October 2020

Synopsis: Approximately 32 million people in the United States have a documented penicillin allergy. However, more than 95% of these patients ultimately are able to tolerate this class of drugs. In addition, most documented penicillin allergies are either “unknown” or cutaneous reactions, which makes it difficult for a clinician to safely prescribe this class of drugs. As it relates to surgery, non beta-lactam perioperative prophylaxis has been associated with increased risk of surgical site infections. Since patient reported penicillin allergies can affect the choice of antibiotics, in both the perioperative and post-operative setting, it is important to determine the validity and severity of patient reported penicillin allergy.

In this paper, Trubiano et al. developed and validated a penicillin allergy decision rule that can be used to risk stratify patients with a reported penicillin allergy. The tool, called PEN-FAST, was derived and validated using a prospective cohort of patients who underwent penicillin allergy testing at two sites in Australia. Patients who reported a penicillin allergy underwent either skin prick testing, intradermal testing, patch testing and/or oral challenging. A patient reported penicillin allergy was defined as a non–immune mediated adverse drug reaction, immune-mediated reaction, or unknown. Univariate analysis was performed using logistic regression to examine the associations between patient/phenotype characteristics with any penicillin-positive allergy test result. In this study, the prevalence of a positive penicillin allergy test was 9.3% (95% CI, 7.2%-11.9%). There were four key features associated with a positive penicillin allergy test result that were included in the clinical decision tool – Penicillin allergy reported by patient, Five years or less since reaction, Anaphylaxis or Angioedema, Severe cutaneous adverse reaction (SCAR), and Treatment required for reaction (hence the mnemonic PEN-FAST). These four criteria are scored as follows: Five or fewer years ago (2 points); anaphylaxis/angioedema or potential SCAR (2 points); treatment required for allergy episode (1 point). Based on the scoring, four risk groups were developed. Very low risk (0 points) group had a 0.6% of having a positive penicillin test result, low risk (1 or 2 points) had a 5% risk, moderate risk (3 points) had a 20% risk, and high risk (4 or 5 points) had a 50% risk. 

Using a cutoff of less than 3 points to classify patients as low risk, the PEN-FAST test had a sensitivity of 70.7% (95% CI, 57.3%-81.9%); specificity of 78.5% (95%C I, 74.9%-81.9%); positive predictive value of 25.3% (95%CI, 18.8%-32.7%); and negative predictive value 96.3% (95% CI, 94.1%-97.8%). These results were consistent across three external validation cohorts. 

Clinical Takeaway
PEN-FAST is a clinical decision rule that demonstrated a high negative predictive value, similar to that found in formal penicillin allergy skin-testing studies. PEN-FAST is a simple, safe and effective tool that can be used to improve antibiotic stewardship by identifying patients with a reported penicillin allergy who is at very low risk of having a true allergy.  

Synopsis: Nonoperative intervention for blunt splenic injury is increasingly common. While 10-20% of blunt splenic injuries present with hemodynamic instability and require immediate operative intervention, hemodynamically stable patients without peritonitis can be managed expectantly in hopes of avoiding splenectomy. However, these patients are still at risk for splenectomy during their hospitalization, often due to delayed hemorrhage. The increased use of splenic arterial embolization in expert trauma centers has helped many of these patients avoid splenectomy, increasing the rate of spleen rescue from 60% to over 80%. In particular, prophylactic splenic artery embolization (pSAE) is recommended for splenic pseudoaneurysms and splenic arteriovenous fistulas due to high rate of secondary rupture associated with the conditions. Large hemoperitoneum and American Association for the Surgery of Trauma (AAST) Organ Injury Score (OIS) >3 are also associated with rupture, but the optimal role of prophylactic embolization in these populations is unknown.

The Splenic Arterial Embolization to Avoid Splenectomy (SPLASH) trial sought to determine if pSAE is superior to surveillance with embolization as needed (SURV) in patients with an AAST OIS grade 3 or greater. This multi-center randomized clinical trial extended across 16 level 1 trauma centers in France. The study population included hemodynamically stable patients who had suffered blunt splenic trauma in the last 48 hours and had a high risk of requiring splenectomy as assessed by CT. Primary outcome was spleen rescue defined as no splenectomy and at least 50% vascularized parenchyma on CT at 30 days post injury. Because the inclusion criteria and outcomes were based on imaging, care was taken to establish guidelines for grading splenic trauma and training participating radiologists on the study’s consensus definitions. Further, each inclusion and outcome CT scan was read by two expert radiologists blinded to the treatment group in order to limit potential biases and interrater variability. 

A total of 140 patients were randomized to pSAE or SURV. Ultimately, the authors found no difference in rates of spleen rescue (56 of 57, 98.2% versus 56 of 60, 93.3%, respectively) or complications at 30 days. They did note there were fewer splenic artery pseudoaneurysms and fewer secondary embolizations in the pSAE group and hospital length of stay was 4 days shorter in the pSAE group (9 versus 13 days) but no difference in the time to return to work.

Additional take homes from the study include that the rate of spleen rescue in the surveillance group was much higher than what the authors had expected based on data in the literature. Because of this, their study may not have been adequately powered to detect differences in splenectomy or complications. Further, although the authors found that only 32.3% of patients in the SURV group ultimately required embolization or splenectomy, when they stratified their patients by injury severity, they found that 71.4% of those with an OIS >4 required embolization or splenectomy. This suggests that there may be a difference in rescue rates if the question was limited to more severe injuries and provides a starting point for future studies seeking to identify additional indications for prophylactic splenic embolization in trauma. In addition, invited commentary led by Dr. Lillian Kao highlighted that such prophylactic use of a definitive therapy may be advantageous in patients at risk for loss to follow up, providing an opportunity for shared decision making in centers with access to capable and safe interventional radiology. 

Synopsis: Carotid endarterectomy (CEA) is the cornerstone treatment of carotid atherosclerotic disease. The most recent guidelines from the Society of Vascular Surgery recommend CEA for symptomatic patients with 50% to 99% internal carotid artery (ICA) stenosis, and asymptomatic patients with 60 to 99% stenosis as long as the perioperative stroke and death rate is less than 3% and the patient’s life expectancy is at least 3 to 5 years. The evidence supporting these guidelines stems from randomized clinical trials dated back to the 1990’s and early 2000’s.

In the absence of more recent trials, the management of asymptomatic carotid disease has become somewhat controversial as the long-term risk of stroke is much lower in asymptomatic patients and it is thought that the natural history of asymptomatic carotid stenosis is improving with recent advances in medical therapy (such as high potency statins, improved antiplatelet regimens, and better blood pressure and diabetes control). 

“Comparative Effectiveness of Carotid Endarterectomy vs Initial Medical Therapy in Patients With Asymptomatic Carotid Stenosis”, published in JAMA Neurology in June 2020, was a retrospective comparative effectiveness study that utilized numerous databases from the Veterans Affairs (VA) between 2005 and 2009 to assess the 5-year stroke rate in patients older than 65, with asymptomatic carotid stenosis (at least 70% on duplex ultrasound) undergoing CEA vs medical therapy (MT). Kaplan-Meier (KM) curves were used to estimate the 5-year risk of fatal and non-fatal strokes, and statistical analysis was performed after constraining the sample to patients who met the strict inclusion/exclusion criteria of the most recent RCT comparing medical therapy to CEA in asymptomatic carotid disease (ACST – Asymptomatic Carotid Stenosis Trial). The authors used an advanced analytic approach, the target trial method, to simulate an RCT, where they randomized patients to one of the 2 treatment cohorts regardless of what treatment (surgical vs medical) the patient actually received. The patient whose actual treatment did not match the randomized treatment was censored on the actual treatment day. They also performed another analysis where they compared patients simply based on the actual treatment received. 

The pragmatic sample consisted of 5221 patients with asymptomatic carotid stenosis >70%, which comprised of 2,712 patients who underwent CEA and a propensity score matched sample of 2509 patients who received MT. The RCT-like sample comprised of 2012 patients receiving CEA and 1890 receiving medical therapy. The perioperative risk of stroke or death with CEA was 2.5%. In the pragmatic sample, the 5-year risk of stroke was 5.6% and 7.8% in the CEA and MT cohorts, respectively, which was statistically significant. When the competing risk of death (deaths due to causes other than stroke) was accounted for, the 5-year risk of stroke was 5.4% vs 6.2% which was not statistically significant. The results for the RCT-like sample with stricter inclusion criteria were similar (5.5% vs 7.6% in CEA vs MT, which was significant; after taking into account competing risk of death, 5-year stroke risk was 5.3% vs 6.2%, not significant). The 5-year risk difference in stroke was 2.3% (to the favor of CEA) in this study, which is half what was reported in prior carotid revascularization RCT’s.

This study suggests the medical therapy may be an acceptable alternative for the treatment of high-grade carotid stenosis in asymptomatic patients, given the perioperative risks and reduced long-term benefit with CEA. This study uses advanced statistical methods to simulate an RCT, an innovative strategy that may pave the way for future studies and can mitigate the cost and organizational challenges associated with performing a “true” RCT. There is no doubt that recent advances in medical therapy have improved cardiovascular outcomes, but there remains no level I evidence to suggest that the risk of CEA outweigh the benefits in asymptomatic patients with high grade carotid stenosis. We hope that the results of the CREST 2 trial (Carotid Revascularization and Medical Management for Asymptomatic Carotid Stenosis Trial), an ongoing RCT, provides more insight into the latter question.