[Clinical and epidemiological cancer registration in Germany]. / Klinisch-epidemiologische Krebsregistrierung in Deutschland.

OBJECTIVE: Function and funding of detailed clinical cancer registries (CCRs) is defined by German Social Code Book V (SGB V) and shall be implemented by the end of 2017. CONTENT: Cancer registration according to regionally defined catchment areas, feedback of results and quality assurance are the basis which determines principles of operation and use of data. Each clinical department delivers only its own findings and therapy, while compilation by the clinical cancer registry describes the patients' way through the regional network of medical care. In this way, oncological centers are not burdened by troublesome documentation of data which originate from other clinics. CONCLUSION: After successful implementation of CCRs, interested physicians and clinics are able to spend time for analysis and use of meaningful data with the objective of improving quality of care within the region, implementing innovative therapies and presenting their results, and generating new hypotheses to stimulate research.

[Changes in therapy for breast cancer patients as a result of mammography screening. An analysis of Bavarian cancer registry data from 2000 to 2008]. / Veränderungen der Therapie des Mammakarzinoms in Bayern seit Einführung des Mammografie-Screenings. Eine Analyse bayerischer Krebsregisterdaten der Jahre 2000 bis 2008.

PURPOSE: A quality controlled mammography screening programme was initiated at the end of 2003 in Bavaria, a region with 12.5 million inhabitants, and transferred over to the national screening programme at the end of 2006. The purpose of this study was to evaluate immediate population-based consequences of mammography screening on breast cancer therapy. METHODS: Data from 75 475 breast cancer cases, diagnosed between 2000 and 2008 and registered in one of the 6 Bavarian clinical cancer registries were analysed. 51.4% of these patients were between 50 and 69 years of age and therefore the target population for screening. Trends of prognostic factors and standard therapies were calculated for 3 age groups (≤49 years, 50-69 years, ≥70 years) by means of annual percentages as well as 95%-confidence intervals for the percent difference between 2000 and 2008 (year of diagnosis). For interpretation of therapy trends, logistic regression models were calculated. RESULTS: Therapy trends showed that the increasingly favourable stage distribution may have resulted in the reduction of more radical surgical methods such as mastectomy (2000: 32.6%; 2008: 19.6%) or axillary dissection (89.0% vs. 37.0%), especially for women aged 50-69. An increase of radiation therapies (59.7% vs. 66.6%) can be explained to some extent by the increase in breast conserving surgeries. The shift to more favourable prognostic factors led, in accordance with the guidelines, to an increase of the proportion of singular endocrine therapies (28.5% vs. 40.7%), a decrease of chemotherapies (20.4% vs. 13.1%) and therefore to more gentle systemic therapies overall. These trends strengthened in the years following the introduction of screening, with a simultaneous rise of screening participants in the target population. CONCLUSION: The introduction of mammography screening in Bavaria has already shown the expected trend towards more favourable prognostic factors. Among other things, this could be a reason for the increasing use of more gentle therapies. Whether the screening in Bavaria leads to a mortality reduction, has to be analysed on the basis of an initial comparison of participation status followed by the trends in mortality thereafter.

Quality of life in localised malignant melanoma.

BACKGROUND: The incidence of melanoma is still increasing in fair-skinned populations. At least 80% of patients have localised disease and expect a 5-year relative survival of >90%. PATIENTS AND METHODS: In 2003-2004, disease-free patients with localised melanoma were recruited from the Munich Cancer Registry to answer quality-of-life (QoL) questionnaires 2 years after treatment. RESULTS: A response rate of 72% was achieved from a total of 1085 distributed questionnaires. Hundred and seventeen questionnaires had to be excluded because of updated information about secondary tumour and progression events. Thus, questionnaires from 664 patients were evaluated. QoL scores in melanoma patients were essentially similar to those of a general population. Differences were detected between women and men concerning emotional and sexual functioning. Age and number of comorbidities were the strongest factors influencing most all aspects of QoL. Fifty percent of patients referred to deficits in communication with their doctors. CONCLUSIONS: Patients who overcome melanoma do not necessarily have a reduced QoL. Strategies used by these melanoma patients resulted in similar levels of coping as previous studies in comparable general populations. Nevertheless, doctor-patient communication was correlated with emotional and social functioning and should be emphasised in treatment and care of melanoma patients.

[Ten years of epidemiological cancer registration in Bavaria]. / Zehn Jahre Krebsregistrierung in Bayern.

For a large territorial state like Bavaria only a decentralised cancer registration structure promises successful results: in the form of regional clinical cancer registries and--using the clinical registration as a base--one population-based registry. After ten years of epidemiological cancer registration in Bavaria it can now be shown that the chosen registration concept has proved itself. Currently the completeness of cancer notifications exceeded the international recommended threshold of 90%. A largely complete data stock is available for the years of diagnosis from 2004 to 2005. The task sharing between clinical and population-based cancer registries avoids double registration of data. Both types of registries are supporting physicians and hospitals with a wide palette of services. Together they enable transparency of cancer occurrence as well as transparency of health care for tumour patients.

[Guideline for the Early Detection of Breast Cancer in Germany 2008. Recommendations from the short version]. / Stufe-3-Leitlinie Brustkrebs-Früherkennung in Deutschland 2008 : Auszug aus der Kurzfassung: Handlungsempfehlungen.

The updated 2008 German Guideline for Early Detection of Breast Cancer provides evidence-based and consensus-based recommendations of the knowledge gained by the German Society for Surgery and the German Society of Plastic, Aesthetic, and Reconstructive Surgeons together with 29 professional societies, associations, and nonmedical organizations. The guideline is meant to assist physicians, healthy women, and patients in medical decisions with recommendations regarding the diagnostic chain in early detection of breast cancer. In addition to these recommendations, the guideline also includes descriptions of quality assurance for resources, procedures, outcomes, and evaluation using a set of quality indicators. It updates the previous version from 2003. The guideline's recommendations are presented. They are described in detail in the full publication (in German) Geburtsh Frauenh 2008; 68:251-261. The long version of the Guideline, methods report, and evidence report are available on the internet at www.awmf-leitlinien.de (reg. no. 077/001) with free access.

Ten-year data on 138 patients with endometrial carcinoma and postoperative vaginal brachytherapy alone: no need for external-beam radiotherapy in low and intermediate risk patients.

OBJECTIVE: To evaluate long-term outcome, risk factors, and causes of death in stage I-IIIA endometrial carcinoma (EC) patients treated only with adjuvant vaginal brachytherapy (VB) and to clarify for which subgroups of patients it is safe to omit external-beam radiotherapy (EBRT). METHODS: Out of 224 EC patients receiving postoperative radiotherapy between 1990 and 2002, 138 had VB alone in curative intent (FIGO [2002]: 85%I, 12%II, 3%IIIA; 18 low risk [IA G1-2, IB G1], 103 intermediate risk [IB G2-3, IC G1-2, IIA-B G1-2], 17 high risk [IC G3, IIIA]). After surgery+/-lymphadenectomy, HDR-brachytherapy prescription (in 95.7% of patients) was 3x10 Gy to the surface or 3x5 Gy at 5 mm tissue depths. RESULTS: Median follow-up was 107 months (range 3-185). Three intermediate and 7 high risk-patients relapsed. The 10-year vaginal control was 99.2%, locoregional control was 95.2% (low/intermediate/high risk: 100%/98.9%/68.8%), and disease-free survival (DFS) was 91.7% (100%/96.8%/55.2%). Risk factors for poor DFS were lymphovascular space invasion, > or = 50% myometrial invasion (univariate, p<0.05), pathological FIGO-stage, and grade 3 (uni-/multivariate, p<0.05). Leading causes of deaths (n=41) were cardiovascular disease (29%) and other malignancies (24%) ahead of EC (19.5%). The 10-year overall survival was 68.5% and the disease-specific survival was 92.4%. Thirty-five secondary tumors in 31 patients led to a higher actuarial death rate (10-year 9.9%, 15-year 17.7%) than EC (7.6%). CONCLUSIONS: Restricting adjuvant therapy to VB alone seems to be safe in low and intermediate risk EC and can be recommended. As death rarely relates to early-stage EC, value of adjuvant therapy is probably better reflected by DFS rather than by overall survival.

Are we wasting our time with the sentinel technique? Fifteen reasons to stop axilla dissection.

Originally, surgery for breast cancer involved removing the pectoral muscles and the regional lymph nodes. This drastic technique was based on Halsted's paradigm of continuous tumour spread via the lymph nodes. In the last century, the amount of surgery has gradually decreased as breast cancer has been recognised as a primary systemic, or partially systemic, disease. Nowadays, breast-conserving therapy is widely used, but axillary lymph node dissection (ALND) and the sentinel technique are still common. Can the patient also be spared such axillary surgery? We have assembled convincing arguments against ALND (and therefore also against the sentinel technique) based on the probability that positive lymph nodes are unlikely to metastasise and that removing them is redundant. At least a discussion of this topic is more than overdue, even if it may be too early to change behaviour.

Influence of hospital volume on local recurrence and survival in a population sample of rectal cancer patients.

AIMS: To investigate the role of hospital volume and individual hospitals on long term outcomes (local recurrence and survival) of rectal cancer patients. METHODS: One thousand thirty-eight patients with rectal cancer were diagnosed between 1996 and 1998. From these, we analysed 884 patients with a resected invasive primary rectal cancer. Median follow-up was 5.7 years. The impact of hospital volume (<10, 10-30 and >30 rectal cancer patients annually) on local recurrence and survival was examined in a Cox model. Differences between the four largest clinics in the high volume group were also investigated. RESULTS: In the multivariate model predicting survival the following variables were significant: UICC stage, grade, age, local recurrence, and (neo-) adjuvant therapy treatment. In the multivariate model predicting local recurrence UICC stage, tumour localisation, and neoadjuvant therapy treatment were significant variables. Hospital volume was not a significant factor for survival or local recurrence. Within the high volume category one hospital showed significantly worse local recurrence rates than all other hospitals, but no survival difference could be seen between the four largest hospitals of the high volume group. CONCLUSIONS: This analysis of a rectal cancer population found that hospital volume did not determine survival or local recurrence. Detailed clinical data with long term follow-up from cancer registries are vital to demonstrate the quality of routine care.

[Aftercare or not--that is the question!]. / Nachsorge oder keine Nachsorge - das ist hier die Frage!

There are about 345,000 Germans who have been diagnosed with urological cancer over the last 5 years expecting help from the health services. Even if aftercare is now a substantial part of cancer treatment, its effectiveness has been proven for only a few cancer types. For the subset of urological cancers, evidence for effective diagnostic measures in the framework of aftercare exists only for cancers of the bladder and testis. In this case, the main objective is early detection of local recurrences and providing a curative chance for small cancers. Lack of effectiveness may result from the logic of cancer itself, and is not the consequence of insufficient study designs. Therefore, the main task of aftercare is talking to the patient, to determine individual needs and to recognise treatment consequences. For asymptomatic patients, there is no indication for routine aftercare diagnostics.

The process of metastasisation for breast cancer.

To investigate the process of metastasis, primary clinical data and disease events such as metastases, local recurrence and survival (median follow-up 9.4 years) from the Munich Cancer Registry from 1978 to 1996 were analysed. Since metastases, even from small tumours, may be initiated before the diagnosis of the primary tumour, the growth of the primary tumour and metastasisation may be two autonomous processes. In our data, survival following metastases was almost unrelated to primary tumour size. However, the number of M1 cases and the time to metastasisation depended on the tumour diameter at diagnosis. The time from initiation of metastases to its diagnosis was estimated as 5.8 years. The growth of metastases was almost homogeneous. However, the growth time following metastasisation-depending on the metastases-free time, receptor status and histological grade-only varied by approximately a factor of 2. Local recurrence, above all, was an indicator of metastases. Furthermore, local recurrence may also have the potential to metastasise. Excess mortality due to local recurrence was estimated up to 9.3 years after diagnosis. Our hypothesised metastases model illustrates the importance of early detection, the concept of breast-conserving therapy and additional metastases from local recurrence. It highlights the benefits of optimal local therapy of the primary tumour and the limitations of systemic therapy. It also questions the use of axilla dissection and lymph node irradiation. Its generalisation to solid tumours may help to clarify many of the current controversial debates.

Moderate progress for ovarian cancer in the last 20 years: prolongation of survival, but no improvement in the cure rate.

Although ovarian cancer treatment has advanced in the last 20 years, long-term survival remains stable. The purpose of this study was to determine whether survival has improved in line with treatment advances in a population-based prospective cohort of ovarian cancer patients (1978-1997, with a follow-up through to 2000). The 10-year overall survival rate for cancer patients was similar before and after 1988: 32.2% (n=1661) and 34.4% (n=2089). For patients after 1988, a 12-month prolongation of median survival was observed. In terms of stage according to the International Federation of Gynecology and Obstetrics (FIGO), only FIGO I and FIGO II patients showed, in addition to a prolongation in survival, an absolute improvement of 12.9 and 12.6% after 5 years and of 13.2 and 8.6% after 10 years. This hardly affected the survival of the total sample. For the most frequent stage FIGO III patients and for FIGO IV patients, a prolongation in survival time, but no improvement in survival rate, was seen after five or 10 years. The progress in FIGO I and II patients may be due to more accurate staging. More effective chemotherapy may also explain some of the improvement. The prolongation in FIGO-stages III-IV may be due to more radical surgery. Patient selection criteria, not only the treatment modalities, may be responsible for the superior results reported in clinical trials. Cancer registries are important for evaluating the quality of healthcare delivery.

Primary breast cancer therapy in six regions of Germany.

Studies from six regions of Germany (Aachen (W1), Dresden (E1), Jena (E2), Marburg (W2), Munich (W3), and Stuttgart (C1)) have been compared to verify and assess the quality of healthcare using breast cancer as an example. All of the data collection was carried out in comprehensive cancer centres and is population-based, with the exception of C1. Classic prognostic factors and the initial treatment of 8661 women with breast cancer, diagnosed between 1996 and 1998, were examined. Primary therapy, breast conserving therapy (BCT), and the use of subsequent local radiation and/or systemic therapy (chemotherapy or hormonal therapy) were analysed. BCT was performed on 39.3-57.7% of patients. By pT-category, the proportion of BCT in the six regions were as follows: for pTis between 37.8 and 64.3%, for pT1 between 51.7 and 71.5%, for pT2 between 25.9 and 51.1%, for pT3 between 0 and 13.1% and for pT4 between 0 and 15.2%. Multivariate analyses, adjusted for age and biological factors, showed a significant influence of the treating hospital on the mastectomy rate. The use of radiotherapy after BCT (80%) was quite homogeneous in the six regions. The application of radiotherapy after mastectomy, however, varied between 10.4 and 32.2%. In all regions, for premenopausal patients, the use of adjuvant systemic therapy almost reflected the St. Gallen-Consensus recommendations. In contrast, post-menopausal women with positive lymph nodes were not always treated according to these standards. In all regions, age had an influence on the administration of treatment: elderly breast cancer patients received less BCT, less radiotherapy and less adjuvant therapy than recommended in the St. Gallen-Consensus. Feedback of the results was made available to each hospital, providing a comparative summary of patient care that could be used by the participating hospitals for self-assessment and quality-control.

Determinants and prognoses of locoregional and distant progression in breast cancer.

PURPOSE: To describe locoregional and distant progression in a population-based breast cancer sample. METHODS AND MATERIALS: Between 1978 and 1998, the Munich Cancer Registry evaluated 14,429 patients. The mean follow-up of survivors was 8.3 years. Metastases (MET), local recurrence (LR), and lymph node recurrence (LNR) were considered as outcome measures. The prognostic factor for, and effects of, LR and MET were assessed multivariately by the Cox and dynamic Aalen models. RESULTS: The LR and MET rate increased with increasing tumor size, with the latter described by pT category. Distant MET occurred earlier than local progression. MET was recorded even earlier for MET alone. The mean time from diagnosis to MET for MET and LR was 54.9, 43.4, 29.4, and 24.7 months and for MET only was 36.5, 31.0, 22.6, and 12.9 months for pT1, pT2, pT3, and pT4, respectively. After MET, survival varied only slightly by pT stage; after LR, a more favorable prognosis, especially for pT1 and pT2, was evident. The prognosis after MET depended mainly on the MET location; 50% of patients with cerebral or nervous system MET survived <1 year and 50% of those with skeletal MET survived >2 years. In the Cox model, the relative risk of LR for MET was 3.0. In the Aalen model, after 30 months, when the hazard rates of MET began to decline, there was still an excess risk of MET after LR. CONCLUSION: This disease description highlights the importance of long-term observational studies. Empiric evidence that LR is both an indicator for, and in part a cause of, MET has been provided. In the future, the MET location should be reported. Variations in guidelines or health care systems that influence the time to MET and survival after MET through different diagnostic procedures should also be considered.

Cancer prevention and the contribution of cancer registries.

"Because they know what they do" should be the contribution of cancer registration to prevention. The public should be informed about the successes and failures of prevention. In addition, each doctor and each hospital should know the long-term results for its patients despite the complex interdisciplinary health care provision. At the same time, the regional results should be available and contrasted with clinical studies and international standards. An important criterion is also the quality of life of the patients, whose cooperation is more than overdue. According to the possibilities of prevention, six important levels can be differentiated. On each level the outcome should be evaluated on the basis of slightly differing criteria. Primary prevention has the largest incidence-, and thus, mortality-reduction potential, essentially caused by the giving up of smoking. However, the primary prevention of the tumor depends upon what kind of tumor it is, as primary prevention is possible to different extents. Often, secondary prevention is the earliest intervention possible, that is, the early detection of cancer in prognostically favorable stages. However, early detection only offers a chance of cure, albeit a smaller one, for some kinds of tumor. The third prevention level comprises primary care according to the state-of-the-art standards, followed by posttreatment care (disease-free phase) even when the disease is running a fatal course (palliative phase). In the terminal phase, adequate tumor-pain therapy and symptom-oriented measures stand for the sixth prevention level. Even the quality of life of the dying can be optimized. At the beginning of treatment at the latest, the quality of life should also be added to the outcome criteria. In this paper, the individual levels of prevention and possible evaluation criteria for successful prevention, which a modern cancer registration should submit, are discussed and examples given. If the evaluation of the quality of treatment and its significance for health care provision becomes the central tasks of the cancer registers, they will need to be transformed into service centers for hospitals, doctors, and patients.

Gastric carcinoma risk index in patients infected with Helicobacter pylori.

Epidemiological data show an association between Helicobacter pylori gastritis and gastric carcinoma. However, most people infected with H. pylori do not develop gastric cancer. We have therefore evaluated histological criteria indicating an increased risk for gastric cancer. H. pylori gastritis was investigated in 117 patients with small (O<2 cm) early gastric carcinomas and in 117 age-matched duodenal ulcer patients infected with H. pylori, who are known to have a low risk for developing gastric carcinoma. The results showed that infiltration with lymphocytes/plasma cells and infiltration with neutrophils predominating in the corpus, and intestinal metaplasia in antrum and corpus were associated with both types of gastric carcinoma (intestinal and diffuse, P<0.0001). If an index is computed by giving one point to each of these features, the predictive value for the presence of gastric carcinoma was 0.79 with a score of 2, and 0.94 with a score of 3 (all points on the index used). Application of this index might be a simple method of identifying patients infected with H. pylori and carrying a higher risk for gastric carcinoma.

Prognostic index for cutaneous melanoma: an analysis after follow-up of 2715 patients.

In 2715 of 4524 patients with cutaneous melanoma treated surgically between 1968 and 1992 prognostic parameters were analysed for their value in predicting the occurrence of first progression. All of the 2715 patients developed only one invasive cutaneous melanoma during the follow-up period. Data concerning tumour thickness and mitotic index (maximum number of mitoses per square millimetre) of the cutaneous melanomas were determined. Between the characteristics age, tumour thickness, mitotic index, prognostic index (PI), sex, site of tumour, melanoma subtype and Clark level, the value of the mitotic index, as a prognostic parameter independent of tumour thickness, and the combination of mitotic index and tumour thickness were evaluated. The development of the first metastases was documented during a mean follow-up of 7.5 years. The majority of first recurrences occurred at regional lymph nodes and attempts have been made to identify those patients at risk of developing metastatic disease. The most effective parameters proved to be tumour thickness and mitotic index. For both parameters an independent prognostic influence was shown. The prognostic index, defined as the product of tumour thickness and number of mitoses per square millimetre, was re-evaluated and confirmed. A new modified prognostic index, defined as the product of square tumour thickness and mitotic index, proved to be even more useful for defining a subgroup of patients who are at risk of developing metastases and, therefore, might benefit from adjuvant therapy.

Design elements for a computerized patient record.

Computerized medical record systems have to present user- and problem-oriented views of a patient record to health-care professionals. Presentation and manipulation of data must be easily adaptable to current and future demands of medical specialties and specific settings. During the definition, development and evaluation of a prototype of a computerized patient record system, design elements were elaborated to support physicians and other health-care professionals. Our approach shows a high degree of flexibility and adaptability to specific needs, problem orientation and connectivity to other systems, via a hospital information network. The explicit description of the contents of a patient record allows to augment the number of items that can be recorded without modifying the data structure. New views on patient data can be added to the system without interfering with the routine use of the system. Application in several medical specialties proved the feasibility of our prototype.

Prospective documentation and analysis of the pre- and early clinical management in severe head injury in southern Bavaria at a population based level.

Treatment of patients suffering from severe head injury is so far restricted to general procedures, whereas specific pharmacological agents of neuroprotection including hypothermia have not been found to improve the outcome in clinical trials. Albeit effective, symptomatic measures of the preclinical rescue of patients (i.e. stabilization or reestablishment of the circulatory and respiratory system) or of the early clinical care (e.g. prompt diagnosis and treatment of an intracranial space occupying mass, maintenance of a competent circulatory and respiratory system, and others) by and large constitute the current treatment based on considerable organizational and logistical efforts. These and other components of the head injury treatment are certainly worthwhile of a systematic analysis as to their efficacy or remaining deficiencies, respectively. Deficits could be associated with delays of providing preclinical rescue procedures (e.g. until intubation of the patient or administration of fluid). Delays could also be associated in the hospital with the diagnostic establishment of intracranial lesions requiring prompt neurosurgical intervention. By support of the Federal Ministry of Education and Research and under the auspices of the Forschungsverbund Neurotraumatology, University of Munich, a prospective system analysis was carried out on major aspects of the pre- and early clinical management at a population based level in patients with traumatic brain injury. Documentation of pertinent data was made from August 1998 to July 1999 covering a catchment area of Southern Bavaria (5.6 mio inhabitants). Altogether 528 cases identified to suffer from severe head injury (GCS < or = 8 or deteriorating to that level within 48 hrs) were enrolled following admission to the hospital and establishment of the diagnosis. Further, patients dying on the scene or during transport to the hospital were also documented, particularly as to the frequency of severe head injury as underlying cause of mortality. The analysis included also cases with additional peripheral trauma (polytrauma). The efficacy of the logistics and organization of the management was studied by documentation of prognosis-relevant time intervals, as for example until arrival of the rescue squad at the scene of an accident, until intubation and administration of fluid, or upon hospital admission until establishment of the CT-diagnosis and commencement of surgery or transfer to the intensive care unit, respectively. The severity of cases studied in the present analysis is evident from a mortality of far above 40% of cases admitted to the hospital, which was increased by about 20% when including prehospital mortality. The outcome data notwithstanding, the emerging results demonstrate a high efficacy of the pre- and early clinical management, as indicated by a prompt arrival of the rescue squad at the scene, a competent prehospital and early clinical management and care, indicative of a low rate of avoidable complications. It is tentatively concluded on the basis of these findings that the patient prognosis is increasingly determined by the manifestations of primary brain damage vs. the development of secondary complications.

Ten-year results following elective surgery for abdominal aortic aneurysm.

OBJECTIVE: 10-year results after elective operation for infrarenal aortic aneurysm considering the influence of risk factors. EXPERIMENTAL DESIGN: Retrospective study with 5-12 year postoperative follow-up. SETTING: University hospital (Klinikum Grosshadern, Munich). PATIENTS: The long-term follow-up was based on 521 (95.6%) out of 545 consecutive patients operated upon electively for abdominal aortic aneurysm between 1978 and 1987. INTERVENTIONS: The infrarenal aneurysms were excluded by aortic tube grafts (314 patients, 59%) or bifurcation grafts (231 patients, 41%). MEASURES: The birthday, operation day and eventually the day of death in the hospital were documented in the charts. The patient's state or cause of death were elicited on the phone 5 to 12 years after the operation. Kaplan-Meier survival curves were calculated based on these data and compared to age-matched normal male populations. RESULTS: Hospital mortality was 6.4%. The cumulative rate of survival following elective surgery was 65% at 5 years and 41% at 10 years, the mean survival time being 95.1 months. Age, coronary artery disease and hypertension had a significant influence on the cumulative survival. Patients with aorto-coronary bypass had a better long-term outcome than those without bypass surgery. CONCLUSIONS: The excellent long-term results within a high-risk population support elective surgery of infrarenal aortic aneurysms. Results of new interventional techniques will have to be compared with this "golden standard" follow-up.

A prototype of a computerized patient record.

Computerized medical record systems (CPRS) should present user and problem oriented views of the patient file. Problem lists, clinical course, medication profiles and results of examinations have to be recorded in a computerized patient record. Patient review screens should give a synopsis of the patient data to inform whenever the patient record is opened. Several different types of data have to be stored in a patient record. Qualitative and quantitative measurements, narratives and images are such examples. Therefore, a CPR must also be able to handle these different data types. New methods and concepts appear frequently in medicine. Thus a CPRS must be flexible enough to cope with coming demands. We developed a prototype of a computer based patient record with a graphical user interface on a SUN workstation. The basis of the system are a dynamic data dictionary, an interpreter language and a large set of basic functions. This approach gives optimal flexibility to the system. A lot of different data types are already supported. Extensions are easily possible. There is also almost no limit concerning the number of medical concepts that can be handled by our prototype. Several applications were built on this platform. Some of them are presented to exemplify the patient and problem oriented handling of the CPR.