Lecture on CKD-MBD

The St John Nephrology Fellowship is excellent. One of the reasons it is good is  we continually use feedback to fix holes. In last year’s in-service exam we found some weakness in calcium so I was asked to buff the fellows calcium knowledge.

Today I started that with a lesson on CKD-MBD where we focused on the KDIGO 2017 update. I pulled what I feel are the the most important articles in this area published recently. Here is what I pulled:

PRIMO. Paricalcitol in pre-dialysis CKD does not magically heal the heart.
Phosphate binder network meta analysis. Concluded that calcium-free based binders are better. All binders have lots of side effects.
The guideline
Executive summary of the guideline with a useful side-by-side comparison of the 2009 and 2017 guidelines.
Phosphorus binders in pre-CKD patients don’t do a lot of good and increase coronary calcification.
Evolve. Prospective data in CKD MBD that produced a nice separation in PTH and not much else.
FGF-23 still has not made it into the guidelines but it is the phantom menace soon to be unveiled.

What did articles did I miss? What are your favorite CKD-MBD articles

P.S. My favorite part of the guidelines, the part that tell you everything you need to know about KDIGO is 4.2.2:

Here it is in 2009:

And here it is in 2017

And the best part is how the level of evidence has not changed. 2C for both. It’s like they just woke one morning on a different side of the bed.

Great update from Swapnil

Hypercalcemia from 1,25 vitamin D toxicity

I received an outpatient consult for acute kidney injury. One of the things that makes Saint Clair Nephrology a remarkable nephrology group is our ability to get patients in quickly. While competing practices in the area have a 3-month wait list to see new patients we get patients in within a week. This patient was seen two days after his doctor called.

The patient was frightened. He had previously been healthy and his doctor told him his kidneys were failing and that he needed to see a nephrologist. He arrived with a creatinine in the high 2s from a base line of 1.2 mg/dL. Along with the AKI his blood pressure was touching 180 systolic, out of character for him. Of note on the initial labs his calcium was 13.6 mg/dL.

The initial work-up showed suppressed PTH. SPEP and UPEP were normal.

On the next visit I checked the 1,25 vitamin D and it was 117 IU. I suspected lymphoma or sarcoidosis but the chest x-ray was unremarkable and the patient did not have any palpable lymph nodes or abnormalities on the CBC. No weight loss, night sweats, or fevers. ACE levels were unremarkable.

On further questioning on his third visit, the patient mentioned he was taking a generic knock off of Mega Red Fish Oil. Fish oils can have significant amounts of vitamin D and the supplement is famously lax with quality control. He stopped the fish oil, we started him on oral prednisone and the 1,25 vitamin D level quickly responded within a couple of weeks. The patient had a full recovery from the hypercalcemia, hypertension, and acute kidney injury.

 

 

Update

Some great comments from Twitter

 

 

The agony and ecstasy of of secondary hyperparathyroidism

Managing secondary hyperparathyroidism in dialysis patients should be a rewarding aspect of nephrology. I thrive on complex management that involves balancing various numbers with clever treatment strategies. It is exactly what I find so exhilarating about a juicy electrolyte case in the ICU.

The principle variables in secondary hyperparathyroidism are:

  • PTH
  • Phosphorous
  • Calcium
And I use one additional lab that is generally ignored in the guidelines, alkaline phosphatase.
To bend these numbers we have a variety of tools with interesting effects, mechanisms of action and side-effects. The principle therapeutics:
  • low phosphorous diet
  • calcium containing binders
  • non-calcium binders
  • calcitriol
  • paricalcitol and doxercalciferol
  • cinacalcet
And additional therapeutics that can be brought to bear in difficult cases or in unusual circumstances
  • dialysate calcium concentration
  • parathyroidectomy

And K/DOQI provided cleanly laid out treatment goals:

  • PTH 150-300
  • Caclium 8.4-9.5
  • Phosphorous 3.5-5.5
  • Calcium x phosphorous product < 55
Patients that achieve those targets have a lower mortality risk than patients that miss these targets:

The numbers (0 of 3, 1 of 3, etc) refer to the number of months a patient is at the K/DOQI target in the quarter, PTH was measured only once a quarter

The problem is that no one has performed a prospective randomized controlled trial showing these targets improve outcomes. We want to believe that the retrospective data showing a survival advantage with cinacalcet and paricalcitol are real and that the observational data showing better calcium and phosphorous (and to a smaller degree, PTH) results in better patient outcomes.

Teng et al. Survival of patients undergoing hemodialysis with paricalcitol or calcitriol therapy. N Engl J Med (2003) vol. 349 (5) pp. 446-56

Block et al. Mineral metabolism, mortality, and morbidity in maintenance hemodialysis. J Am Soc Nephrol (2004) vol. 15 (8) pp. 2208-18
But given nephrology’s previous relationships with retrospective data (see anemia, Kt/V, and statins, and homocysteine) I can’t accept that data. I can’t take these treatment goals seriously. I appreciate that the fresh KDIGO guidelines readily admit that the emperor has no clothes and that the best they can recommend is to generally keep the calcium and phosphorous close to normal (evidence level 2D) and the PTH anywhere from 150 to 600 (evidence level 2c) or roughly wherever the hell you want it.

I love this figure from KDIGO, essentially once the PTH rises over 150 it provides no information. PTH > 300 has a positive predictive value of only 65% for high turnover disease. And don’t miss the laughably small numbers. We are basing global guidelines off of a study of less than 100 patients. From Barreto and Barreto.

It is shameful that Abbott has not done an RCT with survival as an endpoint on Zemplar or Calcijex. They have had 20+ years to do this. Both of the other players in CKD-MBD have taken a chance at building RCT data to support there products:

  • Genzyme took a poke with DCOR (RCT of sevelamer versus calcium based binders) 
  • Amgen is in the final countdown of EVOLVE (RCT of sensipar + usual care vs usual care)
Abbott the oldest player is sitting on the sidelines.
The lack of data, the lack of clarity, and the reliance on observational data muddles the issue enough that I don’t enjoy taking care of secondary hyperparathyroidism. But recently I had a great case, a situation where treating secondary hyperparathyroidism did more than loaded the dice in my patients favor but actually really made a difference.
I have a young dialysis patient who suffers from a horrific trauma a number of years ago. As a result he has profound chronic pain. Much of the pain is back pain but he also complained of diffuse body aches. Earlier this year his PTHs were consistently over a thousand with some over two thousand.
We added 90 mg of cinacalcet daily and the the PTH plummeted to goal. This was in a patient who had not responded to doxercalciferol 10 mcg three times a week. It was nice to see the PTH come down but what made this case standout was that his body aches melted away. We had been sending him to pain clinics and switching narcotics trying to get his pain tolerable and all of a sudden, done. Pain dramatically improved with a log reduction in PTH. 
Sometimes I get so carried away worrying about total mortality that I forget about the direct toxicity of high PTH. 

Mixed acid-base disorder and altered mental status

An 80 year old woman was readmitted to the hospital with mental status changes. She was recently discharged following successful treatment for heart failure and associated fluid overload. Her discharge medications were as follows: (You know the joke: senior asks the intern, “What medications is she on?” And the Intern looks up and says, “uhm, all of them.”)

She was brought to the ED with a week history of increasing confusion and weakness. The patient had some shortness of breath but this was typical for her baseline.

Initial labs:

The ABG showed:

  • pH: 7.47
  • pCO2: 71
  • paO2: 74
  • HCO3: 51

  1. First look at the pH
    1. It’s elevated so this is a alkalosis
  2. Look at the bicarbonate and the pH
    1. If they both are moving in the same direction it is metabolic
    2. If they are moving in opposite directions it is resiratory
    3. Here the pH and bicarb are up, so it is metabolic
  3. Put the two together and identify the primary disorder:
    1. Metabolic Alkalosis
  4. Calculate the predicted pCO2 from the bicarbonate
    1. Calculate how far the bicarbonate has increased, this is the delta bicarbonate
    2. Take two-thirds of the delta and add it to 40, the normal pCO2
    3. In our patient the bicarb has risen from 24 to 51, a delta of 27, two-thirds of that is 18, so the pCO2 should be 58 +/-2
  5. Is there a second primary disorder affecting the pCO2?
    1. Compare the predicted pCO2 to the actual pCO2
    2. If the actual pCO2 is lower than predicted, the patient has an additional respiratory alkalosis
    3. If the actual pCO2 is higher than predicted, the patient has an additional respiratory acidosis
    4. Our patient’s actual pCO2 of 71, is way higher than the predicted 58+/-2.
  6. The complete interpretation of the ABG is: a primary metabolic alkalosis with an additional primary respiratory acidosis

The ED diagnosed her with acute hypercarbic respiratory failure, and blamed the mental status changes on CO2 retention. She was started on bipap and admitted. The following day her pCO2 improved to 50 but she had persistant confusion. At that point we were consulted for acute renal failure, and noted that she had severe hypercalcemia, calcium of 14.7 mg/dl.

Her phosphate was 1.9 and subsequent work-up showed a PTH of 20., with normal 25 OH and 1,25 OH vitamin D.

On the basis of a combined metabolic alkalosis, acute renal failure, normal PTH and elevated calcium we diagnosed her with Milk-Alkali Syndrome, and started her on IV normal saline and SQ calcitonin. We did not give steroids or bisphosphonates. Over the ensuing four days her calcium drifted down to 10.1. On the third day her sensorium cleared.

Our patient seems to perfectly match the modern form of Milk-Alkali Syndrome or Calcium-Alkali Syndrome using Patel and Goldfarb’s suggested nomenclature. The calcium and alkali were both supplied by calcium carbonate. Additionally she was on a thiazide-type diuretic which decreases calcium excretion. The classic 1930’s form of Milk-Alkali Syndrome was associated with high phosphorous levels while the contemporary form has hypophosphatemia. The principle difference comes from the source of calcium:

  • In classic milk-alkali syndrome the patient is calcium loaded from milk, which is very high in phosphorous (370-450 mg per 8 oz)
  • In contemporary milk-alkali syndrome the calcium carbonate provides the calcium and also acts as a phosphorous binder to prevent dietary phosphorous absorption.
According to Patel and Goldfarb, the hypophosphatemia is more than just a spectator, it’s integral to the modern disease. The low phosphorous stimulates conversion of the storage form of vitamin D (25 OH D) to the active form, (1,25 OH D) which further enhances GI calcium absorption:

Low phosphate levels stimulate the renal metabolism of calcitriol and, consequently, absorption of calcium by the gut. Levels of 1,25-hydroxyvitamin D in patients with the calcium-alkali syndrome, of course, are generally low in the setting of hypercalcemia, although some are in the low- normal range and perhaps inappropriately high. These latter levels may depend on previous exposure to vitamin D supplementation, because vitamin D is often added to some over-the-counter calcium preparations, but more epidemiology is needed to clarify this exposure.

Editor snark, I love the sentence: “Levels of 1,25-hydroxyvitamin D in patients with the calcium-alkali syndrome, of course, are generally low in the setting of hypercalcemia, although some are in the low- normal range and perhaps inappropriately high.” So the levels arer either low, normal or high. Thanks for clearing things up.
The contemporary modern patient is typically female, and post menopausal. Other susceptible populations include cardiac transplant patients, pregnant patients, and those with calcium-rich food-fetishes (reported in anorexic nervosa patients).
Though the alkali and calcium are typically exogenous, diuretic-induced alkalosis can contribute to the condition, and doubly so, if the diuretic is a thiazide which decreases renal calcium losses. NSAIDs contribute by lowering GFR.
The Ca sensing receptor (CaSR) in the thick ascending limb of the loop of Henle binds calcium and binds it more avidly with alkalemia. Binding of the calcium sensing receptor shuts down the ROMK channel which decreases sodium reabsorption and increases urinary loss of calcium. Hypercalcemia, by activating the CaSR, acts like Lasix.
The loop-diuretic effect furthers volume deficiency, which, along with direct calcium-induced vasoconstriction, worsens the renal failure. Volume deficiency also stimulate calcium reabsorption in the proximal tubule.
Increased tubular calcium stimulates TRPV5, the principle calcium transporters in the distal nephron, decreasing renal calcium losses and furthering the hypercalcemia. The TRPV5 is also enhanced by the alkalosis.
Volume expansion with sodium chloride is the bedrock of therapy.
Do not miss the excellent and short review in JASN.

Randomized clinical trial of calcium supplementation

Are Calcium supplements good for you?

Maybe not. This article from 2008 shows increased cardiovascular events in woman randomized to calcium supplementation. I had my mom stop her calcium supplement.

women were included in the study if they had been postmenopausal for more than five years, were aged 55 or more, and had a life expectancy of more than five years. We excluded women who were receiving treatment for osteoporosis or taking calcium supplements; those with an other major ongoing disease including hepatic, renal, or thyroid dysfunction, malignancy, or metabolic bone disease; and those with serum 25-hydroxyvitamin D levels less than 25 nmol/l.

Patients were then randomized to a gram of elemental calcium, as calcium citrate as Citracal or matched placebo.

The results were surprising.
Its amazing the study reached clinical significance given the small size of the trial, only 1,400 patients.
A much larger study, The Women’s Health Initiative randomized 36.282 patients to 500 mg of elemental calcium. They just missed significance for the composite outcome of Myocardial infarction/CHD death/CABG/PCI with a confidence interval of 0.99-1.19.
There might be some signal there.

week-end call and a pair of crazy numbers: Glucose and Calcium

Glucose
I saw the highest glucose I can remember in a patient without ESRD. I have seen the glucose go over 2,700 in a patient with the misfurtune to have both DKA and anuric ESRD. Without the osmotic diuresis to lower the glucose the glucose can shoot the moon. This patient had HyperOsmolar Non-Ketotic Coma (or HONK as my fellow calls it, love that) and baseline Cr of 0.83 and a peak glucose of 1,600 mg/dL.
I love the twin graphs showing the falling glucose and the simultaneous resolution of the pseudohyponatremia. The patient had enough pre-existing osmotic diuresis to cause hypernatremia which was masked by the hyperglycemia. As the glucose comes down the sodium goes up from 136 to 162.

Calcium
The other crazy number was the most severe hypercalcemia I have ever seen. The calcium was 18 mg/dL with an albumin of 3.7 g/dL. The patient is a kidney transplant recipient who was recently seen in the outpatient clinic with hypocalcemia. His calcium was 6.5 and his calcitriol was increased from 0.5 mcg to 1 mcg twice daily. He was also continued on his calcium carbonate.

Admission labs:

The other pertinent calcium labs:

  • PTH: 3.2 pg/mL
  • Vit D 1,25 dihydroxy: 36 pg/mL
  • SPEP/UPEP: unremarkable
  • PTHrp: pending
I think this is milk-alkali syndrome from the calcium carbonate exacerbated by the calcitriol. One supporting string of evidence supporting this is the fact that his calcium came down and has not reoccurred. If it was hypercalcemia of malignancy I would have expected his calcium to be resistant to conservative therapy.  

Myles Wolf is coming to speak at Renal Grand Rounds today

Wolf has been everywhere and is one of the premiere scientists elucidating mineral metabolism. He was the senior author on the article in the NEJM on FGF-23 and dialysis survival and the recent article on the survival advantage with phosphorous binders.

Just a quick review of FGF-23 so I’m not an idiot when this rock star nephrologist starts talking. (FYI don’t let the clean cut pic above fool you, he came to the lecture in full rock-star fashion with the long hair, groupies (supplied by Genzyme) and everything)

FGF-23 is produced by osteocytes.

Klotho seems to be a required co-factor for FGF-23, such that mice that are Klotho deficient mimic the phenotype of FGF-23 deficiency.

FGF-23 increases renal phosphorous clearance by blocking Na-Phos reabsorbtion in the proximal tubule. FGF-23 also inhibits 1-alpha-hydroxylase, decreasing 1,25 OH-vitamin D.

Some of the biology is still a mystery. The highest density of fgf-23/klotho receptors are located in the distal tubule but the biologic effects stem from the proximal tubule.

FGF-receptor and Klotho are also found in the parathyroid gland but the exact role it plays is unclear. Some data points to direct stimulation of PTH and both molecules tend to rise together but this may be due to FGF-23 surpressing 1,25 OH D and secondary increases in PTH.

Increased phosphate and 1,25 vitamin D both stimulate the production of FGF-23. [Note Wolf provided data that phosphate levels do not increases FGF-23. He proposed that it is phosphate balance that is important, his supporting data included lupron treated patients bump their phosphorous by half a point but FGF-23 doesn’t budge, I couldn’t find this article on Google]. The Phex endopeptidase cleaves and inactivates FGF-23 so that is another control factor. [Wolf also discussed iron infusions causing phosphorous wasting due to excess FGF-23 ref pubmed related search]