Biomedical subjects

W D Shannon

Publications and source records attributed to W D Shannon.

6 recordsLinked to original sources

PubMed1999

The impact of ganciclovir-resistant cytomegalovirus infection after lung transplantation.

BACKGROUND: Cytomegalovirus (CMV) resistance to ganciclovir has become increasingly common in acquired immunodeficiency syndrome patients but has only rarely been reported in recipients of solid organ transplants. METHODS: A retrospective study of ganciclovir susceptibility testing of CMV isolates recovered from lung transplant recipients was performed. Patients with CMV isolates having partial (1 or =3 microg/ml) to ganciclovir determined by plaque reduction assay were included in a case-control study to identify risk factors for ganciclovir resistance. RESULTS: Between 2/91 and 5/98, 18 patients (5.2% of patients transplanted) were found to have CMV infections with some degree of ganciclovir resistance (4 partially, 14 fully resistant). More positive viral blood cultures (3.2+/-2.5 vs. 1.6+/-1.4 CMV positive cultures, P=0.02) and more episodes of CMV pneumonitis (0.24+/-0.23 vs. 0.10+/-0.17 episodes/bronchoscopy, P=0.02) occurring before the detection of resistance were seen among resistant patients than controls. Ganciclovir-resistant patients received more antithymocyte globulin during induction (70+/-44 vs. 45+/-39 mg/kg, P=0.03) and received ganciclovir for a greater number of days (79+/-52 vs. 64+/-53 days, P=0.005) before the detection of resistance than controls. Ganciclovir-resistant patients had a shorter survival and an earlier onset of bronchiolitis obliterans syndrome compared with patients in the transplant database at Washington University. CONCLUSIONS: Ganciclovir-resistant CMV infection is a serious complication of solid organ transplantation associated with more episodes of viremia, more frequent disease, earlier onset of bronchiolitis obliterans and shorter survival. The use of antithymocyte globulin and prolonged exposure to ganciclovir are risk factors for the development of ganciclovir resistance.

Adult↗

PubMed1999

Combining classification trees using MLE.

We propose a probability distribution for an equivalence class of classification trees (that is, those that ignore the value of the cutpoints but retain tree structure). This distribution is parameterized by a central tree structure representing the true model, and a precision or concentration coefficient representing the variability around the central tree. We use this distribution to model an observed set of classification trees exhibiting variability in tree structure. We propose the maximum likelihood estimate of the central tree as the best tree to represent the set. This MLE retains the interpretability of a single tree model and has excellent generalizability. We implement an ascent search for the MLE tree structure using a data set of 13 classification trees that predict the presence or absence of cancer based on immune system parameters.

Classification↗

PubMed1999

Meta-analysis of randomized controlled trials comparing laparoscopic and open appendectomy.

We performed a meta-analysis to determine whether laparoscopic or open appendectomy gives better outcomes for patients with suspected acute appendicitis. Studies were selected from the MEDLINE database, personal files, and meeting abstracts. Eleven of 21 randomized controlled trials were included in the meta-analysis. Pooled effect size estimates were calculated using a random effects model. Laparoscopic appendectomy reduced time to full functioning by 5.48 days (95% confidence interval [CI] 3.70 to 7.26; p < 0.001), improved postoperative pain at 24 hours measured by a visual analog scale from 0 to 10 by 1.19 points (95% CI -2.14 to -0.24 points; p=0.014), and decreased the absolute risk for wound infection by 3.2% (95% CI -5.6% to -0. 8%; p=0.009). Operating time was increased by 17.12 min (95% CI 14.19 to 20.03; p < 0.0001). There was no difference between the two surgeries for length of hospital stay, readmission rate, and intra-abdominal abscess formation. Laparoscopic appendectomy improves patient outcomes.

Appendectomy↗

PubMed1997

True and false positive peaks in genomewide scans: applications of length-biased sampling to linkage mapping.

Disease-susceptibility loci are now being mapped via genomewide scans in which a linkage statistic is computed at each of a large number of markers. Such disease-susceptibility loci may be identified via a peak in the test statistic when the latter is plotted against the genetic map. In this paper we establish, by appealing to renewal theory, that true positive peaks are expected to be longer than false positive peaks. These results are verified by a realistic simulation of a genomewide linkage study based on the affected-sib-pair design. Since longer peaks are more likely to contain a gene of interest than are shorter peaks, these differences may aid in linkage mapping, justifying assignment of lower priority to shorter peaks. However, since these differences are generally small, statistics based on both peak length and height may not be much more powerful than those based on height alone. The results presented here also provide a theoretical framework for methods that use the length of shared haplotypes in populations to map disease genes.

Chromosome Mapping↗

PubMed1995

An application of decision theory to patient screening for an autologous tumour vaccine trial.

Patients are eligible for accrual onto a phase I autologous tumour vaccine clinical trial if their resected and dissociated tumour achieves a minimum viable cell count. Because tumour pre-processing and cell count determination are expensive, there has been developed a screening procedure based on tumour mass to screen out those tumours unlikely to yield sufficient viable cells. If theta is the ratio of the expected benefit of an accrual onto the study to the cost of tumour pre-processing and cell counting, then we maximize long-run benefit by pre-processing and counting only those tumours whose masses exceed a cutoff mc, such that Pr(sufficient tumour cells masses = mc) = 1/theta. We derive algorithms for estimating mc and evaluate them under a variety of assumptions concerning the cell count/mass relationship. These include explicit equations for mc under parametric assumptions as well as more general algorithms based on non-parametric smoothing techniques. We show that when theta deviates substantially from 2, these methods outperform simple inverse interpolation.

Adenocarcinoma↗

PubMed1991

The effect of graft function on FK506 plasma levels, dosages, and renal function, with particular reference to the liver.

Plasma FK506 was studied in 49 liver, 13 heart, 3 double-lung or heart-lung, and 21 kidney recipients. The levels were correlated with the drug doses used, kidney function, and liver function. In all varieties of recipients, there was an early rise in the FK506 plasma levels that occurred at the time of intravenous administration of the drug. At the same time or shortly after, there were increases in serum creatinine that were transitory except in liver recipients with continuing suboptimal graft function. The quality of hepatic function dominated all aspects of FK506 management in the liver recipients. Those who received well-functioning grafts could be given about the same drug doses as recipients of kidneys and the thoracic organs. Liver recipients with defective grafts had astronomical rises in plasma FK506, a high incidence of renal failure, and probably increased neurotoxicity. In kidney transplant recipients, the FK506 plasma levels and doses were essentially the same in patients with prompt versus delayed renal function. These studies have highlighted the necessity, first, of close pharmacologic monitoring of patients who are given FK506 in the presence of abnormal liver function, and second, of using smaller intravenous induction doses than in past practice.

Adolescent↗