A new generation endovascular graft for repair of abdominal aortic aneurysms.

BACKGROUND: Several endovascular grafts are currently being evaluated for repair of abdominal aortic aneurysms (AAA). The goals of our study were twofold. First was to develop a new endovascular graft with several advantages over previous devices: (1) smaller size (16 fr), (2) recapturability (the device can be partially deployed and then recaptured and moved to a new location or entirely removed if needed), and (3) accuracy and ease of placement. Our second goal was to develop an animal model in which a full-scale prototype of the device could be tested. METHODS: Our final endovascular graft prototype was developed after extensive in-vitro testing, and trials of earlier prototypes in dog, pig, and female sheep models. Uncastrated male sheep, 75 to 100 kg, were chosen as the animal model in which to test the device. These animals had infrarenal aortas that were comparable to that of small humans, with diameters of 12 to 15 mm. Two models were used: (1) native infrarenal aorta, and (2) artificial infrarenal aneurysm. Pre-implant and postimplant angiography and intravascular ultrasound were used to evaluate graft placement, and were repeated prior to euthanasia and necropsy. RESULTS: The final prototype was implanted in 22 animals. Sixteen animals had the device placed in their native infrarenal aorta. Three animals were sacrificed immediately after implantation, and 6 more were euthanized after 2 weeks (n = 2), 6 weeks (n = 2), and 3 to 4 months (n = 2). In 7 animals the device is still in place. All procedures were successful. Pathology confirmed complete exclusion of the aorta and thrombosis of all lumbar branches covered by the graft. There was no evidence of graft malposition, migration, or perigraft leak, and no evidence of significant vessel injury on histology. Six animals had artificial aneurysms surgically created and then repaired with the device. A technical error resulted in a failure in 1 case; the remaining aneurysms were all successfully excluded. CONCLUSIONS: We report the development of a new endovascular prosthesis for the repair of AAA. Newer design features provide for smaller delivery size (16 fr), facilitate accurate placement, and provide the option when the device is partially deployed to recapture and reposition the device if necessary. In addition, we have developed an animal model in which this device, and future endovascular aortic devices, can be tested.

Effects of dietary fish oil on biliary phospholipids and prostaglandin synthesis in the cholesterol-fed prairie dog.

Cholesterol gallstone formation in the prairie dog is accompanied by an increase in the percentage of biliary phospholipids containing arachidonic acid, and an increase in gallbladder prostaglandin (PG) synthesis, but the pathogenetic significance of these changes is unclear. Dietary supplementation with eicosapentaenoic acid (EPA), an omega-3 fatty acid which is commonly found in fish oil, decreases prostaglandin synthesis in some tissues by replacing arachidonic acid, and by competitively inhibiting prostaglandin synthesis. We studied the effect of dietary fish oil on gallbladder PG synthesis, and the relative abundance of various molecular species of phosphatidylcholines and phosphatidylethanolamines in bile and gallbladder epithelium in the cholesterol-fed prairie dog. Prairie dogs were maintained for 4 weeks on one of four diets: i) control, ii) cholesterol-supplemented (0.34%), iii) menhaden oil (50 g/kg chow), or iv) cholesterol plus menhaden oil. Supplementation with menhaden oil resulted in a replacement of arachidonic and linoleic acids with EPA and docosahexaenoic acids in the phospholipids of bile and gallbladder mucosa. In cholesterol-fed animals, supplementation with menhaden oil prevented increased gallbladder PG synthesis. Menhaden oil also reduced the incidence of cholesterol monohydrate crystals among cholesterol-fed animals (9/20 with cholesterol plus menhaden oil vs 21/22 with cholesterol alone), but the improvement could not clearly be attributed to decreased PG synthesis since supplementation with menhaden oil also increased the total phospholipid concentration in bile, and decreased the degree of cholesterol saturation. These results demonstrate that dietary supplementation with omega-3 fatty acids significantly influences biliary phospholipids, and decreases the incidence of cholesterol monohydrate crystal formation in this animal model.

Effect of dietary cholesterol on phosphatidylcholines and phosphatidylethanolamines in bile and gallbladder mucosa in the prairie dog.

Humans with cholesterol gallstones have been reported to have alterations in the molecular species of phospholipids in bile. Both decreases in phospholipids with linoleic acid and increases in those with arachidonic acid have been found. The purpose of this study was to investigate the effect of a lithogenic diet (0.34% cholesterol) on the relative abundance of individual molecular species of phospholipids in the biliary tract of the prairie dog. In hepatic bile, cholesterol feeding resulted in increases in phospholipid species containing arachidonate and decreases in the major species containing its precursor, linoleate. In gallbladder bile of both control and cholesterol-fed animals, phospholipid species containing linoleate were significantly less abundant than in hepatic bile, suggesting that linoleoyl species were selectively absorbed by the gallbladder epithelium. This apparent uptake was significantly increased by cholesterol feeding. Phosphatidylcholines and phosphatidylethanolamines containing arachidonate were also significantly increased in the gallbladder mucosa of the cholesterol-fed animals. These increases in arachidonate-containing phospholipids in the gallbladder mucosa may contribute to the increase in gallbladder prostaglandin synthesis that precedes gallstone formation in this animal model.

Can ventilation-perfusion scans accurately diagnose acute pulmonary embolism?

To examine the accuracy of ventilation-perfusion (V/Q) scanning, we retrospectively reviewed pulmonary angiograms and V/Q scans from 150 patients clinically suspected of having pulmonary embolism. Pulmonary emboli were documented by angiography in 56 patients (37%). In seven patients V/Q scans were interpreted as being normal and these seven patients were angiographically negative for emboli. The remainder of the scans were classified as showing a low, moderate, or high probability of emboli, or as indeterminate scans. Among these four abnormal classifications, pulmonary angiography demonstrated emboli in 13.6%, 62.5%, 85.7%, and 18.4% of these patients, respectively. An analysis of these data demonstrates that an unacceptably high error rate would result if V/Q scans alone were relied upon to establish the presence or absence of pulmonary emboli. In view of the high mortality that results when pulmonary emboli are untreated and the relatively low mortality and morbidity of pulmonary angiography, angiography remains an essential diagnostic procedure in patients suspected of having pulmonary embolism.

Primate model for the study of hepatic metabolism of lipoprotein cholesterol.

We describe a primate model for the study of net cholesterol flux by splanchnic, portal, and hepatic regions concomitant with that of biliary lipid secretion in conscious healthy fed or fasting baboons with exteriorized but physiologically intact enterohepatic circulations. In five fasting baboons studied, we found evidence for low-density lipoprotein (LDL) cholesterol uptake by the liver of 1.0 mg/min, P less than 0.05. Conversely, splanchnic production of LDL cholesterol was found in the fed state (n = 5), amounting to 0.8 mg/min, P less than 0.05. Uptake or secretion of high-density lipoprotein (HDL) cholesterol by either the liver or the intestine could not be detected by transgradient analysis in our animals in either fed or fasting states. Very low-density lipoprotein secretion of 0.4 mg/min by the liver was also found in the fed state. Isotopic biliary cholesterol derived from labeled plasma HDL or LDL cholesterol reflected hepatic LDL cholesterol production in the fed state but LDL cholesterol uptake in the fasting state. Biliary cholesterol secretion amounted to 8.5% of net hepatic lipoprotein cholesterol flux in the fed state and 10% in the fasting state. We conclude that uptake of LDL cholesterol by the liver is appreciable and contributes to biliary sterol secretion in the fasting baboon. Uptake or secretion of HDL cholesterol was not detected in any metabolic beds by transgradient analysis, but in tracer studies appeared to be a major source of lipoprotein cholesterol transferred to bile in the fed state.

The preoperative angiogram as a predictor of peripheral vascular runoff.

The preoperative angiogram is widely used as a means of assessing peripheral vascular runoff before bypass grafting, but the correlation between preoperative angiographic findings and actual measurements of peripheral vascular resistance has not been adequately examined. To test this correlation, we first devised a simple technique for measuring peripheral resistance and validated it in five dogs. Increases in peripheral resistance were artificially produced by temporarily occluding either the deep or superficial femoral artery or by intravenous administration of phenylephrine hydrochloride, a vasoconstrictor. In each instance, significant increases in resistance could be measured. We then used a similar technique to measure resistance in 23 patients undergoing peripheral bypass surgery. In addition, preoperative angiograms for these 23 patients were independently scored by four readers as 0, 1, 2, or 3 based on the number of patent vessels seen below the knee. Variations in scoring from reader to reader suggested that the present criteria for grading angiograms on this basis are unclear. Moreover, the correlation between angiographic score and measured resistance was poor for three of the four scorers (-0.21 to -0.29, p greater than 0.05). The angiographic scores of one reader, however, correlated reasonably well with the peripheral resistance measured at surgery (-0.59, p = 0.01). These findings demonstrate that current criteria for grading the preoperative angiogram are not sufficiently standardized to reliably predict runoff from a preoperative angiogram. However, these findings also suggest that it may be possible to identify angiographic findings that correlate well with changes in measured resistance.

Histamine receptors in primate gallbladder.

The effect of intravenous histamine on intragallbladder pressure (GBp) and subsequent modification of the histamine response by H1- and H2-antagonists was investigated in an awake baboon model. Responses to specific H1- and H2-agonists were also examined in order to further elucidate gallbladder histamine responses. Gallbladder volume (GBv) was arbitrarily set at 60-70% of observed resting GBv, and drugs were then infused intravenously with continuous monitoring of GBp. Histamine administration resulted in a logarithmic dose--response curve with a maximal increase in GBp equal to 31.7 mm Hg at a dose of 0.0625 mg of histamine. Infusion of the H1-antagonist diphenhydramine hydrochloride (Benadryl) resulted in a decreased GBp response to histamine when compared to preblocker response at all doses studied. On the other hand, histamine response following administration of the H2-antagonist metiamide was significantly greater than the preblocker histamine response. Infusion of the H1-agonist 2-pyridylethlamine resulted in a dose-dependent increase in GBp similar to the histamine curve, while infusion of the H2-agonist dimaprit resulted in consistent decreases in GBp. These results extend previous observations in subprimate preparations and demonstrate the presence of both stimulatory H1-receptors and inhibitory H2-receptors in an in situ primate model.

Effects of pH on in vitro gallbladder responses to cholecystokinin octapeptide.

We investigated the effect of pH change on octapeptide cholecystokinin's ability to contract guinea pig gallbladder strips. In the absence of exogenous drugs, pH changes had no demonstrable effect on gallbladder tension. However, when gallbladder strips were contracted with CCK-OP at pH 7.3, increasing the pH in the muscle bath to 7.8 produced further increases in tension at each dose studied (P less than 0.025). Subsequently, decreasing the bath pH to 6.9 decreased the strip tension to less than that observed at pH 7.3 (P less than 0.025). Reversing the order in which these pH alterations were made produced similar results; a pH decrease from 7.3 to 6.9 decreased the response to CCK-OP at the two highest doses (P less than 0.025), while subsequent elevation of pH raised the tension to levels greater than those observed at pH 7.3 (P less than 0.01). Furthermore, contracting the strips at pH 8.0 and slowly decreasing pH at a rate of 0.1 pH units per minute consistently produced an acceleration of tension decay as pH fell. After contracting the strip at pH 6.7, slow increases in pH reversed the tension decay and enhanced the contractile response as pH was increased from 6.7 to 8.0. These results demonstrate that pH changes alter the gallbladder contractile response to cholecystokinin. Although this phenomenon has been reported for a number of other physiologic agents, we believe this is the first such demonstration for a peptide hormone.

pH-dependent activity of H1- and H2-histamine receptors in guinea-pig gallbladder.

Utilizing histamine and selective agonists for H1- and H2-receptors, we examined the pH dependence of histamine-stimulated tension changes in guinea-pig gallbladder, which contains both contracting H1-receptors and relaxing H2-receptors. In muscle strips contracted with histamine and pH 7.3, increasing pH to 7.8 raised tension further (P less than .025), while decreasing pH caused a fall in tension (P less than .025). The H2-agonist Dimaprit relaxed tension at pH 7.3 and increasing the pH decreased the relaxation (p less than .0125). Contractions in response to H1-agonist 2-pyridylethylamine at pH 7.3 were unchanged when pH was elevated but decreased when pH was lowered (P less than .05). Tension changes in response to slow pH alterations suggested that H1-receptor activity is inhibited below pH 7.1 and H2-receptor activity is inhibited above pH 7.6. These reversible changes in activity probably reflect changes at H1- and H2-receptors rather than alterations in the ionic species of histamine.