Persistent urinary podocyte loss following preeclampsia may reflect subclinical renal injury.

OBJECTIVE: Studies have shown that podocyturia, i.e., urinary loss of viable podocytes (glomerular epithelial cells), is associated with proteinuria in preeclampsia. We postulated that urinary podocyte loss may persist after preeclamptic pregnancies, thus resulting in renal injury. This may lead to future chronic renal injury. In addition, we compared the postpartum levels of the angiogenic factors, which previously have been associated with preeclampsia, between normotensive versus preeclamptic pregnancies. STUDY DESIGN: The diagnosis of preeclampsia was confirmed using standard clinical criteria. Random blood and urine samples were obtained within 24 hours prior to delivery and 5 to 8 weeks postpartum. Urine sediments were cultured for 24 hours to select for viable cells and staining for podocin was used to identify podocytes. Serum samples were analyzed for the levels of angiogenic markers using ELISA (enzyme-linked immunosorbent assay) methodology. RESULTS: At delivery, preeclamptic patients (n = 10) had significantly higher proteinuria (p = 0.006) and podocyturia (p<0.001) than normotensive pregnant patients (n = 18). Postpartum proteinuria was similar between these two groups (p = 0.37), while podocyturia was present in 3 of 10 women with preeclampsia and in none of the normotensive controls (p = 0.037). Angiogenic marker levels, including placental growth factor, soluble vascular endothelial growth factor receptor fms-like tyrosine kinase receptor-1 and endoglin, were not significantly different between women with preeclampsia and women with a normotensive pregnancy, either at delivery or postpartum. CONCLUSION: Persistent urinary podocyte loss after preeclamptic pregnancies may constitute a marker of ongoing, subclinical renal injury.

Use of computed tomography in the diagnosis of bowel complications after gynecologic surgery.

OBJECTIVE: To determine factors predictive of bowel complications after gynecologic surgery and establish the added utility of computed tomography (CT) in the diagnostic process. METHODS: Patients who underwent gynecologic surgery between January 2, 2008, and December 30, 2010, who had CT scans of the abdomen, pelvis, or abdomen and pelvis within 42 days for a suspected bowel complication were identified. Logistic regression analysis was used to identify factors predictive of bowel-related complications. The diagnostic accuracy of CT was compared among patient risk groups based on clinical suspicion (pretest probability) of bowel complications. RESULTS: Among 205 eligible patients, 38 (18.5%) patients had a bowel-related complication. Mean time from surgery to CT was 12.4 (10.1) days. Clinical characteristics were used to develop a clinical model that included unexpected drainage from the drain, wound, or stoma (adjusted odds ratio [OR] 26.3, 95% confidence interval [CI] 3.1-224.4, P=.003), coronary artery disease (OR 10.7, CI 1.4-80.9, P=.022), laparotomy (compared with minimally invasive surgery) (OR 4.4, CI 1.1-17.2, P=.032), and age older than 45 years (OR 2.4, CI 0.7-8.8, P=.18). Addition of CT to clinical evaluation increased the predictive ability of the model (area under the curve) from 0.73 to 0.99. Among 57 low-risk patients, three with confirmed bowel-related complications would have been missed if CT was not performed. Among 13 high-risk patients, CT sensitivity was 70%, and it was negative for bowel complications in three patients subsequently confirmed to have serious complications (one anastomotic leak, two bowel perforations). CONCLUSIONS: In patients who have undergone gynecologic surgery and have a high clinical probability of a postoperative bowel-related complication, CT alone may fail to accurately identify patients with serious complications. LEVEL OF EVIDENCE: : II.

Analysis of an environmental exposure health questionnaire in a metropolitan minority population utilizing logistic regression and Support Vector Machines.

The goal of this study was to analyze a 54-item instrument for assessment of perception of exposure to environmental contaminants within the context of the built environment, or exposome. This exposome was defined in five domains to include 1) home and hobby, 2) school, 3) community, 4) occupation, and 5) exposure history. Interviews were conducted with child-bearing-age minority women at Metro Nashville General Hospital at Meharry Medical College. Data were analyzed utilizing DTReg software for Support Vector Machine (SVM) modeling followed by an SPSS package for a logistic regression model. The target (outcome) variable of interest was respondent's residence by ZIP code. The results demonstrate that the rank order of important variables with respect to SVM modeling versus traditional logistic regression models is almost identical. This is the first study documenting that SVM analysis has discriminate power for determination of higher-ordered spatial relationships on an environmental exposure history questionnaire.