Object The authors conducted a study to compare the sensitivity and specificity of helical CT angiography (CTA) and digital subtraction angiography (DSA) in detecting intracranial arterial injuries after penetrating traumatic brain injury (PTBI). Methods In a retrospective evaluation of 48 sets of angiograms from 45 consecutive patients with PTBI, 3 readers unaware of the DSA findings reviewed the CTA images to determine the presence or absence of arterial injuries. A fourth reader reviewed all the disagreements and decided among the 3 interpretations. Sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV) of CTA were calculated on a per-injury basis and in a subpopulation of patients with traumatic intracranial aneurysms (TICAs). Results Sensitivity of CTA for detecting arterial injuries was 72.7% (95% CI 49.8%–89.3%); specificity, 93.5% (95% CI 78.6%–99.2%); PPV, 88.9% (95% CI 65.3%–98.6%); and NPV, 82.9% (95% CI 66.4%–93.4%). All 7 TICAs were correctly identified by CTA. Sensitivity, specificity, PPV, and NPV of CTA in detecting TICAs were 100%. To compare agreement with DSA, the standard of reference, confidence scores categorized as low, intermediate, and high probability yielded an overall effectiveness of 77.8% (95% CI 71.8%–82.9%). Conclusions Computed tomography angiography had limited overall sensitivity in detecting arterial injuries in patients with PTBI. However, it was accurate in identifying TICAs, a subgroup of injuries usually managed by either surgical or endovascular approaches, and non-TICA injuries involving the first-order branches of intracranial arteries.

The synthesis and characterization (X-ray crystallography, UV/vis spectroscopy, electrochemistry, ESI-MS, and (1)H, (13)C, and (59)Co NMR) of the complexes [Co(L)(O(2)CO)]ClO(4)xH(2)O (L = tpa (tpa = tris(2-pyridylmethyl)amine) (x = 1), pmea (pmea = bis((2-pyridyl)methyl)-2-((2-pyridyl)ethyl)amine) (x = 0), pmap (pmap = bis(2-(2-pyridyl)ethyl)(2-pyridylmethyl)amine) (x = 0), tepa (tepa = tris(2-(2-pyridyl)ethyl)amine) (x = 0)) which contain tripodal tetradentate pyridyl ligands and chelated carbonate ligands are reported. The complexes display different colors in both the solid state and solution, which can be rationalized in terms of the different ligand fields exerted by the tripodal ligands. Electrochemical data show that [Co(tepa)(O(2)CO)](+) is the easiest of the four complexes to reduce, and the variation in E(red.) values across the series of complexes can also be explained in terms of the different ligand fields exerted by the tripodal ligands, as can the (59)Co NMR data which show a chemical shift range of over 2000 ppm for the four complexes. [Co(pmea)(O(2)CO)](+) is fluxional in aqueous solution, and VT NMR spectroscopy ((1)H and (13)C) in DMF-d(7) (DMF = dimethylformamide) over the temperature range -25.0 to 75.0 degrees C are consistent with inversion of the unique six-membered chelate ring. This process shows a substantial activation barrier (DeltaG(#) = 58 kJ mol(-1)). The crystal structures of [Co(tpa)(O(2)CO)]ClO(4)xH(2)O, [Co(pmea)(O(2)CO)]ClO(4).3H(2)O, [Co(pmap)(O(2)CO)]ClO(4), and [Co(tepa)(O(2)CO)]ClO(4) are reported, and the complexes containing the asymmetric tripodal ligands pmea and pmap both crystallize as the 6-isomer. The carbonate complexes all show remarkable stability in 6 M HCl solution, with [Co(pmap)(O(2)CO)](+) showing essentially no change in its UV/vis spectrum over 4 h in this medium. The chelated bicarbonate complexes [Co(pmea)(O(2)COH)]ZnCl(4), [Co(pmap)(O(2)COH)][Co(pmap)(O(2)CO)](ClO(4))(3), [Co(pmap)(O(2)COH)]ZnCl(4)xH(2)O, and [Co(pmap(O(2)COH)]ZnBr(4)x2H(2)O can be isolated from acidic aqueous solution, and the crystal structure of [Co(pmap)(O(2)COH)]ZnCl(4)x3H(2)O is reported. The stability of the carbonate complexes in acid is explained by analysis of the crystallographic data for these, and other slow to hydrolyze chelated carbonate complexes, which show that the endo (coordinated) oxygen atoms are significantly hindered by atoms on the ancillary ligands, in contrast to complexes such as [Co(L)(O(2)CO)](+) (L = (NH(3))(4), (en)(2), tren, and nta), which undergo rapid acid hydrolysis and which show no such steric hindrance.

Density functional calculations have been carried out on the experimentally characterized Co(III) [Co(N4)(O2CO)]+ carbonate complexes containing a tripodal tetraamine ligand (N4 = tpa, Metpa, Me2tpa, Me3tpa, pmea, pmap, tepa) and also the model [Co(NH3)4(O2CO)]+ system. Calculations on the model species, performed using both gas-phase and solvent-corrected procedures, have revealed that the inclusion of a condensed-phase environment is necessary to obtain generally satisfactory results for the structural and bonding properties in these systems. Using the solvent-corrected approach, the observed trends in structural parameters for the metal-ligand bonds, 59Co chemical shifts, and changes in visible absorption wavelengths have been satisfactorily reproduced for the [Co(N4)(O2CO)]+ complexes. A time-dependent density functional analysis of the electronic excitations indicates that the overall composition and character of the relevant (d-d) transitions remain similar throughout the series, indicating that the changes in the Co-N interactions, associated with the structural variations occurring as the N-donor ligand identity and size change, appear most likely responsible for the particular spectroscopic features displayed by these species. These observations are further supported by molecular orbital and energy decomposition analyses. The results from the present calculations confirm recent findings that the inclusion of a treatment for solvent effects plays a critical role in the computational modelling of coordination complexes involving mixed (anionic and neutral) ligands.

No abstract