Selection Operator Approach Research Articles

Model Selection of the Effect of Binary Exposures over the Life Course.

Epidemiologists are often interested in examining the effect on a later-life outcome of an exposure measured repeatedly over the life course. When different hypotheses for this effect are proposed by competing theories, it is important to identify those most supported by observed data as a first step toward estimating causal associations. One method is to compare goodness-of-fit of hypothesized models with a saturated model, but it is unclear how to judge the "best" out of two hypothesized models that both pass criteria for a good fit. We developed a new method using the least absolute shrinkage and selection operator to identify which of a small set of hypothesized models explains most of the observed outcome variation. We analyzed a cohort study with repeated measures of socioeconomic position (exposure) through childhood, early- and mid-adulthood, and body mass index (outcome) measured in mid-adulthood. We confirmed previous findings regarding support or lack of support for the following hypotheses: accumulation (number of times exposed), three critical periods (only exposure in childhood, early- or mid-adulthood), and social mobility (transition from low to high socioeconomic position). Simulations showed that our least absolute shrinkage and selection operator approach identified the most suitable hypothesized model with high probability in moderately sized samples, but with lower probability for hypotheses involving change in exposure or highly correlated exposures. Identifying a single, simple hypothesis that represents the specified knowledge of the life course association allows more precise definition of the causal effect of interest.

The changing spectrum of treatment for variceal bleeding.

The objective of this study was to assess the impact of endoscopic therapy, liver transplantation, and transjugular intrahepatic portosystemic shunt (TIPS) on patient selection and outcome of surgical treatment for this complication of portal hypertension, as reflected in a single surgeon's 18-year experience with operations for variceal hemorrhage. Definitive treatment of patients who bleed from portal hypertension has been progressively altered during the past 2 decades during which endoscopic therapy, liver transplantation, and TIPS have successively become available as alternative treatment options to operative portosystemic shunts and devascularization procedures. Two hundred sixty-three consecutive patients who were surgically treated for portal hypertensive bleeding between 1978 and 1996 were reviewed retrospectively. Four Eras separated by the dates when endoscopic therapy (January 1981), liver transplantation (July 1985), and TIPS (January 1993) became available in our institution were analyzed. Throughout all four Eras, a selective operative approach, using the distal splenorenal shunt (DSRS), nonselective shunts, and esophagogastric devascularization, was taken. The most common indications for nonselective shunts and esophagogastric devascularization were medically intractable ascites and splanchnic venous thrombosis, respectively. Most other patients received a DSRS. The risk status (Child's class) of patients undergoing surgery progressively improved (p = 0.001) throughout the 4 Eras, whereas the need for emergency surgery declined (p = 0.002). The percentage of nonselective shunts performed decreased because better options to manage acute bleeding episodes (sclerotherapy, TIPS) and advanced liver disease complicated by ascites (liver transplantation, TIPS) became available (p = 0.009). In all Eras, the operative mortality rate was directly related to Child's class (A, 2.7%; B, 7.5%; and C, 26.1 %) (p = 0.001). As more good-risk patients underwent operations for variceal bleeding, the incidence of postoperative encephalopathy decreased (p = 0.015), and long-term survival improved (p = 0.012), especially since liver transplantation became available to salvage patients who developed hepatic failure after a prior surgical procedure. There were no differences between Eras with respect to rebleeding or shunt occlusion. Distal splenorenal shunts (p = 0.004) and nonselective shunts (p = 0.001) were more protective against rebleeding than was esophagogastric devascularization. The sequential introduction of endoscopic therapy, liver transplantation, and TIPS has resulted in better selection and improved results with respect to quality and length of survival for patients treated surgically for variceal bleeding. Despite these innovations, portosystemic shunts and esophagogastric devascularization remain important and effective options for selected patients with bleeding secondary to portal hypertension.

Transabdominal esophagogastric devascularization as treatment for variceal hemorrhage

During the past 18 years we have used a selective operative approach for variceal bleeders in whom endoscopic sclerotherapy failed or sclerotherapy was not indicated. Esophagogastric devascularization with splenectomy has been reserved for unshuntable patients and for those in whom a shunt was deemed inadvisable. The purposes of this study are to describe the surgical procedure technique and indications for esophagogastric devascularization and to report its long-term results. Thirty-two patients who underwent either a limited (n = 9) or extensive (n = 23) esophagogastric devascularization procedure without esophageal transection for variceal bleeding were retrospectively reviewed. Common indications were thrombosis of all splanchnic veins (n = 12), distal splenorenal shunt thrombosis (n = 7), generalized portal hypertension with isolated splenic vein thrombosis (n = 5), and symptomatic splenomegaly or severe hypersplenism (n = 6). Eighteen patients (56%) had cirrhosis, eleven (34%) received an emergency operation, and eighteen (56%) bled from gastric varices. Three patients with Child's class C disease undergoing emergency surgery died during the early postoperative interval. Rebleeding occurred in nine surviving patients (31%) and was the cause of death in three. Rebleeding rates for the limited and extensive devascularization procedures were 50% and 24%, respectively. Only one of 11 patients with diffuse splanchnic venous thrombosis without liver disease has died. The 5-year survival rate of patients with liver disease was 51%. Only two patients experienced postoperative encephalopathy. When used in selected patients, esophagogastric devascularization without esophageal transection is a reasonably effective alternative to shunt surgery.

Selective Operative Approach for Diagnosis and Treatment of Anterior Mediastinal Masses

We retrospectively reviewed the records of 41 patients with an isolated anterior mediastinal mass to identify whether the traditional surgical approach of resection for both diagnosis and therapy is the optimal approach to these tumors. In addition to plain chest radiographs, clinical evaluation included computed tomography (CT) in 23 and gallium scintigraphy in 23. The initial intervention was attempted excisional therapy in 16 and mediastinoscopy for biopsy in 25. Pathological diagnoses included lymphoma (13), thymoma (11), germ cell tumor (6), carcinoid (2), bronchogenic carcinoma (2), and other benign processes (7). Based on histological findings, nonoperative therapy was the most appropriate treatment in 61%. The clinical selection of the surgical approach was not aided by patient age, symptoms, or findings on CT. The gallium scan was positive in 13 of 15 patients for whom nonoperative therapy was appropriate, and was normal in 7 of 8 for whom excision was indicated. Mediastinoscopy is useful in the diagnosis of anterior mediastinal masses, and may eliminate unnecessary thoracotomy, especially when a gallium scan is positive.

Selective operative approach for variceal hemorrhage

Since 1978, the operation chosen for patients with variceal hemorrhage has been based on preoperative hemodynamic and clinical factors. One hundred sixteen consecutive patients were managed with the following operations: distal splenorenal shunt (75 patients), nonselective shunts (33 patients), and nonshunting operation (8 patients). Emergency surgery was required in 19 percent of patients. The selection criteria used resulted in the majority of high risk patients receiving nonselective shunts. This selective operative approach resulted in an overall operative mortality of 12 percent, a median survival of 3 years, and postoperative encephalopathy, ascites, and recurrent variceal hemorrhage in 20, 23, and 11 percent of patients, respectively. Operative mortality for the total group was closely related to Child's class. Whereas encephalopathy was most frequent after nonselective shunts, ascites was more common after the distal splenorenal shunt. Recurrent hemorrhage rarely occurred after a shunting procedure, but was a frequent complication of nonshunting operations. Neither the type of procedure selected nor the cause of liver disease influenced long-term survival.

Thyroid carcinoma in children: a selective operative approach.

Thyroid carcinoma in children: a selective operative approach.