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).     

Benefits of arthroscopic tuberculoplasty in posterior ankle impingement syndrome

The purpose of this work was to describe the posterior ankle impingement syndrome related to the posterolateral tubercle of the talus bone and to present a retrospective analysis of our results after arthroscopic plasty of the tubercle in 15 ankles with a mean 3-year follow-up. Fifteen cases of posterior ankle impingement (PAI) underwent arthroscopic excision of an impinging bone spur. All the patients (13) were retrospectively evaluated at an average of 36 months after index surgery. There were seven women (bilateral in two of them) and six men. Ten were involved in different kinds of sport and three were professional ballet dancers. Preoperative symptoms included pain localized in the posterior ankle, limitation of motion, weakness and swelling. All patients had failed a course of conservative therapies. Surgery was performed through posterolateral and posteromedial portals as described by van Dijk. After soft tissue debridement, partial resection of the posterolateral process was performed until there was complete plantar flexion without bone impingement. Postoperatively, all patients followed the same rehabilitation protocol. Improvement in their impingement symptoms was recorded in all of them according to AOFAS score. One of them (7%) still had occasional discomfort. The results suggest that arthroscopic bone decompression of the posterolateral tubercle in cases of PAI resistant to non-surgical therapies is an effective treatment.     

Microscopic haematuria as a relative contraindication for tranexamic acid

Tranexamic acid has been advocated for patients with severe bleeding tendency due to thrombocytopenia not responding to platelet transfusions. Macroscopic haematuria is a well-known contraindication for its use in such patients. We present three clinical cases with microscopic haematuria, in whom tranexamic acid caused problems of clot formation in the urinary tract, indicating that microscopic haematuria should also be considered as a contraindication for tranexamic acid.     

Practices of reading and writing: the constitutive role of the patient record in medical work

This paper contributes to a reappraisal of the sociological significance of the medical record. Building upon re-evaluations of the notion of ‘representation’ in ethnomethodological and science and technology studies, it attempts to portray the medical record as an artefact which mediates the social relations that act and work through it. Through practices of reading and writing, it is argued, the medical record figures as a fundamental, constitutive element of medical practice. To address some of the dimensions that are involved, three case fragments are discussed. The way the medical record enters into the process of ‘medical decision making’ in the doctor-patient interaction, and into that interaction itself, is discussed first. Subsequently, some aspects of how the record mediates medical work as it is performed in the wards of a hospital are outlined. A third instance of its prominent role demonstrates how practices of reading and writing tend to produce particular renderings of patients’ histories - including notions of how medical work is structured.     

CAG repeat polymorphism within the KCNN3 gene is a significant contributor to susceptibility to anorexia nervosa: a case-control study of female patients and several ethnic groups in the Israeli Jewish population.

The human small-conductance Ca(2+)-activated potassium channel gene KCNN3 has been involved in mechanisms underlying neuronal function and plasticity. A multiallelic CAG repeat polymorphism within the KCNN3 has been associated with schizophrenia and bipolar disorder. We have previously reported in a family-based study that longer CAG repeats are preferentially transmitted to patients with anorexia nervosa (AN). The present study extends the analysis of KCNN3 allele distribution to a larger series of AN female patients and control groups, incorporating information on ethnicity and co-morbidities associated with AN. The data analysis is presented while considering separately the two alleles of each individual, namely a minor (shorter) and a major (longer) allele. This study has found that the KCNN3 allele distribution in the general Israeli population does not differ significantly in at least four Jewish ethnic groups of Ashkenazi, North African, Iraqi, and Yemenite origin. These have been used as control groups in a matched case-control analysis that has demonstrated a significant over-representation of KCNN3 alleles with longer CAG repeats among AN patients (P < 0.001 for the major allele and P = 0.035 for allele sum). Under dichotomization, a significantly higher prevalence of the L allele (>19 repeats) has been observed among AN patients (P < 0.001). While considering AN and co-morbid phenotypes, a tendency towards longer (L) alleles has been observed in the subset of patients with obsessive-compulsive disorder (OCD) co-morbidity. These findings further implicate KCNN3 as a significant contributor to predisposition to AN.     

Visual system of a naturally microphthalmic mammal: The blind mole rat,Spalax ehrenbergi

Retinal projections and visual thalamo-cortical connections were studied in the subterranean mole rat, belonging to the superspecies Spalax ehrenbergi, by anterograde and retrograde tracing techniques. Quantitative image analysis was used to estimate the relative density and distribution of retinal input to different primary visual nuclei. The visual system of Spalax presents a mosaic of both regressive and progressive morphological features. Following intraocular injections of horseradish peroxidase conjugates, the retina was found to project bilaterally to all visual structures described as receiving retinal afferents in non-fossorial rodents. Structures involved in form analysis and visually guided behaviors are reduced in size by more than 90%, receive a sparse retinal innervation, and are cytoarchitecturally poorly differentiated. The dorsal lateral geniculate nucleus, as defined by cyto- and myelo-architecture, cytochrome oxidase, and acetylcholinesterase distribution as well as by afferent and efferent connections, consists of a narrow sheet 3–5 neurons thick, in the dorsal thalamus. Connections with visual cortex are topographically organized but multiple cortical injections result in widespread and overlapping distributions of geniculate neurons, thus indicating that the cortical map of visual space is imprecise. The superficial layers of the superior colliculus are collapsed to a single layer, and the diffuse ipsilateral distribution of retinal afferents also suggests a lack of precise retinotopic relations. In the pretectum, both the olivary pretectal nucleus and the nucleus of the optic tract could be identified as receiving ipsilateral and contralateral retinal projections. The ventral lateral geniculate nucleus is also bilaterally innervated, but distinct subdivisions of this nucleus or the intergeniculate leaflet could not be distinguished. The retina sends a sparse projection to the dorsal and lateral terminal nuclei of the accessory optic system. The medial terminal nucleus is not present. In contrast to the above, structures of the “non-image forming” visual pathway involved in photoperiodic perception are well developed in Spalax. The suprachiasmatic nucleus receives a bilateral projection from the retina and the absolute size, cytoarchitecture, density, and distribution of retinal afferents in Spalax are comparable with those of other rodents. A relatively hypertrophied retinal projection is observed in the bed nucleus of the stria terminalis. Other regions which receive sparse visual input include the lateral and anterior hypothalamic areas, the retrochiasmatic region, the sub-paraventricular zone, the paraventricular hypothalamic nucleus, the anteroventral and anterodorsal nuclei, the lateral habenula, the mediodorsal nucleus, and the basal telencephalon. These results indicate that the apparently global morphological regression of the visual system conceals a selective expansion of structures related to functions of photoperiodic perception and photo-neuroendocrine regulation. We suggest that the evolution of an atrophied eye and reduced visual system is an adaptively advantageous response to the unique subterranean environment. Factors favoring regression include mechanical aspects, metabolic constraints, and competition between sensory systems. The primary advantage of sensory atrophy is the metabolic economy gained by the reduction of visual structures which do not contribute significantly to the animal's fitness. © 1993 Wiley-Liss, Inc.     

Cocaine tolerance: Interactions among random-ratio and random-interval reinforcement-schedule parameters and repeated exposure to cocaine

Food-deprived pigeons were trained to peck a key under either a three-component multiple random-ratio 5, random-ratio 25, random-ratio 125 schedule or a three-component multiple random-interval 10-sec, random-interval 30-sec, random-interval 125-sec schedule of food presentation. Following determination of acute effects of cocaine (1.0-13.0 mg/kg), are-sponse-rate-reducing dose was given before each daily session. Once performance under conditions of daily administration had become stable, other doses occasionally were substituted for the usual daily dose so that dose effects could be assessed. Tolerance, i.e., a rightward shift in the dose-effect curve was observed in all subjects. For subjects studied under the random-ratio schedules, however, the robustness of the tolerance usually was related to the schedule-parameter value; tolerance was great at lower random values. By contrast, subjects whose responding was maintained by random-interval schedules were less likely to show tolerance that was schedule-parameter dependent. The results also provide suggestive evidence that dose may be an important factor in determining effects of repeated cocaine exposure; repeated administration of larger doses may be less likely to result in tolerance than similar exposure to smaller doses.