Assessing fluid status?

[gtb]

Accurately assessing fluid status remains a huge challenge for me. I'm looking for strategies that have reliably worked for you. Also, please post any papers that you have read which have proven especially useful.

Seems like physican exam by itself produces remarkably variable decisions. Urine lytes help to more accurately predict fluid status, but I'm looking for ways to make a more reliable and quantitative decision at 2 a.m. when urine output has dropped and I get a call from the floor. Lytes take time, and if the patient is not cathed I might not be able to get the results in a reasonable time without straight cathing the patient.

If the patient is hospitalized for a few days and I can get trends in weight, labs, etc., then I've been looking at trends in weight, BUN:Cr, electrolyte concentrations, RBC concentrations (when the patient is not actively bleeding), Ca++ when no active turnover or diet changes are an issue.

For those inobvious cases, what's working for you?

---

Members don't see this ad.

 

[AJM]

Fluid status, especially in complicated medical patients, remains challenging even to experienced physicians. Unfortunately, unlike the surgical patients when a drop in urine output almost always means that they are volume down, the medical patients are not that straightforward.

If there's a question, labs are always helpful, but when you need to make a quick decision, it is always best to examine the patient. It's true that the exam is not always that accurate, but in the middle of the night it's usually the only thing you have to go by. Of course you listen to the lungs, blah, blah, blah, but specifically pay attention to the JVP. Now there have been studies that have shown that JVP is not very accurate, especially when done by inexperienced physicians, but if you practice your measurement will become more accurate. (especially if you practice alongside a cardiologist, who have been shown to be the best at this portion of the exam.) What you are really looking for are extremes, and the extremes are not hard to measure: for example in a patient with no UOP and a JVP to his jaw, you are going to give him Lasix - you don't need to wait for any studies to come back.

What do I do? If it's the middle of the night, I would look at the patients' medical problems, look at their labs from earlier in the day, and will quickly examine the patient. If the JVP is equivocal, then it ends up being a crapshoot. So I take a guess. If the patient does not have CHF or ESRD, I'll give them fluids. If the fluids throw them into pulmonary edema, you can always take off the fluid with diuretics. If the pt has CHF, I'll obviously start with diuresis first. Not that satisfying, but other than putting in a central line and transducing the CVP it's the best you can do. Plus, when you are crosscovering overnight on 50 patients, you don't have time to be ordering urine lytes and CXRs and following up on all of them. You just have to make the decision, and realize that if you're wrong you can easily correct things.

 

[gtb]

JVD is one of the strategies I've used, although I sometimes have no baseline for comparison, especially patients with end-stage COPD, PulmHTN, or left heart problems. Then when I see JVD of 7 or 8, a low urine output, no rales, Cr > 1.2 (baseline 0.7) I get a mixed picture. On one hand, the volume seems high (JVD), but Cr and UOP makes it seem low. That's when I try for existing lab values to help. I'm looking for a better way than my current strategy, which is pretty much as you described. Is there a more evidence based approach?

 

[Tenesma]

the JAMA had an article about 5 -7 years ago that showed that even experienced clinicians couldn't tell 50% of the time what the right fluid status is of a patient based on clinical/physical exam...

I always assume they are dry unless they have underlying pathology that would lead you to think otherwise...

 

[southerndoc]

the JAMA had an article about 5 -7 years ago that showed that even experienced clinicians couldn't tell 50% of the time what the right fluid status is of a patient based on clinical/physical exam...

I always assume they are dry unless they have underlying pathology that would lead you to think otherwise...

One of the things we do in the ED where I am a resident is beside ultrasonography of the inferior vena cava. Have the patient "sniff." If the IVC collapses, the patient is fluid depleted. If the IVC moves slightly, but doesn't collapse, then the patient is euvolemic. If the IVC is plump and doesn't move with sudden inspiration/sniffing, then the patient is fluid overloaded.

I think more specialties will be picking up ultrasonography in the future. It's a great tool... one which has been underutilized in the past.

 

[AJM]

the JAMA had an article about 5 -7 years ago that showed that even experienced clinicians couldn't tell 50% of the time what the right fluid status is of a patient based on clinical/physical exam...

I always assume they are dry unless they have underlying pathology that would lead you to think otherwise...

I remember that article.... I think it might be the same one that I was referring to above about how bad we are with JVP.

Unfortunately, gtb, there's no magic bullet. If there was a more evidence-based approach, everyone would be using it. But this is where the "art" of medicine comes into play. If you have a patient who is acutely sick enough that it's crucial to know their exact volume status, you can always throw in a central line. But by the time you're thinking about doing that, the patient is probably getting to be unit material anyway...

IMO the general strategy Tenesma mentioned of assuming they are dry unless you have reason to suspect otherwise is a good one. If you end up giving a patient too much fluid, you can (almost) always get that fluid off one way or another.

 

[AJM]

One of the things we do in the ED where I am a resident is beside ultrasonography of the inferior vena cava. Have the patient "sniff." If the IVC collapses, the patient is fluid depleted. If the IVC moves slightly, but doesn't collapse, then the patient is euvolemic. If the IVC is plump and doesn't move with sudden inspiration/sniffing, then the patient is fluid overloaded.

I think more specialties will be picking up ultrasonography in the future. It's a great tool... one which has been underutilized in the past.

Yeah, I've done that before as well. Do you know if there's any evidence on bedside ultrasound in this kind of situation? My thinking is that it is highly operator-dependent. Unfortunately, in IM we are not routinely trained in bedside ultrasound. My program recently starting training us in it, but I have yet to personally see a benefit. The ED folks are much better at it IMO, because they get a lot more training and use it much more frequently.

 

[southerndoc]

Yeah, I've done that before as well. Do you know if there's any evidence on bedside ultrasound in this kind of situation? My thinking is that it is highly operator-dependent. Unfortunately, in IM we are not routinely trained in bedside ultrasound. My program recently starting training us in it, but I have yet to personally see a benefit. The ED folks are much better at it IMO, because they get a lot more training and use it much more frequently.

To my knowledge, nothing official has been published. I'm sure it could be studied in individuals who were scheduled to receive PA catheter placements. Do a quick ultrasound, then place the PA.

 

[Annette]

One of the other tricks I learned was to look under the tongue. If it is nice and juicy, lasix, if not, fluids. Also, you can also check urine concentration to give you a guide- faster and cheaper than waiting for electrolytes.

Something else to check in the middle of the night is that the patient does not have a foley. Got called one night for "no urine output." The patient wasn't cathed, and was sleeping!