Sign in

Kaspar Bachmann

@kasparbachmann.bsky.social
681 followers 140 following 119 posts

MD | ICU Physician & Researcher Passionate about gastrointestinal dysfunction, nutrition as well as ECMO and cardiopulmonary physiology. Enthusiast in #MatLab, signal analysis and data-driven research: pubmed.ncbi.nlm.nih.gov/?term=Kaspa…

PostsRepliesMedia
Kaspar Bachmann @kasparbachmann.bsky.social · 08/04/2026
Hi there Thanks for noticing. Here you go: link.springer.com/article/10.1... www.jcvaonline.com/article/S105...
link.springer.com
Mixed venous oxygen saturation cannot be estimated by central venous oxygen saturation in septic shock - Intensive Care Medicine
Objective Central venous oxygen saturation (ScvO2) in initial resuscitation is included in the Surviving Sepsis Campaign guidelines. ScvO2 monitoring has also been suggested to be comparable to mixed ...
020
Kaspar Bachmann @kasparbachmann.bsky.social · 08/04/2026
Problem is twofold: LoA are around +/- 10%. ScvO2 of 55 then is not helpful. Trends do not point in the same direction. We showed this here: pubmed.ncbi.nlm.nih.gov/36511508/ And other studies: doi.org/10.1007/s00134… doi.org/10.1053/j.jvca��� Bottom line: when it counts, it is not helpful.
110
Kaspar Bachmann @kasparbachmann.bsky.social · 08/04/2026
IAP measurement does not require seadtion or neuromuscular blockade. That is a myth that needs to disappear. Measuring pressure is simple and accurate; a measurement of 18 is what it is. Sedation and paralytics can lower that number, but that is a therapy for IAP rather than enabling a measurement.
010
Kaspar Bachmann @kasparbachmann.bsky.social · 07/03/2026
My issue with ScvO2 is, that it can be high while true mixed venous (SvO2) is low. It is a bad surrogate for SvO2. So when in doubt, float a PAC and know for sure.
110
Reposted by Kaspar Bachmann
theelso.bsky.social @theelso.bsky.social · 02/12/2025
LEVOECMO RCT: early administration of levosimendan to facilitate weaning from VA #ECMO in severe but potentially reversible cardiogenic shock 🔍 205 adults/11 ICUs in 🇫🇷 ⚖️ early levosimendan administration did not significantly shorten the time to successful weaning of #ECLS at day 30 🔗 bit.ly/4iAwGwT
031
Kaspar Bachmann @kasparbachmann.bsky.social · 01/11/2025
Thanks for sharing our review! Monitoring GI function when initiating / stepping up nutrition is key to prevent harm from too aggressive nutrition.
020
Kaspar Bachmann @kasparbachmann.bsky.social · 15/08/2025
My question is, does it truly matter? If there’s a strong signal, shouldn’t I be able to detect it using either method (and vice versa)? And if the signal is borderline, shouldn’t I arrive at the same interpretation if my application and understanding of the statistics used are accurate?
010
Reposted by Kaspar Bachmann
Matt Siuba @zentensivist.bsky.social · 21/04/2025
New 📝 We know RV dysfunction occurs on VV-ECMO in ARDS despite ECMO mitigating many of the traditional RV dysfunction risk factors (hypercapnia etc) We don’t understand RV function dynamically over time on ECMO…
3146
Reposted by Kaspar Bachmann
Erin McCreary, PharmD @erinmccreary.bsky.social · 12/04/2025
@gurujosh.bsky.social presents the results of the MSSA domain of the SNAP 🫰 trial at #ESCMID2025 fluclox caused more acute kidney injury than cefazolin; early mortality favored cefazolin Absolutely Herculean effort by the entire global trial team and practice changing data #teamcefazolin
2264
Reposted by Kaspar Bachmann
Wolf Hautz @wolfhautz.bsky.social · 18/03/2025
Have you ever wondered what it takes before a clinical trial ends up in your timeline? Well… we recently published the DDx-BRO trial ⬇️ Here is a 🧵 of its lifecycle: from 💡to 🏁, well, if the latter actually exists.
thelancet.com
Diagnoses supported by a computerised diagnostic decision support system versus conventional diagnoses in emergency patients (DDX-BRO): a multicentre, multiple-period, double-blind, cluster-randomised...
Use of a CDDSS did not reduce the occurrence of diagnostic quality risk compared with the usual diagnostic process in adults presenting to emergency departments. Future research should aim to identify...
131
Reposted by Kaspar Bachmann
Peter Nydahl @nydahlpeter.bsky.social · 23/12/2024
Core outcome set of daily monitoring of gastrointestinal function in adult critically ill patients: a modified Delphi consensus process (COSMOGI) Thanks to @criticalcarereviews.com ccforum.biomedcentral.com/articles/10....
ccforum.biomedcentral.com
Core outcome set of daily monitoring of gastrointestinal function in adult critically ill patients: a modified Delphi consensus process (COSMOGI) - Critical Care
Purpose Gastrointestinal (GI) dysfunction is common in critically ill patients and associated with poor outcomes. There is a lack of standardised methods for daily monitoring of GI function. COSMOGI a...
022
Kaspar Bachmann @kasparbachmann.bsky.social · 25/01/2025
I don’t think that offering specific therapies out of the context of goals of care is helpful and may negatively impact the discussion and patient trajectory.
010
Kaspar Bachmann @kasparbachmann.bsky.social · 25/01/2025
I think we want to define where a patient wants to be in 1/3/6/12 months and with what level of quality of life. We as a team can then offer specific therapies (i.e. Trach/PEG but also other options such as mech. assist devices), if and only if these therapies align with the defined goals of care.
130
Kaspar Bachmann @kasparbachmann.bsky.social · 03/01/2025
#Halo #SwissMountains #Freeride
030
Reposted by Kaspar Bachmann
Nick Mark, MD @nickmark.bsky.social · 31/12/2024
I got tired of opening a dozen PDFs to find the ECMO pressure/flow curve I need. So I made an app! Choose the cannula/size and you can see the max flow for a given pressure drop. Or compare multiple different cannula. Lmk what ya’ll think. Useful? What’s it need? onepagericu.com/ecmo-calcula...
73511
Kaspar Bachmann @kasparbachmann.bsky.social · 01/01/2025
Recently listened to this: soundphysicians.com/podcast-epis... Has some good pearls and pitfalls and some resources in the shownotes.
soundphysicians.com
010
Reposted by Kaspar Bachmann
Maarten van Smeden @maartenvsmeden.bsky.social · 23/12/2024
Let us start 2025 in a positive mood: here are 10 methods things researchers can worry *less* about in 2025
media.tenor.com
a countdown clock with the number 10 in the center
ALT: a countdown clock with the number 10 in the center
15259119
Kaspar Bachmann @kasparbachmann.bsky.social · 23/12/2024
📌
000
Reposted by Kaspar Bachmann
Matt Siuba @zentensivist.bsky.social · 23/12/2024
ICYMI #emimcc
042
Reposted by Kaspar Bachmann
Lars Mølgaard Saxhaug @load-dependent.eurosky.social · 20/12/2024
This was a fun study leveraging social media to demonstrate the high prevalence in Impella malrotation among cases shared on old Twitter. If it's malrotated, but it works well, without suction alarms or other problems, do you change it? #cccsky #cardiosky #emimcc www.pcronline.com/News/Whats-n...
pcronline.com
Impella percutaneous ventricular assist device malrotation in social media: a call to action
This study suggests that correct Impella orientation is often overlooked, and that Impella malrotation is a common issue in real-world clinical practice, potentially associated with relevant complicat...
1135
Kaspar Bachmann @kasparbachmann.bsky.social · 19/12/2024
Do you/does anyone know if there are data repositories with patient data, timepoints and vanco levels as well as reference clearence (maybe from observational studies)? We could then use that to 1) validate and 2) develop new models.
011
Kaspar Bachmann @kasparbachmann.bsky.social · 18/12/2024
#emimcc I plan to do a thread on the paper/topic soon. This project has kept me busy over the last two years and it has been a pleasure working alongside an awesome steering committee and Delphi panel. 🙏
121
Kaspar Bachmann @kasparbachmann.bsky.social · 18/12/2024
Fresh off the press: COSMOGI 🗞️🗞️🔥🔥 What variables should we use to daily monitor #gastrointestinal function in #criticalcare? We have developed a core outcome set and have defined 13 variables for daily GI assessment. Published today in Critical Care: ccforum.biomedcentral.com/articles/10....
ccforum.biomedcentral.com
Core outcome set of daily monitoring of gastrointestinal function in adult critically ill patients: a modified Delphi consensus process (COSMOGI) - Critical Care
Purpose Gastrointestinal (GI) dysfunction is common in critically ill patients and associated with poor outcomes. There is a lack of standardised methods for daily monitoring of GI function. COSMOGI a...
162
Reposted by Kaspar Bachmann
Nate D @natedefelice.bsky.social · 17/12/2024
#MedSky #EMIMCC #IDsky #cardiosky @pulmcrit.bsky.social What is the culture at your hospital for potassium replacement? At most hospitals I'm at it seems to be a culture of replace to K 4.0 and the individual doctor makes the call. This paper confirms my bias... jamanetwork.com/journals/jam...
jamanetwork.com
Potassium Supplementation and Prevention of Atrial Fibrillation After Cardiac Surgery
This randomized clinical trial examines whether a lower serum potassium concentration trigger for supplementation is noninferior to a high-normal trigger for the prevention of atrial fibrillation afte...
1173
Reposted by Kaspar Bachmann
Romain Barthélémy @rombarthelemy.bsky.social · 10/12/2024
Results of the #SAHARAtrial released after #CCRdownunder. A lot to discuss… www.nejm.org/doi/full/10.... #neurocriticalcare
nejm.org
Liberal or Restrictive Transfusion Strategy in Aneurysmal Subarachnoid Hemorrhage | NEJM
The effect of a liberal red-cell transfusion strategy as compared with a restrictive strategy in patients during the critical care period after an aneurysmal subarachnoid hemorrhage is unclear. We ...
1104
Kaspar Bachmann @kasparbachmann.bsky.social · 12/12/2024
No, this was just a (very simple) approximation; imagine a trial that onlynlooked atbthis group and enrolled that number of patients. As the groups are exclusive, this was a first step to get a feeling of how much of a difference there really is…
000
Kaspar Bachmann @kasparbachmann.bsky.social · 12/12/2024
The immunocompromised group was stopped for futility. I think that these groups are not powered enough to draw any conclusion and the difference is just statistical noise. Early intubation may be warranted in these patients who can deteriorate rapidly, also for early diagnosis via bronch/lavage.
010
Kaspar Bachmann @kasparbachmann.bsky.social · 12/12/2024
I agree, for cardiogenic NIPPV treats the not only the lungs but also the heart; this is why the results were suprising to me and I struggle with interpretation.
100
Kaspar Bachmann @kasparbachmann.bsky.social · 11/12/2024
Awesome and interesting work by the late Luciano Gattinoni and colleagues. #emimcc
072
Kaspar Bachmann @kasparbachmann.bsky.social · 11/12/2024
CVP has the same function: it is the intersection of RV function curve and venous return, so it tells you how well the RV handles the VR and if there is RV dysfunction for that given state. It is no measure of volume status or responsiveness (but probably tolerance, as is VEXUS).
000
Kaspar Bachmann @kasparbachmann.bsky.social · 11/12/2024
I think of VEXUS as volume overload combined with RV dysfunction/failure. If you give a healthy person five liters of fluid, the rv might very well handle the venous return and vexus is low; only once RV dysfunction develops (either due to hypervolemia itself or for other reasons), VEXUS goes up.
100
Kaspar Bachmann @kasparbachmann.bsky.social · 11/12/2024
As an additional remark: Figure 2 shows the median estimated OR without borrowing, which line up better with frequentist results, although the superiority threshold was still not met. I wonder whether borrowing is valid for this group, as a cardiogenic pathology will differ from a respiratory one?
100
Reposted by Kaspar Bachmann
Nick Mark, MD @nickmark.bsky.social · 09/12/2024
#ADAPTtrial released & the results are interesting… Compared to SOC, Procalcitonin guided antibiotics in sepsis resulted in: - FEWER antibiotic days: 10.7 vs 9.8 (p=0.01) BUT - HIGHER 28 day mortality: 19.4% vs 20.9% (p=0.02) though this did not meet the pre-specified non-inferiority margin 🤔 1/
63413
Kaspar Bachmann @kasparbachmann.bsky.social · 11/12/2024
I sometimes feel that these patients are sometimes intubated "late", and some of them should be intubated earlier, i.e. the HFNO somewhat masks the severity of the disease. 16/end. That's it, let me know what you think and what your throughts/conclusions of the trial are.
media.tenor.com
a man in an orange suit says it 's a wrap with his hands
ALT: a man in an orange suit says it 's a wrap with his hands
010
Kaspar Bachmann @kasparbachmann.bsky.social · 11/12/2024
In my mind, HFNO is a very potent oxygenation and ventilation support system. This sometimes comes apparent when you remove it for a short time interval, and saturations drop and ventilation work increases rapidly. 15/n
media.tenor.com
a close up of a man 's face with the words `` wind 's in the east '' written above him .
ALT: a close up of a man 's face with the words `` wind 's in the east '' written above him .
110
Kaspar Bachmann @kasparbachmann.bsky.social · 11/12/2024
I will continue using High Flow as my primary approach with everyone other patient group. My personal experience is that if High Flow fails as part of a bundle including everything else we can offer in ARF, NIPPV will not provide benefit (apart from a potential preox tool for intubation). 14/n
110
Kaspar Bachmann @kasparbachmann.bsky.social · 11/12/2024
Crossover between groups (HFNO to NIPPV) was highest in the COPD group with respiratory acidosis (22%). Personally, I think this group (along with prior evidence), as well as the cardiogenic edema group, may benefit from NIV. However, the data in this trial don't support the latter approach. 13/n
120
Kaspar Bachmann @kasparbachmann.bsky.social · 11/12/2024
So, a frequentist approach shows significant differences. I understand that the borrowing in the Bayesian approach moderates this, as the differences in the other groups are much less pronounced & the threshold for superiority was high (>0.992). Still, I wonder how to interpret these results? 12/n.
110
Kaspar Bachmann @kasparbachmann.bsky.social · 11/12/2024
P-value (using a standard two-proportion z-test): Pooled proportion ≈ 43/272 = 0.158, Standard error ≈ 0.044, Difference in proportions = 0.1103 z ≈ 2.5, which corresponds to a p-value around 0.01–0.02. This is statistically significant at the conventional p<0.05 level. 11/n
media.tenor.com
a person is writing a math problem on a piece of paper with the words " when your bored of maths " below it
ALT: a person is writing a math problem on a piece of paper with the words " when your bored of maths " below it
200
Kaspar Bachmann @kasparbachmann.bsky.social · 11/12/2024
Absolute Risk Difference (ARD): 0.110 (11.0% absolute reduction,(95% CI: 2.4%, 19.6%) Relative Risk (RR): RR = (14/136) / (29/136) = 14/29 ≈ 0.48 (95% CI: 0.27 to 0.87) Odds Ratio (OR): Odds(HFNO) = 14/122 = 0.11 Odds(NIV) = 29/107 = 0.27 OR = 0.11475 / 0.27103 ≈ 0.42 [0.21 - 0.84] 10/n
media.tenor.com
a cartoon cat is standing in front of a triangle and a math problem .
ALT: a cartoon cat is standing in front of a triangle and a math problem .
100
Kaspar Bachmann @kasparbachmann.bsky.social · 11/12/2024
First, the numbers surprised me, as they went against what I would have expected. Second, I don't understand how these estimates are calculated. When choosing a simple frequentist approach (via ChatGPT o1) for this, we get significant results. 9/n
media.tenor.com
a baby is sitting on a couch holding a birthday present and eating it .
ALT: a baby is sitting on a couch holding a birthday present and eating it .
200
Kaspar Bachmann @kasparbachmann.bsky.social · 11/12/2024
Looking at the results from the cardiogenic edema group, we can see that 14/136 (10.3%) in the HFNO group reached it, while 29/136 (21.3%) reached it in the NIPPV group. The absolute and relative effect differences for that are non-significante (Table 2). 8/n
100
Kaspar Bachmann @kasparbachmann.bsky.social · 11/12/2024
I do not know much about #bayesian statistics beyond the general concept. However, I had some questions when reading the paper: maybe a #stats pro can enlighten me. I was particularly interested in the cardiogenic edema group, as I think NIPPV makes a lot of sense in this group. 7/n
media.tenor.com
a man with glasses and a beard says " it 's a confusing statistic "
ALT: a man with glasses and a beard says " it 's a confusing statistic "
210
Kaspar Bachmann @kasparbachmann.bsky.social · 11/12/2024
For example, in the largest subgroup (hypoxemic COVID-19), the primary outcome (intubation or death by day 7) occurred in 223 of 435 patients (51.3%) receiving HFNO and 210 of 447 patients (47.0%) receiving NIV. Among these events, the majority were intubations rather than deaths (34 and 17). 6/n
100
Kaspar Bachmann @kasparbachmann.bsky.social · 11/12/2024
I am always a bit sceptical of composite outcomes. It sometimes seems like a statistical Jedi Mind Trick. Intubation and death do not carry the same impact for patients and families, and combining them has some drawbacks. The main driver in this trial was intubation rather than mortality. 5/n
media.tenor.com
a man is saying good morning with his mouth open .
ALT: a man is saying good morning with his mouth open .
100
Kaspar Bachmann @kasparbachmann.bsky.social · 11/12/2024
The trial employed a Bayesian hierarchical model with dynamic borrowing across five patient groups. They used a composite outcome (intubation or death within 7 days) to capture both the need for invasive mechanical ventilation and fatal events as indicators of treatment failure. 4/n
100
Kaspar Bachmann @kasparbachmann.bsky.social · 11/12/2024
HFNO: flow (up to 60) & FiO₂ were adjusted to oxygen saturation (88-92% for COPD, else 92-98%). NIV used face masks with set inspiratory 12-16 & expiratory pressures 4-8 plus FiO₂ to maintain targets. The goal was to apply 24 hours of NIPPV. Both aimed to support breathing without intubation. 3/n
100
Kaspar Bachmann @kasparbachmann.bsky.social · 11/12/2024
The authors show in a large trial involving 4 predefined, exclusive subgroups with ARF (nonimmunocompromised hypoxemia, immunocompromised hypoxemia, COPD exacerbation, cardiogenic edema) the effect of high-flow nasal oxygenation vs. noninvasive ventilation. A COVID group was added later. 2/n
100
Kaspar Bachmann @kasparbachmann.bsky.social · 11/12/2024
The #Renovate Trial was just published. Kudos to the authors for this large trial. HFNO was non-inferior to NIPPV in 1776 patients with acute respiratory failure. I have some comments and discussions; feel free to join in. 👇 1/n #emimcc #ccr
281
Kaspar Bachmann @kasparbachmann.bsky.social · 11/12/2024
🙋 if there‘s a spot left, I would love to join.
000