Estudos de prognóstico clínico consistem na predição do curso de uma doença em pacientes e são utilizados por profissionais da saúde com o intuito de aumentar as chances ou a qualidade de sua recuperação. Sob a perspectiva computacional, a criação de um modelo prognóstico clínico é um problema de classificação, cujo objetivo é identificar a qual classe (dentro de um conjunto de classes predefinidas) uma nova amostra pertence. Este projeto visa a criar modelos prognósticos de lesões traumáticas do plexo braquial, um conjunto de nervos que inervam os membros superiores, utilizando dados de pacientes adultos com esse tipo de lesão. Os dados são provenientes do Instituto de Neurologia Deolindo Couto (INDC) da Universidade Federal do Rio de Janeiro (UFRJ) e contêm dezenas de atributos clínicos coletados por meio de questionários eletrônicos. Com esses modelos prognósticos, deseja-se identificar de maneira automática os possíveis preditores do curso desse tipo de lesão. Árvores de decisão são classificadores frequentemente utilizados para criação de modelos prognósticos, por se tratarem de um modelo transparente, cujo resultado pode ser examinado e interpretado clinicamente. As Florestas Aleatórias, uma técnica que utiliza um conjunto de árvores de decisão para determinar o resultado final da classificação, podem aumentar significativamente a acurácia e a generalização dos modelos gerados, entretanto ainda são pouco utilizadas na criação de modelos prognósticos. Neste projeto, exploramos a utilização de florestas aleatórias nesse contexto, bem como a aplicação de métodos de interpretação de seus modelos gerados, uma vez que a transparência do modelo é um aspecto particularmente importante em domínios clínicos. A estimativa de generalização dos modelos resultantes foi feita por meio de métodos que viabilizam sua utilização sobre um número reduzido de instâncias, uma vez que os dados relativos ao prognóstico são provenientes de 44 pacientes do INDC. Além disso, adaptamos a técnica de florestas aleatórias para incluir a possível existência de valores faltantes, que é uma característica presente nos dados utilizados neste projeto. Foram criados quatro modelos prognósticos - um para cada objetivo de recuperação, sendo eles a ausência de dor e forças satisfatórias avaliadas sobre abdução do ombro, flexão do cotovelo e rotação externa no ombro. As acurácias dos modelos foram estimadas entre 77% e 88%, utilizando o método de validação cruzada leave-one-out. Esses modelos evoluirão com a inclusão de novos dados, provenientes da contínua chegada de novos pacientes em tratamento no INDC, e serão utilizados como parte de um sistema de apoio à decisão clínica, de forma a possibilitar a predição de recuperação de um paciente considerando suas características clínicas. / Studies of prognosis refer to the prediction of the course of a disease in patients and are employed by health professionals in order to improve patients\' recovery chances and quality. Under a computational perspective, the creation of a prognostic model is a classification task that aims to identify to which class (within a predefined set of classes) a new sample belongs. The goal of this project is the creation of prognostic models for traumatic injuries of the brachial plexus, a network of nerves that innervates the upper limbs, using data from adult patients with this kind of injury. The data come from the Neurology Institute Deolindo Couto (INDC) of Rio de Janeiro Federal University (UFRJ) and they are characterized by dozens of clinical features that are collected by means of electronic questionnaires. With the use of these prognostic models we intended to automatically identify possible predictors of the course of brachial plexus injuries. Decision trees are classifiers that are frequently used for the creation of prognostic models since they are a transparent technique that produces results that can be clinically examined and interpreted. Random Forests are a technique that uses a set of decision trees to determine the final classification results and can significantly improve model\'s accuracy and generalization, yet they are still not commonly used for the creation of prognostic models. In this project we explored the use of random forests for that purpose, as well as the use of interpretation methods for the resulting models, since model transparency is an important aspect in clinical domains. Model assessment was achieved by means of methods whose application over a small set of samples is suitable, since the available prognostic data refer to only 44 patients from INDC. Additionally, we adapted the random forests technique to include missing data, that are frequent among the data used in this project. Four prognostic models were created - one for each recovery goal, those being absence of pain and satisfactory strength evaluated over shoulder abduction, elbow flexion and external shoulder rotation. The models\' accuracies were estimated between 77% and 88%, calculated through the leave-one-out cross validation method. These models will evolve with the inclusion of new data from new patients that will arrive at the INDC and they will be used as part of a clinical decision support system, with the purpose of prediction of a patient\'s recovery considering his or her clinical characteristics.

<p>In the demyelinating autoimmune disease multiple sclerosis (MS) there is a great need for validated prognostic biomarkers that can give information about both prognosis and disease course. So far only clinical parameters have been shown to predict future outcome. CXCL13 is a potent B cell chemoattractant that has been suggested to be a potential biomarker candidate. The aim of this study was to investigate the usefulness of CXCL13 as a prognostic biomarker for MS.</p><p>Clinical, paraclinical, laboratory and MRI data about a large group of MS patients and controls were collected. CXCL13 levels in cerebrospinal fluid (CSF) samples from these patients were determined by standard enzymelinked immunosorbent assay (ELISA).</p><p>In general CXCL13 were increased in CSF in MS, especially in relapsing-remitting MS during relapses, i.e. with ongoing inflammations in the central nervous system. CXCL13 is a good candidate prognostic marker for MS, since newly diagnosed MS with high CXCL13 levels showed worsened disease course within five years. Most importantly, MS conversion occurred in higher rate in possible MS patients with high concentrations of CXCL13 in CSF, and in a shorter time point. This observation may support an early treatment decision in these patients.</p><p>In conclusion, this study provides support for an association between CXCL13 levels in the CSF and later development of disease severity in MS.</p>

<p>Bronchial carcinois are subdivided into typical and atypical. Atypical carcinoids are more malignant, but typical carcinoids may also influence survival. In the present study immunohistochemistry was performed to identify prognostic markets in patients with typical bronchial carcinoids. The diagnostic efficacy of octreoscan was evaluated, in comparison with CT and bone scan, and finally our experience of treating patients with metastatic bronchial carcinoids is reported. In an unselected material of 43 patients with typical bronchial carcinoids, metastatic disease was found in 12 patients (28%). Five patients (12%) developed distant metastases and died from their disease. High Ki-67 index, as well as positive staining for bcl-2 or p53 was associated with de- creased survival time. Positive staining for CD44s, v7-8 and v9, as well as positive nuclear staining for nm23 correlated to decreased mortality. Staining for CD44 and Ki-67 should be performed routinely for prognostic evaluation in these patients. </p><p>Octreoscan positive tumors were found in altogether 20/28 patients (71%). The primary tumor was detectable in 81% and intrathoracic metastases in 78% of the patients on octreoscan; the corresponding figures for CT were 94% and 89% respectively. Liver metastases, as shown by CT, were demonstable by octreoscan in 64% of patients. Octreoscan showed 70% and bone scan 90% sensitivity for identification of bone metastases. </p><p>Plasma chromogranin A was elevated in 28/30 patients (94%) with metastatic bronchial carcinoids and was the most sensitive tumor marker. Increased urinary 5'HIAA was found in 68%. </p><p>Biotherapy with α-interferon and Octreotide relieved carcinoid syndrome in 7/16 patients. However, only 4/27 patients showed stable disease during median 15 months, while 23 patients progressed. Treatment with cisplatinum + etoposide resulted in an objective response or stable disease for 6-8 months in 3/8 patients with widespread tumors. Doxorubicin combined with streptozotocin or paclitaxel was associated with stable disease for 9 months in 2/2 patients each. All 7 patients treated with streptozotocin+5-FU progressed. </p><p>Among the 43 unselected typical bronchial carcinoid patients, 5-year and 10-year survival was 95% and 91%, respectively. The prognosis in patients with bronchial carcinoids showing distant metastases was poor: 5-year survival was 70% from diagnosis and 22% from treatment start. </p>

Aim: The overall aim of this thesis was to study prognostic factors for prolonged disability in back pain in the general population and physiotherapists’ screening for prolonged disability, applying a social cognitive learning perspective. Methods and results: Studies I and II were based on a survey in the general population in Sweden. Study I included 1024 individuals aged 20-50 years, reporting non-specific back pain. Four groups (n = 100, 215, 172 and 537) based on duration and recurrence of back pain were formed and compared. After controlling for pain intensity, catastrophising and expectations of future pain were positively correlated to pain duration. Perceived social support was negatively correlated to pain duration. Study II was longitudinal over 12 months and analysed one group reporting first-episode back pain (n = 77), and one group reporting long-term back pain (n = 302). Future pain intensity and disability were predicted by initial levels of pain and disability and pain-related cognitions in both groups. Study III examined the inter-rater reliability of a research protocol for assessment of physiotherapists’ telephone screening for prolonged disability. The results demonstrated sufficient inter-rater reliability. Study IV evaluated the effect of a tailored skills training intervention on physiotherapists’ screening for prolonged disability in back pain. Four physiotherapists in primary care participated in four quasi-experimental single-subject studies. Effects were seen in all participants, with increased screening of prognostic factors and less time spent on detailed discussions about back pain. Conclusions: The identification of mainly cognitive explanatory variables indicates the relevance of a social cognitive perspective of back pain-related disability (studies I and II). Physiotherapists’ telephone screening for prolonged disability in back can be reliably assessed (study III). It is suggested that interventions based on social cognitive theory are effective in producing change in specified clinical behaviours in physiotherapists (study IV).

Bronchial carcinois are subdivided into typical and atypical. Atypical carcinoids are more malignant, but typical carcinoids may also influence survival. In the present study immunohistochemistry was performed to identify prognostic markets in patients with typical bronchial carcinoids. The diagnostic efficacy of octreoscan was evaluated, in comparison with CT and bone scan, and finally our experience of treating patients with metastatic bronchial carcinoids is reported. In an unselected material of 43 patients with typical bronchial carcinoids, metastatic disease was found in 12 patients (28%). Five patients (12%) developed distant metastases and died from their disease. High Ki-67 index, as well as positive staining for bcl-2 or p53 was associated with de- creased survival time. Positive staining for CD44s, v7-8 and v9, as well as positive nuclear staining for nm23 correlated to decreased mortality. Staining for CD44 and Ki-67 should be performed routinely for prognostic evaluation in these patients. Octreoscan positive tumors were found in altogether 20/28 patients (71%). The primary tumor was detectable in 81% and intrathoracic metastases in 78% of the patients on octreoscan; the corresponding figures for CT were 94% and 89% respectively. Liver metastases, as shown by CT, were demonstable by octreoscan in 64% of patients. Octreoscan showed 70% and bone scan 90% sensitivity for identification of bone metastases. Plasma chromogranin A was elevated in 28/30 patients (94%) with metastatic bronchial carcinoids and was the most sensitive tumor marker. Increased urinary 5'HIAA was found in 68%. Biotherapy with α-interferon and Octreotide relieved carcinoid syndrome in 7/16 patients. However, only 4/27 patients showed stable disease during median 15 months, while 23 patients progressed. Treatment with cisplatinum + etoposide resulted in an objective response or stable disease for 6-8 months in 3/8 patients with widespread tumors. Doxorubicin combined with streptozotocin or paclitaxel was associated with stable disease for 9 months in 2/2 patients each. All 7 patients treated with streptozotocin+5-FU progressed. Among the 43 unselected typical bronchial carcinoid patients, 5-year and 10-year survival was 95% and 91%, respectively. The prognosis in patients with bronchial carcinoids showing distant metastases was poor: 5-year survival was 70% from diagnosis and 22% from treatment start.

Dupuytren's contracture (DC) is a fibromatous disease of the palmar fascia of unknown etiology. The present study was undertaken in order to assess pathophysiological mechanisms and consequences. In a cohort study of 2,375 patients operated for DC at the Department of Hand Surgery, Uppsala there was a male: female ratio of 5.9:1. Women had a higher mean age at first operation than men. One-third of the men and one-quarter of the women required repeated surgery. Early age at first operation was associated with recurrent disease. The risk of cancer was determined in 15,212 patients operated on for DC in Sweden. The overall relative risk was increased by 24%. There was a significantly increased risk for buccal, oesophageal, gastric, lung and pancreatic cancers, which indicates that smoking and alcohol abuse are probable risk factors for DC. Furthermore, there was an increased frequency of fibrosarcoma and malignant fibrous histiocytoma, the cause of which is unexplained The causes of death were evaluated in a national cohort of 16,517 patients operated for DC. There was an overall increased mortality (SMR=1.06), inversely related to age and significant for both sexes, in patients under 70 years. The risk estimate was highest for endocrine-, gastrointestinal-, and respiratory diseases, and accidents. There was also an increased SMR for cardiovascular diseases in younger patients more than 10 years after surgery. The most probable mechanism is related to smoking and other lifestyle factors. Outcome after surgery was not related to the immunohistochemical expression of connective tissue activation markers, such as collagen type IV, integrin α5, laminin, smooth muscle α-actin, procollagen type I, and desmin, in surgical specimens in a prospectively investigated group of patients. Furthermore, there were no associations between gender, age at onset of DC, number of operations, heredity, diabetes mellitus, or medication for cardiovascular disease, and the expression of the different markers. The individual characteristics that place a person at high risk are, thus, not obviously related to ongoing connective tissue production at time of surgery or to connective tissue activity in its conventionally used sense.

Chronic Myeloid Leukemia is a kind of blood cancer with around 1 incidence per 100 000 persons/year. After the development of an effective treatment, imatinib, in the late 1990:s, the survival percentage has increased drastically. The high survival has turned the attention to different kinds of treatment responses, which in turn are good prognostic factors to future health status. In this thesis, the focus is on whether or not the patient has achieved a so called major molecular response after 12 month, or not. More precisely, the aim is to find prognostic factors to the 12 month response. In order to find prognostic factors for this binary response variable, a multivariate logistic regression analysis is conducted, with the goal of finding a parsimonious logistic model that describes the data. The analysis is done from a merged dataset from three earlier studies. The prognostic factors in the final model are treatment, 3 month response, and enlarged spleen. However, the residual analysis indicates that the model is incomplete, implying that further research needs to be done.

Pseudomyxoma peritonei (PMP) is a rare disease characterized by mucinous peritoneal metastasis (PM). Different loco-regional treatment strategies, i.e. debulking surgery and cytoreductive surgery (CRS) in combination with hyperthermic intraperitoneal chemotherapy (HIPEC), have changed the prognosis for these patients. CRS is an aggressive surgical procedure with a long learning curve. PMP exists in different types; how many depends on which classification is used. The aims of this thesis were to investigate the time-frame of PMP development from an isolated appendiceal neoplasm, examine the learning process for CRS, evaluate the differences in treatment outcome between debulking surgery and CRS in combination with HIPEC, to evaluate a more detailed PMP classification and to investigate particularly interesting new cysteine-histidine (PINCH) protein as a prognostic factor for PMP. Retrospectively 26 PMP patients were identified as having had an appendectomy with a neoplasm in the appendix but with no evidence of PM at the appendectomy. They were treated for PMP within a median of 13.1 months (3.8-95.3) after the appendectomy. No difference was seen between the types of PMP regarding the time to a clinically significant development of PMP and how much tumour was found at treatment. CRS is a highly invasive treatment and stabilization in the learning curve was seen after 220±10 procedures. Patients treated with CRS+HIPEC had a better 5-year overall survival (OS) than patients treated with debulking surgery, 74% vs. 40%. CRS increased the rate of complete cytoreduction from 25% in patients treated with debulking surgery to 72%. The new four-grade PMP classification showed very good inter-rater agreement between two independent pathologists and a difference in survival rates was observed between the different grades. A positive PINCH staining was recorded in 83% of the tumours and that was associated with poorer survival.

Es wurden 177 Patientinnen mit zwischen 1999 und 2005 an der Universitätsfrauenklinik Leipzig operativ therapiertem Mammakarzinom und vorangegangener Stanzbiopsie, für die vollständige Befunde vorlagen und bei denen keine neoadjuvante Chemotherapie stattfand, hinsichtlich der prognostischen und prädiktiven Faktoren und deren Vergleich zwischen Stanzbiopsie und dem endgültigen Tumorexzidat untersucht. Unsere Daten zeigten, dass die Stanzbiopsie in der Einschätzung des Differenzierungsgrades mit einer Konkordanz von 62,9% und der Lymphgefäßinvasion mit einer Konkordanz in 69,8% keine hohe Genauigkeit besitzt. Bezüglich des histologischen Typs mit einer Übereinstimmung von 77%, der Östrogen- und Progesteronrezeptorbestimmung mit Konkordanzen von 87% und 83% sowie hinsichtlich des Her-2/neu-Status mit einer Konkordanz von 79% fand sich eine moderate bis gute Übereinstimmung mit dem Exzidat, wobei zu diskutieren ist, ob bei initial an der Stanzbiopsie negativem Östrogen- und/oder Progesteronrezeptorstatus oder auch bei positivem Progesteronrezeptor- und negativem Östrogenrezeptornachweis eine erneute immunhistochemische Hormonrezeptoruntersuchung am Exzidat erfolgen sollte sowie ob bei einer Konkordanzrate des Her-2/neu von weniger 95% immer eine zweite Bestimmung am Operationspräparat als Basis einer definitiven Therapieplanung durchgeführt werden muß. In 8,5% wurde an der Biopsie keine Malignität festgestellt. Der Vergleich des Malignitätsgrades mit der Tumorkategorie als auch mit dem Lymphknotenstatus zeigte keine signifikante Korrelation. Eine fortgeschrittene Tumorkategorie war mit dem Vorhandensein von Lymphknotenmetastasen korreliert.

Acute myeloid leukemia (AML) is a highly heterogeneous disorder that results from a block in the differentiation of hematopoietic progenitor cells along with uncontrolled proliferation. In approximately 60% of cases, specific recurrent chromosomal aberrations can be identified by modern cytogenetic techniques, and is an important indicator to classify patients into three prognostic categories: favorable, intermediate, and poor risk. Currently, favorable risk patients are usually treated with chemotherapy while poor risk patients receive allogeneic stem cell transplantation. However, the largest subgroup of AML patients (approximately 40%) has no identifiable cytogenetic abnormalities and is classified as intermediate risk. In this special subgroup of patients, a number of studies have demonstrated the relationship between different translocations involving the mixed lineage leukemia (MLL) gene and patient prognosis. The heterogeneity of MLL-rearranged AML is reflected by the identification of more than 70 different fusion partners of this gene and the panel is continuously increasing. The aim of this study is to develop a sensitive molecular profiling test for relevant risk stratification that can help in the decision of treatment and/or follow-up strategy.