Performance in population models for count data,part II: A new SAEM algorithm

Analysis of count data from clinical trials using mixed effect analysis has recently become widely used. However, algorithms available for the parameter estimation, including LAPLACE and Gaussian quadrature (GQ), are associated with certain limitations, including bias in parameter estimates and the long analysis runtime. The stochastic approximation expectation maximization (SAEM) algorithm has proven to be a very efficient and powerful tool in the analysis of continuous data. The aim of this study was to implement and investigate the performance of a new SAEM algorithm for application to count data. A new SAEM algorithm was implemented in MATLAB for estimation of both, parameters and the Fisher information matrix. Stochastic Monte Carlo simulations followed by re-estimation were performed according to scenarios used in previous studies (part I) to investigate properties of alternative algorithms (Plan et al., 2008, Abstr 1372 []). A single scenario was used to explore six probability distribution models. For parameter estimation, the relative bias was less than 0.92% and 4.13% for fixed and random effects, for all models studied including ones accounting for over- or under-dispersion. Empirical and estimated relative standard errors were similar, with distance between them being <1.7% for all explored scenarios. The longest CPU time was 95 s for parameter estimation and 56 s for SE estimation. The SAEM algorithm was extended for analysis of count data. It provides accurate estimates of both, parameters and standard errors. The estimation is significantly faster compared to LAPLACE and GQ. The algorithm is implemented in Monolix 3.1, (beta-version available in July 2009).     

Differentiation of resistance phenotypes among erythromycin-resistant streptococci

Although macrolide antibiotics have proved to be a valuable alternative to beta-lactam antibiotics in the treatment of respiratory tract infections, resistance to these agents is now becoming established in streptococci, especially among Streptococcus pneumoniae isolates. Of particular concern is the emergence of cross-resistance to 14-, 15- and 16-membered macrolides, licosamides and group B streptogramins (MLSb phenotype). MLS resistance can be expressed either constitutively (cMLS phenotype) or inducibly (iMLS phenotype). MLS resistance is mediated by two classes of methylase genes--the conventional erm(B) and recently described erm(A) determinants. A new macrolide efflux mechanism has been described for streptococci, in which it is associated with a new resistance pattern (M phenotype) characterized by resistance to 14- and 15-membered macrolides, and susceptibility to 16-membered macrolides, lincosamides and streptogramin B. The recognition of the prevalence of M phenotype in streptococci has implications for sensitivity testing and may have an impact on the choice of antibiotic therapy in clinical practice. While M resistance is similar in S. pyogenes and S. pneumoniae being mediated by mef(A) and mef(E), respectively, MLS resistance in both species appears to be genotypically and phenotypically more varied. Differentiation of M and MLS phenotypes of erythromycin-resistant strains can be performed using the erythromycin-clindamycin double-disc method (ECDD). Distinguishing not only M resistance but also constitutively or inducibly MLS phenotype by ECDD in S. pyogenes is easily and reliably achieved. Inducible MLS phenotype S. pyogenes strain is genotypically and phenotypically heterogeneous and is further subdivided into three recently described subtypes, iMLS-A, iMLS-B and iMLS-C, by a triple-disk test with erythromycin plus clindamycin and josamycin. While distinguishing M from MLS resistance in S. pneumoniae by ECDD test is easily and reliably achieved, the differentiation between constitutive and inducible MLS resistance is by far more uncertain. The meaning of inducible MLS resistance appears to be different in S. pneumoniae from that in S. pyogenes. In order to easily differentiate, within erythromycin-resistant pneumococci, not only the strains of the M phenotype from those with MLS resistance but also among the latter, cMLS from iMcLS strains, a triple-disk test has been set up by adding a rokitamycin disk to the conventional     

Recognizing, Scoring, and Predicting Blast Injuries RID="" ID="" This International Association for the Surgery of Trauma and Surgical Intensive Care (IATSIC) article was presented at the 37th World Congress of Surgery International Surgical Week (ISW97), Acapulco, Mexico, August 24–30, 1997.

n = 665, 51%) had an ISS ranging from 0 to 34 (mean 13) had wounds ranging from G1ST (soft tissue wounds caused by low energy transfer) to G3VF (massive wounds with fractures and injury of vital structures) according to the RCWC, with PSS/IS scores from 2 to 105 (mean 60). Statistically significant correlation was found between ISS and PSS/IS as well as RCWC and PSS/IS. Cytokines (IL-1, TNF _alpha ) and amino acids responded to a blast injury in similar manner as to gunshot wounds with a greater ISS or more severe RCWC injury type. The subjective sensations in blasted patients (deafness, thoracic pain, vertigo) and mediators, confirmed in previous experimental investigations as important factors in the pathogenesis of blast injuries (TxA ₂ , sulfidopeptide leukotrienes) were relationed only to the PSS/IS.RID=" ID=" <E5>Correspondence to:</E5> I. Cernak, M.D., Ph.D.     

2. Survival Impact of HER-2/Neu, Cox-2, Urokinase Plasminogen Activator (upa), Cytokeratin 17/5,6 and other Markers with Long-Term Outcome of Early Breast Cancer. Report from the British Columbia Tissue Micro-Array Project (BCTMAP)

Tumor samples are available from over 19,600 Stage I-III breast cancer patients treated according to evolving British Columbia guidelines from 1978 to 1990. A tissue mico-array (TMA) was constructed from 930 of these patients, all of whom participated in randomized or phase II studies. Outcome was defined as 20-year Breast Cancer specific Survival (BrCaSS), with events defined as Breast Ca death. Follow up was median 17.8 years (ranges 11–28). Multiple tumor markers were tested, and results correlated with 20-year BrCaSS for markers expressed versus non-expressed. No difference in BrCaSS was found for aromatase, integrin-linked kinase (ILK), IGF-1 and Topo-isomerase-2. The negative predictive value of IHC versus FISH and ACIS-IHC versus FISH was 96 and 97%, respectively. The positive predictive value of IHC versus FISH and ACIS-IHC versus FISH was 84 and 84%, respectively. All tests, with the exception of HER-2 FISH were done by IHC. Results of other markers (VEGF, ER/PgR, hypoxia markers, etc.), and an interactive multivariate analysis adjusting for conventional prognostic factors and for all above markers, are in progress. Conclusions 1. The TMA is a technique which provides opportunity for rapid screening of multiple genetic markers.2. Expression of Her-2/Neu, uPA, Cox-2 and Cytokeratin 17/5,6 (but not of Aromatase, ILK, TOPO-II and IGF-1) is associated with inferior BrCaSS.3. HER-2 determination by ACIS-IHC provides comparable results to IHC done manually (with a potential for more uniform reporting), and both provide comparable results to Her-2 assessment by FISH. ^** ACIS-IHC:IHC red by Automated Cell Image System (M.L.)     

Hemangiomas of the external auditory canal:a literature review and two new case reports

Aim  

The aim of this paper is to present two case reports of patients with hemangiomas of the external auditory canal, and to overview all cases published in English language literature so far.     

Diagnostic uncertainty and recall bias in chronic low back pain

Patients’ beliefs about the origin of their pain and their cognitive processing of pain-related information have both been shown to be associated with poorer prognosis in low back pain (LBP), but the relationship between specific beliefs and specific cognitive processes is not known. The aim of this study was to examine the relationship between diagnostic uncertainty and recall bias in 2 groups of chronic LBP patients, those who were certain about their diagnosis and those who believed that their pain was due to an undiagnosed problem. Patients (N = 68) endorsed and subsequently recalled pain, illness, depression, and neutral stimuli. They also provided measures of pain, diagnostic status, mood, and disability. Both groups exhibited a recall bias for pain stimuli, but only the group with diagnostic uncertainty also displayed a recall bias for illness-related stimuli. This bias remained after controlling for depression and disability. Sensitivity analyses using grouping by diagnosis/explanation received supported these findings. Higher levels of depression and disability were found in the group with diagnostic uncertainty, but levels of pain intensity did not differ between the groups. Although the methodology does not provide information on causality, the results provide evidence for a relationship between diagnostic uncertainty and recall bias for negative health-related stimuli in chronic LBP patients.     

Intraperitoneal Sodium Thiosulfate for the Treatment of Calciphylaxis

Calcific uremic arteriolopathy (or calciphylaxis) is a severe complication of renal failure characterized by subcutaneous calcification of the small arteries and tissue necrosis. We describe the case of a woman receiving continuous cycling peritoneal dialysis with calciphylaxis involving upper and lower extremities. After intolerance of intravenous sodium thiosulfate and limited intravenous access options, we administered sodium thiosulfate intraperitoneally and quantitated the amount of extra calcium removed. Intraperitoneal administration of sodium thiosulfate was well tolerated and led to removal of extra calcium with peritoneal dialysis.