Case Conference #6

[Kalel]

Advertisement - Members don't see this ad

Here is a case that we have just admitted tonight (5/10):

A 42 yo WM with a PMHx of bipolar d/o, NIDDM controlled with diet, and polysubstance abuse is transferred from an inpatient psychiatric center to our ER with altered mental status X1 day. The transfer summary indicates that the patient was noted to be confused, SOB, lethargic, and ataxic this morning. Yesterday, nursing staff reports that the patient was drinking a large amount of juice and other fluids with polyuria and had a decreased appetite. In the ER, the patient was found to be unresponsive, tachypnic, with a "fruity breath".

PMHx: Bipolar d/o, NIDDM, polysubstance abuse

Meds: Propanolol 10mg bid, Haloperidol 10 mg bid, Valproate 1,250 mg PO qhs

All: NKDA

SHx: Patient has been an in-patient at a local psychiatric center for the past 6 months. He is unemployed, and has no social suppport system. It is not clear what the "polysubstance abuse" consisted of, or how the substances were abused (IV, smoked or snorted)

FHx: Unknown

ROS: Unobtainable at this time. The patient is lethargic and unresponsive to questioning.

PE: Vitals: Tmax : 103.5, Tc: 102.7, BP: 119/78, P:149, R: 50, Pox: 92% of 15 L of Face Mask, Non-Re-Breatherm 100% Oxygen
Ht: 6ft, Wt: 300lbs
HEENT: NC/AT, dry mucous membranes, oropharynx clear. PERRL, non-injected, anicteric.
Neck is supple, with no JVD, no thyroidomegaly
Resp: CTA b/l, but very tachypnic
CVS: Tachycardic, but no murmurs/rubs/ or gallops. No carotid bruits, pedal and radial pulses present bilaterally.
Abd: Obese, Soft/Non-tender, hypoactive bowel sounds
Skin: No cellulitis, within normal limits

Labs:
WBC: 14.4, Hgb: 18.7, Hct: 58.3, Platelets:Aggregates on smear, too many to count
Sodium: 143, Potassium: 6.9, Chloride: 104, Bicarb: 5, BUN: 55, Creatinine: 3.2, Glucose: 1,024, Calcium: 10.5, PO4: 10.5, Mg: 3.5
U/A: pH: 5.5, 2+ blood and protein, 3+ ketones and glucose
Arterial Blood gas: pH: 7.04, pCO2: 12, pO2: 195, HCO3: 3, O2Sat: 99%, base deficit: -28.1

Questions:
-What diagnosis does our patient have?
-What further tests do you want to order to work up our patient?
-How would you begin to treat our patient?
-Feel free to post any comments about the patient or any of the abnormal lab values too.

Will post further diagnostic tests as requested. If you guys post tests that you want that are reasonable and not ordered yet, I can order them now too (you have to post them in the next day or two because the patient will probably be transferred to the floor pretty soon, I'm in the MICU). Thanks for participating.

 

[DrQuinn]

Give her fluids fluids fluids! Watch the K+ drop, be sure to replace it. After a litre or two, give her some insulin IV.

I'd also like to get an EKG, CXR, and a UDS. Check a lithium level too, even though its not on her meds list.

Not sure of the fever (make sure shes' not in NMS...), whatever caused it could have precipitated DKA.

Q, DO

 

[Kalel]

Give her fluids fluids fluids! Watch the K+ drop, be sure to replace it. After a litre or two, give her some insulin IV.

I'd also like to get an EKG, CXR, and a UDS. Check a lithium level too, even though its not on her meds list.

Not sure of the fever (make sure shes' not in NMS...), whatever caused it could have precipitated DKA.

Q, DO

🙂

Yup, our ER agreed with you. The patient was given 4L NS by IV bolus and begun on an insulin drip that was titrated up to 10 units before the patient was transferred to our unit. An ECG was done that was essentially normal (non-specific ST segment changes in lateral leads). What would you give for the fever? Our ER gave my patient a sterilization regimen of Vancomycin, Ceftriaxone, and then Zosyn. Blood and urine cultures were sent and are pending, and a portable CXR only shows 5 ribs (because the patient is so tachypnic), but otherwise shows no acute cardio-pulmonary process.

I don't think that we need to check a litium level because the psych nurse came with our patient, and a complete med list with check-off's for meds ordered was sent over with my patient. Incidentally, the last finger stick my patient had was last friday, when his finger stick was 129.

That's an interesting thought about NMS. I had not considered that diagnosis. He is on a relatively large dose of haloperidol, and does have hyperthermia. Anyways, with NMS, the patient should have muscular rigidity that would be notable on exam. My patient is relatively flaccid right now. NMS is a clinical diagnosis that requires muscular rigidity to be present along side with hyperthermia, plus leukocytosis and laboratory evidence of muscular injury. I don't think that NMS is associated with the development of DKA, but I suppose that it's possible with NMS producing a stressful event. Anyways, to further work that up, here are some more labs that would be ordered with a NMS w/u:
myoglobin is 474 (high)
AST: 22
ALT: 9
CPK: 59
Tbili: 0.9
Ca: 10.5
PO4: 10.5

 

[DrQuinn]

I didn't think she/he had NMS but I am not sure whyt he patient is febrile. No evidence of UTI, and lungs clear. I would consider a head CT and LP actually (forgot to post this). I would be sure to not miss an infection, as it may have precipitated the event.

He's still young, 41, but does have CAD risk factors (a little on hte large size, DM, possible cocaine, possible tabackey)... myoglobin is elevated, be sure the enzymes are negative.

I hope the ER would agree with me! But I'd rather hear what the MICU resident or attending has to say.

Q, DO

 

[Kalel]

Yup, the MICU resident and attending agree with you. MI is one stressor that has been known to throw patients into DKA. A head CT was ordered by the ED which was read preliminary read (per the ER) as having a Left coronal radiata 1 cm cyst vs infarct, no mass effect, MRI may be indicated. Cardiac enzymes are being drawn q8h. First set:
CPK: 59
CK-MB: 0.5
Troponin I: 0.1

After being transferred to the floor:
Second Set:
CPK: 2,853
CK-MB: 59.0
Troponin I: 2.0

3rd set of cardiac enzymes will be drawn at 1am tomorrow.

At this point, a repeat ECG was ordered which showed no change from previous ECG.
I spoke with my attending and resident about these cardiac enzymes that essentially ruled the patient in for having an MI (troponin-I being the most specific), and they informed me that given the overall picture of the patient, it would not be unusual for the patient to exhibit some element of myocardial ischemia/infarction. If he were a simple CP, r/o MI patient, we would ask that he be cathed (can probably wait until tomorrow morning with a troponin I of 2.0 and no ECG changes). But we decided to give our patient 325 mg of Aspirin and 2.5 mg of Metoprolol IV q6h (we didn't want to beta block our patient completely because the thought was that part of his tachycardia was necessary for his sick state). The myoglobin was a good pick-up, as that is usually the first enzyme elevated (and I think the most sensitive enzyme) early in an MI.

The fever does warrant a further w/u. Empiric abx therapy is warranted in a febrile DKA patient with cultures brewing. The LP is a funny story. The ER actually called me and asked if it'd be alright for them to do the LP while the patient was still down there. Of course I said alright, go ahead and do it (nbice of them to offer), and I've been checking for CSF results for the past 3 hrs until finally, my resident told me that they weren't able to do the LP because of the patient's obesity. So I wanted to LP the patient since he does have altered mental status that is not typical with DKA (although can be explained by the glucose of over 1,000 as a hyper-glycemic, hyperosmolar effect), but the fever and infectious work-up all do point to an LP. However, my resident doesn't want to do an LP, and I can't do one unsupervised, so my patient will not receive an LP tonight. We are continuing the IV Zosyn as broad abx coverage, it's an acceptable choice according to Sanford as empiric abx coverage in fever of a non-neutropenic patient where the source isn't known.

Anyways, I've purposely left out a few other very pertinent labs that will help with the w/u of the case.

Here are his most recent labs though:
Sodium: 164, Potassium: 4.7, Chloride: 142, Bicarb: 15, BUN: 26, Cr: 1.6, Glucose: 203, Calcium: 5.9, PO4: 1.0

Right now, the patient is receiving NS at 250 mL/hr, and NS with 20 mEq KCl at 250 mL/hr.
-Should we change the patient's IVF given his hypernatremia?
-Should we replete the patient's phosphorous and why have we been giving the patient potassium when his first potassium was elevated?
-Why did his calcium drop and what should we do about it (I don't know the answer to this question, but I'm going to be looking it up).

 

[MustafaMond]

Sodium: 164, Potassium: 4.7, Chloride: 142, Bicarb: 15, BUN: 26, Cr: 1.6, Glucose: 203, Calcium: 5.9, PO4: 1.0

Right now, the patient is receiving NS at 250 mL/hr, and NS with 20 mEq KCl at 250 mL/hr.
-Should we change the patient's IVF given his hypernatremia?
-Should we replete the patient's phosphorous and why have we been giving the patient potassium when his first potassium was elevated?
-Why did his calcium drop and what should we do about it (I don't know the answer to this question, but I'm going to be looking it up).

Of course, it is always tough to answer these questions over internet, and it would help to see the patient. But here are my thoughts. I don't know what info you are holding back. TSH, culture results would help.

1), I wouldn't change the fluid. He still needs volume, and Its OK to push NS in the setting of hyperNa+, for the sake of volume. The fact he jumped from
143-->164 overnight is funny, and I think you should get it re-checked. Of course, you never want to drop the Na+ too fast in these guys (cerebral edema)

2) Low phos is seen in DKA in the setting of malnutrition. WHy would you want to replace his PO4 when it is already high?? One thing we all know about DKA is that you aggresively replace K+, because it drops fast, as acidosis corrects.

3) Any patient who rapidly corrects his acidosis will drop their Ca+. THink of the hyperventilating pt. who develops respiratory alkalosis, and gets tetanic spasms from hypocalcemia. The Ca+ should should stabilize soon.


4) As far as MS changes, DKA is notorious for producing severe confusion, and fever could be secondary to another source than CNS. If the patient doesnt have meningeal signs, and LP is difficult, its +/- for LP. If it were me, I'd get it done under flouro.

Good Luck with your patient, and its always great to see a medical student so interested in IM!

 

[Kalel]

Hey Mustafa,
Thanks for participating in my case.

-TSH:0.43 (L), T4:4.3(L), Free T4: 0.6(L).

-I discussed the LP with the attending this morning too. Given his improved mental status (patient is uncooperative with MSE, but at least responsive now), we thought that it wasn't necessary. I was thinking and I suppose that his altered mental status could have been accounted for by a co-commitant hyper-osmolar hyperglycemia state, which is now improving. Glucose only has to be above 800 before you see HHNKC.

-Phosphate levels are expected to be elevated in metabolic acidosis, and as expected, fall as insulin drives the phosphate back into the cells. I came across several articles that seemed to suggest that repletion of phosphate was not neccessary and could even be deleterious, however, my patient's phosphate levels dropped to such low levels that my attending wanted me to replete his phosphate anyways. We decided to give him IV KPhos given his NPO status.

-My attending last night agreed with you regarding the patient's hypernatremia. It was thought that the patient was severed volume depleted (as is to be expected in DKA), and therefore, NS was continued throughout the night at a rate of 500 mL/hr with 20 mEq of KCl/L. It was thought that repleting the patient's volume would result in resolution of the patient's hypernatremia (which reached levels of Na 170 last night, kind of scary when you are pushing NS in a patient). It worked to a point, the patient's labs this morning were:
Na: 162 (H), K: 3.3 (L), Cl: 135 (H), CO2: 20 (L), BUN: 20, Cr: 1.8 (H), Glucose: 277 (H).

As you can see, several changes have occurred. First of all, the patient's anion gap has resolved. It's important to remember that when you are treating DKA, the insulin drip is not titrated to bring the patient's glucose down. In fact, when the glucose levels fall too low (eg below 180), you are supposed to run D5 with your NS in to keep your insulin drip doing (which we have just recentaly starting doing). The insulin is to stop the ketogenesis and promote metabolism of the ketones in order to correct the patient's anion-gap metabolic acidosis. It would also be worthwhile to note that we are not monitoring "acetone" levels. A negative acetone test does not exclude the possibility of DKA because hospital acetone tests do not test for the predominant ketone in DKA (beta hydroxybutarate), and the acetone test can remain positive even after the gap and metabolic acidosis has resolved.

Potassium dropped to low levels as expected, because patients in DKA are usually total body potassium depleted despite having elevated potassium levels upon initial presentation. This is secondary to their dehydration as well as their acidosis driving H+ into cells and K+ out of cells. Insulin corrects this by driving insulin back into cells, which can cause dangerous levels of hypokalemia if not monitored. At this point, the patient's volume status was assessed. He had ~2 L urine output overnight, but remained tachycardic (P: 120-140) but normotensive (BP 140/90). It's odd that his hypernatremia had not yet resolved despite receiving another ~4L of fluid last night, so we decided to correct his hypernatremia more quickly by switching to D5 water. I attributed his tachycardia to his fever, and not his volume status. I looked up the dreaded cerebral edema complication of treating DKA because my attendings had not heard of it when I brought it up, and apparently, it occurs almost occlucesively in children for some reason with type I diabetes.

This brings up another point. Our patient has a history of NIDDM, or type II DM. I'm glad that no one thought that this was a contraindication to him developing DKA, because a significant number of the cases of DKA that we see in the hospital these days are actually in type II diabetics. Supposedly, the mechanism is similar to the mechanism in type I diabetics in that type II diabetics eventually stop producing insulin after a period of hyper-insulinemia in the setting of insulin resistance, plus the insulin resistance can get to point where ketoneogenesis occurs because the liver no longer senses the insulin. Hyperosmolar hyperglycemia non-ketototic coma occurs almost exclusively in type II diabetics, but DKA can occur in both type I and type II diabetics.

However, no one has picked up on another key point on my case. My patient had a finger stick last Friday that was ~123. When he came to us, his glucose was over 1,000. It's important to work up all causes of new-onset DKA, even in a patient with probable sources (eg fever, or infection (pneumonia or UTI, two of the most common triggers of DKA)). What about a sudden loss in pancreatic beta cell function (eg pancreatitis)? Our patient initially lacked the mental capacity to complain of abdominal pain, or complain of nausea or vomiting. Fever and tachycardia are common physical findings, as are abdominal tenderness, guarding, and distension. Also of note, our patient is taking valproic acid, a known cause of pancreatitis. The most common causes of pancreatitis are alcohol (~40%) and gall stones (~40%) though, which our patient could have as well (probably not the alcohol cause because of his inpatient status at a psych facility). A blood glucose of greater then 200 is actually one of the Ranson Criteria for staging pancreatitis. Anyways, amylase and lipase were drawn yesterday by our astute ER staff and here are their numbers:

Amylase: 554 (H)
Lipase: 8,909 (H)

Liver enzymes from yesterday were previously posted. However, even given this presentation of hyperglycemia with those really high amylase and lipase values, because of the fact that our patient is in DKA, we cannot make a formal diagnosis of pancreatitis without seeing an abnormal abdominal CT. If the patient were not in DKA, a diagnosis could be made on clinical grounds, but DKA is a known cause of elevated amylase and lipase itself, so we will have to wait to see what a scan shows us. We ordered an abdominal CT with IV contrast yesterday, but our attending last night made us switch it to an abdominal u/s with the thought that anything correctable with surgery (eg gall stones) should be visible by the u/s. I spoke with our different attending this morning though, and given his improved creatinine, he thinks that it's ok for us to get the abdominal CT with IV contrast. So, hopefully, that is being done now or will be done tomorrow. I drew a CRP this morning (to evaluate prognosis in severe pancreatitis) and the value was:
CRP: <0.5
I don't know if that's suggestive of the patient not having pancreatitis, since it's usually such an inflammatory state.

Another odd thing about my patient were some of the labs we got this morning. I don't know why some of them appear the way that they do, but I'm hoping that this picture will be cleared up by tomorrow:
AST: 566 (H, up from 22)
ALT: 278 (H, up from 9)
GGT: 27 (normal)
T Bili: 0.4
Alk Phos: 86 (Normal, down from 133)
T Cholesterol: 247
HDL: 16 (L)
LDL: Unable to calculate with high triglycerides
Triglyceride: 725 (high)
WBC: 10.3, Hgb: 13.6, Hct: 40.0%, Platelets: 89.
RPR: Negative
With an elevated AST and ALT in the setting of elevated amylase and lipase, one would normally assume a diagnosis of gall stone pancreatitis. However, my patient has a normal amylase leve. The triglyceride level is remarkable as well, however, I've been told that triglycerides are typically in the greater then 1,000's range before they can cause pancreatitis. I will have to do more research on this.

Anyways, hopefully, we get some more answers tomorrow.

 

[Seaglass]

2) Low phos is seen in DKA in the setting of malnutrition. WHy would you want to replace his PO4 when it is already high?

I was under the impression that you would want to maintain a normal Phos in the setting of MI/resolving MI.

In our institution we do not routinely replete Phos for uncomplicated DKA unless it is absurdly low or pt. has known cardiac history.

C

 

[MustafaMond]

Elevated liver enzymes could be from shock liver. Is the patient on tpn?

You said in the initial PE, that belly was non-tender, though pancreatitis can present in DKA without abd pain, it is much less common. Good pickup by you guys. Will be waiting to see what CT shows.

YOur attendings point about USG is, "CT showing pancreatitis will not change the management. You are still going to be giving fluids, as primary therapy. An USG will show Gallstones/sludge, which can be invisible on CT scan."

I would agree with the initial attending to get a USG, as it is more cost-effective, and better at demonstrating Gallstones, a reversible cause.

 

[Tenesma]

"we didn't want to beta block our patient completely because the thought was that part of his tachycardia was necessary for his sick state"

what is that supposed to mean??? he is ruling in for MI and you would rather make his heart more ischemic by tolerating tachycardia????

 

[Goofy]

I generally would work up mi in most patients presenting with dka.

Having said that, this seems to be clearly caused by an infectious etiology. Does the patient have any pulmonary symptoms at all? What kind of initial film did you get? Portable studies are an ER favorite and hardly worth the film they are developed on. Get a good pa and lateral.

Here is a good rule of thumb with LPs. If you think of it, you should probably do it. In medicine there seems to be a reluctance in doing this simple and highly useful procedure. Go and do it!

You need to scrutinize for any evidence of infection. The ER regimen is absolutely ludicrous, and reflects a clear lack of understanding of ID. Did you manage to get two sets of cultures from two sites prior to this antibiotic a-bomb? If not, it may complicate the diagnosis.

You shouldn't hesitate to give normal saline in this situation. In fact, it is a breech of standard of care not to. A bag of NSS is about 152, which is actually HYPOTONIC to this patients sodium. It will correct his hypernatremia!

Get the patient out of the ER as quickly as possible and into a unit.

 

[Goofy]

It was thought that repleting the patient's volume would result in resolution of the patient's hypernatremia (which reached levels of Na 170 last night, kind of scary when you are pushing NS in a patient). It worked to a point, the patient's labs this morning were: Na: 162 (H), K: 3.3 (L), Cl: 135 (H), CO2: 20 (L), BUN: 20, Cr: 1.8 (H), Glucose: 277 (H).

Not scary at all. As stated above, the normal saline is hypotonic to this patient's hypernatremia. Although you are thinking about this the wrong way, its nice to see you did the right thing.

 

[rgizmo]

1. "Potassium dropped to low levels as expected, because patients in DKA are usually total body potassium depleted despite having elevated potassium levels upon initial presentation. This is secondary to their dehydration as well as their acidosis driving H+ into cells and K+ out of cells."


This is not entirely true. The reason patients in DKA present with hyperkalemia is not due to acidosis causing protons into the cells and K out (which only occurs with mineral acids, not organic acids as in DKA), but rather secondary to lack of functional insulin. If the former were true, we would not see hyperkalemia in the setting of non-acidotic hyperosmolar coma (which we, of course, do see, although NKHC can have a lactic acidosis).

2. "The myoglobin was a good pick-up, as that is usually the first enzyme elevated (and I think the most sensitive enzyme) early in an MI."

This isn't exactly correct either:

Diagnostic marker cooperative study for the diagnosis of myocardial infarction.
AU - Zimmerman J; Fromm R; Meyer D; Boudreaux A; Wun CC; Smalling R; Davis B; Habib G; Roberts R
SO - Circulation 1999 Apr 6;99(13):1671-7.

BACKGROUND: Millions of patients present annually with chest pain, but only 10% to 15% have myocardial infarction. Lack of diagnostic sensitivity and specificity of clinical and conventional markers prevents or delays treatment and leads to unnecessary costly admissions. Comparative data are lacking on the new markers, yet using all of them is inappropriate and expensive. METHODS AND RESULTS: The Diagnostic Marker Cooperative Study was a prospective, multicenter, double-blind study with consecutive enrollment of patients with chest pain presenting to the emergency department. Diagnostic sensitivity and specificity and frequency of increase in patients with unstable angina were determined for creatine kinase-MB (CK-MB) subforms, myoglobin, total CK-MB (activity and mass), and troponin T and I on the basis of frequent serial sampling for </=24 hours. Of 955 patients with chest pain, 119 (12.5%) had infarction identified by use of CK-MB mass, and 203 (21%) had unstable angina. CK-MB subforms were most sensitive and specific (91% and 89%) within 6 hours of onset, followed by myoglobin (78% and 89%). For late diagnosis, total CK-MB activity (derived from subforms) was the most sensitive and specific (96% and 98%) at 10 hours from onset, followed by troponin I (96% and 93%), but not until 18 hours, and troponin T (87% and 93% at 10 hours). In unstable angina, CK-MB subforms were increased in 29.5%, myoglobin in 23.7%, troponin I in 19.7%, and troponin T in 14.8%. All markers were increased in 99 patients. With each marker as the diagnostic standard, CK-MB subforms and myoglobin remained the most sensitive for early diagnosis. CONCLUSIONS: The CK-MB subform assay alone or in combination with a troponin reliably triages patients with chest pain and should lead to improved therapy and reduced cost.

3. I agree that that pt's with NSTEMI should be rate controlled with a beta-blocker provided there are no contra's. In this case, you're attending must have felt the pt could not tolerate a beta-blocker because of signs of a tenuous circulatory status. However, the pt deserves a trial of rate control titrated up to that point tolerable. In addition, we don't know how much of a role demand ischemia is playing in the pt's troponin leak (in which case, beta-blockers are vital).

 

[johnd]

This guys hypernatremia is pretty interesting. It would be nice to see the urine na/osm on the 2L he put out overnight. could he have DI?

 

[Whisker Barrel Cortex]

I think the CT is critical in this case and ultrasound will not suffice. If you suspect he has pancreatitis, then he could have necrotizing pancreatitis which will show up on a contrast enhanced CT but will very rarely be seen on U/S. Unfortunately, US is terrible for the pancreas. Although I don't know the data behind it, our GI and surgery guys always treat with antibiotics (often imipenem, or liquid jesus as my surgery resident used to call it) if there is any evidence of necrotizing pancreatitis on CT but no antibiotics if there is not.

 

[Kalel]

Thanks for participating in my case everybody. Here are today's labs:
Na: 152 (H), K: 3.3 (L), Cl: 128, CO2: 21(L), BUN: 13, Cr: 1.2, Glucose: 181
Ca: 6.3 (L), Mg: 2.5, PO4: 1.7
T protein: 4.7, Alb: 2.0, AST: 630 (H), ALT: 469 (H), TBili: 0.9, Alk Phos: 64, GGT: 27
LDH: 2,351 (H)
CPK: 2,594 (H), CK-MB: 3.0, Troponin I:1.5
Amylase: 107, Lipase: 458 (H)
CRP: 3.2
WBC: 3.2 (L), HgB: 12.2, Hct: 36.0, Plt: 45 (L)
Abdominal U/S: Unable to visualize the pancreas secondary to bowel gas, fatty infiltration of the liver, normal gall bladder size, no gall stones
TTE: Essentially normal study except for E to A changes consistent with diastolic dysfunction.
A CT with IV contrast of the chest, abdomen, and pelvis has been ordered. There is a back-up at our CT scanner, so it won't be done until later tonight though.
As far as my patient's status, he has no complaints, but is A+OX0, he just says that he wants to go home. I pressed on his abdomen again today and there was no guarding, no tenderness. He continued spiking a fever last night, Tmax ws 103.7.

Regarding the tachycardia: My understanding is that when patients are sick, whether they have hypovolemia or just some sort of infection, the tachycardia is an adaptive response and you should not try to normalize their rate unless you have to. My patient ruled in with cardiac enzymes, but his enzyme appear to be coming down, consistent the 'cardiac leak' theory as opposed to flat out infarction. He will probably need stress or cath before being discharged, particularly given his unfavorable Lipid panel (HDL of 19). Thanks for the corrections regarding myoglobin and hypokalemia/hyperkalemia in DKA.

Regarding the u/s: I read on e-medicine that the sensitivity of the abdominal u/s in pancreatitis is ~60%. In my patient, it was probably less since he is obese. A CT scan is still warranted in most cases, because if you do see things like necrotizing pancreatitis on the scan, then empiric abx therapy would be warranted. Right now, we are using the empiric abx therapy to cover DKA with fever of unknown origin.

Regarding the fever: Possible etiologies that I was thinking about include pancreatitis which we haven't diagnosed in our patient yet, or maybe hepatitis. The normal GGT is confusing and I asked a GI fellow about it today. He wasn't so sure why the GGT would be normal since it's a pretty sensitive enzyme for any hepatic injury. AST can come from a wide variety of cells, and ALt could come from RBC's, but usually, ALT is from liver. Our patient has had a slight drop in Hct, but this could easily be explained by dilution. Anyways, I've ordered a hepatitis panel. Hopefully, this will be back tomorrow. I agree with the LP, although I doubt it will show anything. It would be nice to rule it out though, but both my fellow and resident don't want to do it, so my hands are tied.

Mysteries: Thrombocytopenia, leukopenia? I suppose sepsis or pancreatitis could account for these findings, but there are still a lot of other "abnormalities". I'm glad that the CRP went up, but I would have thought that would have been elevated earlier if our patient did have pancreatitis. Persistent hypernatremia is strange too. DI is always possible, but we are giving our patient a lot of different diseases. I'm going to check a plasma osmolarity (calculated one was high of course), and a urine osmolarity with urine electrolytes. It's weird, I feel like I'm treating a bunch of abnormal computer lab values because as I stated previously, the patient has no complaints. Anyways, I will post the results of the CT scan. The CT of the chest was ordered to be done at the same time in case the patient has a pneumonia not seen on CXR. Well, I'm off to go sign out now. Let me know if you guys have any other thoughts or ideas!

 

[DrQuinn]

Having said that, this seems to be clearly caused by an infectious etiology. Does the patient have any pulmonary symptoms at all? What kind of initial film did you get? Portable studies are an ER favorite and hardly worth the film they are developed on. Get a good pa and lateral.

You need to scrutinize for any evidence of infection. The ER regimen is absolutely ludicrous, and reflects a clear lack of understanding of ID. Get the patient out of the ER as quickly as possible and into a unit.

Ah, we all would love having a perfect PA and lateral film, but alas, this gentleman was unresponsive, tachypneic, and lethargic. "Sir, would you PLEASE hold still and take a deep breath on 1... 2... SIR! Yyou're not listening to me! I need you to hold your arms out......... SIR! Why are you falling forward?! Can't you tell the IM people are telling me that the portable is useless so I need to get a PA/lat? I know that you physically can't do it, but apparently the ER docs are dumb and don't know which CXR to order. SIRRRRRRRRR why can you not hold your breath for me? LISTEN TO ME!"

*sigh*

And I'm not sure what an "ER regimen" is. Do you mean PCXR, BC x 2, U cx? Last time I remember, when I tried to convince hte medicine residents that BC were not needed and are not indicated in teh VAST majority of CAP patients, I get yelled at and to "get it anyways" even though the evidence is clearly not there. I suppose we in the ER just "came up" with this "ID regimen" on a whim, not because its what our ID or IM counterparts ask on all our patients with a possible etiology.

Q, DO

 

[Tenesma]

kalel:

i think this is an important issue re: letting somebody tach away.... sure, if the patient is young without cardiac issues, it isn't a big deal for a few days to be tachycardic in response to fever, trauma, pancreatitis, etc...

However, in somebody who rules in for an MI - a positive troponin is a specific marker for cardiac myocyte DEATH, whether you call it a leak or not. Therefore, your primary role is to cardio protect the heart. There is NO argument in the world not to use beta-blockers and rate control (unless the patient is anaphylactic to beta-blockade or has "demonstrated" worsening reactive airway disease in the setting of beta-blockade).... even if the patient has poor circulatory status, which will respond to volume and if need be pressor support. Not giving a beta-blocker is like saying: i won't give supplemental Oxygen, because 20% oxygen should be enough to help an ischemic/infarcted/"leaking" heart... it just doesn't make sense, and won't hold up in court from a standard of care point of view.

 

[Dubin]

Agree with the above post with regards to B-blockade. I can't imagine that a cardiology service would allow this.

As for the thrombocytopenia, I assume this is a new developement since being in the hospital. If so, consideration to stopping Lovenox/heparin and checking for HIT is necessary. If not, drug-induced thrombocytopenia is a possibility, even after a day or two. For this patient, Plavix or Zosyn might be some likely drugs he received. Sepsis, a consumptive coagulopathy, and DIC must always be in the back of your mind. Follow counts would be a red flag to me that this patient is not as well as he may appear to you.

 

[Kalel]

Hey Guys,
I appreciate your comments, but I strongly disagree with your comments regarding the usage of beta blockers in patient. I looked it up, and actually, the strongest evidence of using beta blockers in acute myocardial infarctions is with significant infarctions (eg ECG changes present, higher levels of troponin, or particularly with wall motion abnormalities). My patient had none of these. According to CMDT, the evidence is sparse when it comes to small MI's and the benefit of the usage of beta blockers.

However, I am not a cardiologist, and since Tenesma was so insistent, I decided to curbside one who I happened to run into in the hall. She agreed with me. The patient presented with a metabolic acidosis and pH of ~7.0. The patient does have CAD (DM history with poor lipid profile), but even though the patient's troponin I did rule the patient in, the patient did not suffer an acute myocardial infarction in the traditional sense. The patient had a few myocardial cell deaths in the setting a acidemia that released a small amount of myocardial enzymes. There is not a strong indication to put a patient on beta blockers in this setting. She told me that she wouldn't even bother stressing or cathing the patient before discharging the patient, because the cardiac enzymes were not released secondary to any plaque rupture, stenosis, or coronary artery occlusion. They were released secondary to the patient's poor metabolic health.

I would encourage you guys to ask cardiologists at your respective schools to see if they disagree with my cardiologist.

I had actually already started the patient on beta blockers before I curbsided this cardiologist (metoprolol 2.5 mg IV q6h), and he remained tachycardic (100's) and hypertensive (140/90s). Today, I asked my pulmonary/critical care attending if we could increased his metoprolol too, and he actually made me d/c it. I agree with his reasoning for d/cing it too. I will explain after I give my patient's update:

New Labs for today:
Na: 151 (H), K: 3.9, Cl: 122 (H), CO2: 20, BUN: 16, Cr: 1.2, Glucose: 271, Anion Gap: 8, Ca: 7.1, PO4: 1.4 (L), Mg: 2.6
Tprotein: 4.8(L), Albumin: 2.2 (L), AST: 436 (H), ALT: 491 (H), Tbili: 1.0, Alk Phos: 64, LDH: 1,942 (H)
CPK: 1,251 (H), CK-MB: 1.2, Troponin I: 0.3
WBC: 5.0, Hgb: 11.5 (L), Hct: 34.4 (L), Platelets: 36 (L)
U/A: +2 blood, trace protein, +3 ketones, +3 glucose, 0-2 Hyaline Casts, 0-2 Fine gran Cast
Uosm: 752, Una: 100, Uk: 29, Ucr: 68
Posm: 320
Blood cultures X4, and UCx remain negative.
Hepatitis viral panel still pending.

CT of the abdomen: the head of the pancreas appeared edematous with a small amount of fluid tracking to the right peri-renal region. These findings were consistent with pancreatitis. I spoke with the radiologist regarding these findings, and she informed me that there was no evidence of necrotizing pancreatitis, and that there was not enough fluid to tap.
CT of the chest: (not officially read yet): shows bilateral lower lobe infiltrate with a small bilateral lower lobe pleural effusion collection.

O'N the patient remained Febrile (Tm: 101.5), but his mental status improved to A+OX2. His psychiatric nurse who stays with him in his room reported that he was not back to his baseline yet. The patient still did not complain of any abdominal pain or nausea, so I took out his NGT and started the patient on a clear liquids diet.

So now there are several new issues to discuss. My patient has a diagnosis of pancreatitis, which probably led to his DKA. By Ranson Criteria:

On admission:
Age >55 years.
WBC count >16,000.
Blood glucose >200 mg/dl.
LDH >350 IU/L.
AST >250 IU/l

At 48 hours:
Fall in Hct >10%.
Rise in BUN >5mg per dl.
Serum calcium <8mg per dl.
Arterial pO2 <60mm Hg.
Base deficit >4 meq/l.
Estimated fluid third spacing of >6 liters.

My patient meets 6 of these criteria (glucose, LDH, AST, Hct, Ca, base deficit). Thus, his predicted mortality for his hospital stay is ~40%. I was very surprised to read this because he looks so healthy right now and essentially has no complaints. I still feel like I am just treating lab abnormalities in him.

Regarding the persistent fever: I was interested in changing his current regimen (which only consists of zosyn right now) since it doesn't appear to be helping his fever. Standard abx therapy for necrotizing pancreatitis is actually imepenam, but my patient does not have necrotizing pancreatitis as evidence by his CT scan which has an accuracy of making this diagnosis close to 90%. We still don't know what we are treating, so I was interested in treating him as if he were hospitalized in the ICU for pneumonia, and thus adding a fluorquinolone (ciprofloxacin) to his zosyn regimen, both of which would provide good coverage for gram negative enteric organisms in case his pancreatitis were to become infected too. However, my attending disagreed, and felt that zosyn would provide sufficient coverage, so that's all he's on right now. His fever could be from several sources, it could be seconary to his acute pancreatitis, his pneumonia, or it could be from an infection that we are missing. The patient had a femoral TLC placed by the ER on admission which we had been conveniently drawing our q2-4 labs from, but my attending made me pull it out because of his persistent fever.

Regarding the patient's elevated LFT's, that's still difficult to explain to me. My attending thinks that it's secondary to the liver being under-perfused or SIRS (Severe Inflammatory Response Syndrome). I'm still confused as to why the GGT would remain normal in this setting, since GGT should be a pretty sensitive marker for any hepatocyte injury. I asked a GI fellow about this, and he didn't know either. I will try to do some more research into it when I have the time. Nevertheless, the hepatitis panel is still pending, even if it may have not been an efficient use of resources.

Regarding the patient's persistent hypernatremia, this kind of confuses me too. Our calculated Uosm suggests that the patient does not have diabetes insipidus and is retaining water well (Uosm>450 suggests appropriate ADH function). My only explanation is that the patient probably has an element of Acute tubular necrosis, with the fine granular casts seen in U/A, presumably acute elevation in Cr, and from his SIRS. This may be impeding his urine concentrating ability. His hypernatremia is improving as we continue to give him fluids (I had him on D5, but his sugars were too difficult to control, so I put him back on 1/2 NS running at 150 cc/hr). I would caculate a water deficit in him and figure out exactaly how much water he needs, but to be frank, I'm not even certain if he's euvolemic, hypervolemic, or hypovolemic. Part of the pathogenesis of acute pancreatitis is sequesteration of water, so it's possible that he is still hypovolemic despite his positive intake by 5-6 liters over the past few days. His tachycardia would also be consistent with that. But then he is also normotensive/hypertensive, his BUN has dropped dramatically, so both of those signs would point to euvolemia to hypervolemia.

Regarding the throbocytopenia, I agree with the heparin thought so I had already d/ced the sub-cut heparin. We are using venodynes for DVT prophylaxis instead now. It may also be secondary to a DIC picture or sepsis too. If his platelets fall below 20, we will have to transfuse him.

I think that d/cing his beta blocker was the right decision. My patient is at high risk for developing sepsis, and will therefore need his reflexive tachycardia and to maintain his blood pressure and prevent further organ damage. Also, if my patient does become septic, he may require pressor support. The pressor of choice for sepsis is Levophed (Norepinephrine), which would be problematic to use in patient who is being beta blockaded, in terms of the having the pressor work, and also in terms of having unopposed alpha constriction. I agree that from a long-term cardiac health standpoint, it would probably be beneficial for the patient's heart if we were to completely beta blockade him with post-MI beta blockers to prevent his heart from working too hard (although the evidence to suggest this is lacking, it makes sense from a physiologic standpoint). However, our main concern right now should be making sure that my patient survives his hospital stay in the ICU.

Regarding the etiology of my patient's pancreatitis, I supsect that it was secondary to his valproic acid. However, we can't rule out something like micro-cholelithiasis or his hypertriglyceridemia (usually they have to be over 1,000 to cause pancreatitis, but 780 is still really high). We have a lipid specialist who I did research with that I will refer the patient to if we discharge him directly rather then transferring him to another service to have his lipids addressed.

Thanks a lot for participating in my case everybody. Even my attending (our university's MICU director) says that my patient is a very interesting patient. I will update you guys if anything changes or if I have any other new thoughts.

 

[Tenesma]

the most common cause of thrombocytopenia in this situation is purely dilutional due to the massive volume resuscitation required... unlikely to be heparin, since it takes a few days before you start seeing thrombocytopenia from HIT (unless it is type II - in which case they would be throwing clots all over the place).

re: tachycardia in the setting of troponin bump. I understand that the patient has a fever-related tachycardia, and you may be right that the stress of the acidemia, etc has caused subendocardial damage. I also agree that this is obviously not an Acute Coronary syndrome and therefore doesn't warrant a cath. He may need a stress test down the line, because it sounds like he definitely has demand-related ischemia.

Beta-blockers decrease myocardial oxygen consumption - which is what he needs. 2.5mg IV q6 is about as homeopathic a dose as you can get... I would target a Heart Rate less than 90. Patients with pancreatitis are notoriously difficult to slow down - or else my goal would have been 80 or less...

If it makes you uneasy to use metoprolol you can use his propranolol (home med) and titrate upwards.

 

[Crypt Abscess]

Good case Kalel. In the setting of unexplained fever and pancreatitis, cholangitis should always be on your differential, especially with the bump in liver enzymes. I know- the normal GGT, alk phos, and US do not support cholangitis, but it is still a possibility. In such a big man, I wonder how much of the CBD and ducts they were able to visualilze on US. If you don't have GI on board I would recomend doing so- he may need EUS or ERCP. A MRCP may also give you a little more info before doing anything invasive such as ERCP. Good luck and please keep us posted. Crypt

 

[Tenesma]

if you want something specific to bile duct you can always check a 5' nucleotidase level...