My SGLT2 inhibitor grand-rounds lecture

This past autumn I was invited to give a lecture at the Michigan ACP. I love that meeting. I decided to talk about the mortality data on SGLT2 inhibitors and how we got that data and how curious that data is. Then, last week I had the opportunity to give the lecture again for grand rounds at St John Hospital and Medical Center. Here it is in 4 chapters:

Chapter 1: The History and top line CV outcomes

Chapter 2: What’s driving the improvement in outcomes?

Chapter 3: Renal outcomes

Chapter 4: Side effects and conclusions

Here is the Keynote presentation for your editing pleasure: Keynote (278 mb)

In response to chapter 2, Matt Sparks had this interesting tweet:

This brings up an important point. One of the most intriguing slides is the one below that looks at how long it takes for the Kaplan-Meier to separate.

With glycemic control and blood pressure interventions, their is notable lag, but with the SGLT2i drugs the lines diverge from the very first dose. We also see that pattern with ACEi in heart failure and aldosterone antagonists in heart failure. This may be a clue of where to look for the cause of the survival advantage.

RALES (3 months)

CONSENSUS (first dose)

I will be adding this slide to the next version of the talk.


List of therapies that reduce cardiovascular mortality in diabetes

I’m giving grand rounds on Tuesday on SGLT2 inhibitors and I’m trying to come up with a list of therapies that lower CV death in diabetes.

Here is my list:

  • Blood pressure control
    • UKPDS
    • ADVANCE All-cause mortality was reduced with a near miss on CV mortality (P=0.041)
  • Empagliflozen
  • Canagliflozin
    • CANVAS Only partial credit here. CV death was part of the composite outcome, but CVD was not significant on its own
  • Semaglutide
    • SUSTAIN-6 Weak. Hit the primary outcome but CV death was explicitly identical between groups
  • Liraglutide

Drugs that have run the FDA CV disease gauntlet and that are non-inferior to standard of care:

  • Exanatide
  • Rosiglitazone
  • Pioglitazone
  • Alogliptin
    • EXAMINE (This is a secondary prevention trial. As far as I can tell it is the only FDA mandated outcome trial that is specifically designed as a secondary prevention. Not sure why.)
  • Saxagliptin
  • Degludec


I’m sure I’m missing some. There must be a statin trial of diabetics. Right?


Swapnil was first with the statin answer:

And Edgar came up with a great visual from a review paper:

And Szymon came up with the Steno trial. I can’t believe I forgot about that one.


NephTalk: A new nephrology podcast by Satellite Health

Satellite Healthcare is a non-for-profit dialysis company. They partnered with NephJC to do Bloggger’s Night the last three years and sponsor the NephJC Kidneys. This year they launched a Podcast, NephTalk. I was lucky to get invited to help out. I have hosted one, an interview with Sumi Sun about preventing blood stream infections. Here is her abstract from Kidney Week:

Background: CVCs are associated with catheter-related bloodstream infection (BSI) resulting in increased morbidity and mortality. Following our report of significantly reduced infection when 320 μg/mL gentamicin in 4% citrate is used as the CVC locking solution (Moran AJKD 2012), this has remained the standard of care in patients dialyzing with a CVC, unless physician order requested otherwise. The infection rates were monitored through an internal QC program developed for National Healthcare Safety Network (NHSN) reporting.

Methods: This study evaluated NHSN data with self-reported infection rates from January 2014 to December 2016 in a non-profit dialysis provider with a total of 57 free-standing dialysis facilities serving more than 5000 HD patients. BSI was reported according to NHSN criteria. Data were audited through comparison to an internal infection control report and discrepancies reconciled prior to final NHSN submission. Blood cultures were mandated before any antibiotic administration for suspected BSI, and 85% or more are sent to one internal lab (Ascend).

Results: The rate of catheter-related bloodstream infection over the three years was 1.00 episodes/100 patient months, 54% lower than the national average of 2.16 for CVC-related BSI (2014 NHSN BSI Pooled Mean Rate/100 patient-months). Monthly BSI rates showed minor fluctuations, however none exceeded the national average in any given month.

Conclusion: Gentamicin 320 μg/mL in 4% sodium citrate as a routine catheter lock demonstrated sustained low CVC-related BSI rates in HD patients, with approximately half the infection rate compared with the national average. Gentamicin-citrate lock should be considered the standard of care in patients with CVC access.

PodCast: Curbsiders #69

I was invited by the Curbsiders to talk about CKD. The discussion went a little long and our discussin got divided into two podcasts, #67 and #69. Here is the second half. I don’t think I made any major mistakes except when discussing combined ACEi and ARB therapy I said ALTITUDE was a study of RAAS inhibtion and endothelin antagonists. Actually ALTITUDE was RAASi and Aliskiren, the direct renin antagonist. And it was stopped not because of hyperkalemia but due to a CV signal (though the combination did have more hyperkalemia). The trial I was thinking about was ASCEND which added the endothelin antagonist avosentin to an ACEi or ARB in patients with diabetic nephropathy. This trial was also stopped early, again for CV issues after only 4 months on the drug!

You can listen to the episode here.

Mistakes in medical education social media

I am in page proof hell.

The two year slog from from gleefully saying “Wow, that sounds like fun” to a published book is wrapping up. I am working with Edgar Lerma and Matt Sparks on Nephrology Secrets fourth edition. It is an amazing amount of work.

Yesterday I was proofing a chapter and found a pretty profound error.

Not a typo.

Not a misspelling.

Not an awkward turn of phrase.

This was a hardcore, error-of-fact that would have confused readers that didn’t know better and cause significant loss of authority for the book by the readers that did.

This error had travelled a long and perilous editing river to finally arrive intact at the final page proof.

I don’t know how the authors proofed their own chapter, but I assume it went through multiple drafts and rewrites. Then the chapter was sent to the editors and for Secrets, each editor read and commented on each chapter. After each editor the chapter was sent back to the authors for revision. This rinse, wash, and repeat went through three cycles. One with each editor. Then the text was turned over to Elsevier and they converted it into a book. The publisher returned proofs to the authors with specific questions that came up during the page layout. Another independent set of eyes. And then the authors signed off on the proofs.

And after all of that I found the error. A significant error.

This error came within one-step of being a permanent, written-in-ink error in the book. That chain of revisions and proofs is what makes books as good as they are. What type of checks are there in social media delivered medical education? How do we assure that the lectures and pearls we push through our blogs and tweets do not contain subtle (or not so subtle) errors. Very little social media has anywhere close to the editorial infrastructure that an Elsevier textbook enjoys.

In my post about Kidney Week I received three different DMs and @s notifying me of 3 different typos and misspelling. Fix and move on.

Typos are easy. There is more embarrassment than ego in those mistakes. Mistakes of content are harder to accept. The instinct is to defend our work, push back against the unsolicited peer reviewer. But we need to keep our ears to the crowd and our minds open so that…

If we are wrong

It is not for long.

Because the strength of social media is using Linus’s Law to uncover mistakes and then it is up to us to put away our egos and make it right.

given enough eyeballs, all bugs are shallow

I really feel the success of FOAMed depends on the crowd notifying authors of mistakes and then the authors fixing those mistakes. A failure on either side of that equation (either side means that if you are reading FOAMed and see a mistake you have an obligation to point it out. Noticing a mistake in medical education material and moving on without notifying the author is like seeing a discarded pistol by the playground and not telling the authorities. That dead seven-year-old is on you) and FOAMed becomes a joke as it morphs into a minefield of crappy, error filled resources.

The Nephrology Social Media Collective internship is accepting applications

The NSMC internship was created a few years ago to assure that nephrology had a surplus of dedicated people creating varied, compelling, and creative FOAMed resources. In the early years of Social Medicine, I believed that digital natives would join #NephTwitter and quickly and effortlessly create original content without guidance.

The medical school class of 2016 began high school the same year Facebook was launched. They literally have been using social media for their entire adult life.

However after a few years I became impatient with how fast young doctors were joining #NephTwitter and wanted a way to incubate these digital natives so that they could become contributors quicker.

The Digital Native Myth is the belief that young people will intuitively understand how to contribute to FOAMed.

The 90, 9, 1 rule is that for any social media experience 90% of people just consume, 9% will comment on other people’s content, and only 1% will contribute content. The idea behind the NSMC Internship is to move as many people as possible to the right. We want to move doctors from the 90% to the 1%.

A few weeks ago in New Orleans we graduated our third class of interns.

This class was our largest yet and some of the graduates are already standouts on #NephTwitter.

The interns provide us with detailed feedback to further develop our internship. Based on that feedback we are going to be adding a new module to the curriculum to teach interns how to build a website as part of a complete social media presence. One of the roles that our graduates have repeatedly been asked to fill, is to develop a social media presence for various projects. This may be for a large study, or an institution like a division or fellowship program. Our feeling is that many programs created twitter handle and that is where they stop. In order to have a real social media presence, Twitter needs to be backed up by by a website and a blog. We are now going to teach our interns how to do that in either WordPress or SquareSpace. Additionally we will teach them how and when to use Medium.

So drop an application and join us. It will be great.

In defense of live Tweeting at meetings

Skeptical Scalpel has re-upped his on going war on Live Tweeting or, as Alex Djuricich used to say, Tweeting the Meeting.

Scalpel and I have sparred about this in the past, and I have flipped and flopped quite a bit. Here is my first post on live tweeting:

And the flip flop

If you look at the blog posts linked in that second post you will see posts written largely from the notes created and archived on Twitter. For a more contemporary example take a look at the post I wrote about the late breaking and high impact trials at this year’s Kidney Week. This post uses the Tweets written during the session to tell the story of the session. I think it does a far better job of conveying the mood of the audience during the session than other posts about the session that I have read.

My other post for KidneyWk (#KidneyWk: Serum and Dialysate Electrolytes and the Risk of Sudden Cardiac Death) was also based on tweets but since there were fewer tweeters in the audience the post relies on their voice less and reads like a more traditional post. But don’t be fooled, it was also largely a product of the notes I took as a damned live tweeter.

Kidney Week is facing an existential crisis

Painting faced an existential crisis with the invention of photography. The platonic ideal of making a picture that fully represented reality was made meaningless by the invention of photography. The goal of painting pivoted from realism to impressionism, surrealism and subsequent movements. Painting faced its crisis by changing the definition of what it strived for. And to be clear artists that spent their career mastering realism didn’t cover to the new movement. New artists came along to fill those niche that had growth and potential despite photography.

Medical education is facing a change in the way students digest it.

Go to a pre-clinical lecture at any medical school in the US and you can find yourself a seat. It doesn’t matter the size of the lecture hall or the size of the med school class, students don’t go to class. Lectures are videotaped and can be streamed. Students gravitate to the most efficient way for them to learn and surprise, it is not sitting in lectures watching slideshows. If the students likes slide shows they will watch it at double speed with the PowerPoint open in front of them, available for them to make notes. It is nothing like the lectures I attended in med school.

Slideshows are tired, allow and inefficient. Students are routing around them. Good for them. I love that students are finding ways to learn in ways most efficient to each individual’s mind.

ASN’s problem is that Kidney Week is slideshow-learning. Selling slideshow-learning to millenials who don’t go to slideshows even when they paid tens of thouands of dollars is going to be like selling sand in the Mojave.

ASN has to make like 19th-century painting and pivot to a different platonic ideal.

What do I mean?

I mean a Shark Tank with real money on the line

I mean TED-like talks

I mean an American Idol-like competition for the best educational lectures.

Have educators around the country compete to make the best lecture on the same subject. Have them compete with the finals at Kidney Week. Rotate the subject every year, but make it a fundamental lecture that everyone needs to give:

  • Acid-base
  • Hyponatremia
  • CKD
  • AKI
  • Dialysis for the internist

Pathology lecture that uses microscopes rather than slides


Biopsy training with cadavers

Ultrasound training with patients

A recreational run

With pre- and post- sodium levels and AKI biomarkers

Cooking classes

Panel discussions with patients


ASN is using a fifty year old model of what makes a medical meeting and if they don’t adapt there will be fewer and fewer, and older and older, people attending. The ER and critical care crew have layed the blue print with SMACC, nephrology should be the first specialty to follow it down the rabbit hole.