Ethical dilemmas in a randomized trial of asthma treatment: can Bayesian statistical analysis explain the results?

OBJECTIVES: The original objective was to determine whether the use of bilevel positive airway pressure (BiPAP) ventilation would reduce the need for endotracheal intubation, the length of hospital stay, and hospital charges in patients with status asthmaticus. The development of physician treatment bias made patient enrollment difficult. The article subsequently describes the use of Bayesian statistics to explain study results when this bias occurs. METHODS: This study was a prospective, randomized controlled clinical trial conducted over a 34.5-month period at an urban university hospital with an emergency department census of 94,000 annual visits. Patients remaining in status asthmaticus after initial standard treatment with inhaled beta-agonists and steroids were randomized to receive BiPAP ventilation plus standard treatment versus standard treatment alone (non-BiPAP), with intubation for either group as needed. Patients with concurrent cardiac or other pulmonary diseases were excluded. The primary outcome measures were endotracheal intubation rate and length of hospital stay. Secondary outcome measures included vital signs (respiratory rate, pulse rate, blood pressure), changes in expiratory peak flow, changes in pulse oximetry values, and hospital charges. Data were analyzed using Fisher's exact test, Mann-Whitney tests, and Bayesian statistics. For patients enrolled in the study more than once, data analysis was performed on the first enrollment only. RESULTS: Nineteen patients were enrolled in the BiPAP group and 16 patients in the non-BiPAP group. Patients were frequently enrolled more than once and the data from the subsequent enrollments were excluded from the analysis. A marked decrease in enrollment, due to physician treatment bias, led to a premature termination of the study. Demographics showed that the groups were similar in age, sex, initial peak flow rate, and arterial blood gas measurements. There was a 7.3% increase (95% CI = -22 to +45) in the intubation rate in the non-BiPAP group (n = 2) compared with that for the BiPAP group (n = 1). No significant difference was seen in length of hospital stay or hospital charges, although there was a favorable trend toward the BiPAP group. Complications encountered in the BiPAP group included one patient with discomfort associated with the nasal BiPAP mask. Bayesian analysis demonstrated that in order for the collected data to be convincing at the 95% confidence level, the prior conviction among treating physicians that BiPAP was a successful treatment modality would have had to be 98.9%. CONCLUSIONS: In this study, BiPAP appeared to have no deleterious effects in patients with status asthmaticus, with a trend toward decreased endotracheal intubation rate, decreased length of hospital stay, and decreased hospital charges. Although further study with more patients is needed to determine the clinical and statistical significance of this intervention, ethical concerns regarding withholding BiPAP treatment from the patients in the control group forced a premature termination of the study in the authors' institution.

Human Schwann cells in vitro. I. Failure to differentiate and support neuronal health under co-culture conditions that promote full function of rodent cells.

Schwann cells (SCs) play critical roles in regeneration after injury to the peripheral nervous system and can also induce axonal regeneration in the central nervous system. Transplantation of purified SCs into sites of neural injury in rodents has confirmed the remarkable ability of these cells to promote axonal regrowth, suggesting that human application of SC transplantation could be valuable. In this report, we have compared the functional capacities of SCs derived from adult human and rodent nerves by of SCs derived from adult human and rodent nerves by maintaining SCs from these two sources in culture with sensory neurons. We noted that techniques commonly in use for maintaining pure rat SC populations are not sufficient to sustain populations of human SCs free of fibroblasts. In these co-cultures, human SCs express a limited profile of characteristic behaviors and they proliferate more slowly than rat SCs in response to axonal contact. Slow SC proliferation, relative to that of contaminating fibroblasts, leads to a high proportion of fibroblasts in the cultures. After 3 to 4 weeks of co-culture with neurons, human SCs express extracellular matrix molecules, but only partially ensheathe axons, whereas rat SCs differentiate, form basal lamina, and ensheathe or myelinate axons. Co-culture of sensory neurons with human (but not rat) SC preparations (or conditioned medium therefrom) leads to a progressive neuronal atrophy characterized by shrinking neuronal cell bodies and a decrease in the density of the neurite network in the culture dish. As the divergent effects of human and rat SCs on neuronal health were also observed in co-cultures with human sensory neurons, these effects reflect differences between the rat and human-derived SC populations, rather than a species mismatch between SCs and neurons. The marked differences in behavior observed between rat and human SCs derived by the same methods requires further exploration if human-derived SCs are to be considered in the treatment of disease. In a companion article we report experiments that define culture conditions more effective in promoting human SC function in vitro.

Human Schwann cells in vitro. II. Myelination of sensory axons following extensive purification and heregulin-induced expansion.

Co-culture conditions are well established in which Schwann cells (SCs) derived from immature or adult rats proliferate and form myelin in response to contact with sensory axons. In a companion article, we report that populations of adult-derived human Schwann cells (HASCs) fail to function under these co-culture conditions. Furthermore, we report progressive atrophy of neurons in co-cultures containing populations of either human fibroblasts or HASCs (which contain both SCs and fibroblasts). Two factors that might account for the insufficiency of the co-culture system to support HASC differentiation are the failure of many HASCs to proliferate and the influence of contaminating fibroblasts. To minimize fibroblast contamination of neuron-HASC co-cultures, we used fluorescence-activated cell sorting to highly purify HASC populations (to more than 99.8%). To stimulate expansion of the HASC population, a mitogenic mixture of heregulin (HRG beta 1 amino acid residues 177-244; 10 nM), cholera toxin (100 ng/mL), and forskolin (1 microM) was used. When these purified and expanded HASCs were co-cultured with embryo-derived rat sensory neurons, neuronal shrinkage did not occur and after 4 to 6 weeks some myelin segments were seen in living co-cultures. This myelin was positively identified as human by immunostaining with a monoclonal antibody specific to the human peripheral myelin protein P0 (antibody 592). Although this is the first reported observation of myelination by HASCs in tissue culture, it should be noted that myelination occurred more slowly and in much less abundance than in comparable cultures containing adult rat-derived SCs. We anticipate that further refinements of the HASC co-culture system that enhance myelin formation will provide insights into important aspects of human SC biology and provide new opportunities for studies of human peripheral neuropathies.

Axon-induced mitogenesis of human Schwann cells involves heregulin and p185erbB2.

The ability of sensory axons to stimulate Schwann cell proliferation by contact was established in the 1970s. Although the mitogen responsible for this proliferation has been localized to the axon surface and biochemically characterized, it has yet to be identified. Recently a family of proteins known as heregulins (HRGs) has been isolated, characterized, and shown to interact with a number of class 1 receptor tyrosine kinases, including the erbB2, erbB3, and erbB4 gene products. These factors include glial growth factor, a Schwann cell mitogen. We have tested the effects of antibodies against components of this system (HRG beta 1 and p185erbB2) in cocultures of rat sensory neurons and human (or rat) Schwann cells to elucidate the role of these proteins in axon-induced Schwann cell proliferation. 2C4, a monoclonal antibody specific for the human p185erbB2 receptor tyrosine kinase, bound to the surface of human Schwann cells and reduced human Schwann cell incorporation of [3H]thymidine by > 90% compared with untreated controls in this coculture system. This antibody had no effect on rat Schwann cell incorporation of [3H]thymidine under similar conditions. A polyclonal antibody raised against HRG beta 1 reduced human and rat Schwann cell incorporation of [3H]thymidine by nearly 80% and up to 49%, respectively, relative to controls. These results imply that a HRG, or a HRG-like molecule, is a component of the axonal mitogen. This mitogen is presented to Schwann cells by axons and induces proliferation through an interaction that involves p185erbB2 on Schwann cells.

A comparison of serum and vaginal dialysate concentration of 13,14-dihydro-15-keto-PGF2 alpha in women undergoing 1st trimester therapeutic abortions.

13,14-Dihydro-15-keto-PGF2 alpha(PGFM) levels in serum and vaginal secretion, before, one hour, and 24 hours post abortion were measured in 13 women from 8 to 12 weeks pregnant. The mean metabolite levels in the blood were 61.2 pg/ml pre-termination, 73.1 pg/ml and 121.4 pg/ml one hour and 24 hours post-termination. The vaginal dialysate means were 136.7 pg/ml, 425.7 pg/ml, and 211.7 pg/ml respectively.

PIP: 13, 14-Dihydro-15-keto prostaglandin F2alpha (PGFM) levels in serum and vaginal secretion before and hour and 24 hours after abortion were measured in 13 women from 8 to 12 weeks pregnant. Patients coming to the Department of Gynecology and Obstetrics at the Johns Hopkins Hospital, Baltimore, United States, were recruited at the time they applied for a first trimester therapeutic abortion (TAB). Prior to the procedure, a blood sample was drawn and a semi-permeable tampon was inserted into the vagina for about 2 hours while the patient waited for her surgery. Upon completion of dilation and suction curettage, a new tampon was placed in the vagina and remain in place for about 2 hours. A blood sample was taken when the tampon was removed. The anesthesia included local 1% lidocaine with IV sedation and 20 units pitocin given during and after the procedure. PGFM was measured by radioimmunoassay. The mean metabolite levels in the blood were 61.2 pg/ml pre-termination. 73.1 pg/ml and 121.4 pg/ml 1 hour and 24 hours post-termination, respectively. The vaginal dialysate means were 136.7 pg/ml, 425.7 pg/ml, and 211.7 pg/ml, respectively. A MANOVA was performed using SPSS/PC+ to determine the significance of changes in vaginal fluid and plasma PGFM levels over time. P-values were 0.0001 for both the dialysate and for plasma. T-test analysis demonstrated significant differences between baseline dialysate levels and levels just after TAB (p 0.0001, two tailed) and between levels just after TAB and 24 hours later (p 0.0001). A similar analysis of plasma demonstrated a significant rise in PGFM levels from baseline to 24 hours later (p 0.0001) and from just after TAB to 24 hours later

Effect of grafting order on innervation of spinal cord transplants by grafted locus coeruleus neurons in oculo.

Fetal brain stem containing locus coeruleus/subcoeruleus (hereafter referred to as LC) and thoracic spinal cord (SC) were sequentially allografted into the anterior eye chamber of adult Sprague-Dawley albino rats creating two groups: (1) LC graft followed after 11 weeks by an SC graft (LC-SC); (2) SC graft followed after 11 weeks by an LC graft (SC-LC). The cografts were allowed to mature in oculo for 15-18 months. After sacrifice, the grafts were processed for the immunohistochemical localization of tyrosine hydroxylase (TH) and dopamine beta-hydroxylase (D beta H) and the extent of fiber ingrowth into the SC graft was measured using computerized image analysis. TH- and D beta H-immunoreactive fibers were found to innervate the SC cograft in five of the six SC-LC graft combinations. The innervation was abundant, but uneven in distribution. The average density of TH-immunoreactive fibers (derived from the LC graft) was 8.29% of the total cross-sectional area of the SC graft neuropil. In contrast, the innervation of the SC graft neuropil was very sparse in the LC-SC graft combinations, with an uneven distribution and an average density of TH-immunoreactive fibers in the SC graft of only 2.28% of the cross-sectional area of the SC graft neuropil. The results support earlier studies of intraspinal grafting of LC neurons, in that embryonic LC neurons upon grafting and during ontogenetic fiber growth are capable of innervating mature spinal cord neuropil devoid of its normal catecholaminergic innervation. However, embryonic spinal cord tissue is a poor stimulant to reinitiate terminal fiber growth from mature LC neurons, in contrast to several other LC target areas such as the cerebral cortex and hippocampus.

Syngeneic Schwann cells derived from adult nerves seeded in semipermeable guidance channels enhance peripheral nerve regeneration.

At present, clinical strategies to repair injured peripheral nerve concentrate on efforts to attain primary suture of the cut nerve ends. If this is not possible, autografts are used to unite the separated nerve segments. Both strategies are based on the recognition that the Schwann cells resident in the peripheral nerve trunk play a crucial role in the regenerative process. Neither strategy may be feasible, however, in extensive or multiple injuries because the amount of autograft material is limited, and allografts are subject to immune rejection. Artificially produced nerve bridges constructed of autologous Schwann cells seeded in guidance channels could be used to overcome these limitations. In the present experiments, the potential of Schwann cells derived from adult nerves and seeded in permselective guidance channels to promote neurite regeneration across an 8 mm nerve gap was evaluated in transected rat sciatic nerves. Immunological sequalae were evaluated by comparing Schwann cells from syngeneic and heterologous rat strains. Schwann cells from either adult outbred (Sprague-Dawley, CD) rats or inbred (Fisher, F) rats were suspended in a Matrigel solution at a density of 80 x 10(6) cells/ml (CD) or 40, 80, or 120 x 10(6) cells/ml (F-40, F-80, and F-120 channels, respectively). Channels containing Schwann cells were compared to sciatic nerve autografts, empty channels, or channels filled with Matrigel alone. One day after seeding permselective synthetic guidance channels with a Schwann cell suspension, a central cable of Schwann cells oriented along the axis of the tube was formed due to syneresis of the hydrogel. By 3 weeks postimplantation, regenerating axons had grown into all channels and autografts. Sciatic nerve autografts supported extensive regeneration, containing 4-5 x 10(4) myelinated axons at the graft midpoint. The ability of channels containing syngeneic Schwann cells to foster regeneration was dependent on the Schwann cell seeding density. At the channel's midpoint, the myelinated axon population in F-120 tubes was intermediate between that in sciatic nerve autografts and F-80 channels, and was significantly higher than in F-40 or control channels. The nerve cable in Schwann cell-containing tubes consisted of larger, more organotypic fascicles than acellular control channels. In contrast, heterologous (CD) Schwann cells elicited a strong immune reaction that impeded nerve regeneration. The present study shows that cultured adult syngeneic Schwann cells seeded in permselective synthetic guidance channels support extensive peripheral nerve regeneration.(ABSTRACT TRUNCATED AT 400 WORDS)

Isolation and functional characterization of Schwann cells derived from adult peripheral nerve.

To facilitate the development of autologous transplantation techniques with which to test the ability of Schwann cell (ScC) implantations to treat nervous system injury, we have developed a method for procuring large, essentially pure populations of ScCs from adult peripheral nerve. By allowing small explants of peripheral nerve trunk to undergo axonal and myelin breakdown in vitro, rather than dissociating the nerve immediately after harvest, we are able to (1) rid the explant of nearly all fibroblasts and (2) capitalize on the intrinsic ScC mitogenic response to peripheral nerve degeneration. Here, we describe a method that yields up to 98% pure ScC populations from adult rat sciatic nerve (based on cell soma and nuclear morphology, S100 staining, and behavior of dissociated cells on neurites) at cell yields of greater than 2 x 10(4) cells/mg of starting nerve weight. The purification technique was successfully applied to human tissue; human phrenic nerve yielded 98% pure ScC populations at cell yields of 2 x 10(4) cells/mg of initial nerve weight. Similar to neonatally derived ScCs, adult rat cells can be expanded in coculture with dorsal root ganglion (DRG) neurons or in isolation in the presence of glial growth factor and forskolin. Cells expanded indefinitely on DRG neurons, or up to 10 weeks on chemical mitogens, return to quiescence following removal of the mitogenic stimulus. Expanded adult-derived rat ScCs retain functional capacity, as evidenced by their ability to myelinate DRG neurites and to support regeneration of processes from embryonic rat retinal explants.

Seemingly paradoxical jumping in cataleptic haloperidol-treated rats is triggered by postural instability.

Paradoxically, animals exhibiting haloperidol-induced cataleptic immobility can be induced to leap vigorously, by pushing them forward from behind. It is shown here that such jumping can also be produced by placing them on a board and tilting it tail-end upward until about 50 degrees above horizontal. In both situations, jumps only occurred when the animal's hindlegs began to slip forward, as they lost their postural stability. As alternatives to jumping from the slope, rats turned to face upwards (negative geotaxis), or adopted a spread-eagled posture during head-first downward sliding, with the body and head flattened against the substrate. All 3 responses to the sloping board were present in some undrugged rats. Such rats, and those given low doses of haloperidol (0.5, 1.0 mg/kg), were more likely to turn upwards than to jump or slide. At high doses (7.5, 10.0 mg/kg), they were more likely to slide downward than to turn or jump. Jumping was most likely to occur at an intermediate dose (5 mg/kg), approximately 60 min after injection. We suggest that in the absence of haloperidol, and at low doses, locomotion is dominant over reflexes defending static equilibrium, and hence rats are more likely to turn upwards (which involves stepping). In contrast, at higher doses, locomotion is more fully suppressed, reducing the likelihood of turning. At very high doses of haloperidol and later in the action of the drug, muscle tonus appears to be weakened, reducing the likelihood of jumping. This possibility was supported by the finding that combined injection of the optimal dose of haloperidol and 2 mg/kg diazepam reduced the ability to cling vertically (suggesting weakness of muscle tone). In such rats, jumping from the sloping board was decreased, and active downward sliding was increased. Thus, different factors influence the occurrence of jumping at different doses of haloperidol. However, these are all active defensive responses to postural instability, and hence are similar to the other reflexes used by haloperidol-treated rats to defend against displacement from static stable equilibrium, such as standing immobile, bracing, clinging, and righting. Jumping in response to loss of stability on the sloping board also occasionally occurred in undrugged rats. Unlike jumps by haloperidol-treated rats, those by undrugged animals only occurred when they could be directed to a safe landing place. Thus, if the board faced the edge of the table, so that the jump would carry the animal into space over the edge, undrugged rats either did not jump or jumped off the side of the board onto the table.(ABSTRACT TRUNCATED AT 400 WORDS)

Visual modulation of vestibularly-triggered air-righting in the rat.

Unlike cats, which can initiate righting in the air either with vestibular or visual input alone, the rat is dependent solely upon the labyrinths to trigger this response. We show, however, that the rat can modulate the onset and speed of its rotation according to the height above the ground from which it is dropped. In the absence of vision, rates initiate rotation with a latency of about 50 ms, irrespective of the height from which they are dropped. With vision, rats can modulate their latency to begin rotation, from about 102 ms at 50 cm, to about 39 ms at 7.5 cm. Similarly, as height of release decreases, the speed of rotation (i.e. degrees/ms) increases. Thus, in rats, even though vision cannot trigger air-righting, it does adaptively modulate this behavior as an allied reflex, increasing the likelihood that the animals will land on their feet.