Category Archives: Complications

Best Of EAST #7: Is There A Relationship Between Number Of Transfusions And Infection?

It has long been known that blood transfusion decreases immunocompetency for a period of time. This has been taken advantage of in transplant surgery for decades. And blood transfusions are used liberally in major trauma. So could blood transfusion make it more likely for a trauma patient to suffer complications such as pneumonia, sepsis, and surgical site infections?

The group at the Massachusetts General Hospital explored this possibility. The analyzed four years of TQIP data, examining patients who received blood transfusions within four hours of arrival. They excluded transfers in, patients with incomplete transfusion counts, and those who died within 48 hours.

They focused on pneumonia, sepsis, and surgical site infections and statistically controlled for demographics, comorbidities, injury severity, and surgical/procedural interventions.

Here are the factoids:

  • A million patients (!) were reviewed and about 41,000 met study criteria
  • The odds ratio of infectious complications increased from 1.23 after 2 units to 4.89 after 40 units
  • Each additional unit after 40 increased the odds of an infection by another 4.9%

The authors concluded that blood transfusion is associated with a dose dependent risk of infectious complications and that patients should be resuscitated to achieve prompt hemorrhage control (really?).

My comment: Well, this certainly looks fairly straightforward. Of course, it suffers from the usual drawbacks of massaging large databases. And remember, it shows an association, not cause and effect. How can we tease out whether the higher infection risk is due to badly hurt patients who need major surgery and prolonged ICU stays with pneumonia? The authors have tried to reduce this as much as possible using logistic regression. Unfortunately, many of the variables are very interdependent and I don’t believe the methods can fully overcome this. And there may be other factors not available for analysis in the TQIP data.

Here is my only question for the authors and presenter:

    • How can you be sure that you have fully controlled for the key variables that might influence your final analysis? Yes, you considered demographics, three listed comorbidities (cirrhosis, diabetes, and steroid use), injury severity, and some interventions. But might there be other factors not listed and maybe not even in the TQIP data? Ideas?

This is one of those papers that makes you say “hmm”. But don’t we always try to stop the bleeding promptly. I’m not sure what alternative we have to giving blood.

Reference: Overtransfusion comes at a significant cost: the dose-dependent relationship between blood transfusions and infections after major trauma. EAST Annual Assembly abstract #26, 2020.

Best Of EAST #4: Cannabis And Venous Thromboembolism

Cannabis and cannabidiol (CBD) are all over the news these days. CBD is legal everywhere, and it seems that more states are legalizing cannabis every few months. There are a few hints in PubMed that cannabinoids (THC) may have some impact on clotting, possibly causing hypercoagulability.

The group at the University of Arizona in Tucson decided to look into this in trauma patients. They did a two year scan of the TQIP database and stratified patients based on their THC status. They matched up THC positive and negative patients and examined thromboembolic events (deep venous thrombosis, pulmonary embolism, stroke, MI) and mortality.

Here are the factoids:

  • Nearly 600,000 patients records were in the database pull, but only 226 patients were THC+
  • They were matched at a 1:2 ratio with similar THC – patients (452)
  • No differences were found in the usual demographics, injury severity, use of DVT prophylaxis, and hospital length of stay
  • The THC+ group had a significantly higher incidence of overal thromboembolic complications (9% vs 3%)
  • Both DVT ( 7% vs 2%) and PE (2.2% vs 0.2%) were significantly higher in the THC+ group
  • No differences were seen in strokes or MI

The authors concluded that THC increases the risk of DVT and PE and that early identification and treatment for thromboembolic complications is required to improve outcomes in this high risk subset of trauma patients.

My comment: Seems compelling, right? But this is one of those abstracts that you have to read really closely. You have two groups of patients that are being compared, and a few statistical differences were found. The groups are small, but even so these differences are great enough to reach statistical significance. Great!

But, now step back and look carefully at the larger patient group. There are almost 600,000 patients there, but am I to believe that only 226 patients (0.04%) were using cannabis? According to recent statistics, approximately 8% of the US population currently uses marijuana. So in theory, about 47,500 patients in the TQIP sample should have tested positive. For whatever reason, this data point was not collected. Could data from the other 47,274 have changed the study result? Probably. 

Here are my questions for the authors and presenter:

  1. What was the impetus for this study? I was not aware of clotting issues due to THC and there is little in the published literature. I’d love to hear some history and be able to read more about this.
  2. What about the long time interval that a patient will test THC+ after partaking? THC remains in body fat for a month or more, and the qualitative test commonly used will provide a positive for weeks after the last use. How long do the thrombogenic effects of THC last? The THC+ result recorded in the dataset could be from THC use well before the traumatic event.
  3. How do you think your small sample of THC+ patients impacts your results given the much larger number of expected marijuana users in the sample?

This is intriguing work. Let’s here more!

Reference: Impact of marijuana on venous thromboembolic events: cannabinoids cause clots in trauma patients. EAST Annual Assembly abstract #4, 2020.

Best Of EAST 2020 #1: Treatment Of Blunt Carotid & Vertebral Injuries

The 33rd Annual Assembly of the Eastern Association for the Surgery of Trauma starts in just two weeks! As usual, I will select several interesting abstracts from the bunch to review. I’ll go over the findings of the research, critique it, and then provide a series of questions for the presenter to consider. These questions are ones that members of the audience may very well ask (hint, hint).

And FYI, I always send a heads-up to the presenters with a link to the post so they can study up beforehand!

So let’s get started with the first abstract that will be kicking off the meeting on January 15. Blunt cerebral / vertebral artery injury (BCVI) is one of those insidious injuries that trauma professionals don’t always think about. But they do occur in about 1% of major trauma patients. It’s one of those injuries that can’t be ignored because very serious complications may occur if it is not treated appropriately (think stroke).

Unless there are extenuating circumstances like bleeding or pseudoaneurysm, treatment is usually pharmaceutical. There are two camps: antiplatelet drugs vs anticoagulant drugs. But there is very little data to determine which one is better.

This abstract is a retrospective review from the National Readmission Database (NRD). This resource is maintained by the US government and provides information on patient readmissions nationally across all payors as well as the uninsured. They included all patients > 18 years old with a BCVI and minor injuries in other body regions. Patients who suffered a stroke complication during their initial hospital stay were excluded.

Patients were divided into two groups: those taking an antiplatelet agent and those prescribed an anticoagulant. Outcomes of interest were readmission with CVA and death, within six months.

Here are the factoids:

  • 725 patients with BCVI were found during the five year study period
  • Patients were propensity matched for a 1:1 ratio of patients taking antiplatelet vs anticoagulant drugs, leaving 370 patients for analysis
  • There was a lower rate of admission in the anticoagulant patients vs the antiplatelet ones (9% vs 26%)
  • There were fewer deaths within 6 months in the anticoagulated patients (1.3% vs 3.9%)
  • Median time to stroke was 6-9 days and was not significantly different between the two groups

The authors concluded that the overall stroke rate after BCVI is 6%. They also found an association with lower rates of CVA within 6 months of discharge in patients on anticoagulants. They recommend further studies to determine which type of chemoprophylaxis is best.

My comments: This is an interesting paper that addresses a problem that we don’t have good answers for. The study was well constructed and simple to follow. The two areas that I have questions about are data quality and statistical power.

The NRD is a powerful tool for research, but does have some shortcomings. It only contains information on readmissions, and may not contain some patients who had asymptomatic strokes or massively stroked and died at home. Not knowing these numbers injects some bias and could change the numbers and findings of the study.

The other issue has to do with statistical power. The overall eligible patient group (725 patients) was small in the first place. Propensity matching for a 1:1 ratio shrunk it to only 370, or 185 in each treatment group. My armchair power calculations show that this study would only be able to detect a 7x difference in mortality, and not the 3x difference seen. I’m glad the authors didn’t claim a “significant decrease in CVA” in the anticoagulated patients vs the antiplatelet drug patients.

Here are my questions for the authors:

  1. What do you see as drawbacks to data quality in your study due to use of the National Readmissions Database? How do you think that patients not included in it impacted your data?
  2. Is there anything you can do to improve the statistical power of the study to see if the mortality difference is truly different? Even though your statistical analysis shows significance, the number of subjects doesn’t allow you to claim this until the mortality in the antiplatelet group reaches 9%. 

This was a simple yet fascinating study, and is a start toward helping us determine which of the two drug classes is most appropriate for patients with BCVI.

Reference: Treatment of blunt cerebrovascular injuries: anticoagulants or antiplatelets? EAST Annual Assembly abstract #1, 2020.

How To Predict Venous Thromboembolism In Pediatric Trauma

As with adults a decade ago, the incidence of venous thromboembolism (VTE) in children is now on the rise. Whereas adult VTE occurs in more than 20% of adult trauma patients without appropriate prophylaxis, it is only about 1% in kids, but increasing. There was a big push in the early 2000′s to develop screening criteria and appropriate methods to prevent VTE. But since the incidence in children was so low, there was no impetus to do the same for children.

The group at OHSU in Portland worked with a number of other US trauma centers, and created some logistic regression equations based on a large dataset from the NTDB. The authors developed and tested 5 different models, each more complex than the last. They ultimately selected a model that provided the best fit with the fewest number of variables.

The tool consists of a list of risk factors, each with an assigned point value. The total point value is then identified on a chart of the regression equation, which shows the risk of VTE in percent.

Here are the factors:

Note that the highest risk factors are age >= 13, ICU admission, and major surgery.

And here is the regression chart:

Bottom line: This is a nice tool, and it’s time for some clinical validation. So now all we have to do is figure out how much risk is too much, and determine which prophylactic tools to use at what level. The key to making this clinically usable is to have a readily available “VTE Risk Calculator” available at your fingertips to do the grunt work. Hmm, maybe I’ll chat with the authors and help develop one!

Reference: A Clinical Tool for the Prediction of Venous Thromboembolism in Pediatric Trauma Patients. JAMA Surg 151(1):50-57, 2016.

Overwhelming Post-Splenectomy Infection (OPSI)

Most trauma professionals have heard of OPSI, but few have ever seen it. The condition was first described in splenectomized children in 1952. Soon after, it was recognized that this infection occurred in asplenic adults as well.

OPSI is principally due to infection by encapsulated organisms, those with a special polysaccharide layer outside of the bacterial wall. This layer is only weakly immunogenic, and confers protection from the normal immune mechanisms, particularly phagocytosis. However, these bacteria are more easily identified and removed in the spleen.

OPSI may be caused by a number of organisms, the most common being Strep. pneumonia, Haemophilus influenza, and meningococcus. For this reason, the standard of care has been to administer vaccines targeting the usual organisms to patients who have lost their spleen.

How common is OPSI? A recent paper from Gernany reviewed comprehensive data from 173 intensive care units over a 2-year period. Here are some of the more interesting factoids:

  • 2,859 ICU beds were screened, but the number of unique patients was not given. This is very disappointing because incidence cannot be calculated!
  •  52 cases of OPSI occurred
  •  Only half of the patients had received vaccines
  •  Pneumococcus was the most common bacterium (42%). There were no H. Flu or meningococcal infections.

Bottom line: Yes, OPSI exists and can occur in your asplenic patients. It is uncommon enough that you and your colleagues will probably never see a case. But proper vaccination remains important. Papers consistently show that we are collectively not very good at ensuring that our splenectomized patients receive all their vaccines, ranging from only 11-50%. We collectively need to make better efforts to provide them to our at-risk patients.

Reference: Overwhelming Postsplenectomy Infection: A Prospective Multicenter Cohort Study. Clin Infec Diseases 62:871-878, 2016.