Telecommunication technology for the management of large scale clinical trials: the Gissi experience

The Coordinating Centre (CC) of the Gruppo Italiano per lo Studio della Sopravvivenza nell'Infarto miocardico (GISSI) used telecommunication technology to develop a computerized network system for the data management of the GISSI studies. Through a personal computer (PC), a communication program, a modem and a telephone line, the investigator in each participating centre can connect with a micro-computer at the CC, to recruit/randomize patients and download reports on the progress of the trial. In the first case, the investigator is required to answer a set of predefined questions, and thereby the system automatically checks eligibility criteria and randomly assigns the patient to a treatment arm. In the second case, once the investigator has made a choice from a list of standard reports and the relative query on CC central database, the generation, the formatting and the transfer of the selected report to the PC are executed automatically on line. The main advantages of this system are a reduction in number of mistakes in data completion and in the human and economic resources required, as well as the real time updating of participating centres. The system was successfully adopted in the GISSI-3 trial (200 Coronary Care Units and 19,394 patients enrolled), in the European arm of the CORE trial and it is currently being used in the GISSI-Prevenzione trial.

Serum homocysteine, MTHFR gene polymorphism, and carotid intimal-medial thickness in NIDDM subjects

We assessed the contribution of serum homocysteine levels, an independent risk factor for vascular disease, and of the methylenetetrahydrofolate reductase (MTHFR) C677T mutation to the variability of carotid intimal-medial thickness (IMT) in patients with non-insulin-dependent diabetes mellitus (NIDDM). Ninety-five patients (33 males and 62 females, mean age 53 +/- 10 years) without nephropathy or other vascular complications were enrolled. Fasting total serum homocysteine and other biochemical analytes were measured. The MTHFR polymorphism was determined by the polymerase chain reaction. Common carotid IMT and plaques or stenoses in the carotid district were measured by ultrasonography. Serum total homocysteine concentrations were higher in subjects with the mutant (Val/Val) genotype than in those with the Ala/Val plus Ala/Ala genotypes (P = 0.02). On univariate analysis, carotid IMT was significantly associated with age, body mass index (BMI), systolic blood pressure, and total cholesterolemia. No significant association was found between IMT and serum homocysteine or the MTHFR polymorphism, although a slightly greater IMT was observed in the homozygous Val genotypes. On multiple regression analysis, only age and BMI were independently associated with IMT and explained about 40% of IMT variability. The results did not change when the analysis was restricted to the subgroups with or without atherosclerotic plaques in the carotid district. In 95 Italian NIDDM patients without nephropathy, neither basal levels of serum total homocysteine nor the MTHFR C677T polymorphism predicted significant changes in common carotid intimal-medial thickness.

Pancreatic resections: early results and functional behaviour of the stapled pancreatic stump

Pancreaticoduodenectomy is the standard surgical treatment for patients with pancreatic head cancer. Morbidity and mortality rates following this procedure have constantly decreased over the past several years. Leakage of the pancreaticoenteric anastomosis is one of the most serious complications, often responsible for a fatal outcome. Several methods for the management of the pancreatic stump have been described in order to reduce the worrisome incidence of this complication, with variable results. In this series, the Authors review their experience of 75 pancreatic resections and analyze the early results and functional behaviour of 6 patients in which the pancreatic stump was stapled without pancreaticoenteric anastomosis.

[Surgical resection of lung metastases from breast cancer. Report of 12 cases]

Thirty-nine patients affected with lung metastases from different primary neoplastic sites have been treated between 1990 and 1998 at 2nd Surgical Division of "Regina Elena" National Cancer Institute of Rome. Among them, 12 were metastases from breast cancer-lung metastases were isolated in 9 cases and multifocal in 3 cases, although always in the same lung. Nine cases underwent a thoracotomic approach: in 6 patients we have performed a wedge resection, in 3 cases a lobectomy. Three patients underwent a wedge resection by means of a video-thoracoscopic approach. We have registered 2 post-operative complications and no deaths. Median survival rate was 40 months and 5 year actuarial survival rate was 42%. Surgery for isolated lung metastases seen to be a safe approach and to improve life expectancy in most of patients.

[Carcinoma of the hypopharynx: reconstruction of the pharyngo-esophageal tract after circular pharyngectomy by the transplantation of a U-shaped jejunal loop to the neck]

Reconstruction of the oro- and hypopharynx has specific difficulties duo to their wide diameters. Thirteen patients underwent reconstruction with a free U-shaped jejunal transplant, after circular pharyngo-laryngectomy for hypopharyngeal cancer invading the oropharynx. This transplant included a side-to-side anastomosis between the two limbs of the jejunal loop and allowed reconstruction of the upper digestive tract after wide carcinologic resection of the pharynx. The U-shaped jejunal transplant facilitated the upper anastomosis, especially the upper part where the resection involved the oropharynx. It formed a reservoir behind the tongue and avoided nasal reflux. Best indication are large resections involving the oropharynx.

Internet for clinical trials: past, present, and future

The Internet and World Wide Web have recently been introduced into the management of some aspects of large-scale clinical trials such as remote randomization and data entry and the distribution of information on trial progress. Electronic mail and websites have also been used to enhance communication among people involved in a clinical trial. The Gruppo Italiano per lo Studio della Sopravvivenza nell'Infarto miocardico acuto (GISSI) used telecommunications and the Internet in some recent large-scale clinical trials. GISSI constructed a website to keep the medical and cardiology community informed about the progress of its studies. Websites for clinical trials could play an important role in the future, especially in international clinical trials. The website could provide information such as study material and study news and tools such as electronic forums on protocol application for use by investigators around the world. This article describes the GISSI experience and outlines an appropriate structure for a clinical trial website.

[Klatskin tumor. A study of 15 resected cases]

STUDY AIM

Klatskin tumors are rare. Prognosis is still poor, and long term survival can be expected only after surgery, which is the treatment of choice. The aim of this study is to report the results of 15 resected cases and, by analysis of the literature, to emphasize the progress of the surgical treatment in hilar cholangiocarcinoma.

PATIENTS AND METHODS

Between 1990 and 1998, 27 patients affected by Klatskin tumor were observed. Eight women and seven men underwent surgical resection. The mean age was 59 years. Thirteen patients (48%) had curative resection (7 hilar resection (HR), 5 HR combined with partial hepatectomy (PH) and 1 HR + PH with portal vein resection). Two patients had palliative resection and surgical drainage.

RESULTS

One in-hospital death occurred right after hepatectomy with portal vein resection (6.6%). Postoperative morbidity was 40%. Patients were regularly followed. Ten patients died and 5 were alive at the time of this study. The 1, 2 and 3-year survival after a curative resection was 84%, 54% and 34%. The median survival was 28.5 months. Lymph node involvement did not show a statistically significant difference on median survival between the positive group and the negative group (26.2 vs 29.8 months) because of the small number of patients. Survival after hilar resection at 1, 2, 3, and 5 years was 100%, 57.1%, 28.6% and 0%. Four out of the 6 patients who underwent hilar resection combined with partial hepatectomy were still alive 1, 23, 29, 38 months after resection. Hepatectomy increased mortality (16% vs 0%). Palliative biliary resection and surgical drainage were successfully performed in 2 patients.

CONCLUSION

Aggressive surgical treatment of Klatskin tumor can improve the survival of patients. Careful pre-operative management has to be carried out by a multidisciplinary approach including surgeons, hepatologists, radiologists and pathologists. Hepatic resection including the caudate lobe is often performed in order to obtain microscopic tumor-free margins and curative resection (R0). Biliary drainage and treatment of cholangitis is mandatory before surgery in order to improve the surgical outcome. Surgical treatment is characterized by high technical difficulties, and better results can be achieved by hepatobiliary surgical teams.

Colorectal carcinoma: laparoscopic versus traditional open surgery. A clinical trial

BACKGROUND/AIMS

The purpose of this perspective study was to define the role of laparoscopic surgery in the treatment of colorectal carcinoma.

METHODOLOGY

One hundred colorectal cancer patients were submitted to surgical treatment between 1993 and 1996. Fifty patients were operated on by videolaparoscopy, the other 50 were operated on according to the standard "open" technique. The two groups had similar demographic (age, gender), pathological (site, stage), and surgical (type and extent of resection) data. Early and late results, benefits and drawbacks of the minimally invasive technique are compared to those of standard open surgery.

RESULTS

No intra-operative complications and no operative mortality occurred in the two groups. Early results (complications within 30 days from surgery) were: 1 pneumonia, 3 wound sepsis, and 3 fistulas (one required a reoperation) in the laparoscopic group; 2 wound sepsis and 5 fistulas (spontaneously recovered) in the open group. Late complications occurred in the laparoscopic group only: 1 bowel bridle occlusion 2 months after surgery (that required a reoperation), and 2 stenoses of the colorectal Knight-Griffen anastomosis, successfully treated by dilatation. Concerning the oncologic results, data were calculated on 40 laparoscopic and 43 open curative resections (stage I, II and III): 20% (8/40) of the laparoscopic and 23% (10/43) of the open group patients resulted in neoplastic progression. The neoplastic recurrences were single site (liver or regional) in 3 laparoscopic and in 5 open patients; multiple sites of relapse were observed in 5 laparoscopic (liver, peritoneum and 1 trocar site) and in 5 open (liver, peritoneum and 1 scar) cases. Five-year disease-free survival rates (Kaplan-Meier method) were similar in the two groups: 73.2% in the laparoscopic and 70.1% in the open.

CONCLUSIONS

Laparoscopic surgery seems to be a feasible and effective treatment of colorectal cancer and, with the improvement of technology and surgeon skill, it will represent an excellent alternative to the more diffuse and consolidated open surgery technique.

Laparoscopic total proctocolectomy with ileal J pouch-anal anastomosis

BACKGROUND/AIMS

Minimally invasive surgery has developed as one of the most important advances in surgical techniques. The laparoscopic procedure has been successfully used to perform colonic resections. Inflammatory bowel diseases like ulcerative colitis (UC) and familial adenomatous polyposis (FAP) appear as a main indication for total laparoscopic proctocolectomy.

METHODOLOGY

At the Second Department of Surgery of the "Regina Elena" Institute for Cancer Research, 5 non-selected patients were submitted within a 3-year period (1993-1996) to a total laparoscopic proctocolectomy with a restorative ileal J pouch-anal anastomosis. They comprised 3 males suffering from UC and 2 females affected by FAP.

RESULTS

No patients undergoing laparoscopic procedure were converted. The average operative time was 364 min (480 min in the first case, 290 min in the fifth case). There were no intra-operative or post-operative complications (except a mild peritoneal bleeding in the first case, spontaneously stopped). Post-operative pain was mild and no analgesics were required. Late results were excellent, with good bowel function within 1 year after the operation, without dietetic, working and sport restrictions and without sexual disorders, mainly in males.

CONCLUSIONS

Laparoscopic total proctocolectomy in the hands of skilled laparoscopic surgeons is a feasible, safe and effective procedure, with early and late results comparable to, and in some aspects better than, those obtained with "open" surgery. Moreover, it does not have the disadvantage of intra-operative fluid loss, prolonged post-operative ileus, pain and, in younger patients, psychological discomfort of the wide scar.

Laparoscopic pancreatic surgery: indications, techniques and preliminary results

BACKGROUND/AIMS

The purpose of this study was to evaluate feasibility, effectiveness and reproducibility of laparoscopy in the diagnosis and management of pancreatic diseases.

METHODOLOGY

At the Second Department of Surgery of the "Regina Elena" Institute for Cancer Research, between June 1995 and June 1997, possibilities of the laparoscopy in diagnosis and staging of pancreatic malignancies were studied. Twenty-four consecutive cases of pancreatic adenocarcinoma were observed: 10 patients were pre-operatively considered unresectable and 14 cases were submitted to a laparoscopic study. Moreover, in the period from January 1996 to June 1997, a study on the therapeutical potentialities of laparoscopy was performed, attempting 5 pure or assisted laparoscopic pancreatic resections.

RESULTS

Three cases (21.4%) out of the 14 evaluated were found to be metastatic (liver and peritoneum); 11 patients were successfully resected with an open standard procedure. Laparoscopy allowed a correct staging in all cases, even in the 21.4% of ultrasonography (US), TC and angiographic false negatives. The 5 operative procedures consisted of one resection of a cystadenoma, one resection of a head cyst, one Wirsung lithotomy with pancreaticojejunostomy for lithiasic chronic pancreatitis, and two body-tail splenopancreatectomies (1 cystadenoma and 1 adenocarcinoma). Satisfactory early results were obtained in all these cases.

CONCLUSIONS

Laparoscopic diagnosis and staging of pancreatic cancer is a valuable method which allows a very high accuracy. Pancreatic resections, even if feasible, are still to be considered an experimental procedure.

[U-shaped free jejunum transplant]

Reconstruction of the oro and hypopharynx has specific difficulties due to their wide diameters. Seven patients underwent reconstruction with a free U-shaped jejunal transplant, after circular pharyngolaryngectomy for hypopharyngeal cancer invading the oropharynx. This transplant included a side-to-side anastomosis between the two limbs of the jejunal loop. This transplant allowed reconstruction of the upper digestive tract after wide carcinologic resection of the pharynx. The U-shaped jejunal transplant facilitated the upper anastomosis, especially at the upper part where the resection involved the oropharynx. It formed a reservoir behind the tongue, and avoided nasal reflux. The best indications are large resections involving the oropharynx.

Diagnostic and surgical features of Klatskin tumors

AIM OF STUDY

Klatskin tumors are rare and their prognosis is poor. Long term survival can be expected only after a surgical resection, the treatment of choice. The aim of this study is to report our single centre experience and, by literature analysis, to define the role of surgery in the treatment of hilar cholangiocarcinoma.

MATERIALS AND METHODS

Between 1990 and 1998, 27 patients affected by Klatskin's tumor were observed. Eight women and seven men (mean age 59 years) underwent surgical resection. Thirteen patients (86%) had curative resection (7 hilar resection (HR), 4 HR combined with partial hepatectomy (PH) and 2 HR + PH with portal vein resection). Two patients (13%) had palliative biliary resection and surgical drainage.

RESULTS

One in-hospital death was recorded after a right hepatectomy with portal vein resection (6.6%). Postoperative morbidity rate was 40%. Patients were regularly followed up to date or to death. Ten patients died and 5 survived. The 1-, 2- and 3-year survival rate after curative resection was 84%, 54% and 34%. The median survival was 28.5 months. Lymph node involvement did not show a statistically significant difference on median survival between the positive group and the negative group (26.2 vs 29.8 months), nor did perineural invasion, because of the small number of patients. The 1-, 2-, 3- and 5-year survival rate after isolated hilar resection was 100%, 57.1%, 28.6% and 0. Four out of 6 patients who underwent hilar resection combined with partial hepatectomy are still alive 1, 23, 29, 38 months after resection. Hepatectomy increased mortality (16% vs 0). Palliative biliary resection and surgical drainage were successfully performed in 2 patients with satisfactory results.

CONCLUSION

Aggressive surgical treatment of Klatskin tumors can improve patients' survival. Careful preoperative management has to be carried out by a multidisciplinary approach including surgeons, gastroenterologists, radiologists and pathologists. Hepatic resection involving the caudate lobe is often performed in order to obtain microscopic tumor-free margins and curative resection (R0). Biliary drainage and treatment of cholangitis is mandatory before surgery in order to improve surgical outcome. Surgical treatment is characterized by high technical difficulties and better results can be achieved by hepatobiliary surgical teams.

[A splenic autograft in posttraumatic splenectomies]

BACKGROUND AND AIMS

The authors report their experience regarding the use of autologous splenic transplantation in post-traumatic splenectomy unable to be treated using conservative surgery. After reviewing the international literature on the subject, they report a retrospective survey of cases treated from January 1992 to December 1996.

METHODS

Owing to the particular logistic location of the hospital in an area with a high density of industry and at the crossroad of major road and rail routes, a total of 56 patients were admitted to the Emergency Ward suffering from abdominal trauma in 4 years. The patients included in this study could not be treated using conservative surgery: the study group included 15 patients aged between 14 and 76 years old. The surgical technique consisted of the graft of sections of splenic pulp in omental pockets, subsequently marked using metal clips. In order to evaluate splenic immunological function a complete hemochromocytometric examination was performed in each patient at the same time as emergency preoperative tests consisting of peripheral blood strip and pitted cells (PC) assay. This was followed by postoperative evaluations at weekly intervals, including platelet count, Howell-Jolly bodies assay (HJb), immunoglobulin M assay and hepatosplenic scintigraphy using erythrocytes marked with 99m-Technetium pertechnetate (99mTc).

RESULTS

An adequate functional recovery of splenic tissue was achieved in all patients with partial recovery of hemocatheretic and immunological function.

CONCLUSIONS

The authors' clinical experience confirmed the data inferred from animal experiments: the simplicity of the preparation technique and the autologous transplantation of splenic pulp in the absence of major complications confirms the possibility of applying this method in all splenectomies performed under emergency conditions.

Ten-year follow-up of the first megatrial testing thrombolytic therapy in patients with acute myocardial infarction: results of the Gruppo Italiano per lo Studio della Sopravvivenza nell'Infarto-1 study. The GISSI Investigators

BACKGROUND

We conducted a 10-year follow-up of the 11 712 patients with acute myocardial infarction randomized in the Gruppo Italiano per lo Studio della Sopravvivenza nell'Infarto-1 study, the first large trial assessing thrombolytic therapy.

METHODS AND RESULTS

Information on survival at 10 years was obtained for the 93% of all randomized patients through the census offices of their towns of residence. The difference in survival produced by streptokinase and sustained up to 1 year was still significant at 10 years (log-rank test, P=0.02), with the absolute benefit of 19 (95% CI 1 to 37) lives saved per 1000 patients treated. The time dependence of the extent of the benefit was confirmed, as the higher mortality rate reductions found in patients treated earlier were still present at 10 years. In the overall population, most of the benefit was obtained before hospital discharge (RR 0.81, 95% CI 0.72 to 0.90), since no difference in survival between thrombolyzed and control patients discharged alive was found at 10 years (RR 0.98, 95% CI 0.90 to 1.06). However, a slight albeit nonsignificant divergence of the survival curves of patients randomized within the first hour was observed [90 (95% CI 34 to 146) lives saved per 1000 at 10 years versus 72 (95% CI 37 to 107) lives saved at hospital discharge].

CONCLUSIONS

The benefits of a single intravenous infusion of 1.5 million units of streptokinase in prolonging survival of patients with acute myocardial infarction is sustained up to 10 years, with a still-evident trend in favor of the patients admitted earlier.

Epidemiology of post-infarction risk stratification strategies in a country with a low volume of revascularization procedures. GISSI-Prognosis Investigators

AIMS

The aims of the GISSI Prognosis Registry were to describe the diagnostic strategies initiated in acute myocardial infarction patients by a representative sample of Italian cardiological centres, and to determine which clinical or hospital characteristics were associated with the initiation of invasive diagnostic or therapeutic procedures.

METHODS AND RESULTS

Baseline characteristics, major in-hospital events and the indication and results of invasive and non-invasive procedures were collected on 1489 acute myocardial infarction patients discharged alive from 65 Italian cardiological centres over a period of 3 months. Twenty-five percent of centres had on site catheterization laboratories while the rest did not. Statistical significance was analysed by chi-square tests for categorical variables. A two-sample Student t-test was used to compare continuous variables. The adjusted analysis was performed utilizing multiple logistic regression models. The most performed procedures were standard, non-invasive: 57.8% of the patients underwent an exercise stress test, 70.8% ambulatory ECG monitoring and 95.6% two-dimensional echocardiography. Nuclear or echocardiographic imaging tests were performed in 40% of acute myocardial infarction survivors. Overall, coronary angiography was planned in 549 patients (36.9%). Variables independently associated with the indication for coronary angiography were residual ischaemia, younger age, contraindication to exercise stress testing, level of patients' education, higher volume of non-invasive diagnostic tests, and male sex. Overall, during a 6-month follow-up period, coronary angiography, percutaneous transluminal coronary angioplasty and coronary artery bypass surgery were performed, respectively in 35%, 10% and 8% of the study population.

CONCLUSIONS

The setting where cardiologists practise determines the patterns of care in acute myocardial infarction patients more than the characteristics of the patient. The absence of evidence-based guidelines on the more complex and expensive procedures favour empirical attitudes and practices. The confirmation in a prospective cohort of patients, which aims to represent the care of a whole country, suggests that more effort should be given to the implementation of controlled studies rather than periodical reformulation of guidelines not supported by hard data.

Neurophysiological evaluation of sexual dysfunction in patients operated for colorectal cancer

Sexual dysfunction after colorectal cancer surgery may be severe and occurs in 25% to 100% of cases. Thirty-eight patients underwent colorectal resection; eight (21%) who were totally impotent and two (5%) who had ejaculatory failure were therefore studied to better understand the neurophysiological alterations related to this type of surgery. The patients were evaluated after surgery with electrophysiological testing, including examination of the sacral reflex (SR), pudendal somatosensory evoked potential (PEP), and motor evoked potential (MEP) responses. Sudomotor skin response (SSR) was also studied in a group of patients. Of the 38 patients studied, 29 showed abnormalities: six of SR, three of PEP, six of MEP, and fourteen of SSR. Only a combination of all these tests permits correct evaluation of the sexual dysfunction.

Loss of FHIT expression in gastric carcinoma

Loss of heterozygosity involving the short arm of chromosome 3 has been reported in gastric and other human tumors. We have cloned and mapped a candidate tumor suppressor gene, FHIT (fragile histidine triad), to this chromosomal region (3p14.2). To investigate the role of FHIT gene alterations in the development of gastric carcinoma, we examined 8 gastric carcinoma-derived cell lines and 32 primary adenocarcinoma samples by Southern blot analysis. We also analyzed the integrity of FHIT transcripts by reverse transcription-PCR. The occurrence of alterations in the FHIT gene and its transcript correlated with the absence of Fhit protein expression by immunoblot analysis in the cancer cell lines. Four of eight cell lines showed deletion or rearrangement within the FHIT gene, together with the absence of the wild-type transcript and the Fhit protein. Among the primary gastric carcinomas, rearrangement of the FHIT gene and/or aberrant reverse transcription-PCR products were detected in 17 of 32 (53%) tumors, and 20 of 30 (67%) samples exhibited an absence of Fhit protein expression. Gastric cancer is thought to develop from carcinogenic exposure, possibly explaining the high frequency of abnormalities in the FHIT gene, a fragile locus exhibiting susceptibility to carcinogen-induced alterations. The consequent absence or reduction of Fhit protein expression is consistent with the proposal that the FHIT gene is a preferential target of environmental carcinogens and that FHIT inactivation plays a role in the development of gastric cancer.

Smoking is not a protective factor for patients with acute myocardial infarction: the viewpoint of the GISSI-2 Study

BACKGROUND

A protective role of smoking in terms of mortality after acute myocardial infarction treated with thrombolytic agents was recently suggested, and this was attributed to the increased chance that smokers will achieve early complete perfusion after thrombolysis. The purpose of the present analysis of the GISSI-2 database was to evaluate the effect of smoking on in-hospital mortality, reinfarction and stroke rates.

METHODS AND RESULTS

This analysis concerns 2611 (26.9%) nonsmokers, 1932 (19.9%) ex-smokers and 5151 (53.0%) active smokers with a first confirmed MI, treated with thrombolytic agents. The relationship between smoking habits and outcome was evaluated by unadjusted and adjusted analysis. Reinfarction and stroke rates were significantly lower in smokers (1.5 and 0.8% respectively) than in ex-smokers (2.5 and 1.1%) or nonsmokers (2.5% and 1.2%). In-hospital mortality significantly increased from 4.7% in smokers, to 7.6% in ex-smokers and 13.8% in nonsmokers. These differences may be due to the different characteristics of the three groups; in particular, smokers were younger than nonsmokers. After adjusted analysis, smoking was not confirmed to be a protective factor for reinfarction, stroke and mortality: OR 1.35 (95% Cl 0.91-2.02), 0.79 (95% Cl 0.58-1.06) and 0.80 (95% Cl 0.60-1.07) respectively.

CONCLUSIONS

Active smokers presented a lower incidence of reinfarction, stroke and in-hospital mortality rates, but after adjustment for other clinical-epidemiological variables, the apparent protective role of smoking was not confirmed.

Gastric cancer. Clinico-biological updating and analysis of 400 operated cases

Gastric cancer is a rather common disease worldwide. In Italy it still accounts for 15,000 deaths annually. A sharp drop in the incidence rate of Lauren's intestinal histotype has been reported, whereas the frequency of the diffuse histotype is relatively steady. If the histogenesis of the latter is still somewhat obscure, the intestinal type confirms the sequence: atrophic gastritis--intestinal metaplasia--dysplasia--neoplasia. These different stages of development can nowadays be singled out through a series of indicators, the most reliable of which are the pepsinogen I/pepsinogen II ratio, the presence of sulphomucins and Lewis antigens in the gastric juices and NOR (Nucleolar Organizer Regions), cell ploidy and oncogenes determination. The genes involved in the neoplastic transformation are mostly oncosuppressors, the most frequent alterations being those relative to the APC gene, p53 and c-myc. In addition to the by now indispensable pathological staging of the disease, the modern prognostic factors are arising great interest: the most significant are the immunohistochemical examination of the peritoneal washing, and cell ploidy. Surgery is still the only potentially curative treatment: the earlier surgery is performed in the course of disease, the greatest the curative potential. The Authors' experience, which includes 400 operated cases with complete follow-up records, is here reported. The resectability rate turned out to be 84%, overall operative mortality was 6.5% with that due to surgical causes along being 3.7%. Overall survival at 5 years was 36%, while that of the curative operations 47%. Good results were obtained with the association surgery + intraoperative radiotherapy which resulted in a significant decrease in local recurrences of the disease.

Predischarge two-dimensional echocardiographic evaluation of left ventricular thrombosis after acute myocardial infarction in the GISSI-3 study

Left ventricular (LV) thrombosis can be found in patients with acute myocardial infarction (AMI). No wide multicenter trial on AMI has provided information about LV thrombosis until now. The protocol of the GISSI-3 study included the search for the presence of LV thrombosis in patients from 200 coronary care units that did not specifically focus on LV thrombosis. We examined the GISSI-3 database results related to 8,326 patients at low to medium risk for LV thrombi in which a predischarge echocardiogram (9 +/- 5 days) was available. LV thrombosis was found in 427 patients (5.1%): 292 of 2,544 patients (11.5%) with anterior AMI and in 135 of 5,782 patients (2.3%) with AMI in other sites (p <0.0001). The incidence of LV thrombosis was higher in patients with ejection fraction < or = 40% (151 of 1,432 [10.5%] vs 276 of 6,894 [4%]; p <0.0001) both in the total population and in the subgroup with anterior AMI (106 of 597 [17.8%] vs 186 of 1,947 [9.6%]; p <0.0001). Multivariate analysis showed that only the Killip class > I and early intravenous beta-blocker administration were independently associated with higher LV thrombosis risk in the subgroup of patients with anterior AMI (odds ratio 1.75, 95% confidence interval 1.28 to 2.39; odds ratio 1.32, 95% confidence interval 1.02 to 1.72, respectively). In patients with anterior AMI, oral beta-blocker therapy given or not given after early intravenous beta-blocker administration does not influence the occurrence of LV thrombosis. The rate of LV thrombosis was similar in patients treated or not treated with nitrates and lisinopril both in the total population and in patients with anterior and nonanterior AMI. In conclusion, in the GISSI-3 population at low to medium risk for LV thrombi, the highest rate of occurrence of LV thrombosis was found among patients with anterior AMI and an ejection fraction < 40%. Killip class > I and the early intravenous beta-blocker administration were the only variables independently associated with a higher predischarge incidence of LV thrombosis after anterior AMI.

Cost-effectiveness analysis of early lisinopril use in patients with acute myocardial infarction. Results from GISSI-3 trial

The cost effectiveness of early treatment with lisinopril in acute myocardial infarction (MI) was estimated using survival and cost data gathered prospectively during the hospitalisation of the overall population of patients enrolled in the third study of the Gruppo Italiano per lo Studio della Sopravvivenza nell'Infarto (GISSI-3), which assessed the efficacy of early (within 24 hours) treatment with an angiotensin-converting enzyme (ACE) inhibitor (lisinopril) for 6 weeks in a group of 19,394 relatively unselected patients with acute MI. A statistically significant reduction in 6-week mortality was achieved among patients treated with lisinopril when compared with patients allocated to the control group (absolute reduction in mortality: 7.5 +/- 3.6 lives saved per 1000 treated patients). The comparative cost-effectiveness ratio for the use of lisinopril, expressed as cost per additional survivor among patients randomised to receive lisinopril, was $US2080 per life saved (1993 values). The sensitivity analysis conducted to examine the effects of varying the estimated absolute reduction in mortality throughout its 95% confidence interval, which ranged from 14.6 to 0.4 lives saved per 1000 treated patients, showed that the cost-effectiveness ratios consequently vary from $US1121 to $US40,910 per life saved. The cost effectiveness of early treatment with lisinopril of a relatively unselected population of patients with acute MI compares very favourably with that of other therapies judged to be worthwhile by the medical community.

Surgical treatment of abdominal aortic aneurysm in association with horseshoe kidney. Three case reports and a review of technique

Horseshoe kidney is a rare congenital anomaly that can create various technical problems during surgery for repair of abdominal aortic aneurysm. The diagnosis of this anomaly should be confirmed preoperatively in order to plan surgical strategy. Nowadays, in more than 90% of all cases, ultrasonography, contrast computerized tomography, urography, and angiography are the best instrumental methods of detecting this anomaly in association with abdominal aortic aneurysm. The transperitoneal approach assures the best exposure of the kidney, the ureters, the aneurysm, and both iliac vessels, but the renal isthmus can pose a problem in reimplanting aberrant renal arteries. When it is known preoperatively that renal revascularization should be performed, the left extraperitoneal approach is a better choice. In any event, the coexistence of horseshoe kidney and abdominal aortic aneurysm does not preclude the treatment of the latter. In elective surgery of abdominal aortic aneurysm, the morbidity and mortality rates in the presence of horseshoe kidney are much the same as those in the presence of normal kidneys. The best results in this kind of surgery are obtained by adapting one's surgical technique to each anatomical variant that is encountered.