Evaluation of different approaches for confounding in nonrandomised observational data: a case-study of antipsychotics treatment.

Although randomised controlled trials are regarded as the gold standard for treatments efficacy, evidence from observational studies remains relevant. To address the problem of possible confounding in these studies, investigators must employ analysis methods that adjust for confounders and lead to an unbiased estimation of the treatment effect. In this paper, the authors describe two relevant statistical methods. The first method represents the classical approach consisting of a multiple regression model including the effects of treatment and covariates. This approach considers the relation between prognostic factors and the outcome variable as a relevant criterion for adjustment. The second method is based on the propensity score, and focuses on the relation between prognostic factors and treatment assignment. These approaches were applied to a cohort of 183 French schizophrenic patients who were followed for a 2-year period (from 1998 to 2000). The probability of relapse according to antipsychotic treatment exposure was modelled using Cox regression models with the two statistical methods. Goodness-of-fit criteria were used to compare the modelling approaches. This study demonstrates that the propensity score, a predicted probability, has an important balancing property that underscores its value in strengthening the results of nonrandomised observational studies.

Circulating lipoprotein-associated phospholipase A2 in high-grade carotid stenosis: a new biomarker for predicting unstable plaque.

OBJECTIVE: To test plasma levels of lipoprotein-associated phospholipase A2 (Lp-PLA2) in patients with high-grade carotid stenosis according to plaque histology. METHODS: This cross-sectional single-centre study included patients with ≥70% North American Symptomatic Carotid Endarterectomy Trial (NASCET) carotid stenosis, who were treated surgically. Serum Lp-PLA2 and high-sensitivity C-reactive protein (hs-CRP) were determined on the day of surgery. Histopathological analysis classified carotid plaque as stable or unstable, according to AHA classification. RESULTS: Of the 42 patients (mean age 70.4 ± 10.5 years; 67% men), neurological symptoms were present in 16 (38%). Unstable plaques were found in 23 (55%). Median plasma level of Lp-PLA2 was significantly higher in patients with unstable plaque compared to those with stable plaque (222.4 (174.9-437.5) interquartile range (IQR) 63.5 vs. 211.1 (174.9-270.6) IQR 37.2 ng ml(-1); p = 0.02). Moreover, median Lp-PLA2 level were higher in asymptomatic patients with unstable plaque (226.8 ng ml(-1) (174.9-437.5) IQR 76.8) vs. stable plaque (206.9 ng ml(-1) (174.9-270.6) IQR 33.7; p = 0.16). Logistic regression showed that only the neurological symptoms (OR = 30.9 (3.7-244.6); p < 0.001) and the plasma Lp-PLA2 level (OR = 1.7 (1.1-12.3); p = 0.03) were independently associated with unstable carotid plaque as defined by histology. CONCLUSIONS: This study showed that circulating Lp-PLA2 was increased in patients with high-grade carotid stenosis and unstable plaque. Lp-PLA2 may be a relevant biomarker to guide for invasive therapy in asymptomatic patients with carotid artery disease.

Imprisonment, alcohol dependence and risk of delusional disorder: a cross-sectional study.

BACKGROUND: Compared to the general population, an excess of psychotic illnesses, major depression and dependence disorders among prisoners has been reported. However, the impact of prison on detainees' psychopathology has rarely been studied. OBJECTIVE: To determine the mental disorders liable to develop or regress on entry into prison and over time. METHOD: Two samples of French prisoners detained in local prisons were interviewed using the same methodology. The first sample consisted of 267 new arrivals. The second was a random sample of 450 prisoners. Diagnoses were assessed using a thorough methodology: each prisoner was interviewed for approximately 2 hours by two clinicians. One of the clinicians used a structured clinical interview, which generates DSM IV diagnoses (MINI plus v 5.0); the second completed the procedure with an open clinical interview. The final DSM IV diagnoses were obtained as a consensus between the two approaches. Multilevel logistic regressions were used to take into account potential confounders. RESULTS: Prevalence rates of mental disorders were substantially higher in prison even for the sample of newcomers (major depression disorder: 24.7%, substance dependence: 17.6% and schizophrenia: 4.1%). Alcohol dependence disorder was significantly more frequent in the sample of newcomers (OR 1.84 [1.01-3.51]). No significant difference was evidenced between samples for substance dependence disorder. Psychotic disorders were significantly less frequent at entry into prison, particularly delusional disorder (OR 0.29 [0.08-0.98]). CONCLUSION: This study shows the contrasted potential effects of prison on psychopathology: alcohol dependence disorders were significantly more frequent for the newcomers, while the frequency of delusional disorders was lower. This evidence is arguing in favour of the validity of the old concept: prison psychosis. Moreover, prisoners should receive relevant help from clinicians to cope with these disorders.

Ultrasound arterial anomalies in patients exposed to nilotinib therapy for chronic myeloid leukemia.

INTRODUCTION: Patients exposed to nilotinib for chronic myeloid leukemia (CML) appear to be at risk of arterial complication. The prevalence and aspect of ultrasound asymptomatic arterial lesions are unknown. OBJECTIVE: To describe prevalence and characteristics of ultrasound arterial anomalies in patients treated with nilotinib for CML. METHODS: Patients treated with nilotinib from 2006 to 2015 in the department of the Paoli-Calmettes Institute, Marseille, were included retrospectively. A vascular ultrasound screening was carried out from 2010. The arterial lesions at the first examination were described: plaque and its echogenicity, stenosis or occlusion. A vascular arterial anomaly (VAA) was defined by the presence of a clinical and/or ultrasound anomaly. Patients with or without VAA at initial vascular examination were compared using bivariate and multivariate analysis. RESULTS: 74 patients were included (51.4% men, mean age 54.5 years); 25 patients had ultrasound arterial anomalies (33.8%). Carotid bulb was the most involved territory (44%). Arterial anomalies were: 88% plaques, 44%>50% stenosis and 12% occlusion. 72.7% plaques were echolucent or hypoechogenic. A VAA was present in 25 patients with initial vascular evaluation (33.8%). Patients with VAA at baseline were significantly older (64.9 vs 49.3, P<0.001), older at nilotinib initiation (60.8 vs 46.5, P<0.001), with more arterial hypertension (40% vs 12.2%, P=0.01), with more cardiovascular risk factors (P=0.03). In patient with no cardiovascular risk factor 12.5% had VAA (n=24). CONCLUSION: Nilotinib seems to be associated to arterial lesions of unstable lipid-like appearance. The most involved arterial territory was the carotid bulb and the most common lesion was echolucent or hypoechogenic plaque. VAA can occur in patients without cardiovascular risk factors. This result encourages us to systematically screen and follow all patients exposed to nilotinib even those without cardiovascular risk factors.

[Endovascular aneurysm repair follow-up by unenhanced and contrast-enhanced duplex ultrasound]. / Surveillance des endoprothèses aortiques abdominales : intérêt de l'échographie-doppler standard et avec contraste.

OBJECTIVE: Long-term surveillance is needed after endovascular aneurysm repair to monitor the aneurysm and search for persistent endoleaks. Our aim was to compare follow-up with duplex ultrasound, with and without a new contrast agent to track endoleaks, versus computed tomography angiography taken as the gold standard. MATERIAL AND METHOD: Patients treated with endograft were included prospectively from December 2005 to July 2006. Aortic duplex ultrasound and computed tomography were used to measure maximal aneurysm diameter and detect endoleaks. Patients with a high risk of endoleaks had a contrast-enhanced ultrasound with Sonovue (Bracco, Milan, Italy). We compared echographic and tomographic diameter and studied the sensitivity of ultrasound endoleak diagnosis. RESULTS: Sixty-seven patients were included. There was a good correlation between maximum anteroposterior diameters (CCI=0.98) measured by ultrasound and tomography, as well as mean maximum cross section diameters (CCI=0.96). Compared to tomography, the sensitivity of ultrasound endoleaks diagnosis was 44% (kappa=0.58). Contrast injection improved this sensitivity significantly (p<0.001) (sensitivity=88%; kappa=0.72). CONCLUSION: These findings confirmed the performance of our ultrasound method for endograft surveillance. Contrast-enhanced ultrasound significantly improves the sensitivity of detection of endoleaks. We suggest alternating ultrasound and tomographic exams. A unique report chart for use nationwide would be useful for standardizing follow-up.

[Payment by activity and financing of innovative hospital procedures: does the new payment system cover endovascular treatment for cerebral aneurysms?]. / Financement des traitements endovasculaires des anévrysmes cérébraux. La tarification à l'activité permet-elle aux établissements de santé de financer la prise en charge des patients?

METHOD: We have calculated all hospital expenses related to treated intracerebral aneurysms for 2005. Catheters, microcatheters, and guidewires as well as coils were included. We have compared these expenses to the payment by activity and fee per service collected for providing these services. RESULTS: Payments received covered only a third of the expenses for the supplies used. Three types of expenses are not reimbursed: the guiding material, the coils used but not released, and the latest generation of coils not yet added to the national list of covered devices. These expenses are also not covered by the payment received for the hospital admission. DISCUSSION: Endovascular management of intracranial aneurysms has become the treatment of choice over the recent years. This treatment is virtually only available in university hospitals. With the current mode of reimbursement, such treatment generates losses to the hospital. CONCLUSION: This example raises the question of financial support for innovative treatments and procedures.