Morphological lymphocytic reaction, patient prognosis and PD-1 expression after surgical resection for oesophageal cancer

BACKGROUND

Immune checkpoint inhibitors, such as antibody against programmed cell death protein (PD-1), have demonstrated antitumour effects in patients with malignancies, including oesophageal cancer. A lymphocytic reaction observed by pathological examination is a manifestation of the host immune response to tumour cells. It was hypothesized that a stronger lymphocytic reaction to tumours might be associated with favourable prognosis in oesophageal cancer.

METHODS

Using a database of resected oesophageal cancers, four morphological components of lymphocytic reactions (peritumoral, intranest, lymphoid and stromal) to tumours were evaluated in relation to clinical outcome, PD-1 expression by immunohistochemistry and total lymphocyte count in blood.

RESULTS

Resected oesophageal cancer specimens from 436 patients were included in the study. Among the four morphological components, only peritumoral reaction was associated with patient prognosis (multivariable P for trend <0·001); patients with a higher peritumoral reaction had significantly longer overall survival than those with a lower reaction (multivariable hazard ratio 0·48, 95 per cent c.i. 0·34 to 0·67). The prognostic effect of peritumoral reaction was not significantly modified by other clinical variables (all P for interaction >0·050). Peritumoral reaction was associated with total lymphocyte count in the blood (P < 0·001), supporting the relationship between local immune response and systemic immune competence. In addition, higher morphological peritumoral reaction was associated with high PD-1 expression on lymphocytes in tumours (P = 0·034).

CONCLUSION

These findings should help to improve risk-adapted therapeutic strategies and help stratify patients in the future clinical setting of immunotherapy for oesophageal cancer.

Minimally invasive esophagectomy may contribute to long-term respiratory function after esophagectomy for esophageal cancer

Evidence suggests that minimally invasive esophagectomy has several advantages with regard to short-term outcomes, compared to open esophagectomy in esophageal cancer patients. However, the impact of minimally invasive esophagectomy on long-term respiratory function remains unknown. The objective of this study is to assess the association between use of the minimally invasive esophagectomy and long-term respiratory dysfunction in esophageal cancer patients after esophagectomy. This retrospective single institution study using prospectively collected data included 87 consecutive esophageal cancer patients who had undergone esophagectomy. All patients underwent a respiratory function test before, and one year after esophagectomy. Logistic regression analysis was used to compute the hazard ratio for long-term respiratory dysfunction. Minimally invasive esophagectomies were performed in 53 patients, and open esophagectomies in 34 patients. The two groups showed no significant differences in terms of postoperative complications and postoperative course. Nor were any differences observed between the two groups in terms of volume capacity (L) and forced expiratory volume 1.0 (L) before esophagectomy (P > 0.34). However, one year after esophagectomy, the decreases in volume capacity and forced expiratory volume 1.0 were significantly less in the minimally invasive esophagectomy group than in the open esophagectomy group (P = 0.04 and P = 0.007, respectively). Multivariate analyses revealed that minimally invasive esophagectomy was an independent favorable factor for maintenance of forced expiratory volume 1.0 (hazard ratio = 0.17, 95% confidence interval 0.04-0.71; P = 0.01). Minimally invasive esophagectomy may be an independent favorable factor for maintenance of long-term respiratory function in esophageal cancer patients after esophagectomy.

Inflammatory pseudotumor of the liver in association with spilled gallstones 3 years after laparoscopic cholecystectomy: report of a case

We report on a case of a female patient diagnosed with inflammatory pseudotumor of the liver in association with spilled gallstones 3 years after laparoscopic cholecystectomy for calculous acute cholecystitis. She was asymptomatic, but CT revealed an intrahepatic mass and two other extrahepatic masses between the liver and the diaphragm. Furthermore, diffusion-weighted MRI and PET suggested all three lesions could be malignant tumors. As the preoperative diagnosis was intrahepatic cholangiocellular carcinoma with peritoneal disseminations, we performed a posterior segmentectomy of the liver combined with partial resection of the diaphragm. Histological examination showed the intrahepatic tumor was an inflammatory granuloma with abscess formations. There were bilirubin stones between the liver and the diaphragm. Therefore, the tumor was diagnosed as inflammatory pseudotumor of the liver in association with spilled gallstones. In conclusion, the liver tumor emerged after laparoscopic cholecystectomy and may involve inflammatory pseudotumor of the liver in association with spilled gallstones.

A case of acute phlegmonous gastritis successfully treated with antibiotics

Acute phlegmonous gastritis is a rare disorder in which bacterial infection occurs in the gastric wall. Gastrectomy involving the affected area has been thought to be an effective form of treatment. The authors report a case of a 32-year-old woman who had severe upper abdominal pain without signs of peritoneal irritation. Endoscopy showed edematous and reddened gastric mucosa with a mass lesion in the gastric antrum. Endoscopic ultrasonography showed thickening of the antral wall and a low-echoic mass in the gastric antrum, thought to represent a fluid collection. White pus was aspirated from the mass. Localized type of acute phlegmonous gastritis with a gastric abscess was diagnosed. Culture of the pus showed Streptococcus pneumoniae. Through early diagnosis without laparotomy, the patient's gastritis was successfully treated with antibiotics alone.

Effects of lumbar sympathectomy on the properties of both endothelium and smooth muscle cells of the canine femoral artery and autogenous vein grafts under poor runoff conditions

The purpose of this study was to determine whether or not a lumbar sympathectomy would modulate the functions of the endothelium, in terms of endothelium-derived relaxing factor (EDRF), under poor distal runoff conditions. First, a poor distal runoff model was developed in the canine right and left femoral arteries. After 5 weeks, a unilateral left sympathectomy was performed from L-3 to L-6. In experiment I, changes in blood flow and endothelium-dependent responses were examined in the canine femoral artery 5 weeks after the lumbar sympathectomy. In experiment II, 5 weeks after the development of a poor runoff model, a unilateral left sympathectomy was performed and both femoral veins were also grafted to the femoral arteries on both sides. Then after 4 weeks, endothelium-dependent responses and intimal thickening of both autogenous vein grafts were examined. The endothelium-dependent responses were examined by mechanical tension recording. In both experiments, the mean blood flow of the left femoral arteries (denervated, 92.2 +/- 47.1 ml/min) and vein grafts (denervated, 100.0 +/- 35.4 ml/min) was significantly higher than that of the right femoral arteries (innervated, 46.9 +/-25.7 ml/min) and vein grafts (innervated, 50.0 +/- 20.1 ml/min) (P < 0.01). In experiment I, the endothelium-dependent relaxations to acetylcholine, ADP, and A23187 were comparable between the right and left femoral arteries. In experiment II, the endothelium-independent contractions to acetylcholine and endothelium-dependent relaxations to ADP and A23187 were all comparable between the right and left vein grafts. In addition, the intimal thickening of the vein graft was comparable between the two groups (denervated, 95.6 +/- 10.8 microm; innervated, 105.0 +/- 15.0 microm). In both experiments, the NE-induced contractions and SNP-induced relaxations were closely similar between the two groups, irrespective of the flow change. These results suggest that lumbar sympathectomy does not alter endothelial function in terms of EDRF, although mean blood flow of the denervated femoral arteries and vein grafts was significantly higher than that of the innervated femoral arteries and vein grafts. These results suggest that continuous vasodilation following sympathectomy may be a more potent factor with respect to regulation of vascular tonus than physiologic regulation of EDRF.

[Surgical treatment of aorto-iliac occlusive disease]

Recently, various endovascular techniques were applied to the short segmental stenosis of aorto-iliac artery. However, bypass operation is the most reliable treatment for multiple stenotic or occlusive lesion. Preoperative systemic evaluation is important for the adequate selection of the patients and the operations. Dipyridamole-thallium scintigraphy is more useful than exercise stress testing to assess cardiac risks of patients with peripheral arterial diseases. Anatomical bypass is the standard operation resulting in favorable patency. Despite of relatively inferior long term patency, extra-anatomical bypass operations are useful alternatives for the patients having high systemic risks or severe sclerosis of aorto-iliac arteries. The improved medical care of the elderly patients with severe systemic diseases results in the increase of extra-anatomical bypass and extended operation combined with infrainguinal revascularization.

Eicosapentanoic acid reduces the intimal thickening of autogenous vein grafts and enhances endothelium-derived relaxing factor

The present study examined the effect of purified eicosapentanoic acid (EPA) on intimal thickening of an autogenous vein graft. In addition, experiments were performed to determine whether EPA supplementation would alter the endothelium-dependent responses of the reversed vein graft. Segments of femoral veins were grafted into the femoral arteries of dogs. Six dogs received regular chow (control group) and six other dogs regular chow with 1500 mg/day 90.0% pure EPA (EPA group). At 6 weeks after surgery, the vein grafts were removed from the dogs, cut into rings, and suspended in organ chambers for isometric tension recording. In some rings, the endothelial cells were removed. When the rings taken from the control group were contracted with norepinephrine, adenosine diphosphate (ADP) and A23187 caused endothelium-dependent relaxations. In the EPA group, the endothelium-dependent responses to ADP were significantly augmented, while A23187 caused comparable endothelium-dependent relaxations. Direct relaxation in response to sodium nitroprusside was comparable between the two groups. Intimal thickening of the grafts in the control group (29.0 +/- 1.8 microns) was significantly (P < 0.05) greater than in the EPA group (12.5 +/- 1.8 microns). These results suggest that EPA enhances the release of endothelium-derived relaxing factor in autogenous vein grafts. This may be one of the effects of EPA in reducing the intimal thickening of autogenous vein grafts.

Late graft failure of autologous vein grafts for arterial occlusive disease: clinical and experimental studies

Late graft failure following arterial reconstructive surgery, especially after infrainguinal reconstruction, remains a major concern for vascular surgeons. To more effectively predict the outcome of reconstructed arteries, we herein propose an intraoperative flow waveform analysis which correlates well with the long-term patency rate of grafts. According to this flow waveform analysis, late graft failure was occasionally seen in grafts with type II waveforms when poor distal runoff vessels had been shown by the preoperative arteriogram. Next, to investigate which events occurring in autologous vein grafts under abnormal hemodynamics may contribute to late graft failure, a distal poor-runoff model was made in the canine femoral artery. In this review, we present the results of our investigation on autologous vein grafts using this poor-runoff model. We also relate our recent findings on the function of regenerated endothelium in autologous vein grafts.

Endothelium-dependent relaxation in response to adenosine diphosphate is impaired under poor runoff conditions in the canine femoral artery

To examine whether poor distal runoff conditions of the canine femoral artery modulates endothelium-dependent relaxations, we developed a poor distal runoff limb in the canine right femoral artery (poor runoff group). The left femoral artery was used as a control. Endothelium-dependent responses were examined in vitro. The rings of femoral artery were obtained from dogs pretreated with 6 weeks of poor distal runoff. In the control, flow rate and tau-variation were 79.1 ml/min and 214.2 dynes/cm2, respectively. In the poor runoff group, the flow rate and tau-variation were 27.4 ml/min and 52.7 dynes/cm2. There were significant differences between the two groups. In the rings taken from the control group contracted with norepinephrine, adenosine diphosphate (ADP) and A23187 caused endothelium-dependent relaxations. In the poor runoff group, the endothelium-dependent relaxations in response to ADP were impaired, while A23187 caused comparable endothelium-dependent relaxations. Direct relaxations in response to sodium nitroprusside were comparable between the two groups. These experiments indicate that under poor runoff conditions, the endothelium-dependent relaxations to ADP are impaired without changing the properties of the smooth muscle cells. This dysfunction of the endothelium under conditions of abnormal flow may accelerate a loss of late patency of the reconstructed arteries.

Comparison of endothelial function between in situ and reversed vein graft: differences in endothelium-dependent responses

BACKGROUND

The endothelium releases endothelium-derived relaxing factors that are potent vasodilators and inhibitors of platelet aggregation. Experiments were performed to determine whether the endothelium-dependent responses differed between in situ and reversed vein grafts. The influence of valve disruption or of dissection of the adventitia was also examined.

METHODS

Segments of canine jugular veins were grafted into the carotid arteries during procedures such as reversed grafting, in situ grafting with valve disruption, in situ grafting without valve disruption, and in situ grafting with dissection of adventitia. After 4 weeks the endothelium-dependent responses of the grafts were examined by isometric tension recording.

RESULTS

In the reversed and in situ vein graft with valve disruption, acetylcholine caused endothelium-independent contractions, whereas in the in situ vein graft without valve disruption acetylcholine-induced endothelium-dependent relaxations were preserved. Adenosine diphosphate (ADP) caused comparable endothelium-dependent relaxations in the in situ vein graft irrespective of valve disruption. In the reversed vein graft ADP-induced relaxations were significantly impaired. In the in situ vein graft with dissection of the adventitia, relaxations in response to acetylcholine and ADP were significantly reduced.

CONCLUSIONS

These results suggest that endothelial function, in terms of endothelium-derived relaxing factors in the in situ vein graft, can be preserved and that adventitial dissection in the in situ vein graft should be minimized to preserve endothelial function.

Intimal hyperplasia of experimental autologous vein graft in hyperlipidemic rabbits with poor distal runoff

Poor distal runoff and hyperlipidemia are factors affecting the fate of an implanted graft. In the present study, combined effects of poor distal runoff and hyperlipidemia on intimal hyperplasia (IH) of the vein graft were examined in a newly developed poor distal runoff model in rabbits. A poor distal runoff model was prepared in the right hindlimb of 30 rabbits. These animals were divided into two groups, depending on the diet provided; normolipidemic diet group (Group NL, n = 14) and hyperlipidemic 1% cholesterol diet group (Group HL, n = 16). Four weeks after preparing the poor runoff model, the femoral vein was implanted into the ipsilateral femoral artery. At 2, 4 and 6 weeks, the grafts were harvested. IH of the graft was measured and macrophages in the IH were examined immunohistochemically. Intimal cell proliferation was also determined by bromodeoxyuridine (BrdU) incorporation. IH of the vein graft was significantly accelerated in cases of poor distal runoff and hyperlipidemia. There were no macrophages in the IH in the NL group. In the HL group, macrophages infiltrated the outer layer of IH, sometimes just above the internal elastic lamina, and increased with time. In the poor distal runoff limbs at 6 weeks, macrophages also appeared in the subendothelial layer but were absent in that layer in the controls. Intimal cell proliferation expressed as the BrdU labeling index (LI) was maximum at 2 weeks. In the HL group, BrdU LI of IH in the poor distal runoff limb was higher than in the control at 2 and 4 weeks. Throughout the experiments, BrdU LIs in the HL group were significantly higher than in the NL. Hyperlipidemia accelerates intimal cell proliferation to a greater extent, then does IH. In cases of a poor distal runoff, the enhancement of cell proliferation by hyperlipidemia is augmented. These responses, in the presence of a hyperlipidemia, may be closely related to the migration of macrophages.

Detection of active cytomegalovirus infection in inflammatory aortic aneurysms with RNA polymerase chain reaction

PURPOSE

We previously reported the possible role of human cytomegalovirus in the pathogenesis of inflammatory aortic diseases. To further analyze the viral cause of human aortic diseases, in this study we examined the presence and the replication of human Herpesviridae in 60 aortic tissues, including 7 inflammatory aneurysms, 37 atherosclerotic aneurysms, and 16 normal aortas.

METHODS

To detect the genome of herpes simplex virus (type 1, type 2), cytomegalovirus, and Epstein-Barr virus, DNA polymerase chain reaction for each virus was performed. To analyze these herpesviral replications, the viral transcript was detected with RNA polymerase chain reaction.

RESULTS

The DNA polymerase chain reaction showed that either herpes simplex virus or cytomegalovirus was present more frequently in inflammatory (29% or 86%, respectively) and atherosclerotic aneurysms (27% or 65%, respectively) than in normal aortic tissues (6% or 31%, respectively), whereas the Epstein-Barr viral genome was not detected in any aortic tissue specimens. By the use of RNA polymerase chain reaction, only the cytomegaloviral transcript was recognized in 71% of the inflammatory aneurysms but was not recognized in any other tissue specimens. No other herpesviral transcripts were detected in any tissue specimens examined in this study.

CONCLUSIONS

Our results thus suggest that the human herpesviruses may play various roles in the pathogenicity of aortic diseases, in particular the replicating infections of the cytomegalovirus might potentially cause the formation of inflammatory aneurysms.

Do patients with intermittent claudication need surgical treatment?

To determine whether surgical intervention is of therapeutic significance for patients with intermittent claudication (IC), 315 patients with IC followed in our outpatient clinics for the past 8 years 3 months (mean of 3 years 10 months), including 142 non-operated and 173 operated patients, were investigated as to changes in symptoms and the prognosis. Changes in symptoms in both groups were analyzed according to the variable distances of claudication and location of the diseased artery. To examine the degree of association between surgery and improvement of symptoms, Goodman and Kruskal's gamma was used. For all subgroups with a variable claudication distance (group A; distance < 100 m, B: 100-500 m, C: > 500 m), surgery inversely correlated with improvement in symptoms (group A; gamma = -0.378, group B; gamma = -0.651, group C; gamma = -0.828). According to location of the diseased artery, surgery correlated with improvement in symptoms in aorto-iliac (gamma = -0.811), aorto-ilio-superficial femoral (gamma = -0.641), and superficial femoral-distal (gamma = -0.533) groups but not in superficial femoral (gamma = -0.427) or infrapopliteal (gamma = -0.194) alone groups. Concerning the prognosis, no significant difference was noted between operated and non-operated groups (p = 0.35). These computed data show that surgical intervention has therapeutic significance for patients with IC, except for cases of claudication with segmental occlusion of the superficial femoral or infrapopliteal artery alone.

A clinical analysis of multiple organ failure following elective surgery

A retrospective analysis was made of 58 patients who unexpectedly developed multiple organ failure (MOF) following elective surgery, and the results were compared with those of 168 control patients who did not develop MOF. In 33 patients with liver cirrhosis, MOF was related to poor liver function, a low albumin level, excessive blood loss, many transfusions, and a high incidence of hypotension. MOF, rather than liver failure alone, was featured by postoperative bleeding and infection. In 15 patients with esophageal carcinoma, MOF was correlated with many transfusions, anastomotic leakage, and postoperative infection. In 10 patients who underwent surgery for an aortic aneurysm, poor renal function and extended anesthesia time were associated with MOF. These results indicate that to prevent MOF following elective surgery, it is important to: (1) Select patients for liver surgery according to their liver function, and minimize the risk of bleeding and infection, (2) avoid too many blood transfusions, and minimize the risk of leakage and infection in esophageal surgery, and (3) select patients for aortic surgery based on renal function and reduce the anesthesia time as much as possible.

Persistent sciatic artery: two case reports and a review of the literature

Persistent sciatic artery (PSA) sometimes shows variant vascular symptoms such as chronic or acute limb ischaemia and a pulsatile mass in the buttocks caused by aneurysmal dilatation. This report presents two cases of surgically treated PSA that caused lower-limb ischaemia. The first patient, a 58-year-old man, showed acute limb ischaemia in the chronic clinical course, and thrombectomy and patch angioplasty were performed. The second patient, a 70-year-old woman, had an incomplete type of PSA without development of the femoral artery while complaining of a 10-year history of coldness and numbness in the left lower limb. A left common iliac to profunda femoris arterial bypass was performed. This is the second case of symptomatic incomplete PSA reported in the literature. This report also reviews 45 cases of PSA published since 1977. The diagnosis of PSA requires arteriography that includes the bilateral iliac systems. Femoropopliteal bypass and, in the case of sciatic artery aneurysm, a ligature of the sciatic artery are recommended for the surgical treatment of PSA.

Surgical strategy of concomitant abdominal aortic aneurysm and gastric cancer

PURPOSE

Selecting the most appropriate surgical approach for patients with abdominal aortic aneurysm (AAA) and concurrent gastric cancer remains controversial. In an attempt to develop guidelines for the management of two concurrent lesions, a retrospective review of patients with concomitant AAA and gastric cancer was undertaken.

METHODS

During the period from January 1985 to December 1992, a total of 222 patients with AAA were admitted to our hospital. Among these, seven patients (3.2%) had gastric cancer and concurrent AAA. Six of the seven patients were treated surgically for both lesions with either a one- or two-stage operation. One patient underwent only an exploratory laparotomy because of the peritoneal dissemination of the gastric cancer. Four of the six patients underwent a two-stage operation. In three cases, the resection of the malignancy was performed first because the gastric cancer was diagnosed as advanced before operation. In one case, the aneurysmectomy was performed first because the aneurysm was more than 6 cm in diameter and the gastric cancer was in an early stage of development. Two of the six patients underwent a one-stage operation and a simultaneous resection was carried out by way of segregated approaches, such as the retroperitoneal approach for AAA and the transperitoneal approach for the malignant lesion.

RESULTS

Five of the seven patients (71.4%) are still alive. The length of follow-up for these patients ranged from 4 months to 4 years.

CONCLUSIONS

The principles of our surgical approaches for concomitant AAA and gastric cancer are as follows. (1) The lesion that absolutely indicates urgent operation should be operated on first. (2) If the malignant lesion is advanced, it is resected first. (3) If the malignancy is not advanced, the AAA should be resected first by the retroperitoneal approach. (4) Simultaneous resection by way of segregated approaches is useful in some patients with early gastric cancer. (5) Both lesions must be resected eventually for improvement of the long-term survival chances.

Blood flow and kinetics of smooth muscle cell proliferation in canine autogenous vein grafts: in vivo BrdU incorporation

The effects of blood flow on the kinetics of smooth muscle cell (SMC) proliferation were examined in canine autogenous vein grafts with a distal poor run-off model. The in vivo bromodeoxyuridine (BrdU, a thymidine analogue) incorporation method was used to label proliferating SMCs in each layer of the vein graft. The BrdU labeling index (LI) was defined as a percentage of labeled cells compared to the total number of SMCs, and BrdU LIs were measured in the media and the intima of the graft. The development of intimal thickening of grafts was accelerated at 2 to 4 weeks after implantation. In poor run-off limbs with an abnormal blood flow condition, as characterized by a low flow and a low shear stress variation, the intima of the graft thickened more progressively than that in control limbs with a normal blood flow. In both groups, the medial BrdU LIs reached a maximum 3 to 5 days after implantation and decreased thereafter. The peak of the intimal BrdU LIs occurred at 1 week in both groups. The intimal LIs of the grafts in poor run-off limbs (6.34% at 1 week and 2.97% at 2 weeks) were significantly higher than those in control limbs (5.34 and 1.98%) for 2 weeks after implantation (P < 0.05). The medial SMC proliferation and the following intimal SMC proliferation of vein grafts were accelerated prior to development of the intimal thickening. SMC proliferation in the intima was prominent in a poor run-off limb with a low flow and a low shear stress variation.

Combination of preoperative lymphangiography using lipiodol and intraoperative lymphangiography using Evans Blue facilitates the accurate identification of postoperative chylous fistulas

In a 61-year-old woman who underwent extensive retroperitoneal lymph node dissection for rectal cancer, postoperative chylous ascites developed. The site of the lymphatic leak was evident on the lymphangiogram with lipiodol taken prior to the surgical repair of the leak, and another site seen on the intraoperative lymphangiogram using Evans Blue. A direct surgical repair led to good results. Therefore, the combination of pre- and intraoperative lymphangiography facilitates the successful surgical management of rare but alarmingly dangerous postoperative chylous ascites.

The reversibility of impaired prostacyclin production of the vein graft

The effects of changes in shear stress (shear stress variation) on production of prostacyclin (PGI2) were examined in canine autologous vein grafts, which were implanted in the poor or normal distal runoff limbs. Four weeks after grafting, the vein grafts were perfused ex vivo and PGI2 was assayed as 6-ketoprostaglandin F1 alpha. The vein grafts were perfused under constant flow for the first 30 min and then under pulsatile flow simulating an abnormal flow with a low shear stress variation or a normal flow with a high shear stress variation for the next 30 min. Basal production rates (30-min cumulative PGI2 production) of vein grafts implanted in the poor runoff and normal runoff limbs were 1.97 +/- 0.71 and 2.19 +/- 0.40 ng/cm2, respectively, with no significant difference. Pulsatile flow effects (increased PGI2 production between 30 and 60 min) of simulated abnormal versus normal flow were 0.50 +/- 0.50 ng/cm2 versus 2.31 +/- 1.87 ng/cm2 in vein grafts implanted in poor runoff limbs (P < 0.05) and 0.48 +/- 0.41 ng/cm2 versus 3.48 +/- 1.08 ng/cm2 in vein grafts implanted in normal runoff limbs (P < 0.01), respectively. There were no significant differences in simulated normal flow effects between poor runoff limbs and normal runoff limbs. The results indicate that the release of PGI2 in vein grafts may decrease in the presence of an abnormal blood flow with a low shear stress variation. In addition, it is suggested that even vein grafts implanted in the poor runoff could increase PGI2 production once the grafts were placed into normal arterial circulation.

Correlation of long-term results of extra-anatomic bypass and flow waveform analysis

To investigate the correlation between the long-term results of an extra-anatomic bypass and operative flow waveform analysis, we retrospectively reviewed 32 axillofemoral and 21 femorofemoral bypasses for between 1 and 5 years after surgical repair. For the grafts with a type O or I flow waveform pattern, the patency at 3 years (83%) and at 5 years (83%) was superior to grafts with a type II flow waveform (69 and 60%). For the axillofemoral bypass grafts with a type O or I waveform, the patency rate was 92% at 5 years. In contrast with the type II flow, the patency rate was 70% at 5 years. For the femorofemoral bypass with a type O or I and a II flow waveform, the patency at 4 years was 67 and 46%, respectively. In cases with concomitant superficial femoral artery occlusion, the flow waveform was type O or I in 23% and type II in 77%. The patency rate at 5 years was 85% for cases with a patent superficial femoral artery, and 51% for an occluded superficial femoral artery (p < 0.05). These results suggest that the long-term outcome of extra-anatomic bypass correlates with the operative flow wave analysis. In addition, good long-term results and an accurate flow wave pattern depend on the distal run-off, particularly the presence of a patent superficial femoral artery.

Operative transluminal laser angioplasty as the sole treatment for late stenoses of femorodistal artery bypass graft: experimental and clinical studies

To determine the role of Nd:YAG laser thermal angioplasty as the sole treatment for late stenoses of femorodistal artery bypass graft, the lasing effect of a larger size of hot-tip probe (3, 4, and 5 mm) was experimentally studied in vitro. For an adequate lasing effect, 30 watts of laser power output for 3 seconds was needed for the 3 mm probe, 40 watts for the 4 mm probe, and 50 watts for the 5 mm probe, respectively. Based on these results, we used Nd:YAG laser thermal angioplasty alone for 25 grafts, including 16 polytetrafluoroethylene (PTFE) grafts, eight saphenous vein grafts, and one externally supported (EXS) Dacron graft in which the stenotic lesions were detected by deterioration of the Doppler flow waveform pattern or a significant fall in the ankle/brachial pressure index (ABPI). Follow-up was from 3 to 24 months (average of 9 months) for PTFE grafts, from 5 to 21 months (average of 11 months) for saphenous vein grafts, and 13 months for the EXS Dacron graft following femorodistal artery reconstructions. Stenotic lesions were most common in the distal anastomotic sites: 11 PTFE grafts, three saphenous vein grafts, and one EXS Dacron graft. Among these, 13 grafts showed a type II flow waveform pattern at the time of surgery. Clinical success was achieved in 12 of the PTFE grafts (75%), in five of the vein grafts (62.5%), and in the single EXS Dacron graft. Four PTFE and three saphenous vein grafts failed subsequent to repeat intraoperative balloon angioplasty in three and graft extension in three and one graft interposition.(ABSTRACT TRUNCATED AT 250 WORDS)

Management of concomitant abdominal aortic aneurysm and gastrointestinal malignancy

Selecting the most appropriate surgical approach for patients with abdominal aortic aneurysm (AAA) and gastrointestinal malignancy remains controversial. In an attempt to develop guidelines for the management of patients with these two simultaneous lesions, a retrospective review of patients who had concomitant AAA and gastrointestinal malignancy was undertaken. During the period from January 1985 to February 1993, 229 patients with AAA were admitted to our hospital. Among these, 19 patients (8%) had a gastrointestinal malignancy together with AAA and were divided into 2 groups. Group I was composed of 11 patients who underwent either a 1- or a 2-stage operation for both lesions. Group II was composed of eight patients who either underwent an operation for one lesion (six patients) or did not have any operation (two patients). Among group I, six patients underwent the two-stage operation. In four of the six patients, the malignancy was resected first. In the remaining two patients, the aneurysmectomy was performed first, because, in one patient, the aneurysm was more than 6 cm in diameter, and, in the other patient, the aneurysm was a saccular type. Among group I, five patients (two patients with gastric cancer, and one patient each with esophageal cancer, rectal cancer, and malignant lymphoma of the stomach) underwent a one-stage operation. In three of the five patients (two patients with gastric cancer and one patient with esophageal cancer), simultaneous resection was carried out by using segregated approaches, namely, the retroperitoneal approach for AAA and the transperitoneal approach for malignancy. Although the clinical characteristics of the patients were different, 8 of the 11 patients (73%) in group I are still alive, whereas only 1 of the 8 patients (13%) in group II is still alive. The principles of our surgical approaches for concomitant AAA and gastrointestinal malignancy are as follows: (1) The lesion that absolutely indicated urgent surgery was resected first. (2) If both lesions were asymptomatic, the malignancy was resected first. (3) Simultaneous resection using different approaches was useful in some patients with concomitant upper early gastrointestinal malignancy. (4) Both lesions need to be resected eventually for better long-term survival.

Impaired endothelial prostacyclin production of the canine vein graft in a poor distal runoff limb

BACKGROUND

Because blood flow modulates endothelial prostacyclin production, the extent of this production in autologous vein grafts implanted in poor distal runoff limbs needed to be examined.

METHODS

Endothelial prostacyclin production in canine autologous vein grafts was measured in poor distal runoff limbs (poor runoff group) and compared with findings in normal runoff limbs (control group). Vein grafts were perfused in a closed circuit at 3 days and 1, 2, 3, and 4 weeks after implantation; after perfusion for the first 30 minutes in a steady flow (basal prostacyclin production), the grafts were exposed to arachidonic acid (stimulated prostacyclin production) for the following 30 minutes. Prostacyclin, as the metabolite 6-keto-PGF1 alpha, was radioimmunoassayed.

RESULTS

Basal and stimulated prostacyclin production increased in both groups during a period of time after implantation. At 2 weeks when endothelialization was complete, prostacyclin production in the poor runoff group was impaired, compared with the findings in the control group, and this difference increased with time. At 4 weeks the stimulated prostacyclin production was 18.91 +/- 4.03 ng/cm2 in the control group and 11.60 +/- 1.67 ng/cm2 in the poor runoff group (p < 0.05).

CONCLUSIONS

We propose that the impaired capacity of the vein graft to produce prostacyclin in a poor distal runoff may lead to loss of graft patency in reconstructed arteries.

Natural course of endothelium-dependent and -independent responses in autogenous femoral veins grafted into the arterial circulation of the dog

We examined the natural course of endothelium-dependent and -independent responses in reversed autogenous vein grafts during regeneration and tissue repair processes after vein grafting in dogs. Vein grafts implanted in the canine femoral artery were removed, cut into rings, and suspended in organ chambers for isometric tension recording at 3 days and 1, 2, 4, and 6 weeks after implantation. Endothelial cells were denuded from some rings. Control veins were taken from nonsurgically treated femoral veins. Acetylcholine caused endothelium-dependent relaxations in the control veins, whereas in the vein grafts there was no evidence of endothelium-dependent relaxations to acetylcholine 3 days after the operation. Acetylcholine caused endothelium-independent contractions throughout the study. The endothelium-dependent responses to ADP and calcium ionophore A23187 were constantly maintained. Three days after the operation, the amplitude of norepinephrine-induced contractions of the vein grafts was impaired, and at 1 week the amplitude was recovered, although it was significantly smaller than the amplitude of contractions of the control veins at any postoperative period. Endothelium-independent relaxations to sodium nitroprusside were maintained throughout the study. Thus, there was a selective loss of acetylcholine-mediated relaxation in vein grafts in the early postoperative stage, a time when the intima is not thickened. These altered responses in vein grafts in the early postoperative period may have a role in graft failure.

Laparoscopic ligation of splenic artery aneurysm

Efforts should be made to avoid splenectomy because host resistance is maintained by the spleen. We describe here a 50-year-old woman in whom laparoscopic ligation of a splenic artery aneurysm 3.0 cm in diameter was successfully performed, without major complications. Computed tomography performed 5 months later revealed no infarction in the spleen. For this patient with liver cirrhosis and cholelithiasis, laparoscopic cholecystectomy was performed at the same operation. Laparoscopic surgery is most feasible when the splenic artery is markedly tortuous and protrudes from the pancreas.

Graft replacement of pararenal inferior vena cava for leiomyosarcoma with the use of venous bypass

Total replacement of the inferior vena cava across the renal vein confluence was successfully performed under venovenous bypass in a patient with primary leiomyosarcoma of the inferior vena cava. Concomitant resection of metastases by right nephrectomy, partial hepatectomy, and cholecystectomy allowed recurrence-free survival for the past 20 months. To the best of our knowledge, this is the first report of successful replacement of the inferior vena cava for a primary tumor.

The potential risk for subclavian vein occlusion in patients on haemodialysis

Subclavian vein (SCV) stenosis or occlusion can be a late complication of temporary haemodialysis or following catheterisation for intravenous hyperalimentation. In five patients with prior catheterisation or trauma of the SCV, incapacitating oedema of the upper extremity developed only after the creation of ipsilateral arteriovenous (AV) fistulas for haemodialysis. The duration of the previous catheterisation was 2 to 4 weeks. Massive upper extremity oedema developed at 10 days to 22 months with an average of 11 months after the establishment of AV fistulas. Subclavian-axillary vein bypass using a ringed polytetrafluoroethylene graft was successful for one patient and ligation of the AV fistulae led to good results for the other three. In planning vascular access procedures in the upper extremity, venography should be mandatory to try to prevent such complications if a previous history of subclavian catheterisation exits.

Simultaneous resection of abdominal aortic aneurysms and early gastric cancer by retroperitoneal and transperitoneal approach

The surgical approach to patients with abdominal aortic aneurysm and gastrointestinal malignancy remains controversial. We experienced two cases with abdominal aortic aneurysm and gastric cancer, which were treated by a one-stage operation using a different approach. At first, the operation for the aneurysm was done through a retroperitoneal approach and then, a partial gastrectomy for gastric cancer was done through a transperitoneal approach. The postoperative course of both cases was uneventful. The patients were discharge on the 19th and 21st postoperative days, respectively. This one-stage operation using different isolated approaches, such as the retroperitoneal approach for abdominal aortic aneurysm and transperitoneal approach for gastric cancer, was useful for the patients with abdominal aortic aneurysm and particularly early gastric cancer in terms of preventing an infection of the prosthetic graft.

Possible role of cytomegalovirus in the pathogenesis of inflammatory aortic diseases: a preliminary report

To search for possible evidence of a relationship between human cytomegalovirus and aortic diseases, we examined 41 aortic lesions excised at surgery and 16 aortic tissues obtained at autopsy for the presence of cytomegalovirus DNA, by use of polymerase chain reaction. Cytomegalovirus DNA was present in seven (88%) of eight lesions of inflammatory aortic diseases with periaortic fibrosis, five of six inflammatory aneurysms, and all of two aortic occlusive lesions with inflammation. Cytomegalovirus DNA was detected in 20 (61%) of 33 atherosclerotic aneurysms, whereas it was detected in only five (31%) of 16 autopsy samples that showed neither inflammation nor atherosclerosis. Thus the possibility that cytomegalovirus may play a role in the pathogenesis of inflammatory aortic diseases warrants further attention.

High aortic occlusion associated with lupus anti-coagulant

Lupus anti-coagulant (LAC) has been recognised in association with venous and arterial thrombotic events. We retrospectively investigated the relationship between LAC and 21 patients with juxtarenal high aortic occlusion. Four of the 21 patients were LAC positive (19%), one of whom was a woman suffering from systemic lupus erythematosus (SLE). The three men had neither SLE nor any other connective tissue disease. The mean age of the four patients was 43.5 years and three of them (75%) were younger than 50 years of age. By contrast, the mean age of the 17 patients without LAC was 59.9 years and 14 of them (82.4%) were older than 50 years. The postoperative courses in all four patients with LAC undergoing arterial reconstructions were uneventful. Two patients were treated by either anti-coagulation or anti-platelet agents, postoperatively. The present paper demonstrates for the first time that, in patients with juxtarenal high aortic occlusion, the incidence of LAC positivity is higher in younger patients (below 50) than in older patients (above 50). In addition, our results show that a preoperative evaluation of the presence of LAC is required in cases with juxtarenal high aortic occlusion, particularly when the partial thromboplastin time (PTT) is prolonged and the patients are younger than 50 years old.

Rupture of embolised coeliac artery pseudoaneurysm into the stomach: is coil embolisation an effective treatment for coeliac anastomotic pseudoaneurysm?

An elderly woman with an anastomotic pseudoaneurysm of the coeliac artery, after previous treatment of a thoraco-abdominal aortic aneurysm, was treated by stainless steel coil embolisation. One year later, the embolised pseudoaneurysm ruptured into the stomach and total gastrectomy and aneurysmorraphy was necessary. She is leading a normal life 6 months later.

Hemodynamics and the development of anastomotic intimal hyperplasia of the polytetrafluoroethylene graft in dogs

To clarify whether or not changes in wall shear stress play a determinant role in the induction of anastomotic intimal hyperplasia of polytetrafluoroethylene (PTFE) grafts, two experimental models were devised. Based on our previous study (J Vasc Surg, 1985), wall shear stress was defined by variation of wall shear stress (tau-variation) in one cardiac cycle. In experiment I, PTFE grafts were implanted into dogs under conditions of 85.2 +/- 36.2 ml/min of low flow rate with 66.7 +/- 31.1 dyne/cm2 of low tau-variation. In experiment II PTFE grafts were implanted under conditions of 10.1 +/- 4.6 ml/min of low flow rate with 120.4 +/- 37.2 dyne/cm2 of moderate tau-variation. The intimal thickness of 581 +/- 127 microns at one month and 1230 +/- 260 microns at three months implantation at the distal anastomosis under an abnormal flow condition in experiment I was statistically significant compared with findings in the case of 124 +/- 36 microns at 1 month and 171 +/- 74 microns at 3 months at the proximal anastomosis under an abnormal flow condition in experiment I, and with those of 164 +/- 68 microns at 1 month and 195 +/- 57 microns at three months at the distal anastomosis in experiment II (p less than 0.01). Under a normal flow condition, there was no evident intimal hyperplasia at proximal and distal anastomotic sites in experiment I. These observations are pertinent evidence that change in wall shear stress and not the rate of blood flow is the determinant factor related to development of intimal hyperplasia of PTFE grafts at the distal anastomosis.

Intraoperative flow waveform analysis aids in preventing early graft failure following reconstruction of arteries of the legs

To enable early detection and treatment of vascular defects leading to early graft failure, intraoperative flow waveform analyses were carried out during lower extremity arterial reconstructions in 226 patients undergoing 102 aortoiliac/femoral and 124 femorodistal bypass grafts. Flow waveform types III or IV indicated early graft failure. These were noted in seven grafts (6.9%) in the aortoiliac/femoral position and in eight grafts (6.5%) in the femorodistal position. The main cause of the abnormal flow waveform pattern was misinterpretation of preoperative arteriographic findings in aortoiliac/femoral reconstructions and technical errors in anastomoses in femorodistal reconstructions. Of 15 grafts with an abnormal flow waveform pattern, 13 were effectively repaired with patch angioplasty, graft extension, or replacement with thrombectomy. In two grafts, the repair failed and amputation had to be done. Thus, intraoperative flow waveform analysis is a simple, useful, and safe method to detect vascular defects leading to early graft failure. Unless assessment of preoperative arteriographic findings in aortoiliac/femoral reconstructions are accurate and anastomotic techniques in femorodistal reconstructions are refined, early graft failure may occur.

Endothelium-derived relaxing factor and vein grafts

Platelet thrombosis, intimal hyperplasia and progression of atherosclerosis are the most important factors determining the patency of vein grafts for arterial occlusive disease. Interactions between aggregating platelets and the vessel wall play an important role in all of these processes. Recently, it has become evident that the endothelium modulates the underlying vascular smooth muscle by releasing an endothelium-derived relaxing factor, a potent vasodilator and anti-aggregating substance. Little information is available, however, on such endothelial functions of autogenous vein grafts. This review focuses on vascular modulation by endothelium-derived relaxing factor in vein grafts.

Laser thermal angioplasty for early repair of anastomotic stenosis after lower extremity arterial reconstruction: initial experience

Laser thermal angioplasty with a sapphire hot-tip Nd-YAG laser system was effectively used for five grafts with anastomotic stenotic lesions which occurred after femoro-popliteal arterial reconstructions. Stenotic lesions were detected by a combination of Doppler flow waveform analysis and ankle/brachial index measurement in 32 femoro-popliteal and five femoro-tibial arterial bypass grafts in the postoperative period. The laser probe used was 3 to 4 mm in diameter. There were no perforations of the arterial wall caused by thermal ablation, even with repeated pulses of 40 watts of laser energy for 3 s. Contact Nd-YAG laser thermal angioplasty using this rounded, 3-4 mm hot-tip facilitated effective widening of the lumen of the anastomotic stenotic lesion. All patients remain well at follow-up from 9 to 20 months after laser thermal angioplasty.

Surgical repair of primary aorto-jejunal fistula associated with non-specific inflammatory abdominal aortic aneurysm

A 28-year-old man with aorto-jejunal fistula associated with primary saccular abdominal aortic aneurysm (AAA) was effectively treated by resection and replacement with polytetraflouroethylene graft. The aetiology was suspected to be a non-specific aortitis, a rare cause of aorto-enteric fistula.

Evaluation of proposed standards for runoff in femoropopliteal arterial reconstructions: correlation between runoff score and flow waveform pattern. A preliminary report

The Ad Hoc Committee on Reporting Standards of the Society for Vascular Surgery and the International Society for Cardiovascular Surgery (SVS/ISCVS) proposed a new angiographic scoring system that grades the quality of vessels distal to the expected bypass site and calculates an overall runoff score. To assess the predictability of the bypass grafts by this runoff scoring system, we compared this runoff score with postoperative increase in the ankle/brachial index (delta ABI) and intraoperatively measured flow waveform patterns in 60 patients undergoing femoropopliteal bypass grafts. No correlation was found between delta ABI and the runoff score. Graft-patent survival time correlated with the runoff score (r = 0.44, p = 0.02). Grafts with type 0 or I flow waveform pattern with a satisfactory long-term patency had a significantly lower runoff score than grafts with type II, III or IV flow waveform pattern (p less than 0.05). When all the grafts were arbitrarily divided into two groups (grafts with a runoff score over 4.5 and those with a score of less than 4.0), a superior long-term patency was obtained in the low runoff score group with polytetrafluoroethylene (p less than 0.01) and saphenous vein grafts (p = 0.04), but not in the group with composite grafts (p = 0.25). Thus, the SVS/ISCVS runoff scoring system can be used to predict the fate of bypass grafts in femoropopliteal arterial reconstructions.

Simulated blood flow and the effects on prostacyclin production in the dog femoral artery

Effects of blood flow on the production of prostacyclin (PGI2) were examined in the canine femoral artery perfused ex vivo. Each artery was perfused in situ with medium 199, under conditions of simulated blood flow. To simulate the normal and abnormal blood flow waveforms at the same mean flow rate, we developed a flow apparatus capable of making various waveforms by changing the duration when the electromagnetic valve in the circuit was open. Group I (n = 7) was exposed to a steep acceleration waveform followed by a steep deceleration, as the normal flow waveform; group II (n = 7) was exposed to a gentle sloping waveform in the deceleration phase, as the abnormal flow waveform. PGI2 was measured as 6-ketoprostaglandin F1 alpha. PGI2 production was estimated as the cumulative production for the first 5 minutes (acute response) and as the production rate after the first 30 minutes (stable production rate). Under conditions of normal flow, the acute response was 5.87 +/- 2.16 ng/cm2/5 min, whereas under conditions of abnormal flow, the rate was 2.20 +/- 0.27 ng/cm2/5 min (p less than 0.01). Stable production rates were 82.5 and 37.5 pg/cm2/min, respectively (p less than 0.05). Both the acute response and the stable production rate of PGI2 production were greater under conditions of simulated normal flow as compared with findings in the case of an abnormal flow. Our working hypothesis is that the decreased production of PGI2, as well as a deterioration in the implanted graft, under conditions of abnormal blood flow leads to a loss of late patency of the reconstructed arteries.

Possible transmission of serum hepatitis in liver surgery with the ultrasonic dissector

When the ultrasonic dissector is used during liver surgery, the possibility exists of the transmission of virus to the attending hospital personnel because of unavoidable splashing of the irrigating solution. It becomes apparent from this study that the titer of hepatitis B surface antigen in the aspirated irrigating solution is almost the same level as that in the serum. We therefore strongly recommend that eye shields should be used in hepatic surgery when the ultrasonic dissector is in use.

Isotopic study of the effects of platelets on development of intimal thickening in autologous vein grafts in dogs

The effects of platelets on the development of intimal thickening of arterially transplanted autologous vein grafts was investigated using canine poor run-off models. A new apparatus consisting of a shielding lead block to measure in vivo platelet adherence only on the intima of the vein graft was developed. In 23 dogs, 51Cr-labelled platelets (20 microCi/kg) were injected and isotope emission over the grafts was counted. Platelet adherence was expressed as the thrombocyte accumulation index (TAI), i.e. the ratio of counts over the graft under an abnormal flow condition in the poor run-off model to those over the graft under normal flow conditions of the contralateral leg. The TAI of the total graft (TTAI) was significantly high immediately and on the first and third days after implantation. The TAIs of the proximal (PTAI) and distal (DTAI) portions of the vein graft were also high at these same times. At 7, 10, 14 and 21 days, the TAI was almost equal to 1.0, a time at which endothelial regeneration was complete as confirmed by scanning electron microscopy. Thus, the prominent intimal thickening of arterially transplanted autologous vein grafts in dogs, induced under an abnormal flow condition, correlates well with the enhanced platelet adherence on the intima of the vein grafts in an early period after implantation.

Ultrastructural evidence of the effects of shear stress variation on intimal thickening in dogs with arterially transplanted autologous vein grafts

Based on our findings that changes in wall shear stress, not the rate of blood flow, were the main hemodynamic factor related to intimal hyperplasia of autologous vein grafts, we further investigated the effect of wall shear stress variation on sequential ultrastructural changes in the intimal hyperplasia of arterially transplanted autovein grafts, using canine models. As noted, wall shear stress variation (tau-variation) could be defined by the variation in wall shear stress within a cardiac cycle, using a desktop flow waveform analyzer. In Group I, which had a high flow rate of 78.4 +/- 4.6 ml/min and low tau-variation of 36.1 +/- 2.2 dynes/cm2, intimal hyperplasia was significant. Ultrastructurally, there was a marked transformation of intimal smooth muscle cells to secretory cells 2 to 4 weeks after implantation. The surface of the intima was lined with modified smooth muscle cells at 2 weeks after implantation. In Group II, which had a low flow rate of 5.6 +/- 2.2 ml/min and normal tau-variation value (174.6 +/- 13.0 dynes/cm2), intimal hyperplasia was minimal, and there were several layers of contractile type smooth muscle cells, with characteristic myofibrillae. The surface of the intima was lined with endothelial cells at 2 weeks after implantation. These findings suggest that, in regions of low wall shear stress variation, intimal smooth muscle cells of autovein grafts may well become secretory cells, and enhanced platelet adherence could occur during early intimal repair, causing intimal hyperplasia to develop.

Computational simulation method of intraluminal velocity profile proved by laser Doppler velocimeter

Measurement of the intraluminal velocity profile within an artery may provide insight into predicting the prognosis of arterial reconstruction. To confirm the results of the computer simulation method a laser Doppler velocimeter using an optical fiber was applied to canine femoral arteries with good or poor runoff, and the intraluminal velocity profile was measured directly. Under normal flow conditions, remarkable fluctuation was obtained at the layer adjacent to the wall, whereas under poor runoff conditions, stagnation of flow was seen at the layer adjacent to the wall. These results showed that our computational simulation method which calculates the intraluminal velocity profile was a reliable, valid method of on-line analysis of the intraluminal velocity profile.

Successful simultaneous repair of coincidental bleeding malignant lymphoma of the stomach and expanding abdominal aortic aneurysm

In a 61-year-old Japanese man, a concomitant bleeding gastric malignant lymphoma and expanding abdominal aortic aneurysm was evident. Bacterial studies were done for a further 32 laparotomy cases. The patient underwent abdominal aortic aneurysmectomy and partial gastrectomy simultaneously. The postoperative course was uneventful except for temporary pneumonia. The positive rate for bacterial culture in cases of gastric or colorectal cancers was significantly high compared with that in the cases of abdominal aortic aneurysm. From the viewpoint of bacterial contamination, these two operations should be done separately. However, simultaneous operations as in the current case are feasible if great care is taken.

Flow waveform analysis and long-term results of autogenous saphenous vein and polytetrafluoroethylene grafts in 140 arterial reconstructions of the lower limbs

In previous work we have found that the outcome of grafts in the lower limbs correlated with the flow waveform pattern of the artery. We have retrospectively reviewed 140 femoro-popliteal bypass operations involving the use of 75 saphenous vein grafts and 65 polytetrafluoroethylene (PTFE) grafts. For grafts with type 0 or I flow waveform pattern the patency at 4 years (56%) was superior to grafts with the type II, III or IV flow waveform (35%) patterns (P less than 0.05). For saphenous vein grafts with type 0 or I flow, the patency rate was 78% at 3 years and 69% at 5 and 8 years. In contrast with type II, III or IV flow the patency rate was 52% at 3 years, 48% at 5 years and 34% at 8 years, with a statistical significance at 4 years (P less than 0.05). PTFE grafts with type 0 or I flow showed a tendency toward an increased patency which was not significant in comparison with the grafts with type II, III or IV flow (P = 0.12). Saphenous vein grafts with type II flow patterns had an increased occlusive rate in the first year whereas PTFE grafts had the same tendency within 2 years. In both types of graft, early occlusions within a month of operation were encountered in grafts with a type III or IV flow waveform pattern. These results indicate that the fate of the reconstructed arteries of the lower limb could be predicted by flow waveform analysis, and a careful and serial postoperative evaluation of the graft should be made, particularly those with type II, III or IV flow waveform patterns.

Recurrent multiple aneurysms in an axillofemoral graft with coagulopathy

An 84-year-old man presented with multiple non-anastomotic aneurysms in a Dacron axillary femoral graft and coagulopathy. Histological examination and the clinical course suggested that a graft versus host reaction might be the main cause of the aneurysms.