Acute RV failure: Stop fearing preload reduction! Aggressive high-dose loop diuretics EARLY → target spot UNa >50-70 mmol/L & UOP 100-150 mL/h. Even if hypotensive-support with pressors. Old dogma is killing patients. Read more here: https://t.co/1sFPhRvXT4 #RVFailure #CriticalCare #FOAMcc #MedTwitter
קינורז… שם מעולה! מאז המתכון הקודם, התחלתי להוסיף גם בורגול וזה יוצא אפילו יותר טוב ואוורירי. היחס הוא 1 קינואה:2 אורז:2 בורגול. כרגיל, קצת שמן זית, קצת מלח וזהו. עכשיו נקרא לזה קינורזגול…
Strongly agree that hallucinations and sycophancy remain serious issues in general-purpose models like ChatGPT.
The future should focus on broader, high-quality retrieval + clear evidence hierarchies + full source transparency, rather than just polishing the LLM layer.
Have you tried combining these tools with direct PubMed/Epistemonikos searches?
@ethanjweiss Classic lock-in strategy: get everyone dependent, then jack up the prices. If you think I’m exaggerating, just look at what Ancestry has done repeatedly with subscription hikes once people were hooked on their data.
You can miss it.
Not all tamponade looks like “textbook” tamponade.
A pericardial effusion went from moderate to large in <48h. No RA or RV collapse on POCUS.
The patient arrested unmonitored before anyone acted.
In patients with high right-sided pressures (RV dilation, severe TR, volume overload), the classic echo signs can be absent.
Tamponade is a physiologic diagnosis, not an echo checkbox.
If you have a growing effusion and the patient isn’t improving:
→ Escalate care immediately
→ Reassess with full Doppler + clinical context
→ Force the decision: drain or observe?
Don’t wait for perfect echo signs.
By the time the textbook picture appears, it may already be too late.
Teens aren’t drinking less because they’re suddenly “healthier”, they’re mainlining dopamine from endless short videos instead.
Alcohol (and real-life socializing) delivers a slower, sustained dopamine release in the brain’s reward circuit (VTA → nucleus accumbens). A 15-second TikTok/Reel gives a rapid spike: novelty, surprise, music and social validation all hit dopamine transporters fast and frequently.
This trains the brain for constant micro-rewards with minimal effort.
Why bother with the delayed, riskier payoff of partying or even hanging out when your phone delivers hits every few seconds? Over time it can blunt sensitivity to natural rewards (anhedonia), reduce motivation for deeper IRL experiences, and contribute to the exact rises in isolation + mental health issues we’re seeing.
The “success” in cutting teen drinking may just be substitution with a more potent, accessible behavioral addiction.
Finally, alcohol makes you do stupid things, and that is something to be considered when everybody around you has a camera and looking for the next viral video. Teens are way more aware of that than us adults.
ארגוני זכויות בעלי חיים כמו PETA שותקים לחלוטין בנוגע להאשמות של קריסטוף. לא נמסרה אפילו הצהרה אחת!
למה? כי הטענות האלה כל כך בדיוניות באופן קריקטורי ובלתי אפשריות מבחינה מכנית, שכל מי שבילה חמש דקות בחברת כלבים יודע את זה מיד. הארגונים האלה יהיו הראשונים להתנפל על ישראל אילו היה בהאשמות האלו אפילו שמץ של אמת.
השתיקה שלהם מהדהדת.
Animal rights groups like PETA have been completely silent on Kristof’s dog-rape allegations against Israel.
Not a single statement. Why? Because the claims are so cartoonishly fictional and mechanically impossible for anyone who’s spent five minutes around dogs. These organizations would’ve been the first to pile on Israel if there was even a shred of truth.
Their silence speaks volumes.
📊 Critical care/intensivists don’t get their own line in the latest AMA benchmark, but they fall under “Other IM Subspecialties” at ~28.7% lifetime risk of being sued.
That’s solidly mid-pack, better than EM (42%), Radiology (38%), and way below the surgical heavyweights (OB/GYN 59.6%, Gen Surg 53%).
Makes sense: we manage incredibly high-acuity patients and tough family conversations, yet the data suggests our risk isn’t as sky-high as the procedural fields. Still, with every code and difficult withdrawal-of-care discussion, that number is a reminder of how litigation shadows even “non-procedural” work.
Anyone else surprised it’s not higher? Or is this what you’d expect?
#MedTwitter #CriticalCare
@BrandonLuuMD I have a couple of small scars on the back of my head from an accident. One day after a short haircut, my chief asked me if it was recently that I got the lobectomy…
@Dr_Chris_Twine Came into the OR as a new surgery intern to operate with my chief for the first time. He asked me if I was right handed and I said I was ambidextrous. He replied “Oh, so you’re equally worthless with both hands”.
Relying heavily on a narrow set of elite journals (NEJM, JAMA, etc.) is dangerous, not a feature.
It misses the true breadth of medical research published across thousands of journals. I recently published a heavily read review on RVF (97% percentile readership) in Lung bc the editor invited it. OE can only access the abstract and none of the 18(!) pages of 1.5 years of my research.
Other researchers choose smaller/specialty journals for faster publication, higher acceptance rates for niche or incremental findings, negative results, regional data, emerging topics, or work that doesn’t fit the broad/high-profile appeal big journals demand.
Limiting AI to “prestige” sources potentially creates large blind spots in real-world evidence.
Relying heavily on a narrow set of elite journals (NEJM, JAMA, etc.) is dangerous, not a feature.
It misses the true breadth of medical research published across thousands of journals.
Researchers choose smaller/specialty journals for faster publication, higher acceptance rates for niche or incremental findings, negative results, regional data, emerging topics, or work that doesn’t fit the broad/high-profile appeal big journals demand.
Limiting AI to “prestige” sources creates blind spots in real-world evidence.