Statistical adjustments for brain size in volumetric neuroimaging studies: Some practical implications in methods

Volumetric magnetic resonance imaging (MRI) brain data provide a valuable tool for detecting structural differences associated with various neurological and psychiatric disorders. Analysis of such data, however, is not always straightforward, and complications can arise when trying to determine which brain structures are “smaller” or “larger” in light of the high degree of individual variability across the population. Several statistical methods for adjusting for individual differences in overall cranial or brain size have been used in the literature, but critical differences exist between them. Using agreement among those methods as an indication of stronger support of a hypothesis is dangerous given that each requires a different set of assumptions be met. Here we examine the theoretical underpinnings of three of these adjustment methods (proportion, residual, and analysis of covariance) and apply them to a volumetric MRI data set. These three methods used for adjusting for brain size are specific cases of a generalized approach which we propose as a recommended modeling strategy. We assess the level of agreement among methods and provide graphical tools to assist researchers in determining how they differ in the types of relationships they can unmask, and provide a useful method by which researchers may tease out important relationships in volumetric MRI data. We conclude with the recommended procedure involving the use of graphical analyses to help uncover potential relationships the ROI volumes may have with head size and give a generalized modeling strategy by which researchers can make such adjustments that include as special cases the three commonly employed methods mentioned above.     

Treatment outcome and survival in participants of phase I oncology trials carried out from 2003 to 2006 at Institut Gustave Roussy.

BACKGROUND: The oncology community usually perceives phase I oncology trials as associated with poor or limited benefits and substantial risks. There is scarce data concerning outcome and survival of patients enrolled in current phase I oncology trials. PATIENTS AND METHODS: We reviewed all phase I oncology trials conducted by investigators from the Adult Phase I Unit at Institut Gustave Roussy from 2003 to 2006. We report data concerning patient demographics, treatment outcome, toxicity, survival and type of care after trial exit. RESULTS: We analyzed 10 trials involving 180 participants. The overall response rate was 7.2%. Disease control (objective response plus stable disease) was achieved in 48.2% of patients. The rate of toxic death was 0.5%. In all, 38% of patients had at least one episode of grade 3 or 4 toxic events. The median progression-free survival and the median overall survival (OS) were 2.3 and 8.7 months, respectively. On multivariate analysis, a time between diagnosis of disease and inclusion in the phase I trial > or =24 months and evidence of disease control were statistically significant predictors of improved OS. CONCLUSION: Current phase I oncology trials are safe and are associated with clinical benefit in a substantial proportion of patients.     

Ethnic ancestry and increased paternal age are risk factors for breast cancer before the age of 40 years.

To study the risk factors associated with breast cancer in women younger than 40 years, a cohort study (The Jerusalem Perinatal Study) of 42 822 female offspring born in hospitals in West Jerusalem during 1964-1976 was carried out. Hazard ratios of potential parental and perinatal risk factors for early breast cancer were measured. The overall incidence of breast cancer was 5.2/100 000 person-years. The highest incidence was found among Jewish women of West Asian ancestry (8.6/100 000 person-years), specifically those whose maternal grandfathers were born in Iraq, Iran or Afghanistan (9.5/100 000 person-years). Using Cox models we found independent risk factors for early breast cancer to be paternal age (relative risk/year=1.06, 95% confidence interval=1.02-1.10, P=0.005), and ancestry from Iraq/Iran/Afghanistan (relative risk=3.1, 95% confidence interval=1.50-6.52, P=0.002). The study confirms a previously observed effect of advanced paternal age on the occurrence of early breast cancer and identifies a novel population group at increased risk for the disease. The excess risk of early breast cancer associated with ancestry from Iraq, Iran and Afghanistan suggests involvement of genetic determinants, environmental exposures and/or lifestyle factors and mandates further investigation.     

Differentiating malignant from benign ovarian tumors with transvaginal color flow imaging.

The impedance to blood flow was examined by transvaginal color flow imaging in 53 ovarian masses before exploratory laparotomy. Serum CA 125 levels were measured in all subjects. Thirty-six had benign ovarian tumors and 17 had malignant ovarian tumors confirmed by histopathologic examination. Intratumoral blood vessels, detected in 16 of the malignant tumors, consistently demonstrated low impedance to flow, with a pulsatility index (PI) always below 1. The PI of the intraovarian or intratumoral blood vessels was greater than 1 in 35 of the 36 benign tumors, although 11 had suspicious sonographic findings (P less than .01) and 14 had elevated CA 125 levels (P less than .001). The sensitivity and specificity of the preoperative PI in detecting malignant ovarian tumors were 94 and 97%, respectively. The sensitivity and specificity of preoperative suspicious sonographic findings in detecting malignant ovarian tumors were 94 and 69%, and those of elevated preoperative serum CA 125 levels were 82 and 61%, respectively. Our results suggest that transvaginal color flow imaging may be a useful clinical tool in the preoperative evaluation of ovarian masses.     

Meningitis and midline facial deformity

A baby with unilateral cleft lip, midline cleft palate and hypertelorism developed meningitis in the first 48 h of life. Examination of the nasopharynx showed a soft tissue mass, which was confirmed as a basal encephalocele by computed tomography. There was also congenital hydrocephalus and the corpus callosum was absent. Surgical treatment included repair of the anterior basal skull defect, repair of the lip and palate, and ventriculo-peritoneal shunt. There is currently evidence of developmental delay and right-sided visual impairment due to Morning Glory syndrome. This case demonstrates that basal encephalocele should be considered in any baby with midline facial deformity who develops meningitis.     

Isolated congenital tuberculosis otitis in a preterm infant

We report an unusual case of localized congenital tuberculosis otitis in a preterm infant. Unlike disseminated congenital cases, the manifestations of localized otitis are associated with a triad of signs: (i) regional lymphadenopathy in the absence of typical systemic features of tuberculosis; (ii) delayed onset of presentation; and (iii) refractory otitis unresponsive to conventional antimicrobial agents. The need for greater diligence in looking for neonatal tuberculosis is emphasized, especially in an ethnic or socioeconomic environment where the disease is prevalent. Congenital tuberculosis, otitis, preterm
PC Ng, Department of Paediatrics, Level 6, Clinical Sciences Building, Prince of Wales Hospital, Shatin, NT, Hong Kong     

A new automated tumor registry and clinical research system and its application to gynecologic oncology patients

Every physician at one time or another wishes that he could quickly recall, group and stratify elements of his past practice and apply that experience toward solving a current problem. In Oncology attempts to use hospital registries for this purpose are often disappointing, as few are computerized and most lack the detail, accuracy, and flexibility to be truly responsive. The development of our current Gynecologic Oncology Tumor registry began in 1982. The final product results from the cooperation of a computer specialist, a registrar, physicians, and nurses. In addition, three independent programs have been incorporated to expand the statistical and reporting functions of the registry. The result is a tumor registry that is the foundation of a clinical research system.     

Post design and the optimally restored endodontically treated tooth

The basics for the sound restoration of endodontically treated teeth have not changed despite the advent of new materials and techniques. Adequate retention for posts and crowns, resistance to cyclic shearing forces, minimal insertional stresses, and even distribution of functional stresses are prime requirements that cannot be compromised without jeopardizing the longevity of the final restoration.     

Operative classification of thromboembolic disease determines outcome after pulmonary endarterectomy

OBJECTIVE: We sought to determine whether type and location of thromboembolic disease in the pulmonary vascular tree predicts the hemodynamic result and clinical outcome in patients undergoing pulmonary endarterectomy. METHODS: From 1998 to 2000, 202 patients with pulmonary hypertension and pulmonary vascular resistance ranging from 194 to 2950 dynes-s-cm(-5) underwent pulmonary endarterectomy. Preoperative and postoperative tricuspid valve function, pulmonary artery pressure, and pulmonary vascular resistance were determined by means of transthoracic echocardiography and measurements with a Swan-Ganz catheter (Edwards Lifesciences, Irvine, Calif), respectively. Patients underwent intraoperative classification of thromboembolism as follows: type 1 (76 patients), fresh thrombus in the main-lobar pulmonary arteries; type 2 (81 patients), intimal thickening and fibrosis proximal to the segmental arteries; type 3 (38 patients), disease within distal segmental arteries only; and type 4 (7 patients), distal arteriolar vasculopathy without visible thromboembolic disease. RESULTS: Overall perioperative mortality was 4.5% (9/202 patients). By means of univariate analysis, patients with type 3 or 4 disease (distal pulmonary vasculopathy) had more residual postoperative tricuspid regurgitation (P <.0001), higher postoperative pulmonary artery systolic pressure (P <.0001), and greater postoperative pulmonary vascular resistance (P <.0001) compared with that seen in patients with type 1 or 2 disease, in whom thromboembolic disease was more surgically accessible. Factors such as severity of preoperative tricuspid regurgitation, patient age, and circulatory arrest time had no correlation with postoperative hemodynamic improvement. Patients with distal thromboembolic disease (type 3-4) had higher perioperative mortality, required longer inotropic support, and had longer hospital stays compared with patients with type 1 or 2 thromboembolic disease. CONCLUSION: The degree of improvement in pulmonary hypertension and tricuspid regurgitation after pulmonary endarterectomy is determined by the type and location of pulmonary thromboembolic disease. Classification of thromboembolism is useful for predicting patient outcome after pulmonary endarterectomy.     

Fluoxetine's effects on cognitive performance in patients with traumatic brain injury

OBJECTIVE: There are preclinical data showing that fluoxetine stimulated expression of Brain Derived Neurotrophic Factor (BDNF) and its specific tyrosine kinase receptor, and caused neuritic elongation and increased dendritic branching density of CA3 hippocampal pyramidal cell neurons in rodents. The latter effect of fluoxetine has been referred to as neuronal remodeling. In view of this preclinical data, we wondered if specific cognitive measures could serve as novel therapeutic targets for fluoxetine in head-injured patients. Theoretically, fluoxetine-induced "neuronal remodeling" might improve cognition, independently of a primary effect on mood. METHOD: In an open-label pilot investigation, fluoxetine hydrochloride (Prozac; 20-60 mg/day) was administered to a heterogeneous group of five head-injured patients with either no or moderate depression for a period of eight months. These patients had no histories of prior treatment with antidepressant medications. They were administered cognitive and memory tests at baseline and after eight months of treatment on fluoxetine. RESULTS: The preliminary results showed that fluoxetine improved mood, in addition to improving performance on the Trail Making Test Part A, an attentional-motor speed task, and the letter-number sequencing subtest of the WAIS-III, a measure reflecting "working memory." CONCLUSIONS: Although fluoxetine had beneficial effects on some measures of cognition, more work is needed to connect these improvements with neuronal remodeling.