Matthieu Resche-Rigon

Propensity scores in intensive care and anaesthesiology literature: a systematic review

IntroductionPropensity score methods have been increasingly used in the last 10xa0years. However, the practical use of the propensity score (PS) has been reported as heterogeneous in several papers reviewing the use of propensity scores and giving some advice. No precedent work has focused on the specific application of PS in intensive care and anaesthesiology literature.ObjectivesAfter a brief development of the theory of propensity score, to assess the use and the quality of reporting of PS studies in intensive care and anaesthesiology, and to evaluate how past reviews have influenced the quality of the reporting.Study design and settingForty-seven articles published between 2006 and 2009 in the intensive care and anaesthesiology literature were evaluated. We extracted the characteristics of the report, the type of analysis, the details of matching procedures, the number of patients in treated and control groups, and the number of covariates included in the PS models.ResultsOf the 47 articles reviewed, 26 used matching on PS, 12 used stratification on PS and 9 used adjustment on PS. The method used was reported in 81% of the articles, and the choice to conduct a paired analysis or not was reported in only 15%. The comparison with the previously published reviews showed little improvement in reporting in the last few years.ConclusionThe quality of reporting propensity scores in intensive care and anaesthesiology literature should be improved. We provide some recommendations to the investigators in order to improve the reporting of PS analyses.

Predictive factors of advanced interventional procedures in a multicentre severe postpartum haemorrhage study

PurposeSevere postpartum haemorrhage (SPPH) is the leading cause of peripartum hysterectomy and maternal death. There are no easily measurable parameters that indicate the failure of medical therapy and the need for an advanced interventional procedure (AIP) to stop genital tract bleeding. The aim of the study was to define factors predictive of the need for an AIP in the management of emergent PPH.MethodsThe study included two phases: (1) an initial retrospective study of 257 consecutive patients with SPPH, allowing the determination of independent predictors of AIP, which were subsequently grouped in a predictive score, followed by (2) a multicentre study of 239 patients admitted during 2007, designed to validate the score. The main outcome measure was the need for an AIP, defined as uterine artery embolization, intraabdominal packing, arterial ligation or hysterectomy.ResultsAbnormalities of placental implantation, prothrombin time <50% (or an International Normalized Ratio >1.64), fibrinogen <2xa0g/l, troponin detectable, and heart rate >115xa0bpm were independently predictive of the need for an AIP. The SPPH score included each of the five predictive factors with a value of 0 or 1. The greater the SPPH score, the greater the percentage of patients needing an AIP (11% for SPPH 0, to 75% for SPPH ≥2). The AUC of the ROC curve of the SPPH score was 0.80.ConclusionsWe identified five independent predictors of the need for an AIP in patients with SPPH and persistent bleeding. Using these predictors in a single score could be a reliable screening tool in patients at risk of persistent genital tract bleeding and needing an AIP.

Propensity score applied to survival data analysis through proportional hazards models: a Monte Carlo study

Propensity score methods are increasingly used in medical literature to estimate treatment effect using data from observational studies. Despite many papers on propensity score analysis, few have focused on the analysis of survival data. Even within the framework of the popular proportional hazard model, the choice among marginal, stratified or adjusted models remains unclear. A Monte Carlo simulation study was used to compare the performance of several survival models to estimate both marginal and conditional treatment effects. The impact of accounting or not for pairing when analysing propensity-score-matched survival data was assessed. In addition, the influence of unmeasured confounders was investigated. After matching on the propensity score, both marginal and conditional treatment effects could be reliably estimated. Ignoring the paired structure of the data led to an increased test size due to an overestimated variance of the treatment effect. Among the various survival models considered, stratified models systematically showed poorer performance. Omitting a covariate in the propensity score model led to a biased estimation of treatment effect, but replacement of the unmeasured confounder by a correlated one allowed a marked decrease in this bias. Our study showed that propensity scores applied to survival data can lead to unbiased estimation of both marginal and conditional treatment effect, when marginal and adjusted Cox models are used. In all cases, it is necessary to account for pairing when analysing propensity-score-matched data, using a robust estimator of the variance.

Prognostic factors in patients with hepatitis C virus infection and systemic vasculitis

OBJECTIVEnHepatitis C virus (HCV)-related systemic vasculitis can cause significant morbidity and mortality. Most studies of the prognosis of patients with HCV-related systemic vasculitis are based on heterogeneous studies performed before the era of antiviral therapy. The aim of this study was to analyze the clinical, biologic, and therapeutic factors associated with prognosis in a homogeneous series of patients with HCV-related systemic vasculitis who were followed up during the era of antiviral therapy.nnnMETHODSnOne hundred fifty-one consecutive HCV RNA-positive patients with vasculitis were prospectively followed up between 1993 and 2009 and were analyzed for clinical, biologic, and therapeutic factors associated with survival.nnnRESULTSnAfter a median followup period of 54 months, 32 patients (21%) had died, mainly of infection and end-stage liver disease. The 1-year, 3-year, 5-year, and 10-year survival rates were 96%, 86%, 75%, and 63%, respectively. Baseline factors associated with a poor prognosis were the presence of severe liver fibrosis (hazard ratio [HR] 5.31), central nervous system involvement (HR 2.74), kidney involvement (HR 1.91), and heart involvement (HR 4.2). The Five-Factors Score (FFS), a vasculitis scoring system, was significantly associated with outcome. In multivariate analysis, severe fibrosis (HR 10.8) and the FFS (HR 2.49) were significantly associated with a poor prognosis. Treatment with the combination of PEGylated interferon plus ribavirin was associated with a good prognosis (HR 0.34), whereas treatment with immunosuppressive agents was associated with a poor outcome, after adjustment for the severity of vasculitis (HR 4.05). Among patients without severe fibrosis, the FFS was a good predictor of outcome, while among those with severe fibrosis, the severity of vasculitis had no prognostic value.nnnCONCLUSIONnAt the time of the diagnosis of HCV-related systemic vasculitis, severe liver fibrosis and the severity of vasculitis were the main prognostic factors. Use of antiviral agents was associated with a good prognosis, whereas treatment with immunosuppressant agents had a negative impact.

Cervical necrotizing fasciitis: 8-years’ experience of microbiology

Cervical necrotizing fasciitis (CNF) is a life-threatening complication of pharyngeal or dental infections. The aim of this paper was to investigate whether dental or pharyngeal source result from different pathogen(s) in CNF and whether antibiotics, given before admission, influence the antimicrobial resistance of pathogens. In 152 CNF patients, Streptococcus milleri group and Prevotella species were the predominant isolates, frequently copathogens, mostly in dental CNF samples. Penicillin and clindamycin resistance were observed in 39% and 37% of cases, respectively, independently of any previous antibiotic therapy. Thus, a combined aerobe–anaerobe infection may have a synergistic effect, which allows the infection to spread in cervical tissues.

Cardiac troponin and skeletal muscle oxygenation in severe post-partum haemorrhage

IntroductionCardiac troponin has been shown to be elevated in one-half of the parturients admitted for post-partum haemorrhage. The purpose of the study was to assess whether increased cardiac troponin was associated with a simultaneous alteration in haemoglobin tissue oxygen saturation in peripheral muscles in post-partum haemorrhage.MethodsTissue haemoglobin oxygen saturation of thenar eminence muscle (StO2) was measured via near-infrared spectroscopy technology. Two sets of StO2 parameters (both isolated baseline and during forearm ischaemia-reperfusion tests) were collected at two time points: upon intensive care unit admission and prior to intensive care unit discharge. Comparisons were performed using Wilcoxon paired tests, and univariate associations were assessed using logistic regression model and Wald tests.ResultsThe 42 studied parturients, admitted for post-partum haemorrhage, had clinical and biological signs of severe blood loss. Initial cardiac troponin I was increased in 24/42 parturients (0.43 ± 0.60 μrg/l). All measured parameters of muscular haemoglobin oxygen saturation, including Srecovery, were also altered at admission and improved together with improved haemodynamics, when bleeding was controlled. Multivariate analysis showed that muscular Srecovery <3%/second at admission was strongly associated with increased cardiac troponin.ConclusionsOur study confirmed the high incidence of increased cardiac troponin, and demonstrated the simultaneous impairment in the reserve of oxygen delivery to peripheral muscles in parturients admitted for severe post-partum haemorrhage.

Computed tomography evaluation of high-grade esophageal necrosis after corrosive ingestion to avoid unnecessary esophagectomy

BackgroundEsophagectomy is the standard of care for high-grade corrosive esophageal necrosis as assessed endoscopically. However, the inaccuracy of endoscopy in determining the depth of intramural necrosis may lead to unnecessary esophageal resection, with devastating consequences. Our aim was to evaluate the use of computed tomography (CT) for the emergency diagnostic workup of endoscopic high-grade corrosive esophageal necrosis.MethodsIn a before (2000–2007)/after (2007–2012) study of patients with grade 3b endoscopic esophageal necrosis, we compared outcomes after routine emergency esophagectomy versus selection for emergency esophagectomy based on CT evidence of transmural necrosis, defined as at least two of the following: esophageal-wall blurring, periesophageal-fat blurring, and the absence of esophageal-wall enhancement. Survival estimated using the Kaplan–Meier method was the primary outcome.ResultsCompared to the routine-esophagectomy group (nxa0=xa0125), the CT group (nxa0=xa072) had better overall survival in the crude analysis (hazard ratio [HR], 0.43; 95xa0% confidence interval [95xa0%CI], 0.21–0.85; Pxa0=xa00.015) and in the analysis matched on gender, age, and ingested agent (HR, 0.36; 95xa0%CI, 0.16–0.79; Pxa0=xa00.011). No deaths occurred among patients managed without emergency esophagectomy based on CT findings, and one-third of CT-group patients had their functioning native esophagus at last follow-up. Self-sufficiency for eating and breathing was more common (84xa0% vs. 65xa0%; relative risk [RR], 1.27; 95xa0%CI, 1.04–1.55; Pxa0=xa00.016) and repeat suicide less common (4xa0% vs. 15xa0%; RR, 0.27; 95xa0%CI, 0.09–0.82; Pxa0=xa00.019) in the CT group.ConclusionThe decision to perform emergency esophagectomy for endoscopic high-grade corrosive esophageal injury should rely on CT findings.

Using the continual reassessment method to estimate the minimum effective dose in phase II dose-finding studies: a case study

Background The Continual Reassessment Method typically is presented as the method of choice for the purpose of dose-finding based on a toxicity scale in phase I clinical trials. However, this adaptive statistical approach also can be applied easily to dose-finding experiments in phase II trials. Purpose To provide a case study from a real clinical trial to illustrate the use of the Continual Reassessment Method in the context of phase II dose finding. Methods The Continual Reassessment Method was used to model the dose-failure relationship in order to estimate the minimal effective dose. This approach was retrospectively used to determine the minimal effective dose of granulocyte colony-stimulating factor for peripheral blood stem cell collection in allografted patients following chemotherapy. Results After the inclusion of 25 patients, the minimal effective dose was estimated to be the third dose level tested in the study. Limitations The main limitation of the Continual Reassessment Method, which is not specific to the method but to the dose-finding setting, is that the empirical choice of the dose range can be either under or over-estimated. The method requires a calibration study prior to trial onset. Conclusions Assuming that a dose-effect relationship is monotonically increasing, the use of the Continual Reassessment Method in phase II dose-finding studies allows the estimation of the minimum effective dose for further studies. Modeling the dose-failure relationship allows the direct use of available software developed for the Continual Reassessment Method in the context of phase I clinical trials.

Oesophagogastrectomy and pancreatoduodenectomy for caustic injury

The justification for pancreatoduodenectomy (PD) for extended duodenal and pancreatic caustic necrosis is still a matter of debate.

Adaptive designs for dose-finding in non-cancer phase II trials: influence of early unexpected outcomes

Background In non-cancer phase II trials, dose-finding trials are usually carried out using fixed designs, in which several doses including a placebo are randomly distributed to patients. However, in certain vulnerable populations, such as neonates or infants, there is an heightened requirement for safety, precluding randomization. Purpose To estimate the minimum effective dose of a new drug from a non-cancer phase II trial, we propose the use of adaptive designs like the Continual Reassessment Method (CRM). This approach estimates the dose closest to some target response, and has been shown to be unbiased and efficient in cancer phase I trials. Methods Based on a motivating example, we point out the individual influence of first outliers in this setting. A weighted version of the CRM is proposed as a theoretical benchmark to control for these outliers. Using simulations, we illustrate how this approach provides further insight into the behavior of the CRM. Results When dealing with low targets like a 10% failure rate, the CRM appears unable to rapidly overcome an early unexpected outcome. This behavior persisted despite changing the inference (Bayesian or likelihood), underlying dose—response model (though slightly improved using the power model), and the number of patients enrolled at each dose level. Limitations The choices for initial guesses of failure rates, the vague prior for the model parameter, and the log-log shape of weights can appear somewhat arbitrary. Conclusions In phase II dose-finding studies in which failure targets are below 20%, the CRM appears quite sensitive to first unexpected outcomes. Using a power model for dose—response improves some behavior if the trial is started at the first dose level and includes at least three to five patients at the starting dose before applying the CRM allocation rule.