All posts by The Trauma Pro

Pet Peeve, Research

The Power Of Trauma Research?

Leave a comment

The veracity of medical research conclusions, specifically trauma research, has always fascinated me. I see so many people who are content to jump to conclusions based on a paper’s title or the conclusions listed in its abstract. However, having read thousands of research papers over the years, I am no longer surprised when these easy-to-digest blurbs don’t entirely match up with the actual details embedded within the manuscript.

The authors of these publications genuinely intend to provide new and valuable information, either for clinical use or to allow future studies to build upon. Unfortunately, mistakes can be made that degrade their value. Problems with research design are among the top culprits that affect the significance of these papers.

I will focus on what are considered “gold standard” studies in this post. The randomized, controlled trial (RCT) is usually considered the big kahuna among research designs. They can provide solid answers to our clinical questions if designed and carried out properly.

In 2010, a checklist of 25 guidelines was created (CONSORT 2010, see reference 1) to guide researchers on the essential items that should be included in all RCTs to ensure high-quality results. A group of investigators from Jackson Memorial in Miami and Denver Health published a paper earlier this month that critically reviewed 187 trauma surgery RCTs published from 2000 to 2021. They analyzed power calculations, adherence to the CONSORT guidelines, and an interesting metric called the fragility index.

Here is a summary of their findings:

  • Only 46% of studies calculated the sample size needed to prove their thesis before beginning data collection. With no pre-defined stop point, the researchers might never be able to demonstrate significant results or may be wasting money on subjects in excess of the number actually needed.
  • Of the 54% that did calculate the needed sample size, two-thirds did not achieve it and were not powered to identify even very large effects. Once again, they are spending research money and will almost certainly be unable to show statistical differences between the groups, even if one actually existed.
  • The CONSORT checklist was applied to studies published after they were developed in 2010; the average number of criteria met was 20/25, and only 11% met all 25 criteria. The most common issue was failure to publish the randomization scheme to ensure no bias was possible due to it.
  • Among 30 studies that had a statistically significant binary outcome, the mean fragility index was 2. In half of these studies, having a different outcome in as few as two patients could swing the final results and conclusion of the study.
  • The majority of the studies (76%) were single-center trials. Frequently, such trial results cannot be generalized to larger and more disparate populations. Larger, confirmatory studies often have results that are at odds with the single-center ones.

Bottom line: What does it all mean? Basically, a lot of well-intentioned but poorly designed research gets published. The sheer volume of it and the work needed to interpret it correctly make it very difficult for the average trauma professional to understand and apply. And unfortunately, there is a lot of pressure to pump out publication volume and not necessarily quality. 

My advice for researchers is to ensure you have access to good research infrastructure, including experts in study design and statistical analysis. Resist the temptation to write up that limited registry review from only your center. Go for bigger numbers and better analysis so your contribution to the research literature is a meaningful one!

References:

  1. CONSORT 2010 Guidelines Checklist for randomized, controlled trials
  2. Statistical Power of Randomized Controlled Trials in Trauma Surgery. JACS 237(5):731-736, 2023.
Philosophy

FAST Is Fast, and FAST Is Last

Leave a comment

Ever been in a trauma activation where it seems like the first thing that happens is that someone steps up to the patient with the ultrasound probe in hand? And then it takes 5 minutes of pushing and prodding to get the exam done?

Well, it’s not supposed to be that way. The whole point of adhering to the usual ATLS protocol is to ensure that the patient stays alive through and well after your exam. And FAST is not part of the primary or secondary surveys, it is an adjunct.

As always, there are a few exceptions to the rule above.

  • If an unstable patient arrives without an obvious source of bleeding, FAST of the abdomen should be able to detect if a large hemoperitoneum is present. This will expedite the patient’s transfer to the OR.
  • A patient in cardiac arrest may benefit from a quick FAST to determine if cardiac activity is present. If not, it may be time to terminate resuscitation.

Many people say that FAST and physical exam can and should happen simultaneously.

In principle, I agree. My previous statements were based on the way that we organize our trauma team and trauma activations at this hospital. The reality is that everyone’s team is different and they may run their trauma activations differently.

The goal is to get all information critical to keeping your patient alive as quickly as possible. In some cases, knowing if there is a significant amount of fluid in the abdomen can be very important. Most trauma resuscitation schemes at trauma centers make use of multiple personnel so that various portions of the patient evaluation can be carried out simultaneously.

But there is also a tradeoff between speed, trauma team size and number of trainees. Centers with fewer or no trainees will have a leaner team with experienced examiners and more room around the patient. At our hospital, we have 8 people clustered immediately around the patient, with half of them being surgery or emergency medicine residents. This means it is more difficult for a physician to step in and do a FAST exam easily. So typically, this physician is the same resident doing the torso portion of the physical exam. This is the main reason for my exhortation to wait until the end of the physical exam and do the FAST quickly.

Bottom line: With the exceptions noted above, always complete the ATLS primary and secondary surveys first. Then pull out the ultrasound machine, but be quick about it. If it takes more than about 60 seconds to do the exam, someone probably needs a little more practice.

Guidelines

Guidelines for Consultants to the Trauma Service

Leave a comment

My last two posts were dedicated to the use of consultants for your trauma patients. Here are my thoughts on best practices in using their services.

Trauma surgeons often rely on consultants to assist in the care of their patients. Orthopedic surgeons and neurosurgeons are some of the more frequent consultants, but various other surgical and medical specialists may be needed. I have found that providing guidelines to consultants helps ensure quality care and good communication between caregivers and patients/families.

We have disseminated guidelines to our colleagues, and I wanted to touch on some of the main points. You can download the full document using the link at the bottom of this post.

To deliver the highest quality and most cost-effective care, we request that the services we consult do the following:

  • Please introduce yourself to our patient and their family, and explain why you are seeing them.
  • Although you may discuss your findings with the patient, please discuss all recommendations with a member of the trauma service first. This avoids patient confusion if the trauma team chooses not to implement any recommendations due to other patient factors you may not be aware of.
  • Document your consultation results in writing (paper or EMR) promptly.
  • If additional tests, imaging, or medications are recommended, discuss with the trauma service first. If appropriate, we will write the orders or clear you to do so and discuss the plan with the patient.
  • We round at specific daily times and welcome your attendance and input.
  • Please communicate any post-discharge instructions to us or enter into the medical record so we can expedite the discharge process and ensure all follow-up visits are scheduled.

Bottom line: A uniform “code of behavior” is essential! Ensuring good patient communication is paramount. They need to hear the same plans from all of their caregivers, or else they will lose faith in us. One of the most important lessons I have learned over the years is that you do not need to implement every recommendation that a consultant makes. They may not be aware of the most current trauma literature and will not be familiar with how their recommendations may impact other injuries.

Click here to download a sample Trauma Services consultant guidelines document.

Tips

Consultant Gives An Unusual Recommendation: What Would You Do?

Leave a comment

I know this has happened to most of you at one point or another:

One of your trauma patients sustains an injury outside of your area of expertise. You engage a consultant to evaluate that condition and manage it. They do so, and it requires some type of invasive procedure. They return from the procedure, and as you are rounding on the patient, you find the consultant has ordered a medication that you have not seen ordered for that procedure before.

What would you do? You are now in an interesting place. Do you discontinue the order? Call up the consultant and ask, what the heck? Might you poison your relationship with them in the process? And what is the impact on your patient?

Lots of questions, but here is what I recommend:

  • Hit the lit! Always assume that they might know something you don’t. They are an expert in their field for a reason, so give them the benefit of the doubt. Thoroughly review the literature to see if this is an approved new practice. But remember, a single interesting paper should never be enough to change your (or their) practice. There needs to be a sufficient body of literature showing that the practice is sound.
  • Talk to the consultant. Now that you are armed with the current thinking, ask them what they were thinking! Let them explain their rationale. Since you have already looked at the available data, you will be able to ask appropriate questions and deflect answers like, “well that’s how we did it where I trained.”
  • Change the orders. Assuming the order was not sound, it’s time to undo the ones that started this entire debate. Get rid of them now so you’re not stepping on any toes. However, if you believed that the order/medication would have been potentially harmful, don’t wait. You should have done it even before the first step!
  • Disseminate the info. Make sure that all of your partners are aware of the issue and the correct course of action (or orders). And send a note to the consultant group summarizing the discussion so none of your consultant’s partners make the same mistake again.

In the next post, a set of guidelines to give all of your consultants to make sure they behave appropriately and interface will with the trauma service.

 

How to

Help! My Consultant Won’t Come In To See A Patient!

Leave a comment

Consultants provide very important services to trauma patients in the ED and inpatient settings. The trauma professionals managing those patients can’t know everything (although we sometimes think we do). But occasionally our patients present issues that require evaluation by other experts in order to guarantee excellent care.

Sometimes our consultants want to do too much, or make recommendations that are not really in their area of expertise (e.g. a cardiologist evaluating a cardiac contusion). See the related post link below for tips on this situation.

But sometimes you know what the patient needs, but the consultant doesn’t agree or doesn’t do what you expect. Or they don’t want to come in when called. What to do?

Here are some tips:

The patient is in the ED and the consultant won’t come in to see the patient.

  • Are they right? Does that problem really need to be dealt with in the ED in the middle of the night? Many simple fractures and wounds do not need immediate attention. They can be dressed/splinted, the patient reassured, and instructed to see the consultant in the clinic the next day.
  • Is your knowledge of current management of the condition correct? Perhaps it has evolved, and it is now commonplace to temporize and deal with the problem as an outpatient during business hours. Make sure you are up on the current literature.

The patient is in the ED and the consultant won’t come in to see the patient, and you are sure that they should! Now what?

  • Call them personally (not a resident, midlevel provider, or any other intermediary) and clearly and concisely explain the situation, and your assessment of why the problem needs their immediate attention.
  • Listen to or elicit their rationale for not seeing the patient. If legitimate, this may help educate you and modify your future management of similar patients. If the rationale is not legitimate, inform them (tactfully) that this is at odds with your education/training/experience with other providers. Ask them to further explain, if they can.

If they still won’t come in despite what you think is a legitimate need, then you must calculate a quick risk:benefit ratio. Will any patient harm occur if the consultant does not see the patient? And what is the professional damage that you will incur if you move on to the next steps. If you believe that harm will occur, here are your options, from least to most damaging to your professional status at the hospital:

  • Contact another consultant in the same or overlapping specialty (if there is one). Apologize for the fact that you know they are not on call, and explain the situation.
  • Appeal to a higher authority. Contact the trauma medical director, service chief, or hospital administrator and see if they can intervene.
  • Explain to the consultant that you truly believe that harm will occur, and you will have to document that fact in the medical record as well as their failure to respond. In some cases, this will shake them loose, but they will certainly be pissed.
  • If all else fails, see if you can find a service that will help you by accepting the patient as an admission so they can be managed appropriately the next day. But then follow through by reporting the event to appropriate people including chief of staff, chief medical officer, VPMA, hospital quality department, and risk management. This is the nuclear option, so be prepared for the fallout.

Bottom line: This is not a fun situation to find yourself in. Good luck!