A 10-year review of 101 patients sustaining an early postoperative small bowel obstruction within 30 days of celiotomy was carried out. Signs, symptoms, lab tests, and x-rays did not indicate which patients required operation. Twenty-three patients were operated on for either failure to resolve their obstruction or because it was feared that ischemic bowel was present. In none of these patients, nor the 78 patients who resolved without reoperation, did dead bowel occur. Early postoperative small bowel obstruction was most often due to adhesions and inflammatory processes. Seven patients died (6.9%), three in the operated and four in the nonoperated group. Because ischemic bowel is very unlikely in patients with early postoperative small bowel obstruction, we advise 10 to 14 days of nasogastric suction initially; after this, improvement is unlikely without reoperation.

The aim of these studies was to characterize the synthesis and secretion of lipoproteins and apolipoprotein B (apo B) and apo A-I by a newborn swine intestinal epithelial cell line (IPEC-1). Differentiated cells exhibited enterocytic features, including microvilli. [3H] oleic acid was taken up and incorporated into cellular lipids and secreted into the basolateral medium in lipoproteins. Total apo B and apo A-I secreted increased with oleic acid incubation. However, cellular apo B and apo A-I content did not change. Whereas undifferentiated cells synthesized and secreted only apo B-100, both apo B-100 and apo B-48 were produced by differentiated cells. The ratio of radiolabeled apo B-48 to apo B-100 in both basolateral medium and cell homogenate increased with oleic acid treatment after 24-h steady-state labeling. However, apo B mRNA editing was unchanged, indicating posttranslational regulation of this ratio. Pulse-chase radiolabeling demonstrated no major changes in cellular or basolateral medium apolipoprotein labeling kinetics with oleic acid or dexamethasone incubation. The dissociation of apo B and apo A-I mass secretion from the secretion of radiolabeled apo B and apo A-I in response to oleic acid absorption suggests the presence of an intracellular pool of apolipoprotein with a slow turnover that is mobilized for secretion in response to fatty acid uptake.

C-->U RNA editing of neurofibromatosis 1 (NF1) mRNA changes an arginine (CGA) to a UGA translational stop codon, predicted to result in translational termination of the edited mRNA. Previous studies demonstrated varying degrees of C-->U RNA editing in peripheral nerve-sheath tumor samples (PNSTs) from patients with NF1, but the basis for this heterogeneity was unexplained. In addition, the role, if any, of apobec-1, the catalytic deaminase that mediates C-->U editing of mammalian apolipoprotein B (apoB) RNA, was unresolved. We have examined these questions in PNSTs from patients with NF1 and demonstrate that a subset (8/34) manifest C-->U editing of RNA. Two distinguishing characteristics were found in the PNSTs that demonstrated editing of NF1 RNA. First, these tumors express apobec-1 mRNA, the first demonstration, in humans, of its expression beyond the luminal gastrointestinal tract. Second, PNSTs with C-->U editing of RNA manifest increased proportions of an alternatively spliced exon, 23A, downstream of the edited base. C-->U editing of RNA in these PNSTs was observed preferentially in transcripts containing exon 23A. These findings were complemented by in vitro studies using synthetic RNA templates incubated in the presence of recombinant apobec-1, which again confirmed preferential editing of transcripts containing exon 23A. Finally, adenovirus-mediated transfection of HepG2 cells revealed induction of editing of apoB RNA, along with preferential editing of NF1 transcripts containing exon 23A. Taken together, the data support the hypothesis that C-->U RNA editing of the NF1 transcript occurs both in a subset of PNSTs and in an alternatively spliced form containing a downstream exon, presumably an optimal configuration for enzymatic deamination by apobec-1.

Decision trees usefully represent sparse, high-dimensional, and noisy data. Having learned a function from these data, we may want to thereafter integrate the function into a larger decision-making problem, for example, for picking the best chemical process catalyst. We study a large-scale, industrially relevant mixed-integer nonlinear nonconvex optimization problem involving both gradient-boosted trees and penalty functions mitigating risk. This mixed-integer optimization problem with convex penalty terms broadly applies to optimizing pretrained regression tree models. Decision makers may wish to optimize discrete models to repurpose legacy predictive models or they may wish to optimize a discrete model that accurately represents a data set. We develop several heuristic methods to find feasible solutions and an exact branch-and-bound algorithm leveraging structural properties of the gradient-boosted trees and penalty functions. We computationally test our methods on a concrete mixture design instance and a chemical catalysis industrial instance.