Prediction of lithium response using clinical data.

OBJECTIVE: Promptly establishing maintenance therapy could reduce morbidity and mortality in patients with bipolar disorder. Using a machine learning approach, we sought to evaluate whether lithium responsiveness (LR) is predictable using clinical markers. METHOD: Our data are the largest existing sample of direct interview-based clinical data from lithium-treated patients (n = 1266, 34.7% responders), collected across seven sites, internationally. We trained a random forest model to classify LR-as defined by the previously validated Alda scale-against 180 clinical predictors. RESULTS: Under appropriate cross-validation procedures, LR was predictable in the pooled sample with an area under the receiver operating characteristic curve of 0.80 (95% CI 0.78-0.82) and a Cohen kappa of 0.46 (0.4-0.51). The model demonstrated a particularly low false-positive rate (specificity 0.91 [0.88-0.92]). Features related to clinical course and the absence of rapid cycling appeared consistently informative. CONCLUSION: Clinical data can inform out-of-sample LR prediction to a potentially clinically relevant degree. Despite the relevance of clinical course and the absence of rapid cycling, there was substantial between-site heterogeneity with respect to feature importance. Future work must focus on improving classification of true positives, better characterizing between- and within-site heterogeneity, and further testing such models on new external datasets.

Anterior pituitary autoantibodies in patients with type 1 diabetes mellitus: methodological problems and clinical correlations.

BACKGROUND: Anti-pituitary antibodies (APA) were described in patients with Type 1 Diabetes (T1D) but their prevalence and relevance remain controversial. MATERIALS AND METHODS: We evaluated the APA prevalence in Sardinian sera from 100 T1D patients, 70 Type 2 Diabetes (T2D) patients and 62 healthy controls, using indirect immunofluorescence on bovine pituitary sections. To compare two different substrates, we tested using bovine sections, further T1D patient sera (n = 11, from Pisa) previously analysed for APA on monkey sections, while some T1D Sardinian patient sera (n = 22) were tested on monkey sections. According to preliminary experiments, positivity were considered ≥1:200 and ≥1:20 for bovine and monkey substrates, respectively. RESULTS AND DISCUSSION: Using bovine sections, APA were detected in 7/100 Sardinian T1D patients (at 1:200 titer) and in none of the other Sardinian sera tested. When the T1D sera from Pisa were tested on bovine and the T1D Sardinian sera were tested on monkey, none of these sera showed corresponding positivity for APA. Pituitary hormone dysfunctions were not found in the 7 APA-positive Sardinian T1D patients. The present study shows that the presence of APA at low-titer is highly related to T1D but not associated with any pituitary dysfunction while the animal species used as substrate appears crucial. CONCLUSION: Further studies are needed to ascertain whether APA detected by different animal species may have different pathological relevance in T1D and/or whether APA in the long run may predict future anterior pituitary dysfunction.

[Neutrophil function in children with acute lyphoblastic leukemia]. / La funzione dei neutrofili nel bambino con leucemia linfoblastica acuta.

Neutrophil function has been studied in 12 patients with acute lymphoblastic leukemia by means of the NBT test and bactericidal assay against E. coli. Tests have been performed before and during treatment. In untreated patients the bactericidal activity of neutrophils was normal, in spite of the decreased number of cells reducing the NBT dye after activation with immuno-complexes. In remission, those patients showed normal reduction. No correlation was observed between the bactericidal power registered after cranial irradiation and the NBT test.