Depakote-induced LFT elevation

[futuredoctor10]

So we all learn in school that Depakote can cause elevated LFTs. Had an interesting patient we were trying to determine if her LFT abnormality was from Depakote versus underlying liver disease:

1. What is the typical extent of the LFT increase (how high do AST and ALT rise) when on Depakote?

2. What is the typical time frame to witness these LFT increases, in other words how soon after starting the drug? Can it happen 2 years, 5 years, or 10 years after starting the drug without any problems?

3. How often do psychiatrists check LFTs on patients on Depakote?

Thanks!

---

Members don't see this ad.

 

[Shikima]

I always check ammonia levels where VPA is used. Also don't forget to monitor LFTs with Zoloft - caught my first case of an irritable liver from this medication, and at a low dosage too.

 

[digitlnoize]

I always check ammonia levels where VPA is used. Also don't forget to monitor LFTs with Zoloft - caught my first case of an irritable liver from this medication, and at a low dosage too.

Agreed about ammonia. We see this a lot with VPA.

 

[PistolPete]

My understanding is that ammonia levels do NOT correlate with any alteration in mental status, no? How high would ammonia have to be to stop depakote? What about LFT's? >2x high end of normal?

 

[Jester25]

So we all learn in school that Depakote can cause elevated LFTs. Had an interesting patient we were trying to determine if her LFT abnormality was from Depakote versus underlying liver disease:

1. What is the typical extent of the LFT increase (how high do AST and ALT rise) when on Depakote?

2. What is the typical time frame to witness these LFT increases, in other words how soon after starting the drug? Can it happen 2 years, 5 years, or 10 years after starting the drug without any problems?

3. How often do psychiatrists check LFTs on patients on Depakote?

Thanks!

1) LFT elevation up to 3x the normal limit may be acceptable depending on risk-benefit ratio. Higher than that, should seriously consider another agent/ tapering off Depakote. Of course, this discussion needs to be had with the patient as well.

2) I am not aware of data that says a specific time frame. However, from clinical experience most LFT elevations occur in the first year. Doesn't mean it can't occur after that and I am sure many have.

3) I would get baseline CMP, then in 6 months, then yearly after that. I would also monitor for clinical signs of hepatitis/ pancreatitis by asking about nausea, vomiting, diarrhea, anorexia, abdominal pain, edema, and weakness as well as visibly looking for jaundice.

 

[Jester25]

My understanding is that ammonia levels do NOT correlate with any alteration in mental status, no? How high would ammonia have to be to stop depakote? What about LFT's? >2x high end of normal?

I treat Depakote like an atypical given it's weight gain properties. Therefore, I get baseline Lipid Panel and A1C, then at 6 months, then yearly after that. I also get yearly CBC to monitor platelets, HFP, and amylase/lipase.

As far as how high Ammonia would have to be, I think that would depend on how well the medication has been working for the patient, how other medications have worked for the patient, has there been an AMS change/ anorexia/ other severe symptoms, and how they respond to something like lactulose. I would also make sure they don't have a urea cycle disorder.

 

[nitemagi]

My understanding is that ammonia levels do NOT correlate with any alteration in mental status, no? How high would ammonia have to be to stop depakote? What about LFT's? >2x high end of normal?

Blood levels don't necessarily correlate with brain levels. Meaning if you are able to bring down an elevated blood level in someone with suspected hepatic encephalopathy, that doesn't mean the brain level has normalized yet.

I'm not sure there's a cutoff for when to stop it. Some like to first try adding L-carnitine (for liver damage), if available. If there's persistent hepatitis, even with that, I'd probably stop it. No point causing ongoing liver damage if something else hasn't been tried. For the ammonia it again depends on the situation (I've seen consult docs add lactulose if the depakote has otherwise been the only effective intervention).

 

[futuredoctor10]

Thanks for all the input! I'm going into a non-psych residency so this is helpful to know 🙂

 

[futuredoctor10]

I always check ammonia levels where VPA is used. Also don't forget to monitor LFTs with Zoloft - caught my first case of an irritable liver from this medication, and at a low dosage too.

Ooh good point on Depakote-induced elevation of ammonia. I saw that in one encephalopathic patient with mild elevation in ammonia, no other source so believed to be Depakote.

Regarding LFT elevation with Zoloft, I have never heard of this side effect... interesting.

Also with SSRIs in general, I was curious about timeline for hyponatremia. From my research, it seems the hyponatremia is typically during the first 2 weeks of treatment. So do you check weekly BMP to monitor the Na+ when starting an SSRI? This is interesting, since we had a patient with hypoxia who was also hyponatremic, and we were not sure if the hypoNa was due to (a) SIADH from lung cancer given her smoking history versus (b) SSRI-induced versus (c) other cause. She'd been on the SSRI for years... so I am guessing the hyponatremia from SSRI should have been much lower on the differential.

 

[Jester25]

i don't think getting routine HFP for a patient on Zoloft is standard of care. if patient complains of side effects that make you think liver, that's a different story.

no you don't check weekly BMPs to monitor for Na. Get it at baseline and yearly at max unless patient complains of side effects that make you think it could be 2/2 hyponatremia. some complaints are so non-specific that it would take a few visits for me to want to get labs on it. i did catch it on an elderly patient thou (more common in elderly) that had been on an SSRI for years.

 

[NJWxMan]

Lots of bad psychiatry in here! Almost everything being mentioned has little to no evidence. If you think you have a trick up your sleeve that few know about, it's probably bunk.

 

[Jester25]

Lots of bad psychiatry in here! Almost everything being mentioned has little to no evidence. If you think you have a trick up your sleeve that few know about, it's probably bunk.

Specifically what?

I think for Depakote the standard of care would be to get LFTs + CBC + VPA level. The Lipid panel and A1C are not standard of care but is something I picked up from a hospital I worked at where it was protocol to do it yearly. I think it's a good idea given the properties of Depakote.

As far as SSRIs, I was trained to get baseline labs. After that it is up to the clinician. I usually wait for symptoms of hyponatermia before ordering BMP or bleeding/ easy bruising before ordering CBC. You don't have to monitor SSRIs with labs unless you suspect these things. Some clinicians feel yearly labs is important, I am not one of them…thus I said "yearly at max".

I don't routinely order Ammonia unless I have high clinical suspicion.

 

[splik]

why check VPA level as standard? what is the point? it's not lithium, we don't go for a target level, and there is so much variability in the serum level it's pretty much useless to check. i only check it if pt appears toxic or to confirm they aren't taking their meds.

 

[notdeadyet]

You can check VPA levels to make sure the patient isn't an odd metabolizer to ensure you're not giving them potentially overly heavy doses that would unnecessarily increase the chance of developing side effects. Where I've worked, we typically check within a few days of dose change. After that, I do an annual check.

 

[Jester25]

why check VPA level as standard? what is the point? it's not lithium, we don't go for a target level, and there is so much variability in the serum level it's pretty much useless to check. i only check it if pt appears toxic or to confirm they aren't taking their meds.

Not sure what you mean by this. I don't like to use Depakote much, but we all know the therapeutic range of 50-150 and ideally in the 75-125 range. If a patient is taking it as prescribed and there are absolutely no changes then the level should stay the same year to year, but patient's med lists change and their health changes. For this reason, the cost benefit ratio of checking it yearly in a stable patient easily warrants one imo.

 

[NJWxMan]

Hgb A1C? Absolutely not the standard of care, expensive, and most insurances don't psych for it unless you have a documented hx of DM.

Ammonia levels? Why? There's no correlation.

L Car. In elevated LFT's? Again, little evidence.

VPA levels should be checked 7 days after dose change. Patient should hold VPA dose morning of draw.

Specifically what?

I think for Depakote the standard of care would be to get LFTs + CBC + VPA level. The Lipid panel and A1C are not standard of care but is something I picked up from a hospital I worked at where it was protocol to do it yearly. I think it's a good idea given the properties of Depakote.

As far as SSRIs, I was trained to get baseline labs. After that it is up to the clinician. I usually wait for symptoms of hyponatermia before ordering BMP or bleeding/ easy bruising before ordering CBC. You don't have to monitor SSRIs with labs unless you suspect these things. Some clinicians feel yearly labs is important, I am not one of them…thus I said "yearly at max".

I don't routinely order Ammonia unless I have high clinical suspicion.

 

[Jester25]

Hgb A1C? Absolutely not the standard of care, expensive, and most insurances don't psych for it unless you have a documented hx of DM.

Ammonia levels? Why? There's no correlation.

L Car. In elevated LFT's? Again, little evidence.

VPA levels should be checked 7 days after dose change. Patient should hold VPA dose morning of draw.

On average patients gain 15lbs on Depakote. For most patients, this is greater than 5% of the body weight. This weight gain predisposes them to increased risks of metabolic syndrome and diabetes different from the mechanism of antipsychotic induced insulin resistance which is independent of weight gain.

I am certainly not calling it the standard of care, but logically it makes sense to me and if I were a patient taking Depakote with weight gain (which happens frequently), I would want yearly Lipid Panels and A1Cs.

Take a look at this article.

 

[Jester25]

Hgb A1C? Absolutely not the standard of care, expensive, and most insurances don't psych for it unless you have a documented hx of DM.

Ammonia levels? Why? There's no correlation.

L Car. In elevated LFT's? Again, little evidence.

VPA levels should be checked 7 days after dose change. Patient should hold VPA dose morning of draw.

Also VPA levels can be checked within 3-4 days given the half life is 16 hours at most. Don't need to wait a full week.

 

[Shikima]

Elevated ammonia levels are bad for the brain.

 

[NJWxMan]

Again, no one is saying that DM and weight gain are not risks with Depakote. That's not the issue. The issue is ordering labs that are unnessecary. The ordering of Hbg A1c's on all patients on SGA's and Depakote started in the VA system. In the oupatient setting, internists screen for DM via fasting blood glucose; they don't order A1C's on everyone. The standard approach is to closely monitor fasting BG.

As for VPA levels being checked 3-4 days out...I agree that one can check 3-4 days after dose change, but clinically, one has to take into account the real world scenarios. Patients may take a few days to fill their new dose prescription; they get confused and get the lab drawn a day early; etc. Lots can go wrong. From where I trained, we would have patients complete lithium or depakote levels 1 week after the dose change.

On average patients gain 15lbs on Depakote. For most patients, this is greater than 5% of the body weight. This weight gain predisposes them to increased risks of metabolic syndrome and diabetes different from the mechanism of antipsychotic induced insulin resistance which is independent of weight gain.

I am certainly not calling it the standard of care, but logically it makes sense to me and if I were a patient taking Depakote with weight gain (which happens frequently), I would want yearly Lipid Panels and A1Cs.

Take a look at this article.

 

[Jester25]

Again, no one is saying that DM and weight gain are not risks with Depakote. That's not the issue. The issue is ordering labs that are unnessecary. The ordering of Hbg A1c's on all patients on SGA's and Depakote started in the VA system. In the oupatient setting, internists screen for DM via fasting blood glucose; they don't order A1C's on everyone. The standard approach is to closely monitor fasting BG.

As for VPA levels being checked 3-4 days out...I agree that one can check 3-4 days after dose change, but clinically, one has to take into account the real world scenarios. Patients may take a few days to fill their new dose prescription; they get confused and get the lab drawn a day early; etc. Lots can go wrong. From where I trained, we would have patients complete lithium or depakote levels 1 week after the dose change.

That is the issue. You are giving patients medications that will increase their risk of having metabolic syndrome and diabetes. Ordering lipid panels and A1C yearly to monitor this makes it necessary and worthwhile. If the A1C is trending in the wrong direction, I would need to have a discussion with the patient and consider different options. FM docs and internists order A1Cs on patients all the time. I see what you are saying about the FBG which offers it's own pluses, but A1C is becoming the standard.

http://care.diabetesjournals.org/content/34/Supplement_2/S184.full

 

[Jester25]

Again, no one is saying that DM and weight gain are not risks with Depakote. That's not the issue. The issue is ordering labs that are unnessecary. The ordering of Hbg A1c's on all patients on SGA's and Depakote started in the VA system. In the oupatient setting, internists screen for DM via fasting blood glucose; they don't order A1C's on everyone. The standard approach is to closely monitor fasting BG.

As for VPA levels being checked 3-4 days out...I agree that one can check 3-4 days after dose change, but clinically, one has to take into account the real world scenarios. Patients may take a few days to fill their new dose prescription; they get confused and get the lab drawn a day early; etc. Lots can go wrong. From where I trained, we would have patients complete lithium or depakote levels 1 week after the dose change.

Also, it's possible to have a normal blood glucose level but an elevated A1C since it measures 3-4 months worth. I prefer the A1C.

 

[NJWxMan]

Jester... I don't know where to begin. Your thinking here is a bit skewed. The article you attached is not a guideline and it clearly identifies many cons of using an A1C for screening. It's widely known that an A1C lack sensitivity. There are also racial disparities, and some medications may affect the result as well. It's also expensive. The guidelines are there for a reason.

That is the issue. You are giving patients medications that will increase their risk of having metabolic syndrome and diabetes. Ordering lipid panels and A1C yearly to monitor this makes it necessary and worthwhile. If the A1C is trending in the wrong direction, I would need to have a discussion with the patient and consider different options. FM docs and internists order A1Cs on patients all the time. I see what you are saying about the FBG which offers it's own pluses, but A1C is becoming the standard.

http://care.diabetesjournals.org/content/34/Supplement_2/S184.full

 

[Jester25]

Jester... I don't know where to begin. Your thinking here is a bit skewed. The article you attached is not a guideline and it clearly identifies many cons of using an A1C for screening. It's widely known that an A1C lack sensitivity. There are also racial disparities, and some medications may affect the result as well. It's also expensive. The guidelines are there for a reason.

Like I have said multiple times already, I am not saying it is the standard of care or part of some guideline to monitor A1C in Depakote patients. I'm using clinical common sense here. When a patient gains weight, they are at risk for metabolic syndrome and diabetes. A1C is part of the standard of care that can be used as both a tool for screening and diagnosis. If you did not know this, see link below. Therefore, getting an A1C in an overweight patient/ someone that has gained a decent amount of weight (>5% of the original weight) is a reasonable decision. I have seen internists do it multiple times on my patients over the winter since a few of them have packed on serious weight. I have also obtained A1Cs on patients in my moonlighting gig and have never had an issues with insurance not clearing it. Not once.

http://care.diabetesjournals.org/co...48e575ecfb9ed63fba7cb8aa&keytype2=tf_ipsecsha

Also, I provided the link to break down the pros and cons of using the A1C. That's it.

 

[Jester25]

Also…for whatever it's worth. Ran this by my wife who is a medicine resident and she said it's a perfectly reasonable thing to do. As I already mentioned, the hospital I work at has it as part of it's protocol. Stahl's Prescriber's guide suggests it. Etc…

 

[NJWxMan]

So basically you don't practice the standard of care nor practice evidence based medicine....that's superb.

You are incorrect about screening with an A1C; it's not very sensitive.

Again, you can past links to any article you want...these are not high impact journals that you are posting. Also, anecdotal practice does not equal good practice.

Like I have said multiple times already, I am not saying it is the standard of care or part of some guideline to monitor A1C in Depakote patients. I'm using clinical common sense here. When a patient gains weight, they are at risk for metabolic syndrome and diabetes. A1C is part of the standard of care that can be used as both a tool for screening and diagnosis. If you did not know this, see link below. Therefore, getting an A1C in an overweight patient/ someone that has gained a decent amount of weight (>5% of the original weight) is a reasonable decision. I have seen internists do it multiple times on my patients over the winter since a few of them have packed on serious weight. I have also obtained A1Cs on patients in my moonlighting gig and have never had an issues with insurance not clearing it. Not once.

http://care.diabetesjournals.org/co...48e575ecfb9ed63fba7cb8aa&keytype2=tf_ipsecsha

Also, I provided the link to break down the pros and cons of using the A1C. That's it.

 

[Jester25]

Hahahaha so guidelines indicate that A1Cs can be used for both screening and diagnosis purposes but since NJwx disagrees, then it's useless. Got it.

 

[splik]

I think this just goes to show that it is easy to become out of touch with medicine as you get further in your psych training. For several years the American Diabetes Association has recommended the use of A1c for diagnosis of diabetes and this is the standard of care these days. Sure there are advantanges and disadvantages, and you could use FPG etc, but it is wrong to say A1c is not used and recommended as a diagnostic tool for diabetes. Also Diabetes Care is the flagship journal for diabetes, and has a fairly high impact factor so you're just making yourself look silly NJwx.

 

[NJWxMan]

The journal has an impact factor of 7 in the endocrine field ....not even a top 30 for all journals, but whatever.

Anecdotally, my patients, both adults and peds, do a good job seeing their PMD's. They usually have their labs completed by their PMD before seeing me. Rarely will they have an A1c. Again, this is anecdotal.

I understand wanting to give patients the most up to date care. I do some CL work and I try to stay at least knowledgable of some of the larger studies. Overall, the various boards have yet to come to a true recommendation for A1c. To not recognize problems with A1c would be ignorant. For example, the AACE updated their guidelines which note that you can screen with A1C, BUT if abnormal, you have to complete a fasting blood glucose. Why not just draw a FBG first?

I think this just goes to show that it is easy to become out of touch with medicine as you get further in your psych training. For several years the American Diabetes Association has recommended the use of A1c for diagnosis of diabetes and this is the standard of care these days. Sure there are advantanges and disadvantages, and you could use FPG etc, but it is wrong to say A1c is not used and recommended as a diagnostic tool for diabetes. Also Diabetes Care is the flagship journal for diabetes, and has a fairly high impact factor so you're just making yourself look silly NJwx.

 

[SteinUmStein]

"Various boards have yet to come up with a true recommendation for A1c"? Are you sure?

http://care.diabetesjournals.org/content/33/Supplement_1/S4.full