Measuring Potassium in Kidney Disease: What You Should Know
Measuring Potassium in Kidney Disease: What You Should Know
George Bakris, MD: Hi. My name is George Bakris. I'm a Professor of Medicine at the University of Chicago Pritzker School of Medicine and the President of the American Society of Hypertension. I'm here with Dr. Dominic Sica, who is a Professor of Medicine at Virginia Commonwealth. We're going to talk today about potassium. Specifically we're going to talk about the measurement of potassium and some important concepts that you should really be aware of in terms of interpreting potassium. Dom, thank you very much for joining me.
Dominic A. Sica, MD: You're welcome.
Dr. Bakris: So I need to know straight up, if you are measuring serum potassium, how accurate is that in terms of your total body potassium? You get a value back from the lab at 3.5 [mEq/L]. What does that mean?
Dr. Sica: Generally 3.5 [mEq/L] is on the low side. Most reference labs have a range of 3.5 to 5.5 [mEq/L], but when you're at the boundaries 3.5 or 5.5 [mEq/L] you probably have an abnormal result to a certain degree. George, you mentioned serum...I would say that for thinking about potassium measurement, serum differs from plasma. For serum in a red top vs plasma in a purple top there may be as much as a 0.2 or 0.3 millimole difference between the two. Sometimes we use that as a tool for so-called pseudohyperkalemia, which is a test tube phenomenon attributable typically to high platelet counts. You separate serum from plasma. Plasma doesn't clot, so the value is even much lower than it would be in the case for a serum value.
Dr. Bakris: Right, so if I had a patient with a platelet count of 700,000/µL, I probably should get a plasma and not a serum measurement.
Dr. Sica: It probably should be a plasma measurement to interpret it correctly.
Dr. Bakris: Right, very good. Let me ask you something. If I have a patient who is diabetic and being treated with an angiotensin-converting enzyme inhibitor , and they come back and their potassium is 5.4 [mEq/L], should I call 911? Should I tell the patient to go to the emergency room? Or, should I say, "That's nice. I'm going to bed"? What are your thoughts?
Dr. Sica: Let me change that. Let me say the value is 6.2 [mEq/L] and you've seen the patient in the clinic and the call comes at 9:00 at night while you're at home. As the physician, the first thing you'd hope is that the blood sugar is 300 [mg/dL]. All kidding aside, therefore, you've got something that you could do at home with a little insulin. The insulin would affect the transcellular flux of potassium into the cell and you drop [the level] as much as 0.8-1 µmol. The 5.4 [mEq/L] is a much safer zone. Generally, for most of us our worry becomes above 6 [mEq/L]. The occasional patient has a problem below 6 [mEq/L], but it's a very small number of occasional patients for whom it's a problem. I think you've got a safe zone and 911 does not have to be called. You then treat proactively, trying to understand the basis for it and whether that's a steady state value. If it had been 4 [mEq/L] and is now 5.4 [mEq/L], then you have to step in reasonably quickly in a day or 2 to modify it.
Dr. Bakris: Now let me ask you something. This is kind of an unfair question, but I'm going to ask it anyway. If you had a choice -- and these are the only 2 choices you had -- you could have a potassium of 2.8 [mEq/L] or you could have a potassium of 5.6 [mEq/L]. Which one of those two would you want and why?
Dr. Sica: I would love the value of 5.6 [mEq/L].
Dr. Bakris: Right. Why?
Dr. Sica: Because I think that's a high normal value. I'd first ask, "Why do I have a 5.6?" Then, if I'm satisfied that the basis for it is something that I have therapy for what caused it, then I've married those 2 concepts: a desired therapy, a potassium value. Okay, I'll live with it. A 2.8 [mEq/L] reflects a pathologic deficiency of potassium with an implied cardiovascular risk that can be rather significant. But even with the 2.8 [mEq/L], I'd say "Why?" Is it diuretic-driven so that it's a change in total body burden? You've lost potassium through the urine. Or is it a transcellular flux problem? If you had a 2.8 [mEq/L level] and it was someone who was taking a beta agonist, for example, albuterol for asthma. Then, 2.8 [mEq/L] around the time of an albuterol nebulization might be perfectly normal for the person. If you reframed it and said 2.8 [mEq/L] as a constant value and 5.6 [mEq/L] as a constant value I'm going to pick that 5.6 [mEq/L] value.
Dr. Bakris: It's interesting. I've polled this around the country and most physicians always pick the low number, the 2.8 [mEq/L]. I just say, "Congratulations, you just died of an arrhythmia." I think they don't appreciate the consequences of the lower potassium. Speaking of low potassium, a hypertensive patient comes to you and they're not controlled at [a systolic level of] 160 [mm Hg]. Their potassium is 3 or 3.2 [mEq/L]. You give them some potassium supplement but nothing that's really going to replete them. How successful are you going to be at treating their blood pressure? Assuming you use the medications appropriately, will you ever get good blood pressure control or will it take more medications?
Dr. Sica: It may take a little more. I mean there are some data suggesting that if you use potassium supplements to bring a 3.2 [mEq/L] up into the 3.8-4.0 [mEq/L] range that you may get some additional blood pressure reduction. On the flipside, those studies -- which are a few not many -- use potassium supplements. As we well know, giving potassium doesn't replace the total body deficit. The cellular complement of potassium is not repleted by potassium supplements. Simply, the blood level comes up. Therefore, you need something like a potassium-sparing diuretic to get some change in the cellular component. I think there is some mysterious witchcraft with potassium supplements and its effect on blood pressure.
Dr. Bakris: Now, in keeping with that, what's the relationship between magnesium and potassium? Because a lot of [physicians] give a lot of potassium but they can't really seem to fix it. Then, all of a sudden they're on a wild goose chase for primary aldosteronism and they can't find anything. It turns out that the [patients] have magnesium levels of 1.1 [mEq/L]. What's that relationship?
Dr. Sica: It looks like when you lose magnesium. So you have external losses, which commonly are stool losses and can be rather dramatic -- you change around 2 other cations. You change around the capacity for the kidney to retain potassium, so patients with low magnesium are basically having an ongoing potassium loss that's continuous. Until you repair the body deficit of magnesium, which can require a substantial repletion in the order of, maybe, as much as 2 mEq/kg body weight, you're really not going to get that much gain in retaining potassium. Flipside, we also see low magnesium levels associated with persistent hypocalcemia that doesn't correct with calcium supplements or vitamin D. Until you give some magnesium, you don't get the desired effect. Low magnesium has an interesting effect on parathyroid hormone. It suppresses the release of parathyroid hormone, and it suppresses the action at the bone level, that is, the osteoclast activation action for a quick efflux of calcium. You've got a twofold problem with magnesium and calcium, that's also there. So it's a triad: low potassium, low calcium, and low magnesium.
Dr. Bakris: So what's the best way to replace magnesium without causing diarrhea?
Dr. Sica: Well, if you have to give it intravenously, you give it intravenously. Magnesium-oxide tends to be a little less diarrhea-provoking than some of the other agents. Then, you try to do your very best to limit the external losses from the body. On the flipside, it looks like all of our potassium-sparing diuretics -- amiloride, triamterene, and spironolactone -- effectively reduce urine magnesium losses. The Bartter syndrome type patients or the isolated magnesium-losing states that some people on cisplatin and other chemotherapy agents will get, they'll get a modest response to a potassium-sparing diuretic. You should say it's potassium-sparing and magnesium-sparing in nature.
Dr. Bakris: Dominic, thank you very much. It has been a pleasure. I think we covered the basics in terms of understanding the range of potassium [levels], what's bad, what's good, and what it really means to supplement and the consequences of not supplementing. With that, thank you for joining us. We hope you got something out of this that would be of practical use for you. Have a good day. Goodbye.
George Bakris, MD: Hi. My name is George Bakris. I'm a Professor of Medicine at the University of Chicago Pritzker School of Medicine and the President of the American Society of Hypertension. I'm here with Dr. Dominic Sica, who is a Professor of Medicine at Virginia Commonwealth. We're going to talk today about potassium. Specifically we're going to talk about the measurement of potassium and some important concepts that you should really be aware of in terms of interpreting potassium. Dom, thank you very much for joining me.
Dominic A. Sica, MD: You're welcome.
Dr. Bakris: So I need to know straight up, if you are measuring serum potassium, how accurate is that in terms of your total body potassium? You get a value back from the lab at 3.5 [mEq/L]. What does that mean?
Dr. Sica: Generally 3.5 [mEq/L] is on the low side. Most reference labs have a range of 3.5 to 5.5 [mEq/L], but when you're at the boundaries 3.5 or 5.5 [mEq/L] you probably have an abnormal result to a certain degree. George, you mentioned serum...I would say that for thinking about potassium measurement, serum differs from plasma. For serum in a red top vs plasma in a purple top there may be as much as a 0.2 or 0.3 millimole difference between the two. Sometimes we use that as a tool for so-called pseudohyperkalemia, which is a test tube phenomenon attributable typically to high platelet counts. You separate serum from plasma. Plasma doesn't clot, so the value is even much lower than it would be in the case for a serum value.
Dr. Bakris: Right, so if I had a patient with a platelet count of 700,000/µL, I probably should get a plasma and not a serum measurement.
Dr. Sica: It probably should be a plasma measurement to interpret it correctly.
Dr. Bakris: Right, very good. Let me ask you something. If I have a patient who is diabetic and being treated with an angiotensin-converting enzyme inhibitor , and they come back and their potassium is 5.4 [mEq/L], should I call 911? Should I tell the patient to go to the emergency room? Or, should I say, "That's nice. I'm going to bed"? What are your thoughts?
Dr. Sica: Let me change that. Let me say the value is 6.2 [mEq/L] and you've seen the patient in the clinic and the call comes at 9:00 at night while you're at home. As the physician, the first thing you'd hope is that the blood sugar is 300 [mg/dL]. All kidding aside, therefore, you've got something that you could do at home with a little insulin. The insulin would affect the transcellular flux of potassium into the cell and you drop [the level] as much as 0.8-1 µmol. The 5.4 [mEq/L] is a much safer zone. Generally, for most of us our worry becomes above 6 [mEq/L]. The occasional patient has a problem below 6 [mEq/L], but it's a very small number of occasional patients for whom it's a problem. I think you've got a safe zone and 911 does not have to be called. You then treat proactively, trying to understand the basis for it and whether that's a steady state value. If it had been 4 [mEq/L] and is now 5.4 [mEq/L], then you have to step in reasonably quickly in a day or 2 to modify it.
Dr. Bakris: Now let me ask you something. This is kind of an unfair question, but I'm going to ask it anyway. If you had a choice -- and these are the only 2 choices you had -- you could have a potassium of 2.8 [mEq/L] or you could have a potassium of 5.6 [mEq/L]. Which one of those two would you want and why?
Dr. Sica: I would love the value of 5.6 [mEq/L].
Dr. Bakris: Right. Why?
Dr. Sica: Because I think that's a high normal value. I'd first ask, "Why do I have a 5.6?" Then, if I'm satisfied that the basis for it is something that I have therapy for what caused it, then I've married those 2 concepts: a desired therapy, a potassium value. Okay, I'll live with it. A 2.8 [mEq/L] reflects a pathologic deficiency of potassium with an implied cardiovascular risk that can be rather significant. But even with the 2.8 [mEq/L], I'd say "Why?" Is it diuretic-driven so that it's a change in total body burden? You've lost potassium through the urine. Or is it a transcellular flux problem? If you had a 2.8 [mEq/L level] and it was someone who was taking a beta agonist, for example, albuterol for asthma. Then, 2.8 [mEq/L] around the time of an albuterol nebulization might be perfectly normal for the person. If you reframed it and said 2.8 [mEq/L] as a constant value and 5.6 [mEq/L] as a constant value I'm going to pick that 5.6 [mEq/L] value.
Dr. Bakris: It's interesting. I've polled this around the country and most physicians always pick the low number, the 2.8 [mEq/L]. I just say, "Congratulations, you just died of an arrhythmia." I think they don't appreciate the consequences of the lower potassium. Speaking of low potassium, a hypertensive patient comes to you and they're not controlled at [a systolic level of] 160 [mm Hg]. Their potassium is 3 or 3.2 [mEq/L]. You give them some potassium supplement but nothing that's really going to replete them. How successful are you going to be at treating their blood pressure? Assuming you use the medications appropriately, will you ever get good blood pressure control or will it take more medications?
Dr. Sica: It may take a little more. I mean there are some data suggesting that if you use potassium supplements to bring a 3.2 [mEq/L] up into the 3.8-4.0 [mEq/L] range that you may get some additional blood pressure reduction. On the flipside, those studies -- which are a few not many -- use potassium supplements. As we well know, giving potassium doesn't replace the total body deficit. The cellular complement of potassium is not repleted by potassium supplements. Simply, the blood level comes up. Therefore, you need something like a potassium-sparing diuretic to get some change in the cellular component. I think there is some mysterious witchcraft with potassium supplements and its effect on blood pressure.
Dr. Bakris: Now, in keeping with that, what's the relationship between magnesium and potassium? Because a lot of [physicians] give a lot of potassium but they can't really seem to fix it. Then, all of a sudden they're on a wild goose chase for primary aldosteronism and they can't find anything. It turns out that the [patients] have magnesium levels of 1.1 [mEq/L]. What's that relationship?
Dr. Sica: It looks like when you lose magnesium. So you have external losses, which commonly are stool losses and can be rather dramatic -- you change around 2 other cations. You change around the capacity for the kidney to retain potassium, so patients with low magnesium are basically having an ongoing potassium loss that's continuous. Until you repair the body deficit of magnesium, which can require a substantial repletion in the order of, maybe, as much as 2 mEq/kg body weight, you're really not going to get that much gain in retaining potassium. Flipside, we also see low magnesium levels associated with persistent hypocalcemia that doesn't correct with calcium supplements or vitamin D. Until you give some magnesium, you don't get the desired effect. Low magnesium has an interesting effect on parathyroid hormone. It suppresses the release of parathyroid hormone, and it suppresses the action at the bone level, that is, the osteoclast activation action for a quick efflux of calcium. You've got a twofold problem with magnesium and calcium, that's also there. So it's a triad: low potassium, low calcium, and low magnesium.
Dr. Bakris: So what's the best way to replace magnesium without causing diarrhea?
Dr. Sica: Well, if you have to give it intravenously, you give it intravenously. Magnesium-oxide tends to be a little less diarrhea-provoking than some of the other agents. Then, you try to do your very best to limit the external losses from the body. On the flipside, it looks like all of our potassium-sparing diuretics -- amiloride, triamterene, and spironolactone -- effectively reduce urine magnesium losses. The Bartter syndrome type patients or the isolated magnesium-losing states that some people on cisplatin and other chemotherapy agents will get, they'll get a modest response to a potassium-sparing diuretic. You should say it's potassium-sparing and magnesium-sparing in nature.
Dr. Bakris: Dominic, thank you very much. It has been a pleasure. I think we covered the basics in terms of understanding the range of potassium [levels], what's bad, what's good, and what it really means to supplement and the consequences of not supplementing. With that, thank you for joining us. We hope you got something out of this that would be of practical use for you. Have a good day. Goodbye.