Posts

Posted on

Must read article on evidence based medicine

This is worst Photoshopped pill bottle I have ever seen.

You don’t see evidence based medicine called out by name often in the lay-press, but Sunday’s Boston Globe dove into the problems with guidelines and quality of evidence in an in-depth analysis. This is a must read.

But evidence-based medicine is only as strong as the evidence used to support it. The stark reality is that evidence can be weak, biased, or even fraudulent. More guidelines stem from expert opinion than from solid clinical trial evidence. And many physicians who write these recommendations have financial ties to drug companies — with no widely accepted policy to correct for conflicts of interest.

More coverage at Cardiobrief.

Posted on

Cardiorenal Syndrome. Revised

I gave the cardiology fellows at St John Hospital and Medical Center a lecture on cardiorenal syndrome this morning. I revised and expanded the lecture I used for the residents:

It could still use a slide or two on the various loop diuretics and their uses.

We also had an interesting discussion on the data suggesting that loop diuretics maybe harmful in acute decompensated heart failure. I should include a couple of slides on that.

Overall a significant upgrade. You can find the lecture in the usual place.

Posted on

Best lecture ever

On the subject of lectures, if you haven’t seen Robert Lustig‘s lecture on Sugar: the Bitter Truth, stop what you are doing and find 90 minutes to watch it. Fantastic lecture. Amazing how good his lecture is with such pathetic slides. It’s not about the deck.

The New York Times is running an excellent profile of the lecture (Really? Sunday New York Times Magazine profile piece on a lecture?)  this Sunday. Gary Taubes wrote the profile, he is the same author who wrote the best article I ever read on the science and politics of the Atkins diet back in 2002.

Take the time to check out the lecture and the profile.

Posted on

IVF and Sodium handout updated

I rewrote large sections if the IVF and Sodium handout. I added Tea and Toast syndrome, more on SIADH and a lot of interactivity. Overall a huge improvement. I used it today with the Beaumont Hospital Family Practice residents and was really pleased. Find the update in the handouts tab.

Posted on

Patient list

Yesterday:

  • My first patient had SIADH and a sodium of 125
  • My last patient had nephrogenic diabetes insipidus and a sodium of 150
The statistician in me, looked at the patient list and concluded, normal sodium.
Posted on

Lecture on how to give a lecture

This post really resonates with me: The Ten Commandments of PowerPoint and I wish that I had included it in my lectuer from last week on how to give a lecture.

The chief resident at St John Hospital and Medical Center asked me to do morning report on giving better presentations. It was an interesting project. I have been pretty busy and didn’t have enough time to put together a really polished presentation, but this is what I came up with.

Here is a link to the PDF and Keynote file (130 mb)

iWork documents are a little wonky if you are not using Safari. So the videos I embedded in the lecture are below if you are having trouble looking at them.

Steve Jobs tells it how it is regarding Microsoft

The birth of a morning report:

Screen captures with command-shift-4:

Smart builds

Highlight text:

Mask an image

Improve a crappy figure:

Posted on

When the GFR is zero how fast does the creatinine rise?

 

How do you get a GFR of zero?
Bilateral bathtub nephrectomy
NephrO-kleptO-mainia

In my clinical experience as the GFR approaches zero the creatinine goes up between 1 and 2 mg/dL everyday.

However I was working out a story problem for an acute renal failure and when I calculated how much the creatinine would rise it was 3.3 per day. Here is how I calculated this:

  • Total body creatinine: 420 mg
    • This assumes that creatinine is equally distributed through out total body water. So 42 liters (60% of 70kg) times 1 mg/dl times 10 dL per liter
  • New creatinine: 1400 mg
    • 20 mg of creatinine generation per kg body weight, 70 kg body weight
  • New total total body creatinine 1820 mg
    • add the first two figures
  • New creatinine: 4.33 mg/dL
    • Divide the total body creatinine (1820 mg) by total body water (420 dL):

Did I do my calculation wrong? The total body creatinine calculation of 420 mg seems awfully low, especially if muscles create 1400 mg of new creatinine everyday.

Picture by The Doctr

Posted on

Index of my posts on membranous nephropathy, renal biopsy and pregnancy

  • The Question: When do you biopsy pregnant patients? A review of Clara Day’s NDT article on renal biopsy in pregnancy.

  • The Survey: Request for doctors to share their practice patterns on decision making regarding pregnant patients with renal disease.

  • The Plea: Request for more participants in the survey practice patterns.
  • The Research: a long, somewhat unstructured, bibiography of the literature.

  • The Conclusions: A summary of my findings and conclusions. This includes the results of the survey.
Posted on

Membranous nephropathy and pregnancy

I promised a full examination of the renal biopsy in pregnancy.

This is the longest post I have ever done and so I have segmented it into two parts:

  1. My conclusions along with  discussion of the survey of clinical decision making
  2. The supporting data, which is structured as a detailed annotated bibliography.

My conclusions:

The renal biopsy is as safe in pregnancy as it is outside of pregnancy. Ultrasound guidance and modern understanding of coagulopathy paired with aggressive use of antihypertensives makes this an acceptable risk in pregnancy.  This is a field of medicine that accepts risk to the fetus to in the name of better diagnostic data, see amniocentesis.

As long as the blood pressure is controlled, and the patient is not coagulopathic, if you need the biopsy, get the biopsy.

However, I have not been fully satisfied that the diagnostic data is that useful in most situations. The therapeutic options in pregnancy seem pretty narrow:

  • IV methylprednisolone with or without azathioprine
  • Therapeutic abortion
  • Delivery of the infant prematurely
  • Symptomatic therapy

Given those options, a biopsy will not make a difference in many situations.

Timing becomes critical as a therapeutic abortion and delivery of the infant both have narrow windows of opportunity. The decision to use symptomatic treatment will not be influenced by histologic findings.

That leaves IV methylprednisolone. In every case report I read where methylprednisolone +/– azathioprine was used I thought the decision to use it coud have been made solely from the patient’s symptoms and that the biopsy results didn’t really influence that decision. In every case the mother had increasing blood pressure, creatinine and proteinuria 20+ weeks into the pregnancy. It was too early to deliver the baby and steroids were used to buy time for the fetus to grow prior to delivery.

Here are the situations I would recommend doing a biopsy:

  • First trimester with early signs that the patient may be in for a rocky pregnancy due to renal disease. There are three risk factors: nephrotic range proteinuria, renal failure with a creatinine over 1.5 and hypertension. Getting a pathologic diagnosis at this point could allow termination of the pregnancy if any frightening diagnosis are made, i.e. lupus nephritis, scleroderma, crescentic GN, possibly MPGN. Outside of those findings I doubt the biopsy will actually change your therapy. I am sympathetic to the argument that each of these diagnosis has pretty good serologic tests that could decrease the need for the biopsy. That said the decisions are weighty enough that I would prefere a tissue diagnosis.
  • The only other situation would be a patient with symptoms suggestive of scleroderma presenting late in pregnancy, 20+ weeks with high blood pressure. The right choice here is probably terminating the pregnancy. The mother’s life is in danger and the condition will not respond to steroids, the only therapeutic option available. I would like to get a tissue diagnosis to confirm prior to proceeding down that road.

In the survey I presented I had 6 different clinical scenarios, they are summarized below:

  1. Gestational age 12 weeks, nephrotic syndrome, normal BP, normal Cr
  2. Gestational age 24 weeks, nephrotic syndrome, normal BP, normal Cr
  3. Gestational age 12 weeks, nephrotic syndrome, normal BP, elevated Cr
  4. Gestational age 24 weeks, nephrotic syndrome, normal BP, elevated Cr
  5. Gestational age 12 weeks, nephrotic syndrome, elevated BP, elevated Cr
  6. Gestational age 24 weeks, nephrotic syndrome, elevated BP, elevated Cr
So I would recommend a biopsy in case 3 and 5, and think long and hard about case 1. I wouldn’t biopsy 2, 4 or 6. I would start steroids in case 6 and consider it in case 4. I didn’t vote this way, as my opinions have changed the more I learned and thought about this. Of note, the only maternal-fetal medicine doctor voted exactly as I just described. 
As of Sunday April 3rd we had 53 votes:
81% are nephrologists. I guess we know who reads PBFluids.
And 76% are from North America. And that compares favorably to the overall traffic to the blog:
from Google Analytics
Mexico represents 0.8% of blog traffic but showed up big time for 4% of the sample.
Here is how the community voted:
  1. Gestational age 12 weeks, nephrotic syndrome, normal BP, normal Cr.
    I voted BIOPSY, the community:

  2. Gestational age 24 weeks, nephrotic syndrome, normal BP, normal Cr.
    I voted NO biopsy, the community:
  3. Gestational age 12 weeks, nephrotic syndrome, normal BP, elevated Cr.
    I voted BIOPSY, the community:
  4. Gestational age 24 weeks, nephrotic syndrome, normal BP, elevated Cr.
    I voted NO biopsy, the community:
  5. Gestational age 12 weeks, nephrotic syndrome, elevated BP, elevated Cr.
    I voted BIOPSY, the community:
  6. Gestational age 24 weeks, nephrotic syndrome, normal BP, elevated Cr.
    I voted NO biopsy, the community:

The thing that confuses me is why the percent of people that want a biopsy at 12 weeks seems to fall as the patient gets sicker: 49% in case 1 to 34% in cases 3 and 5. What happened to 8 of the pro-biopsy doctors?

The same puzzling trend held for the 24 week gestational age with the biopsy rate being highest, 81%, in the healthiest and as the patient developed decreased renal function and hypertension it fell to 64% and 57%.

I broke down the data for attendings versus fellows:

And now you can see where those missing biopsies for the patient at 12 weeks went. Over half the fellows were willing to biopsy the asymptomatic nephrotic syndrome at 12 weeks. But the bravado disappeared if you added a bit of hypertension (scenario 3) with the biopsy rate falling below 20%.

The attendings were more likely to biopsy the patient in just about every scenario with an overall biopsy rate of 59% versus 49% for fellows.

In the U.S. versus the world:

It looks like a general trend for doctors outside of the US to be less likely to biopsy patients, at least in the first trimester.
Here is a link to the raw data in case anyone is interested. And yes I did see the fellow from Lebanon double voted, but given the importance of Chicago in the study of both renal disease in pregnancy and glomerular disease, I figured a vote early, vote often policy was appropriate.
Look here for the link to the second post with the annotated bibliography.