Intensivist ⚕️ POCUS | Hemodynamics | Chief Medical Officer of Deep Breathe AI
Canada
Joined February 2015
- Tweets4.9K
- Following1.1K
- Followers20.5K
- Likes8.6K
Being "intravascularly dry" in the ICU is way rarer than most clinicians think.
Most of the time its mistaking other conditions/circumstances 👇
Being "intravascularly dry" is often suspected when:
1. Furosemide monotherapy for diuresis is used, which produces metabolic alkalosis and hypernatremia irrespective of someone's true volume status.
2. Low MSFP shock due to vasoplegia in conditions like sepsis. The issue isn't about volume, but rather the pressure driving venous return.
3. Oliguria in the ICU. Many cases are 1) not AKI 2) not fluid responsive and 3) congestive
4. The patient has intracavitary gradients or dynamic LVOTo. These patients behave "super dry" . Preload helps temporarily, but afterload, beta blockade etc. accomplishes same goal in more durable fashion.
5. Pancreatitis patients. These patients have third spacing, sure, but actually are often more distributive. They require some fluids, but routine use of 100+ml/hr IV fluids in the absence of objective markers of low stroke volume and fluid responsiveness leads to many of the iatrogenic complications of pancreatitis we see!
I always tell my trainees, get the patient euvolemic and without edema using CRRT and only then switch to IHD. Even 5 days a week IHD in the icu rarely gets patients adequately decongested .. if you have an overloaded patient you switch to IHD in the icu chances are they will remain overloaded!
This is a major trial which will have a big impact on clinical practice. 554 patients with an indication for RRT were randomized to CVVHD vs intermittent HD. Trial stopped early due to higher 90-day mortaslity in the IHD group.
In #neurocriticalcare we always avaid IHD in patients with brain edema, shift or ICP because of the fluid shifts and hypotension. This RCT will further tip the scales toward CVVHD in borderline pts where either approach might seem reasonable.
Or is this the beginning of the end for IHD in the ICU all together?
Intermittent or Continuous Renal-Replacement Therapy in Acute Kidney Injury | @NEJM
nejm.org/doi/full/10.1056/NE…
At @DeepBreatheInc we’re automating point-of-care diagnostics with AI and pushing the boundries of computer vision and ultrasound science.
We’re growing and looking for motivated, hard-working people in regulatory to help support or (with the right experience) lead the regulatory team.
Experience with FDA, Health Canada, and EU markets desired. You'll work closely with me as CMO and help charter some of the first regulatory approvals in the AI SaMD space.
DM me if this sounds like you!
If you are teaching, the ratio of questions you ask to questions you get asked should be at least 4 to 1
Why is it that, despite all clinicians and researchers knowing that hemodynamic interventions across heterogeneous groups of shock patients will not improve outcomes, every trial still essentially adopts that approach?
It comes down to the fact that phenotyping a patient is hard and even harder to operationalize in a trial setting.
Our VESPER RCT led by @john_basmaji randomizing patients with early septic shock to venous congestion-guided resuscitation versus standard of care, along with the logistics of getting the scans done on admission, day 1, and day 2, is immense. I'm optimistic that we may be able to demonstrate that we can modify these patients' trajectories, but I can see why large-scale trials aren't done given how hard it is.
Seeing this is a TEE, the anterior wall is the only one that is working (a bit). You can tell it is TEE because of the omniplane indicator in the top left.
There’s more though! We have a QI project looking at sodium loading in our icu… other culprits behind maintenance fluids. Stay tuned!
Our Division of critical care hired an insanely talented intensivist and biomarker / multi-omics researcher Logan Van Nynatten and we are starting to explore the pathobiology of shock
We are starting with the pathobiology of venous congestion & hemodynamic coherence.. .what else should we do?
Its not about more or less fluids. Its about giving patients the right hemodynamic support based on their physiology!
Some patients in early septic shock are:
1. Fluid responsive with low SV --> fluids
2. Fluid unresponsive with low SV --> pressors, inotropes
3. Fluid responsive with high SV --> pressors
4. Fluid unresponsive with high SV --> pressors
5. Congested --> inotropes + diuresis
6. Impaired microcirculation with hemodynamic coherence --> fix macrohemodynamics
7. Impaired microcirculation with hemodynamic incoherence --> MAP target, avoid edema, other? (anticoagulation)
We need to stop treating "septic shock" as hemodynamic paradigm.
Among adults with septic shock, the use of restricted fluids and early vasopressors did not result in more days alive and out of the hospital through day 90 than the use of liberal fluids and later vasopressors. Full ARISE FLUIDS trial results and Research Summary: nej.md/3PVdwIp
ALT This image contains an excerpt from the Research Summary "Vasopressors or Fluids in Early Septic Shock” published in the New England Journal of Medicine. It includes a depiction of a patient with septic shock. The summary outlines a study with 1000 adults, median age 68, 60% men, and 40% women. Two groups are compared: "Vasopressor Group," with restricted fluids and early vasopressors, and "Fluids Group," with liberal fluids and later vasopressors. There are illustrations representing each group. A prompt to read the full Research Summary at NEJM.org sits at the bottom.
One of the biggest red flags that comes up when I interview trainees for research projects is if they don't use AI in their professional life. I find it correlates strongly with an unwillingness to embrace new ideas and innovation.
Hitting a patient in the head with a rock or displacing the eye with a sharp instrument was used >2500 years ago to treat what common condition?
( described in Sushruta Samhita, an ancient Indian surgical text dating to roughly 600 BCE )..
Journals that don’t adopt a similarly reasonable approach will be left behind
We don't ask whether a model touched their prose. We ask authors to own every word.
@JACCJournals parted w/ICMJE & doesn't require disclosure of AI writing help: the rule is unenforceable, biased, & not good for medical communication.
My Editor's Page: jacc.org/doi/10.1016/j.jacc.…
We need to stop this false human vs. AI dichotomy in research.
If you receive public funding for your research, I think you have an obligation to use or at least explore how AI can enhance your research.
1. Workflow efficiencies
2. Accounting for your lab
3. Writing manuscripts
4. Helping prepare knowledge translation materials.
5. Self peer reviewing your work before you disseminate.
We need to maximize the ROI per dollar of public research funding, appropriate/calibrated AI use is the greatest opportunity to do so in the last 20 years.
Unfortunately, another example of academic virtue signaling.
If AI is disclosed transparently and can help researchers more clearly communicate their science, how is it a negative?
We have Andromeda-VEXUS being submitted imminently - early congestion in septic shock is crazy... really challenges some of our traditional vexus understanding.
Can't wait to share!!!
Beyond Fluid Responsiveness: Why Venous Congestion Matters in Sepsis
CCR Journal Watch
criticalcarereviews.com/late…
Age old debate - how much of the 2.5L pericardial effusion should you drain at once?
Concern is re: precipitating RV failure..
Massive pericardial effusion in a HIV-negative man who presented with marked constitutional symptoms and mild dyspnea. 2500 mL of serous fluid drained. Dx: effusive tuberculous pericarditis.
#POCUS #FOAMed #MedEd
From 🔗Arthur K. et al. Cardiology Clinics, 2017-02-01, Volume 35, Issue 1, Pages 135-144
Daily reminder that fluid responsiveness is a normal physiologic state.
Fluid unresponsiveness is someone tipping (or already in) to heart failure
I love POCUS but feel like airway POCUS is a solution looking for a problem. We already have great tools for ETT confirmation (ETCO2) and no single metric of airway assessment captures predicted difficulty adequately.
Let's put down the probe (just for this indication!)
🤔 Ever wonder about #POCUS for airway management?
If so, ✔️ out @MGottliebMD’s ⭐️HIDI⭐️ in #journal_CHESTCritCare for a detailed & thoughtful approach to:
1. Assess for difficult airway &
2. Confirm ETT placement
chestcc.org/article/S2949-78…
