Rahul Raj

Predicting six-month mortality of patients with traumatic brain injury: usefulness of common intensive care severity scores

IntroductionThe aim of this study was to evaluate the usefulness of the APACHE II (Acute Physiology and Chronic Health Evaluation II), SAPS II (Simplified Acute Physiology Score II) and SOFA (Sequential Organ Failure Assessment) scores compared to simpler models based on age and Glasgow Coma Scale (GCS) in predicting long-term outcome of patients with moderate-to-severe traumatic brain injury (TBI) treated in the intensive care unit (ICU).MethodsA national ICU database was screened for eligible TBI patients (age over 15xa0years, GCS 3–13) admitted in 2003–2012. Logistic regression was used for customization of APACHE II, SAPS II and SOFA score-based models for six-month mortality prediction. These models were compared to an adjusted SOFA-based model (including age) and a reference model (age and GCS). Internal validation was performed by a randomized split-sample technique. Prognostic performance was determined by assessing discrimination, calibration and precision.ResultsIn total, 1,625 patients were included. The overall six-month mortality was 33%. The APACHE II and SAPS II-based models showed good discrimination (area under the curve (AUC) 0.79, 95% confidence interval (CI) 0.75 to 0.82; and 0.80, 95% CI 0.77 to 0.83, respectively), calibration (Pu2009>u20090.05) and precision (Brier score 0.166 to 0.167). The SOFA-based model showed poor discrimination (AUC 0.68, 95% CI 0.64 to 0.72) and precision (Brier score 0.201) but good calibration (Pu2009>u20090.05). The AUC of the SOFA-based model was significantly improved after the insertion of age and GCS (∆AUC +0.11, Pu2009<u20090.001). The performance of the reference model was comparable to the APACHE II and SAPS II in terms of discrimination (AUC 0.77; compared to APACHE II, ΔAUC −0.02, Pu2009=u20090.425; compared to SAPS II, ΔAUC −0.03, Pu2009=u20090.218), calibration (Pu2009>u20090.05) and precision (Brier score 0.181).ConclusionsA simple prognostic model, based only on age and GCS, displayed a fairly good prognostic performance in predicting six-month mortality of ICU-treated patients with TBI. The use of the more complex scoring systems APACHE II, SAPS II and SOFA added little to the prognostic performance.

Hyperoxemia and long-term outcome after traumatic brain injury

IntroductionThe relationship between hyperoxemia and outcome in patients with traumatic brain injury (TBI) is controversial. We sought to investigate the independent relationship between hyperoxemia and long-term mortality in patients with moderate-to-severe traumatic brain injury.MethodsThe Finnish Intensive Care Consortium database was screened for mechanically ventilated patients with a moderate-to-severe TBI. Patients were categorized, according to the highest measured alveolar-arterial O2 gradient or the lowest measured PaO2 value during the first 24 hours of ICU admission, to hypoxemia (<10.0 kPa), normoxemia (10.0 to 13.3 kPa) and hyperoxemia (>13.3 kPa). We adjusted for markers of illness severity to evaluate the independent relationship between hyperoxemia and 6-month mortality.ResultsA total of 1,116 patients were included in the study, of which 16% (nu2009=u2009174) were hypoxemic, 51% (nu2009=u2009567) normoxemic and 33% (nu2009=u2009375) hyperoxemic. The total 6-month mortality was 39% (nu2009=u2009435). A significant association between hyperoxemia and a decreased risk of mortality was found in univariate analysis (Pu2009=u20090.012). However, after adjusting for markers of illness severity in a multivariate logistic regression model hyperoxemia showed no independent relationship with 6-month mortality (hyperoxemia vs. normoxemia OR 0.88, 95% CI 0. 63 to 1.22, Pu2009=u20090.43; hyperoxemia vs. hypoxemia OR 0.97, 95% CI 0.63 to 1.50, Pu2009=u20090.90).ConclusionHyperoxemia in the first 24 hours of ICU admission after a moderate-to-severe TBI is not predictive of 6-month mortality.

Factors correlating with delayed trauma center admission following traumatic brain injury

BackgroundDelayed admission to appropriate care has been shown increase mortality following traumatic brain injury (TBI). We investigated factors associated with delayed admission to a hospital with neurosurgical expertise in a cohort of TBI patients in the intensive care unit (ICU).MethodsA retrospective analysis of all TBI patients treated in the ICUs of Helsinki University Central Hospital was carried out from 1.1.2009 to 31.12.2010. Patients were categorized into two groups: direct admission and delayed admission. Patients in the delayed admission group were initially transported to a local hospital without neurosurgical expertise before inter-transfer to the designated hospital. Multivariate logistic regression was utilized to identify pre-hospital factors associated with delayed admission.ResultsOf 431 included patients 65% of patients were in the direct admission groups and 35% in the delayed admission groups (median time to admission 1:07h, IQR 0:52–1:28 vs. 4:06h, IQR 2:53–5:43, p <0.001). In multivariate analysis factors increasing the likelihood of delayed admission were (OR, 95% CI): male gender (3.82, 1.60-9.13), incident at public place compared to home (0.26, 0.11-0.61), high energy trauma (0.05, 0.01-0.28), pre-hospital physician consultation (0.15, 0.06-0.39) or presence (0.08, 0.03-0.22), hypotension (0.09, 0.01-0.93), major extra cranial injury (0.17, 0.05-0.55), abnormal pupillary light reflex (0.26, 0.09-0.73) and severe alcohol intoxication (12.44, 2.14-72.38). A significant larger proportion of patients in the delayed admission group required acute craniotomy for mass lesion when admitted to the neurosurgical hospital (57%, 21%, p< 0.001). No significant difference in 6-month mortality was noted between the groups (p= 0.814).ConclusionDelayed trauma center admission following TBI is common. Factors increasing likelihood of this were: male gender, incident at public place compared to home, low energy trauma, absence of pre-hospital physician involvement, stable blood pressure, no major extra cranial injuries, normal pupillary light reflex and severe alcohol intoxication. Focused educational efforts and access to physician consultation may help expedite access to appropriate care in TBI patients.

Temporal trends in cardiac arrest incidence and outcome in Finnish intensive care units from 2003 to 2013

PurposeTo estimate temporal trends in incidence and hospital mortality after cardiac arrest in Finnish intensive care units.MethodsUsing a large nationwide intensive care unit (ICU) database we identified patients suffering from cardiac arrest following ICU admission (ICU-CA) during the study period (2003–2013). ICU-CA was defined as need for cardiopulmonary resuscitation and/or defibrillation (non-arrest cardioversions were excluded) according to the Therapeutic Intervention Scoring System-76. Patients admitted with an admission diagnosis of cardiac arrest were excluded. We determined crude incidence and risk-adjusted hospital mortality (based on a customized severity of illness model) for all ICU-CA patients, and for predefined admission diagnosis subgroups. Temporal trends for the observed period were calculated for crude incidence and risk-adjusted hospital mortality.ResultsCrude incidence for all ICU-CA patients was 29/1,000 ICU admissions, with the highest incidence 118/1,000 in the non-operative cardiovascular subgroup. Overall hospital mortality for ICU-CA patients was 55.5xa0% [95xa0% confidence interval (CI) 54–57xa0%]. Hospital mortality was 53.1xa0% (95xa0% CI 50.4–55.8xa0%) for non-operative cardiovascular ICU-CA patients, 32.9xa0% (95xa0% CI 26.9–38.9xa0%) for post cardiac surgery ICU-CA patients, and 56.3xa0% (95xa0% CI 51.2–61.3xa0%) for neurological/neurosurgical ICU-CA patients. There was a significant reduction in the overall ICU-CA incidence and in the risk-adjusted hospital mortality of ICU-CA and non-cardiac arrest cases (non-CA) over the observed study period (pxa0<xa00.001).ConclusionOur data suggest that the incidence of ICU-CA has decreased in Finnish ICUs between 2003 and 2013. Similar reduction in hospital mortality over time was observed for both ICU-CA and non-CA populations.

Delayed Migration of Fractured K-wire Causing Vertebral Artery Invagination After Anterior Atlantoaxial Fixation: A Case Report

BACKGROUNDnMost of the physicians attention during spinal surgery, when using wires and screws, is toward the avoidance of injuries of critical structures (nerves and vessels). When such wires are broken during surgery, the most important point is to take them out safely or, if it is impossible, to leaf them in secure place and follow the patient closely. Migrations of broken Kirschner wire (K-wire) are well known in literature; however, to the best of our knowledge, migration of a fractured K-wire during anterior atlantoaxial fixation of cervical spine has not been reported in the literature.nnnCASE DESCRIPTIONnWe report a case in which a fractured K-wire was imbedded in the lateral mass of C1 for 3 years and then migrated to endanger the dominant right vertebral artery. By using posterior approach and drilling right part of posterior arch of C1, we manage to secure the vertebral artery. The broken K-wire was extracted successfully. In our case, with optimal follow-up, the burred wire inside hard bone was moved in delayed fashion to come out of the bone, grooving the dominant vertebral artery.nnnCONCLUSIONSnOur recommendation is to inspect the K-wire before using it and to try retrieve as much as possible when removing it.

The incidence and outcome of in-ICU cardiac arrest.

Dear Editor, We would like to thank Eastwood and Bellomo [1] for their interest in our paper on intensive care unit cardiac arrest (ICU-CA) in Finland between 2003 and 2013 [2]. We agree that ICU-CA incidence varies greatly. In a recent systematic review we found incidence rates to vary between 6 and 78/1,000 ICU admissions [3]. We think that the most important reason for this, apart from obvious differences in end-of-life care and allocation of ICU beds, is indeed the definition for cardiac arrest used in the different studies. In our study we defined cardiac arrest cases on the basis of the TISS item ‘‘cardiac arrest and/or countershock within past 48 h’’, according to which even brief arrests that required only defibrillation would result in a positive TISS item. A more commonly used definition ‘‘the need for chest compressions’’ might exclude brief episodes of ventricular fibrillation or pulseless ventricular tachycardia treated with defibrillation only. Additionally cardiac arrest occurring in the areas where all expertise and resources needed for cardiac arrest management are readily available would not necessarily lead to rapid response team activation. The noted decrease in ICU-CA incidence and ICU-CA survival is in line with studies showing a decrease in in-hospital cardiac arrest incidence and improved survival overall [4]. We agree that these changes are multifactorial. Our findings should, as we have stated in our paper, be tested in other health care settings. Finally we think that adult ICU-CA has been somewhat neglected by the scientific community and there is a clear need for prospective studies looking at the etiology, management and outcome of ICU-CA.