Posts

Posted on

Michelle Bachmann taking up the anti-vaccine crusade

Earlier this week Congressperson Bachmann took a crack at the HPV vaccine, saying that she heard from some person that her daughter had become a retard after getting vaccinated.

“There’s a woman who came up crying to me tonight after the debate,” Bachmann said after the debate, where she had told Perry on stage that she was “offended” by his decision. “She said her daughter was given that vaccine. She told me her daughter suffered mental retardation as a result of that vaccine.” The Telegraph

It’s unfortunate that people who don’t die of cervical cancer never realize it and unable to talk with presidential hopefuls. I dream of day when nephrology fellows won’t get calls in the middle of the night to handle potassiums of 8 caused by cervical cancer associated bladder outlet obstruction.
Interesting infographic on the HPV vaccine and cervical cancer. Click here for a readable version.
Posted on

Outpatient hyponatremia

Almost all of the hyponatremia I see is inpatient, but this week a woman was referred to my clinic with a sodium of 128. She has a sharp family doctor who ordered all the right tests. Here are the key pieces:

  • Plasma sodium 128
  • Plasma osmolality 277
  • Urine osmolality 180
  • Urine SpGrav 1.005
  • Urine sodium 14
She has a history of hypothyroidism but her TSH was over suppressed, no hint of hypothyroidism. She was not taking any diuretics. She was on an SSRI that could cause SIADH but the low urine osmolality and low specific gravity argue against excessive ADH activity.
The case hinges on the low urine osmolality. This is a rare case of ADH-independent hyponatremia. All of the major causes of hyponatremia (volume depletion, diuretics*, heart failure, SIADH, etc) are driven by ADH which prevents the kidney from clearing free water. The low urine osmolality indicates the kidney is not under the influence of ADH and doing what it needs to in order to correct the sodium, i.e. excreting excess water in the form of dilute urine.

I believe there are only a few causes of ADH-independent hyponatremia, and only two occur with any regularity:
  • Psychogenic polydipsia
  • Tea and toast syndrome
  • Reset osmostat (rare)
When the kidney is making dilute urine and the patient has hyponatremia the problem is not in the quality of the urine, which is appropriate, but in the quantity of urine, which is inadequate.

She described her diet as a peanut butter and jelly sandwich for breakfast, some fruit and juice with some nuts for lunch and nothing for dinner. Her fluid intake was high (4-5 cups of coffee, a large water bottle of water, some juice, some soda, and a couple of additional glasses of water) but no where near enough for psychogenic polydipsia.
I suspect she has tea and toast syndrome. I am checking a 24-hour urine osmolality to gauge her daily osmolar load and then plan to have her increase the amount of protein and minerals in her diet while trying to taper her fluid intake.
More on tea and toast syndrome here
A well done case report on a healthy patient wih reset osmostat can be found here (pdf)
* if the diuretics are still active the urine osmolality could be low but once the drug wears off the urine osmolality will climb
Posted on

New kid on the block: The Kidney Doctor

I first met Ajay Singh when he came to St John Hospital as part of a symposium on chronic kidney disease in 2004 or 2005. It was a great meeting and Singh gave two memorable lectures.

The first was a dismantling of the MDRD equation as an accurate measure of GFR. He was speaking against an equation that was way better than a simple creatinine but had some real problems, especially when used in patients without kidney disease. It was a inflammatory and a bit wonky for a conference directed to primary care doctors. Here we, the local nephrologists, were trying to get our doctors to recognize occult CKD by abandoning serum creatinine in favor of the superior eGFR and then the invited expert comes in and tells them how stupid this is.

His second lecture was the correction of anemia dog-and-pony show. He gave an amazing and persuasive presentation in favor of correcting of anemia in renal disease. Though the data was all retrospective and observational it was clear that Dr. Singh was personally a few steps past equipoise. At the time CHOIR was in full swing recruiting and retaining patients and my group was part of that process as a research site for CHOIR.

Five or so years later he returned to talk with our fellows and staff regarding anemia. This was after the publication of CHOIR, but I believe before the release of TREAT, though my memory is a bit hazy on the timing.

What I do remember is that he talked about the dangers of correcting anemia and the lack of data supporting its use. I remember being so angry. I felt that for the last half dozen years I had worked to convince my CKD patients that they needed to enroll in our anemia clinic, needed to come to our office for EPO shots and iron infusions, and that all this would make them feel better, protect their heart and prolong their life, all purported advantages of ESA therapy. And now Mr. Harvard returns and tells us that this is wrong, without ever apologizing, without even mentioning how he’d jumped the fence.

I stopped him mid-lecture and told him that the last time he’d been in Detroit he’d been telling us how important it was to treat anemia and now he had completely changed positions. Dr. Singh paused, looked at me, and gave the best answer possible. I can’t remember his exact words, so I’m paraphrasing here,

“The data has changed. Now we know more and what I’m telling you is what is currently correct. In medicine, there is no room for intellectual loyalty. We must be loyal to our patients not our theories. The reason my position has changed is that I am following the data. Would you want me to do anything else?”

His answer completly satisfied me and it extuinguished my rage. I was better able to deal with my regret and embarrassment at having to abandon a long held belief and practice pattern at the feet of new data.

His new blog is off to a flying start with a productivity that hasn’t been seen since Nate Hellman and quality that, to my eyes, no one can match.

Thanks Ajay, I’m looking forward to following your blog.

Posted on

HIPAA violations go mass production

Stanford let demographic information on 20,000 ER visits escape onto the web.

A billing contractor created a spreadsheet containing names, account numbers, diagnosis, and length of stay for every admission to the ER for 6 months. Somehow, this spreadsheet was uploaded as an attachment to a post asking for help making bar-charts (you can’t make this up) to the tutoring website Student of Fortune. The information remained on the site freely available from September 2010 to August 2011.

Get all the details at the New York Times.

Maybe the next time we get alarmed about the threat to privacy from a case report or educational lecture we should think about the losers that are releasing medical data by the bucketful.

Posted on

Hyperkalemia or not

A patient came to the hospital with a swollen arm. The ED suspected a DVT and ordered a doppler ultrasound which confirmed their suspicion. The admission labs included a chem-7 which revealed a potassium of 7. Her creatinine was 1 and she wasn’t taking an ACEi, ARB, aldactone, ketoconazole, or potassium supplements. The ER was surprised and repeated the study and checked an EKG:

Narrow QRS and unimpressive T-waves

The EKG gave no hint of hyperkalemia, though EKG changes are not a sensitive marker for hyperkalemia. The ED gave insulin, glucose and Kayexalate for the lab finding of hyperkalemia. We were consulted to determine the cause of the hyperkalemia. The patient’s past medical history was significant for primary thrombocytosis and during the hospital stay her platelet count rose to over a million.

dats a lot o’platelets

We presumed that his hyperkalemia was actually pseudohyperkalemia due to the high platelet count. Platelets release potassium when they clot and the risk of pseudohyperkalemia rises as the platelet count approaches a million.

You remember this classic NEJM article from 1962. 

We then sent the patients blood to the ABG lab in a heparinized syringe rather than a red top and the potassium normalized. Platelets release potassium when they are activated. By measuring the potassium in whole blood rather than serum, the contribution of platelet activation is prevented. The ABG results are the electrolytes to the far left in the screen-grab below (click to enlarge).

– Posted using BlogPress from my iPad

Posted on

Great animated gifs from Dr. Strangelove

General ‘Buck’ Turgidson listening to Premier Kissoff

Mr. President, we are rapidly approaching a moment of truth both for ourselves as human beings and for the life of our nation. Now, truth is not always a pleasant thing. But it is necessary now to make a choice, to choose between two admittedly regrettable, but nevertheless distinguishable, postwar environments: one where you got twenty million people killed, and the other where you got a hundred and fifty million people killed. 

Peter Seller’s best movie?

From If We Don’t Remember Me.

Posted on

Hypertensive emergency with thrombotic microangiopathy

We admitted a patient with a rip-roaring hypertensive emergency and microangiopathic hemolytic anemia (MAHA). On admission his creatinine was 4. The creatinine kept creeping up despite controlling the blood pressure. Additionally his LDH, platelets and schistocytes either normalized or were heading towards normal, yet the creatinine continued to rise. Early on in his hospitalization we started spironolactone with minimal effect on his blood pressure and no improvement in renal function. As the patient spiraled toward dialysis we decided to attempt to block aldosterone secretion responsible for the thrombotic microangiopathy. We started captopril 50 mg tid. I planned to titrate it up to scleroderma renal crisis-like doses.
Murphy must have been working the night shift because the following morning he was complaining of a swollen lip. We stopped the ACEi due to angioedema. We started aliskiren instead. She tolerated that and for the first time in a week her creatinine leveled off and started to come down.
The biopsy is pending.
Was his renal function about to turn around anyways? Who knows?
The team was conflicted on starting the ACEi. So many times we try to minimize exposure to ACEi in acute kidney injury to maximum glomerular perfusion. He was admitted with severe hypokalemia so aldosterone and renin levels were drawn on admission. Those are send-out labs at our institution and we didn’t get results until into the second week. Aldo 60, plasma renin activity 5. That’s among the highest aldo I have ever seen.
There is experimental evidence pointing to aldosterone having direct vasculopathic activity driving microangiopathic hemolytic anemia.
Chander et al, took 28 Stroke-prone spontaneously hypertensive rats and then divided them into four groups:
  1. Sham surgery
  2. Bilateral adrenalectomy
  3. Bilateral adrenalectomy + angiotensin 2
  4. Bilateral adrenalectomy + aldosterone
Dexamethasone was provided to adrenalectomy rats to prevent Addison’s disease.
Blood pressure was elevated in all rats but was less severe in the adrenalectomy group.
Proteinuria was worse in the control group and when aldosterone was restored.
The experimental conditions resulted in malignant hypertension with MAHA as seen in the pathology from the control animals (panel A). Adrenalectomy resulted in sparing of the renal architecture (panel B) and the addition of angiotensin did not induce MAHA (panel C). The addition of aldosterone did induce similar pathology as seen in the control animals implicating aldosterone as the bad actor in this animal model of malignant hypertension and MAHA.
  1. MAHA is seen in panel A, control animals
  2. Adrenalectomy (panel B) prevents the characteristic lesions of malignant hypertension
  3. Likewise with adrenalectomy + AT2 (panel C)
  4. The fibrinoid necrosis returns with the aldo infusion in panel D.
From Chander’s data it looks like aldosterone is the bad actor, I was unable to find information on using spironolactone to reverse MAHA. There is data showing a protective effect of ramipril to prevent MAHA, again using an animal model of malignant hypertension.
I look forward to see how much renal recovery he gets with aliskiren, though it will be impossible to prove that it was aliskiren, I for one believe.
Another point we debated was the use of aliskiren after ACEi induced angioedema. We decided to use it.  Here is what the package insert has to say:

Angioedema: Two cases of angioedema with respiratory symptoms were reported with Tekturna use in the clinical studies. Two other cases of periorbital edema without respiratory symptoms were reported as possible angioedema and resulted in discontinuation. The rate of these angioedema cases in the completed studies was 0.06%. In addition, 26 other cases of edema involving the face, hands, or whole body were reported with Tekturna use including 4 leading to discontinuation. In the placebo controlled studies, however, the incidence of edema involving the face, hands or whole body was 0.4% with Tekturna compared with 0.5% with placebo. In a long term active control study with Tekturna and HCTZ arms, the incidence of edema involving the face, hand or whole body was 0.4% in both treatment arms [see Warnings and Precautions (5.2)].

++++++++
Below is an annotated biography of additional information, mainly focussing on the poor renal outcomes associated with MAHA in malignant hypertension.
JNCVII defines hypertensive emergencies as:

Hypertensive emergencies are characterized by severe elevations in BP (>180/120 mmHg) complicated by evidence of impending or progressive target organ dysfunction… Examples include hypertensive encephalopathy, intracerebral hemorrhage, acute MI, acute left ventricular failure with pulmonary edema, unstable angina pectoris, dissecting aortic aneurysm, or eclampsia.

Strange that acute renal failure is not mentioned as a complication. The recommendation is to reduce the blood pressure by no more than 25% in the first minutes to an hour and subsequently shoot for 160/100 for the next 2-6 hours. The authors point to 2 exceptions: aortic dissection where the SBP should be less than 100 and in acute stroke where the data is less clear. (a moment of clairvoyance for the JNCVII crew as they correctly predicted the lack of benefit from aggressive blood pressure control in the midst of an acute stroke. This was confirmed with 2011’s SCAST study)

Meta-analysis showing the lack of benefit from blood pressure treatment in acute stroke
IV blood pressure agents helpful in hypertensive emergency from JNCVII

The first article we looked at was Bert Jan van den Born’s retrospective review. These authors looked at patients with malignant hypertension. Cases were identified by looking at every hospital admission with the diagnosis of hypertension and then screening the charts for an ophthmology exam showing:

  • bilateral flame-shaped retinal hemorrhages 
  • bilateral linear “splinter” retinal hemorrhages 
  • or “cotton-wool” exudates.

Wonderful pic of cotton-wool exudates and splinter hemorrhages. Thanks ACP
Flame hemorrhages. Without permission from AAO

If I had to wait for an ophtho consult to make a diagnosis, half my patients would be ready for discharge with the diagnosis still pending.

After patients were deemed to have hypertensive urgency, they were categorized as having MAHA. This was defined as a low platelet count with either an elevated LDH or schistocytes. Additionally the LDH/schistocytes and the platelets had to recover following recovery from the hypertensive crisis.

The endpoints were creatinine and proteinuria at admission and follow-up creatinine.

The study found 110 patients that met the criteria, and 97 were ultimately available for analysis.

  • 4 were excluded because the retinal changes were due to intracranial masses rather than hypertension. 
  • 5 were excluded because of a lack of platelet count 
  • 4 were excluded because they had an alternative explanation for thrombocytopenia
26 of the patients had MAHA. The MAHA cohort was more:

  • black (73 vs 35%)
  • hypertensive (242/150 vs 225/145
  • uremic (Cr 7.8 vs 1.4)
  • proteinuric (88% vs 41%)
To convert creatinine from micromol to mg/dl, divide by 88
Dialysis was needed in 15 of the 26 patients with MAHA and 2 of the 71 without MAHA. They had renal recovery in 6 of the dialysis patients: four within 2 months, one at 8 months and the last at two years!
Then the authors drop what might be the worst line of logic ever published in Hypertension:

Improvement of renal function, defined as a reduction of serum creatinine >50% compared with baseline, was noted in 17 patients during follow-up. Cox regression analysis showed that MAHA and systolic blood pressure at admission were the most powerful indicators of renal improvement with an HR of 0.24 (95% CI, 0.08 to 0.75) and 1.02 per mm Hg increase in systolic blood pressure (95% CI, 1.01 to 1.05; Table 4). Improvement of kidney function over time in patients with and without MAHA is shown in Figure 4.

The authors try to make the case that there is more renal healing with MAHA than in the absense of MAHA without specifically addressing that the patients with MAHA had 20x the rate of dialysis of those without it or that the average Cr of MAHA-absent patients was only 1.4, making a 50% reduction in serum creatinine require the growth of a third kidney. Ridiculous.
The second article was Gonzalez’s study on the long term renal outcome following hypertensive emergency. The introduction begins with a nice history of hypertensive urgency. They state that patient survival was very poor before the 1970 and the development of effective anti-hypertensive drugs. After 1970 renal and patient survival climber to 50 and 75% respectively at five years. A decade later, with the introduction of calcium channel blockers and ACEi in 1980, survival rose to 81% renal and 90% patient survival at 5 years.
Gonzalez retrospectively looked at patients with a diagnosis of malignant hypertension. They also required retinal changes (hypertensive retinopathy grade 3 or 4). Strangely, they excluded a third of patients who had secondary hypertension behind the malignant hypertension. This, along with a few other exclusions due to lack of follow-up for various reasons, left 197 in the cohort. The cohort was divided by ultimate renal outcome. Group 1 had stable or improved renal function while group 2 had worsening renal function.
They found that patients who developed worsenning kidney function had at baseline: 
  • pre-existing CKD
  • higher creatinine on presentation
  • more proteinuria 
  • more microscopic hematuria.
125 (85%) had acute kidney injury on admission and 15 required dialysis. 8 of these had no prior CKD. Only 2 patients recovered sufficient renal function to stop dialysis.
On multivariate analysis only proteinuria in follow-up was associate with a lack of renal stabilization or improvement. 
83% of the patients with worsening kidney function (group 2) ultimately were started on dialysis. Remember, they had over 30 years of follow-up for some of these patients and there is some selection bias as group 2 was defined by a lack creatinine stabilization/improvement.
Since the cohort was collected over 30 years (1974-2007) the authors made some observations regarding the change in malignant hypertension over time. There was no change in the incidence, however there were fewer people needing dialysis. Treatment evolved from diuretics to calcium-channel blockers and RAS blocked.

The third article was a case report and lit review by Shavit et al. The article defines malignant nephrosclerosis as renal failure due to malignant hypertension. They describe three cases with varying outcomes:

  1. 55 year old admitted with a blood pressure of 220/130, Cr 11, normal platelet count, LDH 1,100. A renal biopsy showed concentric intimal hyperplasia, fibrinoid necrosis of arteriolar wall, shrinkage of the glomerular tufts. The patient remained dialysis dependent 2 years after presentation.
  2. 55 year old admitted with 240/125, Cr 13, LDH 1,430, normal platelet count. Kidney biopsy revealed intimal thickening, luminal narrowing, fibrinoid necrosis. His creatinine improved over 2 months and he remained dialysis free with significant CKD 2 years after presentation.
  3. 28 year old admitted with a blood pressure of 210/135, Cr 4.5, K 2.9, normal platelet count. Kidney biopsy showed severe intimal thickening, and fibrinoid necrosis. Creatinine improved over 2 weeks to 1.8 and remains stable at 3 years of follow-up.
The authors then review the literature with an interesting discussion that renin and aldo levels are commonly elevated in malignant nephrosclerosis and that there is a strong correlation between renin activity and the presence of MAHA, suggestive of renin mediated vascular damage.

They mention research finding low levels of ADAMTS13 in malignant hypertension. ADAMTS13 level fall as LDH levels rise and platelet count fall. A follow-up study of 21 patients failed to confirm these findings.

The next article we pulled was by Akimoto et al, Clinical Features of Malignant Hypertension with Thrombotic Microangiopathy. This was retrospective review of 16 cases of malignant hypertension. MH was defined as an elevated blood pressure with retinal changes. MAHA was defined by an increase in LDH, low hemoglobin and low haptoglobin. Additionally to meet the definition of MAHA patients needed to normalize these indices after correction of the blood pressure. Of note 7 patients had biopsies. Five of those biopsies showed evidence of malignant nephrosclerosis (fibrinoid necrosis) but only 3 of them met the authors’ clinical definition of MAHA. Interesting that those reasonable clues could be missing 40% of cases.

They found higher aldosterone levels in patients with MAHA than in patients without. They found a tight correlation between aldosterone levels and LDH, R2 of 0.4 (p=0.0096).

Four of the seven patients with MAHA required dialysis, however 2 were able to come off. Three of the nine patients without MAHA required dialysis and none recovered renal function.