with Rasha Sawaya, Children's National, on the Hot Seat
The Case
Here's how you answered our questions
How much fluid do you order for your initial bolus in this baby with cardiac history?
LESS than 3 years PEM experience (n=13)
MORE than 3 years PEM experience (n=10)
Everyone either wanted to give 10 ml/kg or 20 ml/kg fluid bolus. No one wanted to give 5 ml/kg. A larger percentage of those with less than three years of PEM experience would give a small bolus to start whereas the greater than 3 years PEM experience group had a larger percentage who would start with the standard 20 ml/kg bolus. This discussion is reflected in the comments below in the section on "Fluids in cardiac patients."
What is your NEXT priority in this patient?
LESS than 3 years PEM experience (n=11)
Greater than 3 years PEM experience (n=8)
No one was interested in stopping the NS bolus to wait for repeat lab results. Otherwise, there was a fairly even split in both groups between getting a STAT EKG next or not worrying about the EKG and going to the bolus next. The question is which will kill first, the hypovolemia or potential for arrhythmia? Of note, one person with more than 3 years PEM experience noted that he or she would get urine electrolytes, which include urine Na and osm.
Clinical Reasoning
About hypernatremia
"I hope everyone made the diagnosis of hypernatremia before the critical lab results were revealed. Decreased activity and "doughy" skin together in an infant is pathognomonic for hypernatremia until proven otherwise." -Dewesh Agrawal, Children's National
"In excess solute load, urine Na and Osms will both be high. In hypernatremic dehydration (from lack of sufficient free water intake), urine Na will be low but Osms will be high. In diabetes insipidus, uine Na and Osms will both be low." -Dewesh Agrawal, Children's National
"I'd ask them how they did the formula, but hold off on the formula demonstration for now but pass it on to the inpatient team to assure it's being done right. Let's say you do the demo, you find out and tell them they've done it wrong, and the child dies in the next couple hours during the resuscitation - the parents are always going to think the death was their fault." -Paul Mullan, Children's National
Fluids in cardiac patients
"Keep in mind, most of the cardiac lesions require preload and volume so I would err on the side of giving fluid rather than restricting it. The best way to know is looking at the child's medicines. Unless the child is on obvious afterload reduction (diuretics, calcium blockers, etc)...then odds are that fluid will help and not hurt the child." -Dave Mathison, Children's National
"1st 10 minutes: line, 20/kg bolus push pull, labs including stat VBG with lytes, portable CXR, EKG, call cardiology down with echo machine." -Paul Mullan, Children's National
"He's post TGA - so he should have "normal anatomy" now as his atrial septostomy should have been closed and everything reattached nicely. Except - these TGA kids can have trouble at all of the "connection points" that were repaired - coronaries (can be low flow/stenosed as these were reconnected), aorta (can get AI), thromboses (virchow's triad = stasis + hypercoagulable state + endovascular integrity not perfect; he's got 3 of 3 right now)." -Paul Mullan, Children's National
The differential diagnosis
"Usually diarrhea causes hypernatremia not vomiting. So that's a little suspicious to me.....vomiting and hypernatremia make me think of intracranial things and DI rather than routine hypernatremic hypovolemic dehydration." -Dave Mathison, Children's National
"The money is in asking the caregiver/s to demonstrate how they actually mix the formula -- not just asking how they do it, but to demonstrate it." -Dewesh Agrawal, Children's National
"If he is having any of those cardiac problems, even with perfect formula mixing, he could go into a low cardiac output state --> low GI flow with vomiting --> low renal flow with resultant bump up in K, crt, Na, and bun --> low brain flow with lethargy, etc." -Paul Mullan, Children's National
From the Hot Seat
by Rasha Sawaya, Children's National
First time on the hot seat, “hot” describes it well!
This is how I would run through this case:
With that initial history and physical exam provided I notice the following: A 4 month old, not healthy child, s/p cardiac surgery with complications, still on cardiac medications (hence does not have a perfect cardiac function), who has vomiting, decreased activity, decreased urine output. On exam, VS wnl, definite weight loss, and signs of severe dehydration.
Note: 1st sign of dehydration or shock in kids is tachycardia, this kid is not, but she is on propranolol, which would mask that. I’m not sure what “doughy” skin means. It makes me think of severe decreased skin turgor that I have seen with severe dehydration and malnutrition in underdeveloped countries.
Missing information: what’s the capillary refill (I find it often changes to the worse, before any peripheral pulses changes), and type of feeds and mixing as suggested in above comments with a bit more details about the past 11 days.
What am I thinking at this point? This child is sick and needs immediate resuscitation. I’m not sure what caused this illness: is the vomiting a cause or a symptom, but either way she is severely dehydrated and not in cardiogenic shock (based on exam) and therefore I would say fluids first. I don’t think there is a harm in starting 20cc/kg NS bolus (child is not on meds for heart failure at this point) however I have learnt that everyone else in the ED is scared of this and arguing about 10 vs 20 is not usually fruitful and a waste of time. So I usually go ahead and start with 10cc/kg NS bolus, but given fast, over 15min (if need be faster), while I remain at the bedside if possible, assessing the respiratory status of the child, the liver, the perfusion and if all well, will give another 10cc/kg NS bolus immediately after… and continue as such. I found this puts up less of a fight and everyone is more comfortable with it, without it affecting my management.
Get a D-stick (hypoglycemia is more common than we think in dehydration) and i-stat lytes (if the child is as sick as described).
While all this is taking place, I would wonder, what is causing all this?
This child can still have what any other child can: Gastroenteritis (diarrhea will soon start), gastritis, a UTI (though no fevers) and pyloric stenosis (though it’s a late presentation); More severely volvulus (though abdomen is soft, not distended and vomiting non bilious), intussusception (I’ve seen it at this age, though the child does not seem to be in pain or discomfort).
So what work-up will I start while starting fluid resuscitation?
CXR, EKG to r/o signs of heart failure as they are not always present on physical exam alone (plus every cardiac kid gets these anytime they appear sick), BMP, UA/Ur Cx.
Finally, what will I do with that BMP?
To me, this still looks like hypernatremic dehydration: elevated Na, BUN, Creat (for age, usually it is 0.3-0.4 in infants). The vomiting could have started it, and in a cardiac child, generally not that healthy, and with poorer reserves, it could be enough to tip her off. I would go with my 1st instinct, and treat with more fluids. That would be the treatment of any hypernatremia at this point.
Also to remember: With this Na, she is at risk for seizures. Keep on monitors, have Non-rebreather and BMV at the bedside, and send to the PICU. You will not fix this Na fast enough for the floor!
Denouement
by Liz Quaal Hines, Johns Hopkins
Upon review of the discharge paperwork from the facility, the recipe for the formula was for 34 kcal/oz rather than 24 kcal/oz. Also in consideration was an atypical presentation of pyloric stenosis , and the baby had a pyloric US which was normal. Lab findings were consistent with hypernatremic dehydration. Urinalysis had a high specific gravity (vs low as would be expected in DI—also in the differential for severe hypernatremia dehydration). The hypernatremia was due to inappropriately concentrated formula coupled with vomiting. The hypercaloric formula was sludge-like consistency, making intake difficult as well as causing emesis. The child was admitted and placed on NS fluids with close monitoring of Na levels. She was transitioned to 24kcal/oz formula, patient did very well without additional emesis or hypernatremia.
As a reminder about management of hypernatremic dehydration: Using NS is appropriate as maintenance IV fluid. NS has 154meq/L of Na, which is significantly lower than the patient’s current Na. As the patient’s sodium falls, it would be appropriate to transition to 1/2NS. The rate of fluid administration depends on the chronicity of the hypernatremia. It should be slow when it's chronic and can be faster when it is acute.
Correction: The above denouement was corrected to clarify the lack of a relationship between correction of hypernatremia and central pontine myelinolysis. In addition, we clarified that the rate of hypernatremic correction partially depends on the chronicity of the hypernatremia. Thank you to our readers for your post-hoc commentary. The Hot Seat is a recurrent online case series aimed at facilitating asynchronous sharing of PEM knowledge and experience. Faculty are placed on The Hot Seat and publish their opinions without knowing the case or its outcomes. Cases are based on real cases and written by PEM fellows at Children's National in DC, Johns Hopkins in MD, and INOVA Fairfax in VA to highlight teaching points. Emily Willner is the faculty advisor for the Hot Seat.
I just wanted to add in a few tidbits about neurologic changes in hypernatremia, as well as sodium correction.
ReplyDeleteIn general, central pontine myelinolysis is seen with rapid correction of HYPOnatremia. Rapid correction of HYPERnatremia may have neurologic consequences, but not specifically central pontine myelinolysis. Instead, what happens is that when you first become hypernatremic, fluid shifts out of your brain cells to balance out all that extra sodium floating around (acute brain shrinkage has been documented). This can cause altered mental status, lethargy, coma and seizures. If this lasts more than 2-3 days, the brain cells actually extract fluid from the CSF so they can be in better intracellular fluid balance. Then, when you rapidly correct the extracellular hypernatremia, the fluid flows back into the brain cells - and now they have too much fluid - resulting in cerebral edema.
As for the rate of correction of hypernatremia, this is going to depend on how long the patient has been hypernatremic. In general, if it has been going on for > 24-48h, you consider it chronic. In chronic hypernatremia, most studies will suggest correcting at about 0.5meq/hour, or a MAX of 10-12 meq/day. In acute hypernatremia (<24h), you want to re-establish normal sodium levels within 24 hours to prevent all of the fluid shifts in the brain in the first place. Realistically, patients with true acute hypernatremia are at extremely low risk for overly rapid correction because the fluid shifts haven't happened yet.
Hope this helps!
I agree completely with Priya. Rapid correction of HYPERnatremia leads to cerebral edema. Rapid correction of HYPOnatremia leads to osmotic demyelination syndrome (formerly known as central pontine myelinolysis).
DeleteAlso, one more correction from the Denouement. I do not think it is correct to state that "lab findings were consistent with hypernatremic dehydration" without knowing the urine [Na]. If low (in the setting of a concentrated urine), then yes, you would be correct. If high, then the hypernatemia would be due to excess solute load -- which is what the history indicates.