for the theory that extrahepatic liver cells are susceptible to local and systemic growth factors in a similar way to homotopic hepatocytes in the intact liver.

Radiological definition of a new model of portal diversion in the rat.

Diversion of portal blood away from the liver can be accomplished in the rat by two straightforward surgical procedures: subcutaneous transposition of the spleen followed later by portal vein ligation. This experimental model has great potential value in the study of liver cell transplants, porto-systemic shunts and hepatic porto-privation. Contrast medium radiology has been used to clarify and define the model further. Splenography, achieved by direct percutaneous puncture, demonstrated the developing spleno-subcutaneous vessels. Collaterals are seen as early as 3 days and appear extensive by 21 days after transposition. Almost all emanate from the convex outer surface of the spleen and course cranially and caudally in the subcutaneous tissues to drain into the subclavian and iliac veins. The appearance of these collaterals correlates well with survival after portal vein occlusion subsequent to splenic transposition. Direct portography demonstrates that, following portal vein ligation, contrast medium is diverted away from the liver into the splenic veins, through and around the spleen, before draining into the systemic circulation through spleno-subcutaneous collaterals. These radiological studies have demonstrated the changed circulatory pathways of this model of portal diversion and have confirmed that it is the splenic veins and the spleno-subcutaneous collaterals which are fundamental to its successful outcome.

Dynamic pressure measurements in the evaluation of a pancreaticolienal portal diversion model in the rat.

Diversion of portal blood away from the liver has been accomplished, in the rat, transposition of the scarified spleen, followed by later portal vein ligation. ligation. The resulting dynamic portal pressure changes have been measured. In 7 control (sham-transposed) rats, portal ligation produced a 7-fold sustained and fatal pressure increase. In 7 previously transposed rats, the increase was only 3-fold and was followed by a slow fall to a lower plateau pressure, with 100% 3 h survival; clamping the splenic pedicle converted the pressure trace of that of controls. These results support the view that splenosubcutaneous collaterals, stimulated by scarification, are essential for successful portal diversion.