OBJECTIVEModern surgical planning and prognostication requires the most accurate outcomes data to practice evidence-based medicine. For clinicians treating children following traumatic brain injury (TBI) these data are severely lacking. The first aim of this study was to assess published CT classification systems in the authors’ pediatric cohort. A pediatric-specific machine-learning algorithm called an artificial neural network (ANN) was then created that robustly outperformed traditional CT classification systems in predicting TBI outcomes in children.METHODSThe clinical records of children under the age of 18 who suffered a TBI and underwent head CT within 24 hours after TBI (n = 565) were retrospectively reviewed.RESULTS“Favorable” outcome (alive with Glasgow Outcome Scale [GOS] score ≥ 4 at 6 months postinjury, n = 533) and “unfavorable” outcome (death at 6 months or GOS score ≤ 3 at 6 months postinjury, n = 32) were used as the primary outcomes. The area under the receiver operating characteristic (ROC) curve (AUC) was used to delineate the strength of each CT grading system in predicting survival (Helsinki, 0.814; Rotterdam, 0.838; and Marshall, 0.781). The AUC for CT score in predicting GOS score ≤ 3, a measure of overall functionality, was similarly predictive (Helsinki, 0.717; Rotterdam, 0.748; and Marshall, 0.663). An ANN was then constructed that was able to predict 6-month outcomes with profound accuracy (AUC = 0.9462 ± 0.0422).CONCLUSIONSThis study showed that machine-learning can be leveraged to more accurately predict TBI outcomes in children.

Using human autoimmune sera as molecular probes, we previously described the association of phosphorylated serine/arginine splicing factors (SR splicing factors) with the U1-small nuclear ribonucleoprotein (U1-snRNP) and U3-small nucleolar RNP (snoRNP) in apoptotic cells. SR proteins are highly conserved autoantigens whose activity is tightly regulated by reversible phosphorylation of serine residues by at least eight different SR protein kinase kinases (SRPKs), including SRPK1, SRPK2, and the scleroderma autoantigen topoisomerase I. In this report, we demonstrate that only one of the known SRPKs, SRPK1, is associated with the U1-snRNP autoantigen complex in healthy and apoptotic cells. SRPK1 is activated early during apoptosis, followed by caspase-mediated proteolytic inactivation at later time points. SRPKs are cleaved in vivo after multiple apoptotic stimuli, and cleavage can be inhibited by overexpression of bcl-2 and bcl-xL, and by exposure to soluble peptide caspase inhibitors. Incubation of recombinant caspases with in vitro–translated SRPKs demonstrates that SRPK1 and SRPK2 are in vitro substrates for caspases-8 and -9, respectively. In contrast, topoisomerase I is cleaved by downstream caspases (-3 and -6). Since each of these SRPKs sits at a distinct checkpoint in the caspase cascade, SRPKs may serve an important role in signaling pathways governing apoptosis, alternative mRNA splicing, SR protein trafficking, RNA stability, and possibly the generation of autoantibodies directed against splicing factors.

Proteins cleaved by apoptotic caspases are commonly recognized by autoantibodies found in the serum of patients with rheumatic disease. We report that the 72-kDa signal recognition particle (SRP) protein, a rare target of autoantibodies found in the serum of patients with dermatomyositis and systemic lupus erythematosus, is rapidly cleaved in Jurkat T cells treated with apoptotic (i.e. Fas ligation, treatment with ␥ or ultraviolet radiation, or co-culture with anisomycin or staurosporine) but not proliferative (CD3 cross-linking) stimuli. Cleavage of SRP 72 produces a 66-kDa amino-terminal fragment and a 6-kDa carboxyl-terminal fragment that is selectively phosphorylated on serine residues. Cleavage of SRP 72 is prevented by chemical and peptide caspase inhibitors, and by overexpression of bcl-2, an inhibitor of apoptotic cell death. Analysis of the carboxyl terminus of SRP 72 has identified a putative cleavage site (SELD/A) for group III caspases, and carboxyl-terminal serine residues that are highly conserved in phylogeny. Both serine phosphorylation and caspase cleavage of SRP 72 are observed in cells derived from human, dog, rat, and mouse. Canine SRP 72 is cleaved in vitro by recombinant caspase 3 but retains the ability to mediate transport of a signal peptide-containing protein into the endoplasmic reticulum lumen. The 72-kDa component of the SRP joins a growing list of autoantigens that undergo post-translational modifications during programmed cell death.Proteins modified by the proteases and kinases that are activated during apoptosis are often involved in both the execution phase of cell death and in the development of autoantibodies in patients with systemic lupus erythematosus and mixed connective tissue disease (reviewed in Ref. 1). For example, at least 17 proteins that are known to be cleaved by caspases during apoptosis are autoantigens, including the 70-kDa component of the U1-small nuclear ribonuclear protein complex (U1-70 kDa) (2), poly(A) ribose polymerase (3), DNAdependent protein kinase (DNA-PK) (4), hnRNP C1 and C2 (5), lamins A, B, and C (6), the nuclear mitotic apparatus protein (NuMA) (7, 8), topoisomerases 1 and 2 (8), the nucleolar protein UBF/NOR-90 (8), and ␣-fodrin (9, 10) (reviewed in Ref. 1). In addition, phosphorylated serine/arginine splicing factors have recently been shown to specifically associate with the U1-small nuclear RNP autoantigen complex during apoptosis (11, 12). These observations have led to the hypothesis that proteins modified during apoptosis can be presented to the immune system in a way that bypasses tolerance to self proteins. Although the molecular mechanisms by which this occurs are not known, the data suggests that patient-derived autoantisera may be useful in the identification of proteins that contribute to the execution phase of apoptosis.While screening a panel of human autoantisera for their ability to precipitate new phosphoproteins from apoptotic Jurkat cell lysates, we serendipitously identified several sera that precipitated phosphoproteins fr...

Objectives: With decreasing mortality in PICUs, a growing number of survivors experience long-lasting physical impairments. Early physical rehabilitation and mobilization during critical illness are safe and feasible, but little is known about the prevalence in PICUs. We aimed to evaluate the prevalence of rehabilitation for critically ill children and associated barriers. Design: National 2-day point prevalence study. Setting: Eighty-two PICUs in 65 hospitals across the United States. Patients: All patients admitted to a participating PICU for greater than or equal to 72 hours on each point prevalence day. Interventions: None. Measurements and Main Results: The primary outcome was prevalence of physical therapy– or occupational therapy–provided mobility on the study days. PICUs also prospectively collected timing of initial rehabilitation team consultation, clinical and patient mobility data, potential mobility–associated safety events, and barriers to mobility. The point prevalence of physical therapy– or occupational therapy–provided mobility during 1,769 patient-days was 35% and associated with older age (adjusted odds ratio for 13–17 vs < 3 yr, 2.1; 95% CI, 1.5–3.1) and male gender (adjusted odds ratio for females, 0.76; 95% CI, 0.61–0.95). Patients with higher baseline function (Pediatric Cerebral Performance Category, ≤ 2 vs > 2) less often had rehabilitation consultation within the first 72 hours (27% vs 38%; p < 0.001). Patients were completely immobile on 19% of patient-days. A potential safety event occurred in only 4% of 4,700 mobility sessions, most commonly a transient change in vital signs. Out-of-bed mobility was negatively associated with the presence of an endotracheal tube (adjusted odds ratio, 0.13; 95% CI, 0.1–0.2) and urinary catheter (adjusted odds ratio, 0.28; 95% CI, 0.1–0.6). Positive associations included family presence in children less than 3 years old (adjusted odds ratio, 4.55; 95% CI, 3.1–6.6). Conclusions: Younger children, females, and patients with higher baseline function less commonly receive rehabilitation in U.S. PICUs, and early rehabilitation consultation is infrequent. These findings highlight the need for systematic design of rehabilitation interventions for all critically ill children at risk of functional impairments.