Month: September 2013

Every year the National Kidney Foundation of Michigan’s Scientific Advisory Counsel puts together the Ronald Easterling Memorial Lecture. Ronald Easterling was a Michigan nephrologist who was instrumental in the establishment of dialysis. He worked for years at the University of Michigan and kept meticulous records of dialysis patients which became the prototype of the USRDS. His namesake’s lecture is a fund raiser for the NKF of Michigan.

This year I had a role in lining up the speaker and we hooked Dave deBronkart, i.e. ePatient Dave. Dave just recently spoke at MedX and you can get a feel for his view on medicine from his TEDx talk (yeah, we got a TED-level speaker for the Easterling!)

It should be hugely entertaining. If you are interested in the intersection of social media, patients and medicine I would highly recommend you come to the Easterling at Townsend in Birmingham, Michigan.

When the economic stimulus package passed and the basic blue print for meaningful use started to be hammered together, my practice committed to making the jump from paper to an EMR. Through blood sweat and tears (mostly tears) we implemented an EMR program in the first year that funds were available and we successfully qualified for meaningful use. We threw off the shackles of our paper chart and embraced the shackles of our EMR system, Allscripts (in case you wanted to follow us down this boondoggle).

It is hard work moving a large, multi-site practice across this chasm and the federal money really helped because it was expensive, more expensive and more complex than just about anything else our practice has accomplished, i.e. we did a heart transplant on our out-patient practice without taking taking a day off work.

After qualifying for Meaningful Use Part One for two years we were well on our way to qualifying for meaningful use Part Two when we hit a massive speed bump.

I maintain the medical records on the patients I see in clinic. I contribute to but do not maintain the medical records of the patients I see in the hospital. Those are maintained by the hospital. It makes sense for the hospital to take charge of the medical record and because I am only one of many doctors taking care of each patient. In addition to doctors there are pharmacists, nurses, physical therapists, etc that all share the medical record. So when meaningful use tracks how compliant I have been in using my EMR, they wisely don’t include the patients I see in the hospital as part of the denominator:

I also see a lot of my patients in the dialysis unit. Part of the reason so many of the visits are in the dialysis unit is that the federally implemented a payment structure encourages four visits a month. So I strive to go four times a month. These dialysis visits represent a large proportion of my monthly non-hospital-based patient encounters. Dialysis is truly the third leg that supports our practice.

Note: the feds look like they are right on this one with improved outcomes seen with longer time spent with the patient.

Just like in the hospital, I contribute to the medical record in the dialysis unit but I do not control it. It is a shared resource used by the nurses, dietitians, social workers and technicians, who all contribute to patient care and the medical record. This medical record is newly electronic. I work in Davita facilities
and I have been blown away by the simplicity, speed, and ease of use of their Falcon EMR. (I was not paid to say that, though I jointly own a dialysis unit with Davita.)

Unfortunately, that EMR is not certified for Meaningful Use, so not only do I not get credit for using it, all of those patient encounters are counted in my denominator against my office use of the EMR dragging me down so even with 100% compliance in the clinic I am unable to qualify for Meaningful Use. I not only forfit the incentive money but I will start to be penalized for not adopting an EMR. I will be punished despite, investing in and using an EMR for all of the outpatient visits in my clinic. The only outpatient visits that I don’t use the EMR for, are the dialysis visits, and even in that situation I use an EMR.

This feels incredibly stupid, mean spirited and counter productive. Please Davita, get Falcon approved for Meaningful Use ASAP so this mindless punnishment does not befall all neprhologists. Please CMS grant nephrologists hardship exemptions so that dialysis unit visits are excluded from the denominator calculation just like hospital and emergency room encounters are.

BTW: I have been told that Fresenius’s in-house EMR is likewise not approved for meaningful use. Lunacy.

Awhile ago I posted this tweet to twitter:

RT if you have ever given a patient money so they can buy their medications.
— Joel Topf (@kidney_boy) March 7, 2013

It received 24 retweets and a lot of twittversation around it. A number of doctors give money to patients when they are in need. Too many times the safety net is porous or too difficult for unsophisticated patients to navigate. This week-end I saw this tweet:

Great article in JAMA by physician reprimanded for giving $ for Rx to patient who couldn’t afford to pay http://t.co/7zIchqb1W7 #meded
— Margaret Chisolm (@whole_patients) September 28, 2013

The article is really good and is worth a read.

When I was just out of fellowship I saw a young patient in the emergency department with hypokalemia and severe hypertension. We sent an aldo and renin and treated her blood pressure with a number of medications but not spironolactone. I was sure he was going to have primary hyperaldosteronism and starting an aldosterone antagonist would complicate getting selective adrenal vein sampling to diagnose an adrenal adenoma. I set her up for a follow-up appointment and discharged him. He never made it to the follow-up visit. I didn’t see him again for years but when he re-presented he still was hypertensive but his kidney’s were thrashed. Half a decade of uncontrolled hypertension had taken their toll.

I asked him about why he didn’t follow-up. Turns out he lost his job a few weeks after being discharged and didn’t have any medical insurance. I regret not giving him the spironolactone prescription.
I know that for $0.13 a day I could have had better blood pressure control than with whatever concoction anyone else threw at him.

He eventually went become one of my dialysis patients. He was morbidly obese and was rejected by the transplant team for that reason. We were discussing what to do next. I was trying to convince him to take charge of his life and lose the weight. He kept making excuses about why he couldn’t. At one point he said that he didn’t have enough money for a treadmill. I pulled up Craig’s List and found a an $80 dollar treadmill for sale I told her her could probably buy it for $40 and gave him $20 to get him halfway there. I know my guilt from the earlier miss influenced my behavior but mainly I was trying to shake him out of dialytic nihilism.

Soon after that, either I switched shifts or he changed units and we lost contact.

The other day I was talking to our transplant nephrologist and he told me that the patient was touched by that gesture. The act of pulling out my wallet and throwing down some cheddar made him feel that he needed to recommit and give a full throated effort to lose weight.

He was successful. He lost the weight, received the kidney and is off dialysis.

Best $20 I ever spent.

I gave my fructose and uric acid lecture yesterday. Here are the slides from that lecture.

Is it just me or does the science seem to be moving slow in this field. Where are the interventional trials with febuxostat? Where are the larger trials on allopurinol with more than 50 people in each arm?

PDF (47 mb)
Keynote presentation (78.1 mb)

Nice update from twitter:

2. @kidney_boy. ..The CKD-FIX trial will have N = 620; https://t.co/snyFTHyxxm
— swapnil hiremath (@hswapnil) September 28, 2013

For the second year I attended Med 2.0. Just like last year I covered the conference for eAJKD. It is a fascinating look at the influences of social media, the internet and mobile technology on medicine and health. Please take a look at these posts:

• Why Medicine 2.0
• Susan Michie and the science of changing behavior
• Twitter: listening to a million conversations at once
• My very short time on stage at Med 2.0
• Technology for the elderly
• Closing Keynote, the present and future of sensors

Great blog post with video of Sherry Pagoto’s talk, Hashtag your way to Health.

The best way to revisit the conference is through the twitter stream. Symplur has all the tweets archived here.

The twitter analytics are mind blowing:

Compare that to Kidney Week ’12 which felt about 10 times the size of Med 2.0:

Gr8 quest! Lets ask @kidney_boy @deanoburns “When following UO in morbidly obese&using 0.5ml/kg/hr do U use actual,ideal or adjusted weight?
— Haney Mallemat (@CriticalCareNow) August 29, 2013

The short answer is just use the patients mass. No need for ideal or adjusted weight.

The long answer follows: To answer this I’m gonna to proceed from first principles. Why do we measure urine output? It is the most fundamental and basic measures of renal function. Before there were MDRD, creatinine or a Chem-7 we could measure the urine output. A biochemical urinalysis meant tasting it.

Oliguria means the urine output has fallen below the minimal healthy amount. (NEJM Review) So how much urine do people need to make to stay healthy? The flippant answer is enough, enough for what? Enough to stay in balance. Enough that every drop of water that is ingested is then excreted. Enough that every microgram of solute is excreted.

In regards to water, some will be lost with respiration, some in stool and some through the skin, but whatever is left over must be excreted by the kidneys. If water is not excreted and accumulates patients develop hyponatremia.

In regards to solute, proteins, electrolytes and minerals are constantly being absorbed by our indiscriminate digestive tracks. All of them most be accounted for and must either be incorporated into the body (growth) or cleared out, most of this work devolves on the kidney. If solute is not excreted it will accumulate and we will see increases in BUN from protein, edema from salt accumulation and other symptoms from the infinite variety of solutes we take in.

Obviously, accumulation of solutes is the factor whose symptoms better correlate with out definition of renal failure, so oliguria should be defined as the urine output, below which, people begin to accumulate solute. This volume is highly dependent on the diet. If the patient is eating a tiny amount of solute they will be able to have a very low urine output without beginning to accumulate solutes.

But the definition of oliguria is not built for these edge cases, it is designed to answer the question, what is the minimum amount of urine a person could make and still get rid of all the solute of a regular diet.

This math is pretty straight forward:

The definition of oliguria is 400 or 500 mL per day. The 583 on a normal diet is pretty close. If acutely ill patients decrease solute intake their minimal urine volume will creep down to the established 400-500 mL limits.

But this sets up the question, “Does urine output matter? Is it superior to biochemical assessments of renal function? Does the fact that I can order a serum creatinine every 12 hours make following hourly urine output unimportant?”

The answer is No. Urine output provides additional information on renal function that can not be seen with serum creatinine. Urine output increases the sensitivity of the RIFLE and AKIN definitions of AKI without compromising specificity. Increased sensitivity means that when combing through a cohort of patients, the addition of urine output will roughly double the number of patients flagged with AKI. Maintaining specificity means that clinical outcomes of these patients will be roughly identical to patients who meet AKI criteria due to changes in creatinine.

Sensitivity data: the use of oliguria roughly doubled the incidence of AKI from 24 to 52% of this cohort. While specificity was maintained:

In this multi-center, prospective study, however, they found differing conclusions. Though oliguria was associated with AKI from Cr, the relationship was to lose to be clinically useful. Oliguria was frequent, they found it in one third of ICU days. This resulted is a high false positive rate of AKI. They also found that half of patients who develop AKI by creatinine did not have oliguria for 4 or more hours the day prior and a quarter did not have oliguria at all. Only when they increased the duration of oliguria to at least 12 hours did the likelihood ratio rise above 10 (necessary for a clinically useful test) but this dramatically dropped the specificity of oliguria to the point that it would miss over half the cases of AKI.

So the addition of oliguria does add sensitivity without additional loss of specificity but in no way can oliguria displace creatinine as a reliable and early detector of AKI.

Looking through the data collected on using urine output to define AKI and predict outcome, they do not use ideal weight or adjusted weight, just straight mass. It would be interesting for someone to re-run the analysis on the previously collected data using ideal or adjusted weight and see if oliguria became a better biomarker of acute kidney injury.

What a crappy text book, it doesn’t even have “oilguria” in it. Who wrote this piece of…? oh. pic.twitter.com/U1JRBLji89
— Joel Topf (@kidney_boy) September 6, 2013