Search

Objectives There is no consensus for the specific management of elderly patients presenting with oral cavity squamous cell carcinomas (OC SCC). We report our findings in the treatment of primary OC SCC, for patients of 70 years of age or more, in a French university hospital center. Patients and methods One hundred and twenty five patients diagnosed between 2000 and 2010, were included retrospectively. Independent risk factors of post-operative complications were identified using a logistic regression. The overall survival (OS) was estimated using the Kaplan Meier method. Independent factors of survival were calculated using a Cox model. Results The patient's median age was 78. Women presented significantly more premalignant lesions, less alcohol intoxication, and less tobacco consumption. Half of the population sample was staged T4 in the TNM classification. Eighty eight percent of the patients received a curative treatment. The independent risk factors for post-operative complications were T3/T4 stages (OR 4.3 [1.3–14.4]), lymph node metastasis (OR 6.9 [2.1–22.7]), and alcohol abuse (OR 3.5 [1.1–11.0]). The median OS was 14.0 months. The independent negative prognostic factors for OS for patients treated curatively were: age >79 years (HR 1.9 [1.2–3.2]), stage T2/T3/T4 tumor vs. T1 (HR = 3.0 [1.5–6.0], P = 0.001) and substandard surgery (HR = 1.8 [1.1–2.9], P = 0.03). Conclusions The management of OC SCC in elderly patients is complex and requires collaboration among gerontologists, surgeons and oncologists. The treatment choice is related to the disease extent and preoperative morbid conditions.

International audience; Objective: To examine the relationship between a Parkinson's disease (PD) polygenic risk score (PRS) and impulse control disorders (ICDs) in PD. Background: Genome wide association studies (GWAS) have brought forth a PRS associated with increased risk of PD and younger disease onset. ICDs are frequent adverse effects of dopaminergic drugs and are also more frequent in patients with younger disease onset. It is unknown whether ICDs and PD share genetic susceptibility. Methods: We used data from a multicenter longitudinal cohort of PD patients with annual visits up to 6 years (DIG-PD). At each visit ICDs, defined as compulsive gambling, buying, eating, or sexual behavior were evaluated by movement disorders specialists. We genotyped DNAs using the Megachip assay (Illumina) and calculated a weighted PRS based on 90 SNPs associated with PD. We estimated the association between PRS and prevalence of ICDs at each visit using Poisson generalized estimating equations, adjusted for dopaminergic treatment and other known risk factors for ICDs. Results: Of 403 patients, 185 developed ICDs. Patients with younger age at onset had a higher prevalence of ICDs (p < 0.001) as well as higher PRS values (p = 0.06). At baseline, there was no association between the PRS and ICDs (overall, p = 0.84). The prevalence of ICDs increased over time similarly across the quartiles of the PRS (overall, p = 0.88; DA users, p = 0.99). Conclusion: Despite younger disease onset being associated with both higher PRS and ICD prevalence, our findings are not in favor of common susceptibility genes for PD and ICDs.

ObjectiveTo assess the cost-effectiveness of acupuncture for pelvic girdle and low back pain (PGLBP) during pregnancy.DesignPragmatic-open-label randomised controlled trial.SettingFive maternity hospitals.PopulationPregnant women with PGLBP.Method1:1 randomization to standard care or standard care plus acupuncture (5 sessions by an acupuncturist midwife).Main outcome measureEfficacy: proportion of days with self-assessed pain by numerical rating scale (NRS) ≤ 4/10. Cost effectiveness (societal viewpoint, time horizon: pregnancy): incremental cost per days with NRS ≤ 4/10. Indirect non-healthcare costs included daily compensations for sick leave and productivity loss caused by absenteeism or presenteeism.Results96 women were allocated to acupuncture and 103 to standard care (total 199). The proportion of days with NRS ≤ 4/10 was greater in the acupuncture group than in the standard care group (61% vs 48%, p = 0.007). The mean Oswestry disability score was lower in the acupuncture group than with standard care alone (33 versus 38, Δ = 5, 95% CI: 0.8 to 9, p = 0.02). Average total costs were higher in the control group (€2947) than in the acupuncture group (€2635, Δ = -€312, 95% CI: -966 to +325), resulting from the higher indirect costs of absenteeism and presenteeism. Acupuncture was a dominant strategy when both healthcare and non-healthcare costs were included. Costs for the health system (employer and out-of-pocket costs excluded) were slightly higher for acupuncture (€1512 versus €1452, Δ = €60, 95% CI: -272 to +470).ConclusionAcupuncture was a dominant strategy when accounting for employer costs. A 100% probability of cost-effectiveness was obtained for a willingness to pay of €100 per days with pain NRS ≤ 4.

International audience; UNLABELLED: The criteria for defining failure to control bleeding in cirrhosis patients were introduced at the Baveno II/III meetings and were widely used as endpoints in clinical trials. Because they lacked specificity, the Baveno IV criteria were proposed in 2005 and slightly modified in 2010 (Baveno V). These criteria included a new index for patients undergoing transfusion, called adjusted-blood-requirement-index (ABRI=number of blood units/(final-initial hematocrit+0.01)), with a cutoff value of 0.75. In this multicenter prospective study, we sought to 1) validate the Baveno IV/V criteria; 2) compare them to the Baveno II/III criteria; 3) assess ABRI performance using a standardized calculation. The key inclusion criteria were: 1) variceal bleeding; 2) cirrhosis; 3) no need to modify the transfusion policy. The patients were classified according to the Baveno IV, V, and II/III criteria. The gold standard for failure during a 5-day period was the clinical judgment of three independent experts, blinded to the Baveno assessments. A total of 249 patients were included. The experts' agreement in clinical judgment of the failure was 80%. Failure occurred in 20.5% of patients; the c-statistics were 0.72 versus 0.64 and 0.65 for Baveno IV versus Baveno II/III and Baveno V criteria (P=0.001 for both). ABRI did not improve the diagnostic performance of the Baveno IV criteria. The Baveno IV, but not Baveno II/III, criteria independently predicted survival.CONCLUSION: The Baveno IV criteria demonstrated a higher accuracy than the Baveno II/III and Baveno V criteria for assessing failure to control bleeding and predicted survival independently. Together, our results show that ABRI is not a useful metric, and the Baveno IV criteria should replace the Baveno II/III criteria.

Radar altimetry was, very early in the development of space technology, identified as a key technique to provide essential information on solid Earth and ocean dynamics. Satellite altimetry calculation results from the combination of two measurements. From a different perspective, "Satellite with Argos and AltiKa" is an innovative radar altimetry mission that utilizes a new Ka-band altimeter concept (AltiKa). The Geodynamic Experimental Ocean Satellite inherited radar technology from the Skylab and provided more accurate altimetric measurements. New altimetry techniques require short-term accuracy characteristics. The mission requirement is thus expressed using a spatial frequency spectrum. While on its path between the satellite and the Earth's surface, the radar signal is impacted by various geophysical phenomena that must be accounted to provide a precise range measurement. In order to reach the expected range measurement accuracy, the effects of the atmosphere on the propagation of the altimeter microwave radar pulse must be accounted for.

BACKGROUND Circulating proteins released by tumor cells have recently been investigated as potential single surrogate biomarkers for glioblastoma multiforme (GBM). The aim of the current hypothesis-generating study was to evaluate the diagnostic and prognostic role of preoperative insulin-like growth factor-binding protein 2 (IGFBP-2), chitinase-3-like protein 1 (YKL-40), and glial fibrillary acidic protein (GFAP) plasma levels in patients with GBM, both as single markers and as a combined profile. METHODS Plasma samples from 111 patients with GBM and a subset of 40 patients with nonglial brain tumors were obtained preoperatively. Plasma from 99 healthy controls was also analyzed. IGFBP-2, YKL-40, and GFAP levels were determined using enzyme-linked immunoadsorbent assay tests. Their association with histological and radiological variables was assessed. RESULTS Circulating levels of all 3 proteins were found to be significantly higher in patients with GBM compared with healthy controls (P < .01). Only YKL-40 and GFAP were found to demonstrate significant differences between patients with GBM and nonglial brain tumors (P = .04). GFAP was undetectable (

Highlights • 2 doses of vaccine induces a humoral response comparable to that triggered by natural infection after allogeneic transplant. • Seroprotection rate significantly increases after the second dose of adjuvanted vaccine. • Adjuvanted vaccine was safe with low rate of graft-versus-host disease., Background The present study evaluated immunogenicity and tolerance of two-dose influenza A/H1N1pdm09 vaccination in allogeneic hematopoietic stem cell transplantation (HSCT) recipients, and compared the vaccine-induced humoral response to that triggered by natural infection in another group of HSCT patients. Methods Adult allogeneic HSCT recipients vaccinated with two doses of influenza A/H1N1pdm09 vaccine, separated by 3 weeks, and patients with proven influenza A/H1N1pdm09 infection were included. Antibody responses were measured by hemagglutination-inhibition assay 1) on days 0, 21, 42 and 6 months after the first vaccine injection in vaccinated patients and 2) before pandemic and after influenza A/H1N1pdm09 infection, in patients presented natural infection. Results At baseline, 3% of 59 recipients of adjuvanted vaccine and 0% of 20 infected patients were seroprotected (antibody titer ≥ 1/40). Seroprotection rate observed 42 days after vaccination was not different from that observed after natural infection (66% and 60% respectively, p = 0.78). In vaccinated patients, seroprotection rate increased significantly from 54% to 66% between day 21 and 42 (p = 0.015). Moreover, after 6 months, seroprotection rate in 21 vaccinated patients was similar to that observed in 10 infected patients evaluated at least 76 days after infection (D76–217) (60% and 81% respectively, p = 0.2). In multivariate analysis, no immunosuppressive treatment or chronic graft-versus-host disease (GVHD) and longer time between transplantation and vaccination/infection were associated with a stronger humoral response. The adjuvanted vaccine was safe with low rate of GVHD worsening. Conclusion In HSCT recipients, two doses of influenza A/H1N1pdm09 adjuvanted vaccine were safe and induced a humoral response comparable to that triggered by natural infection in these patients.

International audience; Objective: To evaluate the efficacy and safety of zonisamide in patients with myoclonus-dystonia. Methods: We conducted a randomized, double-blind, placebo-controlled crossover trial of zonisamide (300 mg/d) in 24 patients with myoclonus-dystonia. Each treatment period consisted of a 6-week titration phase followed by a 3-week fixed-dose phase. The periods were separated by a 5-week washout period. The co–primary outcomes were action myoclonus severity (section 4 of the Unified Myoclonus Rating Scale [UMRS 4]) and myoclonus-related functional disability (UMRS 5). Secondary outcomes included dystonia severity, assessed with the movement and disability subscales of the Burke-Fahn-Marsden-Dystonia Rating Scale (BFM), the Clinical Global Impression–Improvement scale (CGI), and safety measures. Wilcoxon signed-rank tests for paired data were used to analyze treatment effects. Results: Twenty-three patients (11 men, 12 women) were analyzed in the intention-to-treat analysis. Zonisamide significantly improved both action myoclonus (median improvement [95% confidence limits] −5 [−9.25 to −1.44], p = 0.003) and myoclonus-related functional disability (median improvement [95% confidence limits] −2 [−2.58 to −2.46], p = 0.007) compared to placebo. Zonisamide also significantly improved dystonia (BFM movement) compared to placebo (median improvement [95% confidence limits] −3 [−8.46 to 0.03], p = 0.009). No difference was found between zonisamide and placebo with respect to the CGI (median improvement [95% confidence limits] −1 [−1.31 to 0.09], p = 0.1). Zonisamide was well-tolerated. Conclusions: Zonisamide is well-tolerated and effective on the motor symptoms of myoclonus-dystonia. Classification of evidence: This study provides Class I evidence that zonisamide improves myoclonus and related disability in patients with myoclonus-dystonia

Des etudes realisees dans differentes pays ont montre que la mortalite par tuberculose etait plus faible dans les populations juives que dans les autres communautes vivant dans les memes regions. Cette situation a egalement ete constatee dans la ville de Tunis. Notre etude a pour but de mieux analyser ce fait a partir des statistiques de mortalite de la ville de Tunis. Elle montre que dans la premiere moitie du XXe siecle, c’est-a-dire avant l’apparition des traitements antibiotiques, la mortalite par tuberculose etait plus faible dans la population juive que dans les populations musulmane, francaise, italienne et maltaise ; en revanche, le taux de mortalite infantile etait plus eleve dans la population juive que dans les populations italienne ou francaise. Par exemple, au cours de la periode 1919-1939, en moyenne annuelle, ces deux taux s’etablissent respectivement a 81 pour 100 000 et 164 pour 1 000 dans la population juive, contre 193 pour 100 000 et 107 pour 1 000 dans la population francaise. Cette situation ne peut s’expliquer par les conditions de vie ; on ne peut cependant exclure le role d’une alimentation differente, la consommation de viande casher dans la population juive pouvant eviter la contamination par le bacille tuberculeux bovin. Des facteurs genetiques sont egalement avances, mais leur role n’a pu etre mis en evidence avec certitude.

The feedforward multi-layer perceptron (MLP) type neural network (NN) presented in this paper has been developed for on-board applications of high-speed signal processing. It is fully analog in order to avoid analog---digital conversions and to limit chip size and power consumption. It is constituted by a single input, ten neurons in the hidden layer and a single output. The MLP-NN has been implemented in a 84 pins (0.6 μm CMOS ASIC) and has a power consumption of 600 mW. The NN layout size is 1.8 mm × 0.7 mm. This paper reminds the design and the simulations of each implemented cell and details the different experimental tests achieved.

REPLY: We thank Dr. Ninio for his interest in our 25-yr-old publications and for his attempt to analyze our data and interpretations using a strategy different from the one we used for this pioneering work. However, his approach is exclusively concerned with curve fitting. It avoids the more demanding problem of the recovery of parameters of explicit statistical models of biological processes. That is, he decided to discriminate between fits of our published histograms visually, or “by eye.” The field long ago progressed to the point where the standard for confronting data with models is statistical analysis, not visual fitting. Reevaluation of old methods can be a good thing (Lowen et al. 1997), provided there is a sound scientific questions at stake. Yet, we are surprised that Dr. Ninio raises concerns about the validity of aspects of these seminal papers but does not want to review the subsequent literature that appeared about our results and conclusions. This is difficult to understand because the aim of a meaningful scientific approach is to progress in the understanding of natural phenomena and processes. In this respect, it is worth noting that subsequent research carried out independently by a number of eminent scientists supported our proposal of the “One Vesicle Hypothesis” of transmitter release at several synapses in the CNS. In this response we address the flaws in the logic of Ninio’s analysis and his criticisms and misunderstandings of our published work. Because of the progress in this field, and given that Dr. Ninio is not an expert in this area, we believe our response here will be of interest to other investigators who are concerned with using rigorous methods to extract information from experimental distributions. BACKGROUND Experimental context We performed simultaneous intracellular recordings and dye injections, in vivo, to collect data sets of fluctuating inhibitory postsynaptic potentials (IPSPs) evoked in the Mauthner cell of teleost fish by single presynaptic action potentials and used an optimization procedure to determine the best-fitting binomial and Poisson descriptions of the data. We then compared those results with histological reconstructions of the presynaptic cells, including the axonal branching pattern and the identification of contacts with the postsynaptic neuron. This was one of the earliest attempts to obtain such structure‐function correlations. We were particularly interested in two questions that had been raised previously: 1) Were the fluctuations in PSP amplitude better described by a binomial or a Poisson distribution? and 2) Could the statistical parameters be correlated with physical structures?

Anemia remains frequent in the first month following renal transplantation and is a risk factor for cardiovascular accidents. The purpose of this study was to analyze the prevalence of anemia during this period notably among different immunosuppressive treatment groups.In this study, we entered the patients who had received a renal allograft in our transplant unit from 1993 to 2003, including patients who had received azathioprine (AZA) from 1993 to 1996 and mycophenolate mofetil (MMF) from 1996 to 2003. No patient received rHu-erythropoietin after transplantation during that period. A mathematical model normalized the hemoglobin (Hb) threshold level at which blood transfusion was decided and Hb on admission.One hundred and eighty-eight patients on AZA and 223 on MMF were included in the analysis. The mean age +/- SD was 41 +/- 12 years in the AZA group, and 45 +/- 12 years in the MMF group (P.006). Before the transplantation, Hb was higher in the MMF group (11.4 +/- 1.9 vs 10.2 +/- 2 g/dL, P.0001). After normalization at a threshold level of transfusion of 7 g/dL, transfusions were more frequent among the MMF group (72% vs 48%, P.0001). Double therapy with MMF (1500 mg/d) + steroids or therapy with MMF (750 mg/d) + tacrolimus + steroids increased the risk of transfusion compared to the AZA group. MMF (750 mg/d) + cyclosporine conferred a similar risk of transfusion compared with the AZA group.MMF alone or in association with tacrolimus is associated with an increased risk of anemia and transfusion in the immediate posttransplantation period.

Background. Renal outcome after ciclosporin (CsA) is not clear in most studies involving patients with many renal comorbid conditions. We first report on renal function recovery after CsA in previously healthy kidney patients. Methods. Uveitis patients, enroled in a unique single centre cohort follow-up study initiated in 1987, were prospectively evaluated for plasma creatinine and glomerular filtration rate (GFR) before, during (>2 years) and after (>6 months) CsA therapy. We hypothesized that CsA alters renal function progressively over time according to two additive exponential components (irreversible and reversible) and used a mixed linear model with exponential speed parameters maximizing the likelihood. Results. Twenty-seven patients treated for 60 � 34 months (CsA 5.1 � 2.5 mg/kg/day) were followed up for 56 � 42 months after CsA withdrawal. Baseline creatinine was 0.92 � 0.15 mg/dl. The reversible effect of CsA was quantified as a 0.11 � 0.07 mg/dl increase in creatinine/100 mg CsA/day (P < 0.001) and a 6.0 � 3.7 ml/min/1.73 m 2 decrease in GFR/100 mg CsA/day (P < 0.0001). The irreversible effect was quantified as a 0.03 � 0.05 mg/dl increase in creatinine/100 g cumulative CsA received (P < 0.007) and a decrease of 3.3 � 3.9 ml/min/1.73 m 2 GFR/100 g CsA. Conclusions. Although significant decrease in GFR is induced by low-dose CsA therapy in previously healthy kidney patients, renal function recovery is possible after CsA withdrawal and best predicted by CsA daily dosage. Irreversible loss in GFR is correlated to cumulated CsA exposure. The lowest CsA dosage and shortest exposure time effect as well as unlimited renal monitoring are required in order to provide the best long-term renal outcome.

Comment in High-Volume Hemofiltration in Post-Cardiac Surgery Shock. A Heroic Therapy? [Am J Respir Crit Care Med. 2015]; International audience; RATIONALE:Post-cardiac surgery shock is associated with high morbidity and mortality. By removing toxins and proinflammatory mediators and correcting metabolic acidosis, high-volume hemofiltration (HVHF) might halt the vicious circle leading to death by improving myocardial performance and reducing vasopressor dependence.OBJECTIVES:To determine whether early HVHF decreases all-cause mortality 30 days after randomization.METHODS:This prospective, multicenter randomized controlled trial included patients with severe shock requiring high-dose catecholamines 3-24 hours post-cardiac surgery who were randomized to early HVHF (80 ml/kg/h for 48 h), followed by standard-volume continuous venovenous hemodiafiltration (CVVHDF) until resolution of shock and recovery of renal function, or conservative standard care, with delayed CVVHDF only for persistent, severe acute kidney injury.MEASUREMENTS AND MAIN RESULTS:On Day 30, 40 of 112 (36%) HVHF and 40 of 112 (36%) control subjects (odds ratio, 1.00; 95% confidence interval, 0.64-1.56; P = 1.00) had died; only 57% of the control subjects had received renal-replacement therapy. Between-group survivors' Day-60, Day-90, intensive care unit, and in-hospital mortality rates, Day-30 ventilator-free days, and renal function recovery were comparable. HVHF patients experienced faster correction of metabolic acidosis and tended to be more rapidly weaned off catecholamines but had more frequent hypophosphatemia, metabolic alkalosis, and thrombocytopenia.CONCLUSIONS:For patients with post-cardiac surgery shock requiring high-dose catecholamines, the early HVHF onset for 48 hours, followed by standard volume until resolution of shock and recovery of renal function, did not lower Day-30 mortality and did not impact other important patient-centered outcomes compared with a conservative strategy with delayed CVVHDF initiation only for patients with persistent, severe acute kidney injury. Clinical trial registered with www.clinicaltrials.gov (NCT 01077349).

Background: The short term benefits of bilateral stimulation of the subthalamic nucleus (STN) in patients with advanced levodopa responsive Parkinson's disease (PD) are well documented, but long term benefits are still uncertain. Objectives: This study provides a 5 year follow up of PD patients treated with stimulation of the STN. Methods: Thirty seven consecutive patients with PD treated with bilateral STN stimulation were assessed prospectively 6, 24, and 60 months after neurosurgery. Parkinsonian motor disability was evaluated with and without levodopa treatment, with and without bilateral STN stimulation. Neuropsychological and mood assessments included the Mattis Dementia Rating Scale, the frontal score, and the Montgomery-Asberg Depression Rating Scale (MADRS). Results: No severe peri- or immediate postoperative side effects were observed. Six patients died and one was lost to follow up. Five years after neurosurgery: (i) activity of daily living (Unified Parkinson Disease Rating Scale (UPDRS) II) was improved by stimulation of the STN by 40% ("off" drug) and 60% ("on" drug); (ii) parkinsonian motor disability (UPDRS III) was improved by 54% ("off" drug) and 73% ("on" drug); (iii) the severity of levodopa related motor complications was decreased by 67% and the levodopa daily doses were reduced by 58%. The MADRS was unchanged, but cognitive performance declined significantly. Persisting adverse effects included eyelid opening apraxia, weight gain, addiction to levodopa treatment, hypomania and disinhibition, depression, dysarthria, dyskinesias, and apathy. Conclusions: Despite moderate motor and cognitive decline, probably due to disease progression, the marked improvement in motor function observed postoperatively was sustained 5 years after neurosurgery.

Aims To describe the effect of nicotine replacement therapy (NRT) on the risk of relapse as a function of time since the quit date. Data sources Meta-analysis of 21 published, randomized, controlled clinical trials, comparing NRT to placebo. Data extraction A total of 6644 smokers were treated with NRT and 2766 smokers treated with placebo. Data synthesis During treatment with the medication, NRT reduced the hazard ratio (HR) significantly compared with placebo [early HR = 0.62 (95% CI: 0.58–0.67)]. At the end of the average treatment duration (145 days), the HR was 0.81 (95% CI: 0.71–0.94), showing that the benefit was still present at this time. After stopping treatment, the HR increased progressively up to a value of 1.44 (95% CI, 1.18–1.76) showing that the risk of relapse was higher after stopping NRT than after stopping placebo. If NRT and placebo had not been stopped, the HR of smoking relapse would have been established at 0.95 (95% CI: 0.76–1.18, P = 0.64), indicating a similar risk of relapse with NRT and placebo. Moreover, the observed HR of smoking relapse was significantly higher than the expected HR of smoking relapse if NRT had been continued: the difference in HR is 1.51 (95% CI: 1.16–1.98, P

A dynamical system modeling the HIV infection, including a mutation occurrence process, is used, after simplifications, to show the impact of the viral diversity on the immune response and disease dynamics, by introducing an indicator of the immune system behavior, the immunological recognition efficacy (IRE) index. The existence, the expression and the stability of the endemically infected steady state of the IRE index-based model, as function of this index, are mathematically analyzed. The monotony of the steady state with respect to the IRE index is studied and an asymptotic analysis of the dynamical system performed. It is shown that the IRE index-based model provides a bound to the responses of the initial complex dynamical system. The biological interpretation of these mathematical results is the exhaustion of the immune system as a consequence of the continuous generation of viral mutants.

Because treatment failure in many HIV-infected persons may be due to multiple causes, including resistance to antiretroviral agents, it is important to better tailor drug therapy to individual patients. This improvement requires the prediction of treatment outcome from baseline immunological or virological factors, and from results of resistance tests. Here, we review briefly the available clinical factors that have an impact on therapy outcome, and discuss the role of a predictive modelling approach integrating these factors proposed in a previous work. Mathematical and statistical models could become essential tools to address questions that are difficult to study clinically and experimentally, thereby guiding decisions in the choice of individualized drug regimens.

Aims For ethical and economic reasons, interim analysis of phase III clinical trials is essential. This study was conducted to compare the efficiency of two interim analysis procedures, which could be used to allow early termination of a clinical trial. Methods We made a post hoc application of two interim analysis methods (Lan & DeMets with O’Brien–Flemming modification, and the triangular test according to Whitehead) by using individual patient data from four published placebo-controlled survival trials. We determined the date the trial would have been stopped had each method been used, and we estimated consequent results in terms of events and patient numbers included in the trial, the duration of the trial, and on treatment effect. Results The triangular test provided the lowest number of events required to reach a conclusion of the trials while providing an accurate estimate of experimental treatment effects. The triangular test thus indicated the smallest number of patients that would have been enrolled, and the shortest trial duration. The difference between the methods was most important with a detrimental effect of experimental treatment: the number of required events was reduced by 75% and the trial duration was shortened by 48% with the triangular test compared to the Lan & DeMets method. Conclusions Stopping a trial early must depend on the clinical context. It is most important to stop a placebo-controlled trial as soon as possible when the experimental treatment can be shown deleterious. In such a situation the triangular test is more appropriate than the Lan & DeMets method. When a treatment effect is no different from, or better than, placebo the triangular test is also superior but the importance of premature termination of the trial in such cases has to be balanced against the inevitable reduction of information that the trial can provide.

Cyclosporine (CsA), a widely used immunosuppressive drug, is an effective treatment of sight-threatening posterior idiopathic uveitis. CsA's main side effect is nephrotoxicity. The aim of this single-center prospective cohort study (conducted in a tertiary care teaching hospital in Paris, France) was to assess the long-term renal tolerance of a low-dose CsA treatment in patients with previously healthy kidneys on clinical, biologic, and pathologic criteria. Forty-one patients treated with 4.3 +/- 1.6 mg/kg body wt per day CsA for 44.9 +/- 3.6 mo were included. Mean follow-up was 55.4 +/- 0.2 mo. BP, CsA trough level, and renal function were prospectively monitored together with blood urea, creatinine clearance, GFR, and effective renal plasma flow. Eleven patients underwent serial kidney biopsies before and after 2 yr of a 4 +/- 0.9 mg/kg daily CsA treatment. Sustained low-dose CsA treatment induced a significant increase in plasma creatinine (P < 0.0001), a significant decrease in creatinine clearance (P < 0.0001), and isotopic GFR (P < 0.0001) over time. The highest dose induced more severe alterations in any of the renal parameters than the lowest dose. Prevalence of hypertension was particularly high. Histopathologic data showed significant interstitial fibrosis (P < 0.003) and tubular atrophy (P < 0.003) after 2 yr. Low-dose long-term CsA treatment induces significant renal impairment and a high incidence of hypertension. Our study suggests that lowering daily dosage may prevent CsA-induced nephrotoxicity if a daily dose of < or =3 mg/kg is used. Whether once established it is reversible is still prospective, although the occurrence of interstitial fibrosis in the kidney would argue against reversibility.

Considerable experience has been gained with endoluminal stent-graft techniques for the treatment of infrarenal aortic aneurysms. The inclusion of uncovered suprarenal stents eases the treatment of cases with a short, angulated, or bottlenecked proximal neck. However, few data have been published on the effect of a stent-graft crossing renal artery ostia on long-term renal function. We therefore retrospectively analyzed the influence of intra-aortic stent-grafts, including suprarenal fixation on long-term renal function. Thirty-nine patients who received a Gianturco Z stent across renal artery ostia as part of a made-to-measure stent-graft treating an infrarenal aortic aneurysm were evaluated at 6 (all patients) and 30 months (15 patients) after surgery. Renal function evaluation included serum creatinine level, creatinine clearance using the Cockcroft-Gault formula, and renal tomography. Initial technical success was achieved in all patients. There were no minor or major complications. Mean serum creatinine levels were 1.16 ± 0.25, 1.12 ± 0.28, and 1.24 ± 0.25 mg/dL before and 6 and 30 months after endoluminal stent-graft performance, respectively ( P = not significant [NS]). Mean creatinine clearances were 60.3 ± 19.7, 63 ± 23, and 60.6 ± 17.5 mL/min before and 6 and 30 months after endoluminal stent-graft performance, respectively ( P = NS). Only two patients (5%) had a decrease in glomerular filtration rate greater than 20% at 6 months. Both patients had renal insufficiency before endovascular grafting. Renal function was stable in all patients at 30 months' follow-up. We suggest that renal function is unaffected by coverage of renal arterial ostia with Gianturco Z stents. © 2002 by the National Kidney Foundation, Inc.

International audience; Background : Catecholamine-O-methyl-tranferase (COMT) initiates dopamine degradation. Its activity is mainly determined by a single nucleotide polymorphism in the COMT gene (Val158Met, rs4680) separating high (Val/Val, COMTHH), intermediate (Val/Met, COMTHL) and low metabolizers (Met/Met, COMTLL). We investigated dopaminergic denervation in the striatum in PD patients according to COMT rs4680 genotype.Methods : Patients with idiopathic PD were assessed for motor severity (UPDRS-III rating scale in OFF-state), dopaminergic denervation using [123I]-FP-CIT SPECT imaging, and genotyped for the COMT rs4680 enzyme. [123I]-FP-CIT binding potential (BP) for each voxel was defined by the ratio of tracer-binding in the region of interest (striatum, caudate nucleus and putamen) to that in a region of non-specific activity. Genotyping was performed using TaqMan® SNP genotyping assay. We used a regression model to evaluate the effect of COMT genotype on the BP in the striatum and its sub-regions.Results : Genotype distribution was: 11 (27.5%) COMTHH, 26 (65%) COMTHL and 3 (7.5%) COMTLL. There were no significant differences in disease severity, treatments, or motor scores between genotypes. When adjusted to clinical severity, gender and age, low and intermediate metabolizers showed significantly higher rates of striatal denervation (COMTHL+LL BP=1.32 +/- 0.04) than high metabolizers (COMTHH, BP=1.6 +/- 0.08; F(1.34)=9.0, p=0.005). Striatal sub-regions showed similar results. BP and UPDRS-III motor scores (r=0.44, p=0.04) (p

The biology of fatal pandemic influenza infection remains undefined.To characterize the virologic and immune parameters associated with severity or death in patients who required mechanical ventilation for A(H1N1) 2009 pneumonia of various degrees of severity during the two waves of the 2009-2011 pandemic in Paris, France.This multicenter study included 34 unvaccinated patients with very severe or fatal confirmed influenza A(H1N1) infections. It analyzed plasma A(H1N1) 2009 reverse-transcriptase polymerase chain reaction, hemagglutinin 222G viral mutation, and humoral and cellular immune responses to the virus, assessed in hemagglutination inhibition (HI), microneutralization, ELISA, lymphoproliferative, ELISpot IFN-γ, and cytokine and chemokine assays.The patients' median age was 35 years. Influenza A(H1N1) 2009 viremia was detected in 4 of 34 cases, and a 222G hemagglutinin mutation in 7 of 17 cases, all of them with sequential organ failure assessment greater than or equal to 8. HI antibodies were detectable in 19 of 26 survivors and undetectable in all six fatal fulminant cases. ELISA and microneutralization titers were concordant. B-cell immunophenotyping and plasma levels of immunoglobulin classes did not differ between patients who survived and died. After immune complex dissociation, influenza ELISA serology became strongly positive in the bronchoalveolar lavage of the two fatal cases tested. H1N1-specific T-cell responses in lymphoproliferative and IFN-γ assays were detectable in survivors' peripheral blood, and lymphoproliferative assays were negative in the three fatal cases tested. Plasma levels of IL-6 and IL-10 were high in fatal cases and correlated with severity. Finally, a negative HI serology 4 days after the onset of influenza symptoms predicted death from fulminant influenza (P = 0.04).Early negative A(H1N1) 2009 HI serology can predict death from influenza. This negative serology in fatal cases in young adults reflects the trapping of anti-H1N1 antibodies in immune complexes in the lungs, associated with poor specific helper T-cell response. Clinical trial registered with www.clinicaltrials.gov (NCT 01089400).

Circulating proteins released by tumor cells have recently been investigated as potential single surrogate biomarkers for glioblastoma multiforme (GBM). The aim of the current hypothesis-generating study was to evaluate the diagnostic and prognostic role of preoperative insulin-like growth factor-binding protein 2 (IGFBP-2), chitinase-3-like protein 1 (YKL-40), and glial fibrillary acidic protein (GFAP) plasma levels in patients with GBM, both as single markers and as a combined profile.Plasma samples from 111 patients with GBM and a subset of 40 patients with nonglial brain tumors were obtained preoperatively. Plasma from 99 healthy controls was also analyzed. IGFBP-2, YKL-40, and GFAP levels were determined using enzyme-linked immunoadsorbent assay tests. Their association with histological and radiological variables was assessed.Circulating levels of all 3 proteins were found to be significantly higher in patients with GBM compared with healthy controls (P.01). Only YKL-40 and GFAP were found to demonstrate significant differences between patients with GBM and nonglial brain tumors (P = .04). GFAP was undetectable (0.02 ng/mL) in all patients without GBM. A receiver operating characteristic analysis accounting for a 2-step diagnostic procedure including the 3 biomarkers afforded an area under the curve of 0.77 for differentiating patients with GBM from those with nonglial brain tumors. There was a significant correlation between tumor volume and plasma IGFBP-2 level (Spearman Rho correlation coefficient, 0.22; P = .025) and GFAP (Spearman Rho correlation coefficient, 0.36; P.001) among patients with GBM. Preoperative plasma IGFBP-2 levels were found to be independently associated with worse overall survival among patients with GBM (hazard ratio, 1.3; P = .05).A combined profile of preoperative IGFBP-2, GFAP, and YKL-40 plasma levels could serve as an additional diagnostic tool for patients with inoperable brain lesions suggestive of GBM. In addition, IGFBP-2 levels appear to constitute an independent prognostic factor in patients with GBM.

Background —β-Blockade–induced benefit in heart failure (HF) could be related to baseline heart rate and treatment-induced heart rate reduction, but no such relationships have been demonstrated. Methods and Results —In CIBIS II, we studied the relationships between baseline heart rate (BHR), heart rate changes at 2 months (HRC), nature of cardiac rhythm (sinus rhythm or atrial fibrillation), and outcomes (mortality and hospitalization for HF). Multivariate analysis of CIBIS II showed that in addition to β-blocker treatment, BHR and HRC were both significantly related to survival and hospitalization for worsening HF, the lowest BHR and the greatest HRC being associated with best survival and reduction of hospital admissions. No interaction between the 3 variables was observed, meaning that on one hand, HRC-related improvement in survival was similar at all levels of BHR, and on the other hand, bisoprolol-induced benefit over placebo for survival was observed to a similar extent at any level of both BHR and HRC. Bisoprolol reduced mortality in patients with sinus rhythm (relative risk 0.58, P P =NS). A similar result was observed for cardiovascular mortality and hospitalization for HF worsening. Conclusions —BHR and HRC are significantly related to prognosis in heart failure. β-Blockade with bisoprolol further improves survival at any level of BHR and HRC and to a similar extent. The benefit of bisoprolol is questionable, however, in patients with atrial fibrillation.

Background: A combination of low-dose aspirin with anticoagulants may provide better protection against thromboembolic events compared to anticoagulants alone in high-risk patients with atrial fibrillation. Objective: Evaluation of the preventive efficacy against nonfatal thromboembolic events and vascular deaths of the combination of the oral anticoagulant fluindione and aspirin (100 mg) in patients with high-risk atrial fibrillation. Methods: A multicenter, placebo-controlled, double-blind, randomized trial was conducted at 49 investigating centers in France. Atrial fibrillation patients with a previous thromboembolic event or older than 65 years and with either a history of hypertension, a recent episode of heart failure or decreased left ventricular function were included in the study. Patients were treated with fluindione plus placebo (i.e. anticoagulant alone) or fluindione plus aspirin (i.e. combination therapy), with an international normalized ratio target of between 2 and 2.6. The combined primary endpoint was stroke (ischemic or hemorrhagic), myocardial infarction, systemic arterial emboli or vascular death. The secondary endpoint was the incidence of hemorrhagic complications. Results: The 157 participants (average age 74 years; 52% women; 42% with paroxysmal atrial fibrillation) were followed for an average of 0.84 years. Three nonfatal thromboembolic events were observed (1 in the anticoagulation group, 2 in the combination group) and 6 patients died (3 in the anticoagulation group, 3 in the combination group), none of them from a thromboembolic complication. However, 3 deaths were secondary to severe hemorrhagic complications (1 in the anticoagulation group, 2 in the combination group). Nonfatal hemorrhagic complications occurred more often in the combination group (n = 10, 13.1%) compared to the anticoagulation group (n = 1, 1.2%) (p = 0.003). Conclusion: The combination of aspirin with anticoagulant is associated with increased bleeding in elderly atrial fibrillation patients. The effect on thromboembolism and the overall balance of benefit to risk could not be accurately assessed in this study due to the limited number of ischemic events.

Nosocomial (hospital-acquired) infections are very frequent in intensive care units (ICU). The risk of death after severe infection is high, but the precise rate of death in ICU attributable to nosocomial infection is not known. The goal of this project was to build a statistical model to predict the occurrence of nosocomial infections in ICU and the outcome of the patients. We collected data on 676 consecutive patients admitted to an ICU for more than 24 hours between 1993 and 1996. The following data were collected for each patient: history; clinical examination at entry; subsequent infections; outcome. A multi-state heterogeneous semi-Markov model was determined and then validated; the initial data set was randomly split into two groups: two-thirds (450 patients) to build the model and one-third (226 patients) to validate it. The model defined five states: ICU admission; first simple infection; first complicated infection; death, and discharge from the ICU. Transitions between these states determined nine different events. The global model of patient histories can be divided into nine survival models, each corresponding to one of these events. The possible events from a given state were considered to be competing. Since many risk factors induced non-proportional hazard functions, piecewise exponential models were used to model event occurrence. The effect of continuous covariates on hazard functions has been described with a non-parametric method that enables non-linear relations to be shown. Among other things, the model allows patients' post-admission histories to be predicted from data available at ICU admission. The bootstrap estimator of the attributable risk of death due to simple or complicated nosocomial infections is 44.2 per cent (95 per cent CI 26.0-61.6 per cent). We were also able to characterize the most highly exposed patients, those who comprise the high-risk group on whom prevention efforts must be focused.

The aims of the study were to analyze the clinical features, the penetrance and the mode of inheritance of 13 French families with dilated cardiomyopathy using diagnostic criteria recently established by a European collaboration. Methods: Screening consisted of physical examination, ECG and Echo of all the probands first degree relatives (n=118). Using major Echo criteria [ejection fraction (EF) 117% of the predictive value], or combined minor Echo/ECG criteria, relatives were classified as affected, unknown or healthy. Results: (1) Adult affected relatives (n=31) were identified with major Echo criteria in 74% of cases, and with combined minor Echo/ECG criteria in 26% of cases. (2) In the unknown relatives (n=21), the most common abnormality was an isolated left ventricular dilation (67%). (3) Mode of inheritance was autosomal dominant (AD) in 11 families and possibly autosomal recessive in two. (4) In AD families, the penetrance was incomplete in adults (72%), age-related (O.R.: 1.3 per 10 years; 95% CI 1.03–1.56) and sex-related [greater in men (87%) than in women (61%), actuarial survival curve: P < 0.002]. (5) Mortality related to end stage heart failure was 2.2 times as high as mortality related to sudden death (11% vs. 5%). Conclusions: (1) In the absence of a specific phenotype of FDC, the characterization of relatives appears more accurate when minor criteria were added. (2) Since high mortality (16%) and incomplete penetrance frequently give rise to small nuclei of clinically affected and alive relatives per family, the accurate model of penetrance that we proposed might be helpful in the future to enhance the statistical power of linkage analysis in this disease.

The aim of this study was to assess the factors, including surgical portosystemic shunts, which affect survival in adults with Budd-Chiari syndrome. Multivariate retrospective analysis was performed using characteristics recorded at the time of diagnosis in 120 patients admitted from 1970 to 1992, of whom 82 were treated with surgical portosystemic shunts and 38 received only medical therapy. The 1-, 5-, and 10-year survival rates were 77 +/- 4%, 64 +/- 5%, and 57 +/- 6%, respectively. Survival was significantly better in the subgroup of patients diagnosed after versus before 1985. In both subgroups, and in patients with, as well as in patients without surgical shunts, 4 factors were found to be inversely and independently related to survival: age, response of ascites to diuretics, Pugh score, and serum creatinine. In patients diagnosed since 1985, an index combining these 4 factors allowed to differentiate patients with a good outcome (5-year survival 95%) from those with a poor outcome (5-year survival 62%; P

Objectives: We evaluated the parameters influencing CD4 cell reconstitution after the introduction of highly active antiretroviral therapies in real life, as well as the frequency and the determinants of the discrepancies occurring between virus and CD4 cell count evolution. Design and methods: A total of 317 pre-treated patients starting a protease inhibitor (PI)-containing regimen were prospectively followed for 2 years on an intent-to-treat basis for CD4 cell counts and viral loads. Results: The CD4 cell counts rapidly increased from baseline (50/mm 3 ) by a median of 50/mm 3 at month 2 (+0.72 CD4 cells/mm 3 /day) and up to 137/mm 3 at the last follow-up (second slope: +0.16 CD4 cells/mm 3 /day). Two independent major factors among five parameters tested significantly affected the first phase, which was negatively correlated to the slope of CD4 cell decline before PI initiation, and was positively correlated to baseline CD4 cell counts (P = 0.0001); the second phase was mostly affected by the mean viral load reduction over time (P = 0.0001). Paradoxical CD4 cell reconstitution (15% of cases) was defined by a rapid or slow CD4 cell increase contrasting with a minor or strong viral reduction, respectively. The role of previous CD4 cell decline and the low effect of viral load reduction during the first 2 months explain the early paradoxical CD4 cell responses. The major influence of viral load reduction on the long-term reconstitution, however, reduces such paradoxical responses at 2 years. Conclusions : Early paradoxical CD4 cell reconstitution after the introduction of a PI are explained by the major influence of previous disease progression on the early CD4 cell increase, whereas the magnitude of viral load reduction over time reduces such paradoxical evolutions in the long term.

Population approaches are appealing methods for detecting then assessing drug-drug interactions mainly because they can cope with sparse data and quantify the interindividual pharmacokinetic (PK) and pharmacodynamic (PD) variability. Unfortunately these methods sometime fail to detect interactions expected on biochemical and/or pharmacological basis and the reasons of these false negatives are somewhat unclear. The aim of this paper is firstly to propose a strategy to detect and assess PD drug-drug interactions when performing the analysis with a nonparametric population approach, then to evaluate the influence of some design variates (i.e., number of subjects, individual measurements) and of the PD interindividual variability level on the performances of the suggested strategy. Two interacting drugs A and B are considered, the drug B being supposed to exhibit by itself a pharmacological action of no interest in this work but increasing the A effect. Concentrations of A and B after concomitant administration are simulated as well as the effect under various combinations of design variates and PD variability levels in the context of a controlled trial. Replications of simulated data are then analyzed by the NPML method, the concentration of the drug B being included as a covariate. In a first step, no model relating the latter to each PD parameter is specified and the NPML results are then proceeded graphically, and also by examining the expected reductions of variance and entropy of the estimated PD parameter distribution provided by the covariate. In a further step, a simple second stage model suggested by the graphic approach is introduced, the fixed effect and its associated variance are estimated and a statistical test is then performed to compare this fixed effect to a given value. The performances of our strategy are also compared to those of a non-population-based approach method commonly used for detecting interactions. Our results illustrate the relevance of our strategy in a case where the concentration of one of the two drugs can be included as a covariate and show that an existing interaction can be detected more often than with a usual approach. The prominent role of the interindividual PD variability level and of the two controlled factors is also shown.

The expectation of the determinant of the inverse of the population Fisher information matrix is proposed as a criterion to evaluate and optimize designs for the estimation of population pharmacokinetic (PK) parameters. Given a PK model, a measurement error model, a parametric distribution of the parameters and a prior distribution representing the belief about the hyperparameters to be estimated, the EID criterion is minimized in order to find the optimal population design. In this approach, a group is defined as a number of subjects to whom the same sampling schedule (i.e., the number of samples and their timing) is applied. The constraints, which are defined a priori, are the number of groups, the size of each group and the number of samples per subject in each group. The goal of the optimization is to determine the optimal sampling times in each group. This criterion is applied to a one-compartment open model with first-order absorption. The error model is either homoscedastic or heteroscedastic with constant coefficient of variation. Individual parameters are assumed to arise from a lognormal distribution with mean vectorMand covariance matrixC. Uncertainties about theMandCare accounted for by a prior distribution which is normal forMand Wishart forC. Sampling times are optimized by using a stochastic gradient algorithm. Influence of the number of different sampling schemes, the number of subjects per sampling schedule, the number of samples per subject in each sampling scheme, the uncertainties onMandCand the assumption about the error model and the dose have been investigated.

A population analysis of the kinetics of mizolastine was performed from concentrations on 449 allergic patients, using the nonparametric maximum likelihood method (NPML). A two-compartment open model with zero-order absorption was used to describe the kinetics of mizolastine after oral administration. A heteroscedastic variance model was assumed for the error. To explain the kinetic variability, eight covariates were introduced in the analysis: gender, pharmaceutical dosage form, age, body weight, serum creatinine concentration, creatinine renal clearance, plasma levels of hepatic transaminases ASAT and ALAT. Their relationships to the kinetic parameters were studied by means of the estimated distribution of each kinetic parameter conditional on different levels of each covariate. An important interindividual kinetic variability was found for all parameters. Moreover, several kinetic parameters among which the duration of absorption were found to be influenced by pharmaceutical dosage form and gender. Body weight and creatinine renal clearance were found to have a little influence on the oral clearance and the smallest disposition rate constant. This population analysis was validated on a separate group of 247 other patients. For each observed concentration of this sample, a predictive distribution was computed using the individual covariates. Predicted concentrations and standardized prediction errors were deduced. The mean and variance of the standardized prediction errors were, respectively, 0.21 and 2.79. Moreover, in the validation sample, the predicted cumulative distribution function of each observed concentration was computed. Empirical distribution of these values was not significantly different from a uniform distribution, as expected under the assumption that the population model estimated by NPML is adequate.

Aims To evaluate the distribution of population kinetic parameters for clozapine and their relationship to age and gender in patients on continuous treatment with the drug. Methods Retrospective therapeutic drug monitoring data (391 samples from 241 patients) were evaluated using the nonparametric maximum likelihood method. Patients treated concomitantly with drugs known to interact with clozapine were not included. The distribution of clozapine clearance was compared with the distribution of the activity of the drug metabolic enzyme CYP1A2 found in other populations, as recent studies indicate that CYP1A2 is a major determinant for clozapine elimination. The kinetic linearity for clozapine was studied in 41 patients who each provided data from more than one dose level. Results Clozapine clearance was highly variable in the population and skewed towards high values. Men had higher clearances CL/F (median with 25% and 75% quartiles 38.2 (22.0, 60.0) vs 28.3 (15.2, 48.6) l h−1 ) and a larger volume of distribution V /F (694 (224, 970) vs 401 (189, 932) l) than women. Clearance did not decrease with age in any gender. Clozapine clearance was similarly distributed as the indices of CYP1A2-activity found in other populations by other authors. Evidence of nonlinear kinetics was not found. Conclusion The large kinetic variability for clozapine found in this study implies that the dose of clozapine needs to be individualised over a wide dose range. The similarity of the distribution of clozapine clearance in this study and the CYP1A2-activity in other populations support the assumption that CYP1A2 is a major determinant for clozapine elimination.

Statistical modeling methods have had increasing use in drug disposition studies, both to estimate pharmacokinetic parameters and to develop regression models that relate these parameter estimates to patient characteristics. These methods are particularly flexible as they allow non-linearity and sparse within-patient information. In the past few years, multiple analysis methods have become available, but there is a lack of systematic comparisons of their estimates on the same data sets. Two simulated data sets were therefore developed by the Population Modeling Workgroup of the Biopharmaceutical Section of the American Statistical Association. We analysed these data sets using seven population modeling programs, some of which contain multiple analysis methods. Although each data set represents a single replicate from a given model and data collection design, the results suggest that the behaviour of some methods differs from that of the others.

SUMMARY An approach is proposed to optimal design of experiments for estimating randomeffects regression models. The population designs are defined by the number of subjects and the individual designs to be performed. Cost functions associated with individual designs are incorporated. For a given maximal cost, an algorithm is proposed for finding the statistical population design that maximises the determinant of the Fisher information matrix of the population parameters. The Fisher information matrix is formulated for linear models and normal distributions. The approach is applied to the design of an optimal experiment in toxicokinetics using a first-order linearisation of the model. Several cost functions and designs of various orders are studied. An example illustrates the optimal population designs and the increased efficiency of some optimal designs over more standard designs.

The population approach was developed to estimate population characteristics of the pharmacokinetics (PKs) of a drug from a sparse-sampling designed experiment in a set of individuals. This approac...

The most widely used model for multidimensional survival analysis is the Cox model. This model is semi-parametric, since its hazard function is the product of an unspecified baseline hazard, and a parametric functional form relating the hazard and the covariates. We consider a more flexible and fully nonparametric proportional hazards model, where the functional form of the covariates effect is left unspecified. In this model, estimation is based on the maximum likelihood method. Results obtained from a Monte-Carlo experiment and from real data are presented. Finally, the advantages and the limitations of the approacha are discussed.

Population models were developed to analyze processes, described by parametric models, from measurements obtained in a sample of individuals. In order to analyze the sources of interindividual variability, covariates may be incorporated in the population analysis. The exploratory analyses and the two-stage approaches which use standard non-linear regression techniques are simple tools to select meaningful covariates. The global population approaches may be divided into two classes within which the covariates are handled differently: the parametric and the non-parametric methods. The power as well as the limitations of each approach regarding handling of covariates are illustrated and compared using the same data set which concerns the pharmacokinetics of gentamicin in neonates. With parametric approaches a second-stage model between structural parameters and covariates has to be defined. In the non-parametric method the joint distribution of parameters and covariates is estimated without parametric assumptions; however, it is assumed that covariates are observed with some error and parameters involved in functional relationships are not estimated. The important results concerning gentamicin in neonates were found by the two methods.

A computer-assisted Bayesian individual estimation of radioiodine thyroid uptake kinetics for patients suffering from Grave's disease is proposed. The program provides a fast computation of the activity to be administered to a given patient to achieve a target thyroid absorbed dose. This determination relies upon the patient biological covariates and upon a small number of measurements performed during a preliminar kinetic study of radioiodine thyroid uptake. Our results indicate that a two-sample Bayesian approach is reliable when external thyroid counts are performed at 2 h and 168 h after a test dose and has advantages over conventional kinetic experiments in terms of patients acceptability. This method is implemented on widespread computers and interfaced with a patient database. An interactive user interface with in-line data checking is provided. The program could be also a tool to better study the relationship between the absorbed dose and the clinical effect.