Posts

Posted on

What causes hyponatremia in marathon runners

Me, running the NYC Marathon

One of the first blog posts ever on PBFluids was a review of Almond Et al’s study of hyponatremia. With this year’s Boston Marathon now complete I have re-reviewed the subject.

The Almond study was high profile and did a good job of demonstrating the risk factors for marathon induced hyponatremia. (See this post for a review) However some of the findings were self evident: increased weight gain was associated with hyponatremia. What is not answered is, why those who developed hyponatremia gained 3 liters of water. Why didn’t these patients just urinate the excess water? Normally, a falling sodium, shuts down ADH like a bordello on Easter. The retention of water is indirect evidence of ADH. Could it be that marathon running and ultra-endurance events could be added to the list of causes of the Syndrome of Anti Diuretic Hormone (SIADH).

It would have been nice to see a U/A or urine osmolality in Almond’s data to confirm this.

Siegel et al. (PDF) has done the most detailed study I am aware of on exercise induced hyponatremia. They did detailed biochemical assessments on 39 runners in the 2001 Boston Marathon. They drew pre-race (day before) and post-race (within 2 hous of finishing) samples for:

  • CPK
  • IL-6
  • ADH (vasopressin)
  • cortisol
  • prolactin
  • CRP

They also looked at 308 runners who collapsed during the 2004 Boston Marathon and measured:

  • IL-6
  • ADH (vasopressin)

Additionally they did some blood tests on 2 runners who had died of cerebral edema from exercise induced hyponatremia. One from the 2002 Marine marathon and the other from the 2002 Boston Marathon.

The normal patients had spikes in their CPK from 150 to 2,323. They also had a doubling of cortisol and prolactin but no change in ADH levels. The rise in CPK was matched by increases in IL-6 followed by an increase in CRP.

Of the 308 collapsed runners only 16 had hyponatremia. All of the hyponatremic runners reported a lack of urination during the race. 7 of the 16 had inappropriately high ADH levels in the blood. The authors concluded that lack of urination (though only driven by ADH in half the patients) rather than fluid loading was the predominant cause of hyponatremia.

The article then describes the laboratory and clinical scenario surrounding the two deaths in 2002. The data is summarized in the following table:

Importantly, both patients were initially treated with 150 mL/hr of normal saline without improvement. Two years later, two patients presented with similar symptoms and responded well to 3% saline:
The primary conclusions from this study, which admittedly is a bit schizophrenic with numerous anecdotal reports from various populations, is that exercise induced hyponatremia is not due to sodium loss but rather from fluid retention. Some of this fluid retention is driven by ADH and hence introduces exercise induced hyponatremia as a novel cause of SIADH. The diagnosis of SIADH is backed up by elevated urine sodium, elevated urine osmolality and normal (or high in the case of cortisol) cortsiol and TSH levels.

The elevated urinary sodium levels (consistent with SIADH) are a critical fact in the etiology of hyponatremia. If we were dealing with hypovolemia (commonly, but erroneously, referred to as dehydration), a cause of hyponatremia, one would expect a low urine sodium (usually less than 10 but always less than 20). The high urine sodium means that these patients were not volume depleted, It was not loss of sodium through the sweat which lead to the low sodium. This means that changing the sodium content of sport drinks is unlikely to prevent the complication.

The authors point out NSAIDs (ibuprofen, Motrin, Advil, naproxen) enhance renal response to ADH and should be avoided in the 24-hours prior to a race.

The authors recommend treating acute symptomatic hyponatremia from a marathon with 3% saline 1 mL/kg/hr to raise serum sodium 4-6 mEq/L and then to slow the rate to target 12 mEq/L in the first 24 hours of therapy. Just as is in all cases of SIADH 0.9% saline may not improve the serum Na.

Posted on

Nephrology myths: drink a lot of water

I am on the twitter and I came across this tweet:


The tweet reads, “It annoys us at The Kidney Group when so-called experts claim being well-hydrated is overrated and without much merit. Completely untrue.”

The fact is this is total bullshit. Outside of patients with kidney stones or pre-renal azotemia, I am aware of no human data showing improved kidney function from increased fluid intake. In fact in the MDRD study they found an association with high fluid intake and faster progression to dialysis. Having a 24-hour urine volume of 2.4 liters was associated with a loss of kidney function of 1 ml/min/year faster than patients with a urine output of 1.4 liters.

In a comprehensive study on the risk factors for the development of ESRD (PDF) (27+ years of follow-up of 177,570 patients) having nocturia (HR 1.36) was about as important a risk factor as anemia (HR 1.33) or family history of kidney disease (HR 1.40) on multivariate analysis.

The authors take on the significance of nocturia:

It is interesting that nocturia (defined herein as self-report of “always having to interrupt sleep to urinate”) emerged as an independent risk factor for ESRD because it is a widely held clinical belief that nocturnal polyuria is an early sign of chronic kidney disease due to decreased urinary concentrating ability, although some data suggest that increased salt, not water excretion, is more important. Therefore, nocturia may reflect subtle early renal disease not captured by serum creatinine level or urine dipstick analysis. We also cannot exclude the possibility that nocturia reflects undiagnosed DM. An alternative hypothesis is that nocturia reflects a high volume of ingested fluid that is detrimental (especially among patients with existing kidney disease), as high urine volume increases intratubular volume and pressure and these stretch forces induce fibrosis. Practically speaking, our data lend no support to the notion that a high volume of water intake should be recommended in clinical practice as being beneficial to kidney function.

For a summary of the myth of water intake and kidney health look at this excellent review by Wenzel et al in CJASN (PDF).

Posted on

Introducing the Acid-Base Machine

Last month I realized that in order for students to really learn acid-base interpretation they need practice. Lots of practice. So I started everyday’s teaching rounds by assigning each student four ABGs to interpret.

Creating all of those ABGs became pretty tedious so I started fiddling around with an excel spreadsheet to automate the process. I used ABG Machine 1.0 for that month but, unfortunately, it created too many respiratory problems and too few metabolic disorders. I completely re-crafted the randomization algorithm so that it should provide a balanced distribution of ABG problems (a quarter metabolic acidosis, a quarter metabolic alkalosis, a quarter respiratory alkalosis and a quarter respiratory acidosis).

Introducing The Acid-Base Machine 2.0

The spreadsheet is made up of 12 individual sheets. The first one is the Question sheet, you should print one copy for every student or resident. The second sheet is the Answer sheet, I printed one for me and one for each student and I would pass them out after the excercise so if they wanted to brush up on some additional questions they had the answers. The next ten sheets are the guts of the machine and you can ignore them unless you want to tinker with how the randomization works.

Download the Excel file

Posted on

Is hypertension in octagenarians a characteristic or a disease

I was at the NKF Spring Clinical meeting and I saw variations of the two following slides in at least three different lectures:

The slide on the left comes from the AHA Heart Disease and Stroke Statistics from 2007 (PDF). It shows the increasing prevalence of hypertension as people get older. The slide on the right is from the Framingham Heart Study (PDF) and shows the 20 year risk for normotensive adults for developing hypertension. Both figures are repeated below:


The part of these figures that kept gnawing at my brain was putting them together, stacking them if you will. 60-70% of the sixty-year olds have hypertension and the minority that are normotensive have a 90% chance of developing hypertension by the time they are 85. So essentially every 80 year old has hypertension.

Does hypertension cease to be a disease when everybody has it? Is hypertension less pathology and rather part of the normal physiology of aging?

Speaking against the idea of geriatric hypertension being normal physiology is the powerful survival advantage gained by treating the high blood pressure. This data comes from HYVET published last spring in the NEJM (PDF). Prior to HYVET there was retrospective data pointing to better survival with higher blood pressures (Oates 2007) and a meta-analysis of 80+ year olds enrolled in RCTs showed a reduction in cardiovascular evens but a trend to increased total mortality.

The HYVET ransdomized 3,845 octagenarians with blood pressures 160-190 with a diastolic of less than 110 to either placebo or indapamide (thiazide-like diuretic) with additional perindopril if the systolic blood pressure remained over 150. The primary end-point was number of strokes (fatal + non-fatal)

Results. The investigators achieved good blood pressure separation between the control and experimental groups with a 15 mmHg difference in the systolics and 6 mmHg difference between the diastolics.

The effect on morbidity and mortality were dramatic (all results expressed as intension-to-treat) with active treament resulting in:

  • 30% reduction in the rate of fatal or nonfatal stroke (95% confidence interval [CI], –1 to 51; P=0.06)
  • 39% reduction in the rate of death from stroke (95% CI, 1 to 62; P=0.05)
  • 21% reduction in the rate of death from any cause (95% CI, 4 to 35; P=0.02)
  • 23% reduction in the rate of death from cardiovascular causes (95% CI, –1 to 40; P=0.06)
  • 64% reduction in the rate of heart failure (95% CI, 42 to 78; P<0.001)
  • Fewer serious adverse events (358, vs. 448 in the placebo group; P=0.001).
Click on the graphs to see larger versions.

So even if hypertension in the elderly is not a disease, treating it seems to have dramatic benefits for patients. Vote your opinion!

Posted on

Its not the sodium intake its the sodium:potassium ratio

Don’t worry only about sodium intake (NYC, I’m looking at you) and its not just potassium intake (DASH diet in the cross-hairs). It’s all about the sodium potassium ratio. This is shown by Cook et al (PDF). during reanalysis of the Trial of Hypertension Prevention I and II. This trial had serial 24-hour urine collections done in 2,275 patients with pre-hypertension in the late 80’s and 90’s. The investigators looked at that data through the lens of 15 years of follow-up to determine the risk of cadiovascular events:

In observational analyses of the mean urinary excretion during 11⁄2 to 3 years, we found a suggested positive relationship of urinary sodium excretion and a suggested inverse relationship of urinary potassium excretion with risk of CVD, but neither was statistically significant when considered separately. Both measures strengthened when modeled jointly, with opposite but similar effects on risk. However, the sodium to potassium excretion ratio displayed the strongest and statistically significant association, with a 24% increase in risk per unit of the ratio that was similar for CHD and stroke and was consistent across subgroups.

Here is the key figure. Note in the graph the rate of events is presented on a log scale so the 2 indicates a rate 100 times the rate at zero.

Posted on

What’s new in Potassium: sudden cardiac death

As the Nephrology Fellow Network recently covered the etiology of cardiovascular disease in dialyzors is unique from the general public. Use of statins, the foundation of preventative cardiology, has repeatedly failed to prevent cardiovascular vascular disease (CVD) among dialyzors. One reason for this, is the propensity for these patients to die of sudden cardiac death (a lethal heart rhythm requiring a shock of electricity or luck to reverse) rather than acute myocardial infarction (heart attacks). In this study (PDF), from Italy, the investigators found that nearly half of the cardiovascular deaths were due to sudden cardiac death (SCD). The authors retrospectively looked at their data to find risk factors for SCD.

They prospectively looked at 476 patients in 5 Italian hemodialysis units. The cohort was tracked for 3 years and had 167 deaths (35%), 32 due to SCD and 35 due to other CVD. On multivariate analysis they found the following risk factors for SCD:
As important as what was significant, is what was not significant. Left ventricular hypertrophy, heart failure and valvular heart disease, all important risk factors for SCD among non-dialysis patients were not associated with SCD in their cohort.

The most interesting analysis was when they parsed out the day of the week the patients died of SCD. Instead of looking at the absolute day they related the day to the patients dialysis schedule. I have modifed their chart to reflect this, with twin X-axis: one for MWF and another for TTS patients.
The red line indicates how high the bars would be if there was no relationship to the dialysis schedule. The highest risk periods were the 24 hours before dialysis at the beginning of the week and the 24 hours after the dialysis at the beginning of the week. Not dialyzing for the two days over the week-end put patients at risk for SCD both before and after subsequent dialysis.

This sounds like an electrolyte associated complication rather than a uremic toxin because of the risk after dialysis, indicating the change in the toxin, not just the high level, is a risk-factor. This is supported by studies (1, 2) of potassium modeling in which the potassium in the dialysate is lowered sequentially during dialysis. By modeling the potassium, the speed of potassium removal is decreased. This has been shown to decrease pre-mature ventricular contractions (a benign momentary disturbance in the heart rhythm that is being used as a proxy for more serious arhythmias, like SCD. Medicine has gotten in trouble with this proxy in the past so it may not be appropriate.).

Summary: modestly high potassiums are associated increased SCD and the two day dialysis holiday on traditional three day a week dialysis is likewise associated with SCD. Hello daily dialysis!

The lecture on Potassium that this entry was drawn from: