Infection after penetrating brain injury – An Eastern Association for the Surgery of Trauma multicenter study oral presentation at the 32nd annual meeting of the Eastern Association for the Surgery of Trauma, January 15-19, 2019, in Austin, Texas. Harmon LA, Haase DJ, Kufera JA, Adnan S, Cabral D, Lottenberg L, Cunningham K, Bonne S, Burgess J, Rehbein JL, Semon G, Noorbakhsh MR, Cragun BN, Agrawal V, Truitt M, Marcotte J, Goldenberg A, Behbahaninia M, Keric N, Hammer PM, Nahmias J, Grigorian A, Turay D, Chakravarthy V, Lalchandani P, Kim D, Chapin T, Dunn J, Portillo V, Schroeppel T, Stein DM. J Trauma Acute Care Surg. 2019 Jul;87(1):61-67.
Each year, more than a million Americans receive care for traumatic brain injuries, and among those injuries due to penetrating trauma, mortality rates have historically been as high as 85%. The incidence of central nervous system (CNS) infections in the setting of penetrating traumatic brain injury (pTBI) have previously been reported to be up to 25%, even when considering military and civilian injures separately. Additionally, while many providers use of prophylactic antibiotics in pTBI, there is no consensus on the duration of therapy. The purpose of this study was to determine the incidence of infection after pTBI and to determine predictive factors for infection after pTBI.
This study was conducted as a multicenter, retrospective chart review of 16 Level I and 1 Level II trauma centers. Inclusion criteria consisted of adult patients (17 years and older) with a pTBI and survival longer than 72 hours, who were admitted to the various centers between 2006 and 2016. The study included all patients with dural penetration confirmed either by CT imaging or an operative report, and regardless of other organ system injury. Importantly, the effects of retained foreign bodies, surgical intervention (craniotomy or craniectomy), intracerebral pressure (ICP) monitoring, natural orifice involvement, and degree of penetration were analyzed. The primary endpoint was incidence of CNS infections after pTBI. Infection was defined by cultures, imaging, or clinical presentation. Extracalvarial infections were excluded. Also, patients were classified based on the presence of CNS infection and use of prophylactic antibiotics (given in the first 24 hours of admission).
Over an 11-year period, 763 patients were included in this study. Most were young men (mean age 35) and the primary mechanism in 90% of patients was a gunshot wound. For all patients, there was a 7% infection rate. Sixty two percent received prophylactic antibiotics. Sixty percent had extended penetration (>1 cm of intrusion, or involving multiple lobes). Forty percent had a trajectory through the oropharynx, and 63% had a retained foreign body. Forty eight percent underwent surgery and 29% had ICP monitor use. Importantly, there was no difference in the infection rate in comparing patients who did or did not receive antibiotics. Also, there was no significant difference in infection rates based on the degree of dural penetration, oropharyngeal trajectory, and the presence of retained foreign bodies. Patients who underwent operative interventions or had ICP monitoring, however, had a significant increase in the infection rate. A multivariable logistic regression model supported the findings that prophylactic antibiotic use had no impact on the secondary infection rate, and that operative interventions and ICP monitor use in patients were associated with CNS infections. Moreover, those patients with CNS infections had a significantly longer hospitalization than those without infections. Finally, the most common organisms isolated were Staphylococcus and Enterococcus.
The key take away points of this important study are that the rate of infection after pTBI is low at 7%, and there is no reduction in the secondary infection rate with the use of prophylactic antibiotics for those patients not undergoing operative intervention. Also, risk factors for CNS infection following pTBI include operative intervention and ICP monitor use and are not significantly affected by prophylactic
Earlier time to hemostasis is associated with decreased mortality and rate of complications: Results from the Pragmatic Randomized Optimal Platelet and Plasma Ratio Trial. Chang R, Kerby JD, Kalkwarf KJ, Van Belle G, Fox EE, Cotton BA, Cohen MJ, Schreiber MA, Brasel K, Bulger EM, Inaba K, Rizoli S, Podbielski JM, Wade CE, Holcomb JB, and PROPPR Study Group. J Trauma Acute Care Surg. 2019 Aug;87(2):342-349.
Effective management of traumatic hemorrhage involves a balanced blood product resuscitation and timely interventions to achieve hemostasis. Surgeons know that early hemorrhage control reduces physiologic derangements and leads to improved survival and morbidity. Also, recent reports have demonstrated that minimizing time to laparotomy or interventional radiology procedures, and using a balanced blood product resuscitation lead to improved clinical outcomes. This study aims to specifically evaluate the impact of time to hemostasis on the risk of complications and mortality.
This study is a secondary analysis of the Pragmatic Randomized Optimal Platelet and Plasma Ratios (PROPPR) trial which involved the randomization of 680 patients among 12 trauma centers with traumatic hemorrhage potentially requiring massive transfusion to resuscitation strategies of 1:1:1 versus 1:1:2 of plasma to platelets to packed red blood cells from 2012 to 2013. Patients that were included were those who underwent an emergent procedure (in the operating room (OR) or interventional radiology (IR) suite) within 90 minutes of hospital arrival.
Of the original subjects of the PROPPR trial, 468 patients (69%) underwent an emergent procedure and these patients had a significantly higher incidence of severe abdominal injury, lower incidence of severe head injury, and lower incidence of traumatic brain injury-related death compared with patients not requiring an emergent procedure. Hemostasis was achieved in 87% of cases (408 patients), and time to hemostasis was higher in blunt trauma cases when compared to penetrating trauma cases (median, 139 min vs. 119 min, p<0.01) and after IR versus OR interventions (median, 184 min vs. 127 min, p<0.01). Moreover, the time from OR/IR to hemostasis accounted for the majority of the overall time to hemostasis (median, 77%, IQR 63-87%).
Based on time to hemostasis, patients were divided into quartiles or included in a fifth group if hemostasis was not achieved: group 1, <90 minutes; group 2, 90 to 129 minutes; group 3, 130 to 183 minutes; group 4, >183 minutes; group 5, exsanguination. Among these groups, no difference existed in terms of age, admission systolic blood pressure or heart rate, or treatment arm. Importantly, the higher numbered group (with patients requiring longer time to hemostasis) was associated with more derangements with base excess on admission, increased Injury Severity Scores, and a higher incidence of blunt trauma. Also, as the group number increased, there were significant increases in the transfusion of blood products and the incidence of acute kidney injury (AKI), acute respiratory distress syndrome (ARDS), multiple-organ failure (MOF), and sepsis as well as a decrease in ICU-free days. A regression analysis demonstrated that every 15-minute decrease in time to hemostasis was associated with decreases in 30-day mortality and the incidence of AKI, ARDS, MOF, and sepsis but not 24-hour mortality or venous thromboembolic events.
The key takeaway points are that time to hemostasis is a valuable end point for hemorrhage control; it should supplant the previous focus on time from hospital arrival to OR/IR arrival for future traumatic hemorrhage studies since the vast majority of time in achieving hemostasis is spent doing the hemorrhage control procedure itself. Also, decreased time to hemostasis was associated with decreased 30-day mortality and decreased incidences in AKI, ARDS, sepsis, and MOF.
Effect of Low-Dose Supplementation of Arginine Vasopressin on Need for Blood Product Transfusions in Patients with Trauma and Hemorrhagic Shock. Sims CA, Holena D, Kim P, Pascual J, Smith B, Martin N, Seamon M, Shiroff A, Raza S, Kaplan L, Grill E, Zimmerman N, Mason C, Abella B, Reilly P. JAMA Surg. 2019 Nov 1;154(11):994-1003.
Blood product replacement and aggressive resuscitation are the mainstay of the management of hemorrhagic shock; however, “over resuscitation” has serious sequelae including coagulopathy, acute lung injury and abdominal compartment syndrome to name a few. Arginine vasopressin (AVP) is a hormone which has neuroendocrine effects on vasomotor tone, and pituitary stores can be rapidly depleted in hemorrhagic shock potentially leading to refractory hypotension. This study sought to determine the impact of administering vasopressin during the initial resuscitation of patients in hemorrhagic shock. This was a single-center, double-blind, randomized clinical trial at an urban Level 1 trauma center evaluating two groups: one group was randomized to AVP administration during the first 48hrs of resuscitation and one group was randomized to placebo. The primary endpoint was the cumulative blood volume transfused within 48hrs. Secondary endpoints included total volume of crystalloid, estimated blood loss, overall fluid balance, and total vasopressor requirement within the first 48hrs. Their power analysis estimated they would need 50 patients per group to find a 50% reduction in the amount of cumulative blood products transfused. The study was not powered for the secondary endpoints or post-hoc analysis. One hundred patients were enrolled from May 2013 to May 2017 and were eligible for inclusion. They performed an Intention-To-Treat and Per-Protocol analysis. In both analyses, in the AVP treated group versus the non-AVP treated group, the use of total blood products (median, 1.4L vs. 2.9L, p=0.01), FFP (median, 0.9L vs. 1.0L, p=0.03), platelets (median, 200ml vs. 300ml, p = 0.02) and cryoprecipitate (median, 12.6ml vs. 34.7ml, p = 0.04) were significantly lower in the AVP treated group. In the Per-Protocol group, the same results were found, with the addition that statistically fewer PRBCs were used in the AVP group versus the non-AVP group (median, 0.6L vs. 1.2L, p = 0.02). Of the secondary endpoints, the AVP treated group had a better fluid balance (median, 5.0L vs. 6.7L, p =0.03) and increased urine output at the end of 48hrs (median, 5.0L vs. 6.7L, p =0.03). There was no difference in total vasopressor requirements (norepinephrine equivalents, median, 581µg vs. 1536µg, p=0.40) or the total volume of crystalloid administered (median, 9.7L vs. 10.7L, p=0.25) Complications (including ARDS and acute kidney injury) were common in both groups and the incidence of any complication was not significantly different between the two groups (29 (59%) vs. 34 (67%), p=0.44). There was no significant difference between groups for median hospital length of stay (16 days vs. 22days), ICU length of stay (5 days vs. 9 days) or mortality (12% vs. 12%). Post-hoc analysis found significantly fewer DVTs in the AVP group (20% vs. 39%, p=0.05).
This is a well-designed and implemented study. It demonstrated that administration of AVP during initial resuscitation does reduce the need for blood products. There was no significant difference in complication rates or mortality, however, the study was not powered to detect these differences and therefore, care must be taken when interpreting the secondary endpoints, including the post-hoc finding of reduced DVTs in the AVP group. The use of AVP in the initial resuscitation of hemorrhagic shock demonstrates the potential to improve outcomes and more effectively utilize resources, and therefore, warrants further investigation.
Lead toxicity from retained bullet fragments: a systematic review and meta-analysis. Apte A, Bradford K, Dente C, Smith RN. J Trauma Acute Care Surg. 2019 Sep;87(3):707-716.
Firearm injury is a public health epidemic, with over 100,000 non-fatal firearm injuries in 2016 alone. Many of these patients have retained bullets or fragments, and this study sought to “systematically examine and aggregate existing studies” to gain a better understanding of the sequelae of retained bullets. The authors found over 2,000 articles concerning this subject, but after removing duplicates and screening each one, they found only 12 articles over 30 years (1988 to 2018) meeting criteria for inclusion (RCTs, prospective or retrospective cohorts, case-control and cross-sectional observational studies). Of these 12, only five were included for meta-analysis. There was a significant level of heterogeneity of the studies (I2 of 74%), leading them to use a random effects model. In their pooled sample of 462 patients, they found significantly higher blood lead levels (BLL) in those with retained bullets versus those without (3.70 vs 7.24 ug/dL, p<0.01). In the systematic review, they focused on four main things: location, duration, number of fragments and symptoms. Two of these were independent risk factors for elevated BLL: the number of retained bullets/fragments and the presence of bony fracture. The duration of retention demonstrated conflicting results, and this remains poorly understood though some data suggest that levels may peak at 3 months and steadily decrease thereafter. An intra-articular location of the bullet did not tend to relate to blood lead levels, which as the authors point out is an important take-away since the only truly accepted indications for removal have traditionally been intra-articular location or when exposed to cerebrospinal fluid (CSF). It has traditionally been thought that exposure to synovial fluid or CSF might lead to quicker absorption of lead, though these data seem to refute that. Lead toxicity symptoms are vague and therefore, difficult to capture. Only two studies reported positive symptoms. One study reported gastrointestinal and neurological disturbances, and another memory loss, irritancy and bad mood all indicative of psychological disturbances.
This was an excellent systematic review, collating a large amount of information into one succinct review and meta-analysis. The authors point out that one of the weaknesses of the study is the lack of randomized trials evaluating retained bullets, thus limiting the conclusions that can be drawn, and highlighting the great need to study the long-term effects of fire-arm injury.