July 2021

Synopsis: There is no approved pharmacological treatment for acute respiratory distress syndrome (ARDS). The efficacy and role of corticosteroids in the treatment of ARDS remains unclear.

The DEXA-ARDS study was a multicenter, randomized, controlled trial across 17 intensive care units (ICUs) in Spain that enrolled 277 patients between March 2013 and December 2018 with established moderate-severe ARDS (defined as P:F ratio ≤ 200mmHg with PEEP ≥ 10 cmH2O and FiO2 ≥ 0.5 at 24 h after ARDS onset). Patients in the dexamethasone group received 20 mg IV daily from day 1-5 and 10mg IV daily from day 6-10. The treatment was continued for a maximum of 10 days or until extubation (if before day 10). Both groups were treated with lung protective ventilation and other standard ARDS care. The two most common causes of ARDS were pneumonia (53%) and sepsis (25%) across each group. The majority of patients in both groups had moderate ARDS at enrollment (dexamethasone N=118/139, control N=121/138). Due to low enrollment rate, the trial was stopped after enrolling 88% (277/314) of the planned sample size. 

The primary study outcome was ventilator-free days at 28 days and the secondary outcome was all-cause mortality at 60 days. The mean number of ventilator-free days was higher in the dexamethasone group (between-group difference 4.8 days (95% CI 2.57-7.03); p<0.0001). The mortality at 60 days was significantly lower in the dexamethasone group (21%) compared to the control group (36%) (between-group difference -15.3% (95% CI -25.9- -4.9); p=0.0047). The use of neuromuscular blockade, recruitment maneuvers and extracorporeal lung support did not differ between the two groups. However, patients in the control group were more likely to be treated with prone positioning (between-groups difference 10.3% (95% CI 4.0%-20.3%); p=0.0492). There were no significant differences in adverse events between the two groups, including hyperglycemia, new infection and barotrauma. 

Take home point:
In patients with early, but established, moderate-severe ARDS, the use of dexamethasone, compared to standard of care, significantly increased the number of ventilator-free days at 28 days and significantly reduced 60-day mortality, without an increased risk of adverse events.  

Synopsis: Hirschsprung disease (HD) is a congenital disorder of neural crest cell migration within the colon, leading to impaired motility. Please see Dr. Johnston and Dr. Swendiman’s full review of management and outcomes of long-segment HD in the June 2021 edition of PEBLR. Surgical treatment of HD involves resection of the aganglionic segment of colon and is usually undertaken within the first year of life. HD-associated enterocolitis (HAEC) is an inflammatory disorder characterized by abdominal pain, fever, and bloody bowel movements and can occur either before or after surgery. These symptoms overlap with those of inflammatory bowel disease (IBD), of which the incidence of childhood and very early onset are increasing. Given the overlap in entities, this two-part study sought to determine the frequency with which IBD is diagnosed in patients with HD.

In Part 1, 716 children diagnosed with HD were followed until IBD diagnosis or end of the study period (1991-2016) with 18 (2.5%) ultimately developing IBD compared to 0.2% of children without HD (RR 12; 95% CI 7.5-19). In Part 2, a case-control replication study was designed in which patients with a diagnosis of IBD were each matched with 10 age- and sex-matched controls using the parameters developed in Part 1, and the frequency of HD within the IBD cohort and matched controls was subsequently determined. The odds of having a previous diagnosis of HD in cases of IBD was significantly higher (OR 24 and OR 40 in two different Canadian provinces). Additionally, this study determined that in patients with HD, males were three-fold more likely to develop IBD (compared to a 2-fold higher male-to-female incidence in the general population). In those subsequently diagnosed with IBD, Crohn’s was more likely than ulcerative colitis. 

Ultimately, this study found that IBD can emerge in >2% of patients diagnosed with HD. Though the age of diagnosis, the presentation of disease, and the treatment modalities vary between HD/HAEC and IBD, the link between the two entities is important when treating patients with either diagnosis. One of the limitations of this study is the relatively short follow up time. The authors tracked patients through childhood but given the bimodal distribution of onset of IBD, it would be of particular interest to determine whether there is increased risk of late-onset IBD in HD patients. 

Synopsis: The rupture of abdominal aortic aneurysms (AAA) is the fifteenth leading cause of death in the country. AAA repair is indicated for asymptomatic aneurysms larger than 5.5cm, rapidly expanding aneurysms, and symptomatic aneurysms. With the evolution of endovascular technology over the last three decades, EndoVascular Aneurysm Repair (EVAR) has gained tremendous popularity as a minimally invasive treatment option; however, open AAA repair is seldom necessary in patients with aortic anatomic constraints and is often preferred in younger patients with AAA’s.

Open AAA repairs can be performed via two approaches: retroperitoneal or transperitoneal. A retroperitoneal approach is thought to give better exposure to the proximal abdominal aorta, and requires less bowel manipulation; while the transperitoneal approach is thought to provide better exposure to the right iliac and renal arteries, and allows for inspection of the viscera. There is limited high-quality data to support either approaches for better patient outcomes and the existing trials show conflicting results. 

"Retroperitoneal versus transperitoneal approach for elective open abdominal aortic aneurysm repair", is a meta-analysis published by Mei et al in the Cochrane Database of Systematic Reviews earlier this year that compares the two approaches. This review searched for randomized control trials (RCT) comparing the two approaches in patients undergoing elective open aortic surgery for aneurysmal disease; articles including redo aortic procedures, repairs for occlusive disease, and emergent repairs were excluded. Five RCTs met the inclusion criteria, and a total of 152 patients were included in the analysis. The primary endpoints examined were in-hospital and long-term mortality, hematoma, abdominal wall hernia, and chronic wound pain; secondary outcomes examined included intensive care unit (ICU) length of stay, hospital length of stay, intraoperative blood loss, aortic cross clamp time, and operative time. The study found no statistically significant difference in mortality, hematoma, abdominal wall hernia, chronic wound pain, aortic cross-clamp or operative time between the two approaches. However, patients undergoing retroperitoneal approach had a shorter length of ICU stay (mean difference – MD ~19 hours) and overall hospital stay (MD ~3 days), as well as less intraoperative blood loss (MD~500 cc). 

This systematic review suggests that both retroperitoneal and transperitoneal approaches have comparable outcomes when considering open AAA repair, although a retroperitoneal approach may reduce the ICU and overall hospital length of stay for patients. Unfortunately, the trials included in this review had small sample sizes and poor long-term follow up. In the absence of well-designed, large-scale randomized trials, it is difficult to determine whether either approach is ‘superior’, and the decision is often based on the surgeon’s preference and comfort level with each approach. 

Synopsis: If you have rotated through the trauma bay, you have certainly heard and likely even used a variation of the Seuss-inspired phrase “GCS 8, intubate”. This dogma is supported by both ATLS as well as the Eastern Association for the Surgery of Trauma practice management guidelines. In this presented study, Dr. Hatchimonji and his colleagues asked a beautifully simple question – does a GCS of 8 really require intubation? In other words, do patients who are recommended to undergo intubation actually do better than those that do not. Using the 2016 National Trauma Data Bank, they examined outcomes of 6676 patients presenting to trauma centers across the United States with a GCS of 6 to 8.

Overall, intubation was associated with an increased odds of mortality (1.05, p < 0.001). This odds ratio was stable across multiple subsets and sensitivity analyses. Both ICU and overall length of stay were significantly increased in the intubated group versus the non-intubated group. 

One of the strengths of this study is the rigorous statistical methods utilized. The treatment in question, intubation, is dependent on factors which also contribute to the primary outcome, mortality. While logistic and linear regression requires one to correctly model the outcome variable as a function of both the treatment variable as well as all covariates (usually a tall task, especially with multiple covariates that likely have differing effects on different populations of study), methods utilizing propensity scores only require correctly modeling the effect of covariates on just the treatment variable. This particular study utilized inverse probability weight regression adjustment, which uses propensity scoring to weight all the individuals of a study corresponding to their likelihood of receiving the treatment. This is an extension of traditional propensity matching, where “treated” and “untreated” patients are matched based on their propensity for receiving the treatment, creating a poor man’s randomized control trial where covariates are not different between the two treatment groups of a retrospective observational study (similar to the typical “Table 1” of a randomized control trial). 

In addition to the statistical methods, the authors perform several subgroup and sensitivity analyses, all producing similar results with regards to their primary outcome of mortality. Comparisons were performed on the entire cohort, the cohort minus those who proceeded directly to the operating room, only head injured patients, only non-head injured patients, intoxicated patients, and non-intoxicated patients. As the authors point out, the primary takeaway of the small increase in mortality risk with intubation “is less impactful for practice than the knowledge that there is no decrease in mortality risk” with intubation 

Interestingly, a similar study was recently published by Jakob et al. in the Journal of Trauma and Acute Care Surgery, “Isolated traumatic brain injury: Routine intubation for Glasgow Coma Scale 7 or 8 may be harmful!”¹, that I would urge you to read and compare. Besides the questionable use of logistic regression to control for complex confounders (see above explanation) resulting in questionably large treatment effects of non-intubation on survival and complications, I would like to point out one aspect of the manuscript submitted by Hatchimonji et al. that is subtly superior – the title. Just like the headshot of a fellowship application, an effective title encapsulates the essence of an article in terms of its content, goals, and tone. Effective scientific literature is not about edicts, definitive statements, or proclamations ending in exclamation marks. It’s about asking the right question in the right manner, and as Hatchimonji et al. have done well (starting with their title), leading the reader to their own conclusions. 

¹ Jakob DA, Lewis M, Benjamin ER, Demetriades D. Isolated traumatic brain injury: Routine intubation for Glasgow Coma Scale 7 or 8 may be harmful!. Journal of Trauma and Acute Care Surgery. 2021 May 1;90(5):874-9.