To the Editor A recently published randomized clinical trial 1 studied the effect of high-flow oxygen vs conventional oxygen on invasive mechanical ventilation and clinical recovery in patients with severe COVID-19. It is interesting to contrast the results of this study with those of a large randomized clinical trial posted as a preprint, 2 which reported similar intubation rates in the high-flow oxygen and conventional oxygen groups and similar implementation of awake prone positioning in both groups. We note that in the JAMA trial, 1 when compared with the conventional oxygen group, a larger percentage of patients in the high-flow oxygen group underwent prone positioning (75% vs 64%), and they had a longer total duration of prone positioning (median of 21 hours vs 18 hours, respectively).Prone positioning while awake has been reported to reduce intubation rates for patients with acute hypoxemic respiratory failure due to COVID-19 who are supported with highflow oxygen therapy. 3 Moreover, patients receiving high-flow oxygen therapy who underwent prone positioning for a duration of 8 hours per day or longer were less likely to require intubation than patients who underwent prone positioning for a shorter duration. 3 To better delineate the benefit derived from high-flow oxygen therapy instead of its benefit in combination with awake prone positioning, we hope that the authors can provide data about the cumulative intubation rate in the following subgroups: (1) patients who received any amount of awake prone positioning vs those who did not and (2) patients who received awake prone positioning for at least 8 hours per day vs those who underwent prone positioning for less than 8 hours per day.

BACKGROUND This study aimed to determine the critical threshold of systolic blood pressure (SBP) for aortic occlusion (AO) in severely injured patients with profound hemorrhagic shock. METHODS All adult patients (>15 years) undergoing AO via resuscitative endovascular balloon occlusion of the aorta (REBOA) or thoracotomy with aortic cross clamping (TACC) between 2014 and 2018 at level I trauma center were included. Patients who required cardiopulmonary resuscitation in the prehospital setting were excluded. A logistic regression analysis based on mechanism of injury, age, Injury Severity Score, REBOA/TACC, and SBP on admission was done. RESULTS A total of 107 patients underwent AO. In 57, TACC was performed, and in 50, REBOA was performed. Sixty patients who underwent AO developed traumatic cardiac arrest (TCA), and 47 did not (no TCA). Penetrating trauma was more prevalent in the TCA group (TCA, 90% vs. no TCA, 74%; p < 0.05) but did not modify 24-hour mortality (odds ratio, 0.51; 95% confidence interval, 0.13–2.00; p = 0.337). Overall, 24-hour mortality was 47% (50) and 52% (56) for 28-day mortality. When the SBP reached 60 mm Hg, the predicted mortality at 24 hours was more than 50% and a SBP lower than 70 mm Hg was also associated with an increased of probability of cardiac arrest. CONCLUSION Systolic blood pressure of 60 mm Hg appears to be the optimal value upon which AO must be performed immediately to prevent the probability of death (>50%). However, values of SBP less than 70 mm Hg also increase the probability of cardiac arrest. LEVEL OF EVIDENCE Therapeutic study, level IV.

Background The goal of our study was to evaluate the differences in care and clinical outcomes of patients with chest trauma between two hospitals, including one public trauma center (Pu-TC) and one private trauma center (Pri-TC). Methods Patients with thoracic trauma admitted from January 2012 to December 2018 at two level I trauma centers (Pu-TC: Hospital Universitario del Valle, Pri-TC: Fundacio ´n Valle del Lili) in Cali, Colombia, were included. Multivariable logistic regression was used to assess for differences in in-hospital mortality, adjusting for relevant demographic and clinical characteristics. Results A total of 482 patients were identified; 300 (62.2%) at the Pri-TC and 182 (37.8%) at the Pu-TC. Median age was 27 years (IQR 21-36) and median Injury Severity Score was 25 . 456 patients (94.6%) were male, and the majority had penetrating trauma [total 465 (96.5%); Pri-TC 287 (95.7%), Pu-TC 179 (98.4%), p 0.08]. All patients arrived at the emergency room with unstable hemodynamics. There were no statistically significant differences in post-operative complications, including retained hemothorax [Pri-TC 19 vs. Pu-TC 18], pneumonia [Pri-TC 14 vs. Pu-TC 14], empyema [Pri-TC 13 vs. Pu-TC 13] and mediastinitis [Pri-TC 6 vs. Pu-TC 2]. Logistic regression did, however, show a higher odds of mortality when patients were treated at the Pu-TC , p \ 0.001]. Conclusions Our study found significant statistical differences in clinical outcomes between patients treated at a Pu-TC and Pri-TC. The results are intended to stimulate discussions to better understand reasons for outcome variability and ways to reduce it.

BackgroundDengue fever, a major public health problem throughout tropical and subtropical regions, has often unpredictable clinical evolution and outcomes. Thrombocytopenia is a common laboratory finding in dengue fever and severe dengue during the dengue critical phase. To the best of our knowledge, there is no clinical data about patient and disease factors that could predict in a short time the platelet recovery. Mean platelet volume (MPV), a measurement of platelet size, has a strong inverse correlation with platelet count and could indirectly reflect bone marrow activity. The aim of this study was to describe the behavior of MPV during the platelet count nadir and recovery.MethodsAn observational prospective study was conducted. We included patients with confirmed dengue virus infection with SD BIOLINE Dengue Duo kit (Abbott, Santa Clara, USA; former Alere Inc., Waltham, USA) attended at Fundación Valle del Lili, Cali - Colombia. Blood count was analyzed by xn-3000 system impedance method (Sysmex, Kobe, Japan). Laboratory and clinical data were recollected from clinical charts and clinical laboratory database. Platelet count (PC) and MPV were measured repeatedly during clinical management. Time was measured from the first blood count. A non-parametric analysis with a cubic smoothing spline was performed for platelet count and MPV.ResultsA total of 54 patients were analyzed from April 2016 to January 2016. 50% of patients had at least three blood counts. The median of the lowest PC was 112,500/L (IQR = 67,000–148,500), and the median of the highest MPV was 11. 25 fL (IQR = 10. 42–12. 15). MPV increased from the first blood count until day six, while platelets presented slight fluctuations. On the sixth day after first blood count, MPV presented a high peak that suggests an inverse relationship with a platelets decrease (Figure 1).ConclusionMPV increased with thrombocytopenia during the critical period and its decline precedes platelet count recovery. MPV could be useful to predict the platelet count recovery. Disclosures All authors: No reported disclosures.

Introduction: Although the peace process in Colombia resulted in a significant reduction in the number of anti-personnel mines across the country, there are no reliable data on the effects of this phenomenon on outcomes for patients who were victims of these devices. Objective: The objective of this study was to assess mortality from landmine injuries before and during the Colombian peace process. Furthermore possible associations between peace negotiations and mortality were explored. Methods: For this study, we used the "Colombian Victims of Antipersonnel Mines Injuries registry" (MAP/MUSE database) data from 2002 to 2018. This registry was launched in 2001 by the government of Colombia with the aim of prospectively and systematically collect information on all the cases of anti-personnel mine injuries in the country. The period between 2002-2012 was classified as the pre-negotiation period (war period), and 2014-2018 as the peace negotiations period, with 2013 classified as a washout year. Multivariate logistic regression was used to explore the association between peace negotiations and mortality among anti-personnel landmine injured individuals. Results: A total of 10306 landmine injury cases were registered. Of these, 1180 (11.4%) occurred in the peace-negotiation period. Mortality was significantly lower during the period of peace negotiations. After adjusting for sex, age group, race, active duty soldier status, rural area, and geographic departments case volumes, the peace negotiation period was found to be associated with lower risk-adjusted odds of mortality after suffering a landmine injury (OR= 0.6, 95% CI, 0.5-0.7; p<0.001). Conclusion: Our findings suggest an association between the period of peace negotiation and a lower likelihood of mortality among victims of anti-personnel landmines.