Skip to content

Amylase Level for Pancreatitis: Stop doing it

2016-11-11T19:19:23+00:00

BloodTestTubeA patient actively vomiting is wheeled into your ED. Within minutes IV access is obtained, and your nurse asks what tests and medicines are wanted. A liter of normal saline, ondansetron, and an H2 blocker are easy, but what labs to order? I think we can all agree on a metabolic profile to look at electrolytes and liver function tests, and a lipase level to assess for pancreatitis. But what about an amylase level?

Originally from Clinical Monster blog

It is known to increase in pancreatitis, but we are already ordering a lipase level. Are we just ordering two tests for the same purpose? Which is better and does it help to order both?

Sensitivity of Amylase For Pancreatitis

In one study [1], among patients diagnosed with alcohol pancreatitis, 32% had a NORMAL amylase level!

A second paper by Clavien et al [2], found that 20% of people with pancreatitis on CT had normal initial amylase levels in the ED upon their arrival. However of those 20%, 88% had an elevated lipase. Here, amylase levels were much less sensitive if the patient presented with:

  • A longer duration of symptoms
  • Alcoholic pancreatitis
  • Chronic pancreatitis

More studies have supported the above findings. Amylase [3] levels returned to normal much faster than lipase levels in acute pancreatitis, some within 24 hours. Amylase [4] levels rise less high and less often, and return to normal sooner in alcohol pancreatitis compared to gallstone pancreatitis. Amylase [5] levels are often spuriously normal in cases of hypertriglyceridemia induced pancreatitis. Lastly, one study [6] found a 91% decrease in amylase production vs a 26% decrease in lipase production in patients with chronic pancreatitis.

Conclusion: Amylase does not have a high sensitivity in pancreatitis, except in the specific case of gallstone pancreatitis presenting less than 24 hours after symptom onset.

Specificity of Amylase for Pancreatitis

According to a review article [6], serum amylase is elevated often in:

  • Celiac disease
  • HIV
  • Lymphoma
  • Ulcerative colitis
  • Rheumatoid arthritis
  • Monoclonal gammopathy
  • Renal failure
  • Ectopic pregnancy
  • Salpingitis
  • Anorexia/bulimia
  • Parotitis
  • Bowel obstruction/infarction
  • Tumors involving/inflitrating the salivary glands
  • Amylase producing tumors

Unfortunately in comparison lipase is also elevated in HIV, diabetic ketoacidosis, bowel obstruction/infarction, and renal failure – a shorter list, but still not 100% specific. In the review, the specificity of lipase was quite different across all the papers reviewed. Possible sources of this difference may have been the cut-off used for an abnormal value, the type of assay used, and other differences possibly related to the pancreatitis’ etiology.

Conclusion: Specificity is not great for amylase. It may not be great for lipase either. Lipase is at least as specific, possibly more so.

Comparison of Amylase Versus Lipase

A retrospective chart review [7] examined 1,000 pancreatitis patients. Receiver operator curves showed improved diagnostic accuracy of lipase over amylase using lipase >208 U/L, (normal< 190 U/L). 20% of patients did not have elevated amylase on their initial ED visit, while only 3% did not have an elevated lipase.

A review compared various papers’ sensitivities of lipase vs amylase – 95% vs 79%, 100% vs 72%, 100% vs 55%. And concluded that:

“The cumulative literature supports the replacement of amylase with lipase [6].”

So lipase is better than amylase. But are both better than lipase alone?

One study [8] that validated lipase’s increased sensitivity compared to amylase also discovered that evaluating the amylase level in addition to the lipase level “does not enhance discrimination of acute pancreatitis”.

A review [9] declared that 6 different papers have concluded that there is no benefit to adding amylase to lipase and declare that “the obtaining of both serum amylase and lipase levels is not warranted.” 

So… STOP ORDERING AMYLASE

This information is apparently so common knowledge that there is actually a paper [11] published on the best way to stop EM doctors from ordering amylase. One analysis showed 93% of patients with lipase ordered had amylase ordered as well pre-intervention. An educational intervention was unsuccessful at decreasing the concomitant ordering of amylase! Only by removing amylase from the order sets, did they find a significant decrease in amylase orders – down to 25%. Removing it from the pre-set “trauma labs” labs lowered it even more – to 14%. 

Conclusion

Stop ordering an amylase level. Of the two tests for pancreatitis, lipase is better. Adding an amylase to the lipase does not help. Education interventions such as lectures (or writing a blog about it), does not seem to reduce ED physicians ordering amylase. A more effective stratagem seems to involve removing from the electronic medical order sets.  

Reprinted with permission from clinicalmonster.com/blog

References

  1. Spechler SJ, Dalton JW, Robbins AH, Gerzof SG, Stern JS, Johnson WC, Nabseth DC, Schimmel EM. Prevalence of normal serum amylase levels in patients with acute alcoholic pancreatitis. Dig Dis Sci. 1983 Oct;28(10):865-9. PMID: 6193932.
  2. Clavien PA, Robert J, Meyer P, Borst F, Hauser H, Herrmann F, Dunand V, Rohner A. Acute pancreatitis and normoamylasemia. Not an uncommon combination. Ann Surg. 1989 Nov;210(5):614-20. PMID: 2479346
  3. Ventrucci M, Pezzilli R, Naldoni P, Platè L, Baldoni F, Gullo L, Barbara L. Serum pancreatic enzyme behavior during the course of acute pancreatitis. Pancreas. 1987;2(5):506-9. PMID: 2444967.
  4. Hiatt JR, Calabria RP, Passaro E Jr, Wilson SE. The amylase profile: a discriminant in biliary and pancreatic disease. Am J Surg. 1987 Nov;154(5):490-2. PMID: 2445215.
  5. Toskes PP. Hyperlipidemic pancreatitis. Gastroenterol Clin North Am. 1990 Dec;19(4):783-91. PMID: 2269517.
  6. Yadav D, Agarwal N, Pitchumoni CS. A critical evaluation of laboratory tests in acute pancreatitis. Am J Gastroenterol. 2002 Jun;97(6):1309-18. PMID: 12094843.
  7. Smith RC, Southwell-Keely J, Chesher D. Should serum pancreatic lipase replace serum amylase as a biomarker of acute pancreatitis? ANZ J Surg. 2005 Jun;75(6):399-404. PMID: 15943725.
  8. Werner M, Steinberg WM, Pauley C. Strategic use of individual and combined enzyme indicators for acute pancreatitis analyzed by receiver-operator characteristics. Clin Chem. 1989 Jun;35(6):967-71. PMID: 2471605.
  9. Vissers RJ, Abu-Laban RB, McHugh DF. Amylase and lipase in the emergency department evaluation of acute pancreatitis. J Emerg Med. 1999 Nov-Dec;17(6):1027-37. PMID: 10595892.
  10. Keim V, Teich N, Fiedler F, Hartig W, Thiele G, Mössner J. A comparison of lipase and amylase in the diagnosis of acute pancreatitis in patients with abdominal pain. Pancreas. 1998 Jan;16(1):45-9. PMID: 9436862.
  11. Volz KA, McGillicuddy DC, Horowitz GL, Wolfe RE, Joyce N, Sanchez LD. Eliminating amylase testing from the evaluation of pancreatitis in the emergency  department. West J Emerg Med. 2010 Sep;11(4):344-7. PMID: 21079706
Andrew Grock, MD

Andrew Grock, MD

Lead Editor/Co-Founder of ALiEM Approved Instructional Resources (AIR)
Assistant Professor of Emergency Medicine
UCLA Emergency Medicine Department
Andrew Grock, MD

Latest posts by Andrew Grock, MD (see all)

  • Guest

    Nice post–our group published a quality project in the Harvard Business Review recently demonstrating the cost of this practice

    http://blogs.hbr.org/2013/11/reducing-patient-charges-by-eliminating-amylase-from-lab-panels/

  • Jeff Wiswell

    Nice post–our group recently published a quality project in the Harvard Business Review exploring the cost of ordering amylase in the ED

    http://blogs.hbr.org/2013/11/reducing-patient-charges-by-eliminating-amylase-from-lab-panels/

  • Jason West

    This is a really nice topic for departments that have order sets, whether ordered by the triage nurses or the residents.

    So, this business review he is talking about took notes from reference 11 and did the same thing at their own institution. They implemented education on reducing amylase ordering, removed it from order sets, and removed it from a psych clearance panel. Both institutions expect to reduce patient charges by over 300k per year by doing this.

    And this is a completely different topic, but I was surprised to see that psych clearance included an amylase level. Seems like a waste.

  • Guest

    If I had a dollar every time GI asked me to order an amylase/lipase, as if it was a bundle order, or surgery asked me to get a CBC to look for a white count in a slam-dunk appy…well, I’d still be a resident, but at least I would be able to afford lunch in our hospital cafeteria.

  • Andrew Grock

    I appreciate your post Dr. Wiswell! It’s good to know that removing it from the order set works at multiple institutions. I’ve also been recently amused that my hospital re-structured the EMR for “efficiency” and guess what? Amylase and Lipase, which were previously ordered separately, are now ordered together with a single click! We’ve gone backwards!
    Also, in reference to Dr West’s response – for us, psych clearance does not include an amylase, but our trauma lab bundle does.

  • Chris Cole @DocOnSkis

    This is mildly confusing. Is it still standard practice in the US or UK to order an amylase for pancreatitis? Phased out in Australia > 10 years ago. ?

    • Andrew Grock

      Hi Dr. Cole. According to this guideline recommendation by the American College of Gastroenterology:
      1. The diagnosis of AP is most often established by the presence of two of the three following criteria: (i) abdominal pain consistent with the disease, (ii) serum amylase and / or lipase greater than three times the upper limit of normal, and / or (iii) characteristic fi ndings from abdominal imaging
      (strong recommendation, moderate quality of evidence).
      2. Contrast-enhanced computed tomographic (CECT) and / or magnetic resonance imaging (MRI) of the pancreas should be reserved for patients in
      whom the diagnosis is unclear or who fail to improve clinically within the fi rst 48 – 72 h after hospital admission (strong recommendation, low quality of
      evidence). ”

      So I guess may still be “standard practice”

      http://d2j7fjepcxuj0a.cloudfront.net/wp-content/uploads/2013/09/ACG_Guideline_AcutePancreatitis_September_2013.pdf

  • Hamhock

    Although I haven’t ordered amylase for non-traumatic pancreatitis in years, I have often paused when presented with blunt abdominal trauma.

    “Experts” have recommended it for blunt small bowel and pancreatic injury.

    Anyone have any comments?

    (shamefully, I haven’t had the chance to do a lit search yet)

    HH

    • Andrew Grock

      Hi Hamhock. I’m afraid my literature search did not include traumatic pancreatic injury. To start, I took a look at the EAST guidelines. They report that CT scan is not very sensitive in diagnosing pancreatic injuries and that delay to diagnosis is associated with increased mortality. Here the studies on CT scans were from the 1990s. It is unclear if our newer generation scanners are better?
      Without a robust imaging modality, perhaps labs are more important. The studies reviewed by EAST are all known pancreatic injury where they retrospectively analyzed amylase levels only on those patients. One study found a sensitivity of 83% and a positive predictive value of 4%. The amylase level does not seem to correlate to level of injury either.

      They conclude that “repeated amylase determinations for 48 hours might be helpful in diagnosing major duct injuries. The implications of delayed intervention in a pancreatic duct injury must be weighed against the morbidity of a non-therapeutic laparotomy as well”

      https://www.east.org/resources/treatment-guidelines/pancreatic-trauma-diagnosis-and-management-of

      • Michelle Lin

        Hi HH: Agree with Andy that there’s no great literature on this. EAST puts it as Level 3 (poorest) level of evidence in blunt abdominal trauma.

        “Amylase/Lipase levels are suggestive but not diagnostic of pancreatic injury”

        With such a poor test characteristic, this makes it a poor screening test. Don’t get it. I too think that with higher generation CT scanners, this will be non-issue in the coming years (at least in the ED). I haven’t seen an amylase/lipase ordered in traumas in the past 10 years at my institution.

  • Dean Burns

    This is all well & good. However, if the institution in which you work doesn’t have lipase as an option, you’re left with no choice. If amylase a worthwhile test in the absence of lipase, or not worth bothering with? I’d be delighted to save healthcare funds if this test is felt to be completely hopeless. You’re thoughts?

    • Andrew Grock

      Thanks for the interesting question. From my reading, it seems like amylase is an adequate test, but has limits.
      My main findings were that lipase is better (more sensitive and specific), and that amylase and lipase combined are no more helpful than a lipase alone. If your shop doesn’t have lipase, I personally would order amylase for pancreatitis.
      Keep in mind that amylase is more likely negative in pancreatitis with delayed (3 or more days) presentation, alcoholic pancreatitis, or an acute episode of chronic pancreatitis. Also, amylase can be elevated for many other reasons in patients without pancreatitis.