Urinary amylase / urinary creatinine ratio (uAm/uCr) - a less-invasive parameter for management of hyperamylasemia
© Terui et al.; licensee BioMed Central Ltd. 2013
Received: 5 March 2013
Accepted: 9 December 2013
Published: 13 December 2013
The serum level of amylase (sAm) is commonly used as a biochemical marker for diagnosis and management of pancreatic disorders. However, the use of the urine level of amylase (uAm) is limited in practice, because the diagnostic ability of uAm is inferior to that of sAm. In the present study, the possible concordance of uAm-rerated parameters with sAm was investigated, and evaluate the usefulness of uAm for management of hyperamylasemia.
From June 1995 to October 2009, 804 samples of both urine and blood were collected from 128 patients in order to measure the serum level of amylase (sAm) and the urine level of amylase (uAm) and creatinine (uCr). Concordance of parameters using uAm compared to sAm was assessed. Parameters used were uAm, amylase creatinine clearance ratio (ACCR), and the ratio of uAm to uCr (uAm/uCr).
uAm/uCr had the best correlation with sAm (r = 0.779, p < 0.001) compared to uAm (r = 0.620, p < 0.001) and to ACCR (r = 0.374, p < 0.001), when sAm was over the standard level. The area under the receiver operating characteristic curve of uAm/uCr (0.884) was significantly higher than that of uAm (0.766) and of ACCR (0.666) (p < 0.001 for each). The cutoff value of uAm/uCr was 569.8, with a sensitivity of 81.0% and a specificity of 83.1%.
The uAm/uCr ratio correlated with sAm, and may be an alternative to sAm for prediction of hyperamylasemia. Use of urine samples results in a decreased need for blood sampling, which is especially beneficial in pediatric patients.
KeywordsHyperamylasemia Pancreatitis Choledochal cyst Amylase Creatinine Urine
The serum level of amylase (sAm) is commonly used as a biochemical marker for diagnosis and management of pancreatic disorders [1–3]. Amylase is also one of only a few serum enzymes that are detectable in urine due to its small size and glomerular filtration . However, the use of the urine level of amylase (uAm) is limited in practice , because the diagnostic ability of uAm is inferior to that of sAm [6, 7]. In a few reports, uAm has been used as a marker after endoscopic retrograde cholangiopancreatography or pancreas transplantation [8, 9]. The amylase creatinine clearance ratio (ACCR) is an index that uses uAm. ACCR is known to increase during pancreatitis; however, it has little diagnostic value because of its low specificity and sensitivity [5, 7].
Since 1990s, we have pursued possibility of using urine samples for management of pancreatitis, in order to decrease occasions of blood sampling. Index using uAm, however, had little scientific basis. In the present study, the possible concordance of uAm with sAm was investigated. If a urine sample is equally or more useful and reliable to measure a biochemical parameter, then patients can be subject to fewer blood draws in the management of hyperamylasemia. This is especially relevant in pediatric patients. For parameters possibly correlated with sAm, the ratios ACCR and uAm/uCr (the corrected value of uAm divided by the urine level of creatinine (uCr)) were adopted in the present study. The concept of uAm/uCr is based on other methods that are used to evaluate excretion of various enzymes into the urine [10, 11].
Urine samples were collected with the spot collection technique. The timings of sample collection were various and also independent from those of blood sampling. In neonates and infants, urine was sampled by using disposable Pediatric Urine Collector (Atom®).
Both sAm and uAm were determined by an enzymatic method using blocked p-nitrophenyl-α-maltoheptaoside as a substrate. uCr was determined by an enzymatic method involving creatinase, sarcosineoxidase and peroxidase. Standard levels of sAm and uAm in our institution are set at 35–110 IU/L and <450 IU/L, respectively.
ACCR was calculated in 604 samples (75% of all samples), because not all blood samples were tested for the serum level of creatinine (sCr). ACCR was determined independently of urine volume and time of sampling by using the formula (uAm / sAm) / (uCr / sCr) ; uAm/uCr was calculated simply by dividing uAm by uCr.
Statistical analyses were performed by using software (statistical computing; R Foundation for Statistical Computing, Vienna, Austria). Differences between groups were tested by means of the Student’s t-test or the Mann–Whitney U test. Associations were assessed by Pearson’s correlation coefficient test or Spearman’s rank correlation and expressed as the corresponding correlation coefficient, r. Simple regression analysis was used to evaluate correlations between sAm and parameters of urinary amylase: uAm, ACCR, and uAm/uCr. The abilities of the three parameters of urinary amylase to predict abnormalities of sAm were assessed with receiver operating characteristic (ROC) analysis. The area under the curve of each ROC was calculated, and the difference between each pair was assessed using the critical ratio. The cutoff value was derived by the point with shortest distance to the point with a true positive fraction of 100% and a false positive fraction of 0% on the ROC curve. All results are expressed as mean ± standard deviation. P values of <0.05 were considered statistically significant.
Informed consent for participation in the study was not obtained because all data were collected as part of routine diagnosis and treatment, and analyzed retrospectively. The study protocol was approved by the institutional ethics committee of Chiba University.
The level of amylase in age-groups
Serum amylase (% of normal level)
33.5 ± 33.6
56.6 ± 47.0
1 – 5
178.1 ± 248.9
600.6 ± 938.7
6 – 12
183.2 ± 269.5
755.8 ± 2253.9
105.6 ± 91.2
438.8 ± 616.7
The level of amylase in different clinical conditions
Serum amylase (% of normal level)
125.0 ± 185.5
485.8 ± 1547.4
196.1 ± 216.9
598.0 ± 1084.5
370.9 ± 429.7
1435.8 ± 2195.4
Pancreatic duct obstruction
178.7 ± 177.8
686.5 ± 1824.4
138.2 ± 62.4
725.6 ± 845.5
115.7 ± 110.7
576.1 ± 766.0
Association between sAm and urinary amylase parameters
With regards to correlation analysis, uAm/uCr had the best correlation with sAm (r = 0.667, p < 0.001) compared to uAm (r = 0.508, p < 0.001) and ACCR (r = 0.167, p < 0.001). In 273 samples (34.0% of all samples), sAm was over the standard level (>110 IU/L). When sAm was over the standard level, uAm/uCr also had the best correlation with sAm (r = 0.779, p < 0.001) compared to uAm (r = 0.620, p < 0.001) and ACCR (r = 0.374, p < 0.001). When sAm was normal, all the three parameters did not correlate well with sAm (uAm, r = 0.245, p < 0.001; ACCR, r = −0.218, p < 0.001; uAm/uCr, r = 0.289, p < 0.001).
In 11 samples (1.4% of all the samples), sCr was over the standard level (>1.0 mg/ml). When sCr was over the standard level, uAm/uCr was 1402.0 ± 2854.1 and correlated with sAm (r = 0.699, p = 0.025). When sCr was normal, uAm/uCr was 181.0 ± 50.3 and also correlated with sAm (r = 0.665, p < 10-12).
Correlation between sAmy and uA/uC by different age-groups and diseases
Pancreatic duct obstruction
Prediction of sAm abnormalities with urinary amylase parameters
In the diagnosis and management of pancreatic disorders, measurement of sAm has played a central role since Elman et al. demonstrated the value of sAm in diagnosis of pancreatitis [5, 14]. However, it is also true that sAm has problems with respect to diagnostic ability. Estimates of the sensitivity of amylase range from 60% to 98% [15, 16], although it is difficult to estimate because elevated sAm itself is frequently used as a diagnostic criterion [5, 17, 18]. Furthermore, almost 40% of patients under 3 years of age with a diagnosis of acute pancreatitis were reported to have normal amylase but elevated lipase levels . This erratic low sensitivity of sAm has been explained as resulting from the short half-life of amylase in the serum [1–3, 15]. The specificity of sAm is also not high enough to use as a screening parameter for pancreatic disorders. While hyperamylasemia is often seen in various pancreatic diseases, extrapancreatic factors also result in hyperamylasemia (salivary disease, gastrointestinal disease, gynecologic disease, extra-pancreatic tumors, and others) [1–3]. Hyperamylasemia was found in 9.7% of patients with appendicitis , 9.3% of patients with acute cholecystitis , and 16.0% of patients with a perforated peptic ulcer . To compensate for the incomplete diagnostic ability of sAm, other pancreatic enzymes have been proposed. In particular, the serum level of lipase is a comparable marker of pancreatic disease, but rapid and cost-effective measurement systems are not broadly available [1–3]. Urinary trypsinogen-2 strip test has also drawn attention recently for early detection of pancreatitis, but is not quantitative and less sensitive [23, 24]. Therefore, sAm has some issues concerning its use, but has been used broadly as practical marker of pancreatic disorders.
Amylase has a relatively small molecular weight of about 50,000 Daltons; therefore, it passes through the glomerulus and is one of only a few serum enzymes detectable in urine . Therefore, uAm can potentially be used as a marker of hyperamylasemia. The purpose of the present study was to assess the usefulness of urine samples in this regard. The leading reason for conducting this research was that if urine sample is more useful and reliable, then patients can undergo a urine test as an alternative to a blood test. Collection of urine is less invasive and generally leads to less suffering than blood sampling, which can be especially stressful in pediatric patients.
The parameters investigated from urine samples in the present study were uAm, ACCR and uAm/uCr. First of all, uAm can be easily obtained from urine samples. By using a 1-hour value, a superior sensitivity of uAm was reported in 1963 . However, uAm has recently been reported to have no further advantages in the diagnosis of pancreatitis compared to sAm [7, 16]. Even if amylase activity per hour is used as normalization, the diagnostic performance of uAm is worse than that of sAm . Second, ACCR was developed to differentiate pancreatitis from hyperamylasemia caused by other conditions. The concept was based on the fact that amylase clearance increases significantly in acute pancreatitis, while creatinine clearance does not [26–28]. However, ACCR also has problems with specificity and sensitivity, and is considered to be less important than sAm [7, 15, 28–30]. Third, uAm/uCr is the corrected value of uAm divided by uCr. The concept of uAm/uCr is based on the evidence that the variabilities of enzyme excretion into the urine are the smallest when the results are expressed as the ratio of enzyme activity / uCr [31, 32]. Amylase passes through the renal filtration circuit primarily without passive elimination or reabsorption [33, 34]. Therefore, uAm satisfies the criteria for the use of the correction by uCr . Another merit of uAm/uCr is that it can be obtained from a urine sample, and a blood sample is not required.
In the present study, among the three parameters of urinary amylase (uAm, ACCR, and uAm/uCr), uAm/uCr had the best correlation with sAm especially in the patients with hyperamylasemia, and the best ability to predict an abnormal elevation of sAm. This result suggests the potential use of uAm/uCr as an alternative for sAm. Furthermore, details of the correlation between sAm and uAm/uCr were evaluated in order to elucidate the characteristics of uAm/uCr. The correlation between sAm and uAm/uCr was low in babies, and was significant in infants and schoolchildren. This indicates that the level of amylase itself cannot be used in babies, as discussed later. The low correlation in adolescents can be explained by the high rate of normal sAm in adolescent samples in the present study. In the range of normal sAm, correlation between sAm and uAm/uCr tended not to be significant. A number of normal samples were also collected during follow-up for a choledochal cyst, and also resulted in a low correlation. Low correlation in pancreatitis probably results from a low mean sAm (115.7 IU/L) despite the low rate of normal sAm in this population. Except for these conditions, correlations were relatively high in a number of various other conditions, including tumor, trauma, pancreatic duct obstruction, and non-pancreatic disease. Also, in renal failure with elevated sCr, the correlation between sAm and uAm/uCr was significant. Therefore, uAm/uCr could be applicable for various conditions of hyperamylasemia after the first year of life, and does not appear to be influenced by elevated sCr. In the management of hyperamylasemia, uAm/uCr can potentially be used not for diagnosis but as a marker for following up on levels of amylase.
In the pediatric population, the level of amylase activity increases with age [1, 15]. Amylolytic enzyme activity was found to be feeble in children under 1 year of age, and exceeded the average range for adults beyond the first year . The adult level of pancreatic amylase activity in duodenal juice is reached at the age of 18 months . In serum, the amylase level reaches the normal adult level by the age of 8 months ; however, the major component of amylase originates from the salivary glands. Pancreatic amylase activity in serum reaches the normal adult level by 7 to 10 years of age [38, 39]. In contrast, the pancreatic proportion of urinary amylase approaches the adult level by 16 months of age . These age-dependent specificities should be considered when interpreting data surrounding sAm and uAm/uCr in a pediatric population.
A limitation of this study is the characteristics of study population; because the subjects were patients from our department of pediatric surgery, the study population does not match common etiologies of pediatric hyperamylasemia in the greater population . In the present study, the number of the patients with choledochal cyst was quite high, and only a few cases of pancreatitis from systemic causes were investigated. Thus, further investigations are needed in different study populations. The method of urine sampling is also a limitation. The recommended urine sampling technique is the second morning urine after voiding the night urine, in order to avoid the influences of diurnal rhythms of enzyme excretion and collection errors . In the present study, however, the time of day of urine sampling was not constant, and was also not constant in relation to the time of blood sampling. These issues of sample collection caused unavoidable measurement error, but they were considered acceptable by limiting the time of urine and blood sampling to the same day and by including a large number of samples. Furthermore, uAm tends to peak later and maintain high level longer than sAm [15, 41]. These physiological differences also should be considered in interpreting the data.
uAm/uCr was found to significantly correlate with sAm, and therefore may be of use as an alternative to sAm during management of patients with hyperamylasemia. Use of urine samples results in a decreased number of blood samples, which is especially important in pediatric patients. This retrospective preliminary report will need to be supported by further studies in order to elucidate the practical usefulness of uAm/uCr in management of hyperamylasemia.
The serum level of amylase
The urine level of amylase
The urine level of creatinine
The serum level of creatinine
The amylase creatinine clearance ratio
The corrected value of uAm divided by uCre
Receiver operating characteristic.
We would like to thank Dr. Yasuyuki Higashimoto (Department of Pediatric Surgery, Chiba Children’s Hospital) for his helpful suggestions and encouragement.
- Vissers RJ, Abu-Laban RB, McHugh DF: Amylase and lipase in the emergency department evaluation of acute pancreatitis. J Emerg Med. 1999, 17: 1027-1037. 10.1016/S0736-4679(99)00136-5.View ArticlePubMedGoogle Scholar
- Matull WR, Pereira SP, O’Donohue JW: Biochemical markers of acute pancreatitis. J Clin Pathol. 2006, 59: 340-344. 10.1136/jcp.2002.002923.View ArticlePubMedPubMed CentralGoogle Scholar
- Harper SJ, Cheslyn-Curtis S: Acute pancreatitis. Ann Clin Biochem. 2011, 48: 23-37. 10.1258/acb.2010.010196.View ArticlePubMedGoogle Scholar
- McClatchey KD: Diagnostic enzymology. Clinical laboratory medicine. Edited by: McClatchey KD. 1994, Baltimore: Williams and Wilkins, 279-286.Google Scholar
- Ranson JH: Diagnostic standards for acute pancreatitis. World J Surg. 1997, 21: 136-142. 10.1007/s002689900205.View ArticlePubMedGoogle Scholar
- 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, 35: 967-971.PubMedGoogle Scholar
- Gwozdz GP, Steinberg WM, Werner M, Henry JP, Pauley C: Comparative evaluation of the diagnosis of acute pancreatitis based on serum and urine enzyme assays. Clin Chim Acta. 1990, 187: 243-254. 10.1016/0009-8981(90)90109-6.View ArticlePubMedGoogle Scholar
- Hegewald MJ, Isenberg G, Sterling RK, Cooper GS, Chak A, Sivak MV: Evaluation of a rapid urine amylase test using post-ERCP hyperamylasemia as a model. Am J Gastroenterol. 2001, 96: 2640-2645. 10.1111/j.1572-0241.2001.04118.x.View ArticlePubMedGoogle Scholar
- Prieto M, Collins W, Scott MH, Sells RA: Method for home monitoring of urinary amylase after pancreas transplantation. Diabetes. 1989, 38: 68-70. 10.2337/diab.38.1.S68.View ArticlePubMedGoogle Scholar
- Jung K: Enzyme activities in urine: how should we express their excretion? A critical literature review. Eur J Clin Chem Clin Biochem. 1991, 29: 725-729.PubMedGoogle Scholar
- Boeniger MF, Lowry LK, Rosenberg J: Interpretation of urine results used to assess chemical exposure with emphasis on creatinine adjustments: a review. Am Ind Hyg Assoc J. 1993, 54: 615-627. 10.1080/15298669391355134.View ArticlePubMedGoogle Scholar
- Dreiling DA, Leichtling JJ, Janowitz HD: The amylase-creatinine clearance ratio. Diagnostic parameter of physiologic phenomenon?. Am J Gastroenterol. 1974, 61: 290-296.PubMedGoogle Scholar
- Terui K, Yoshida H, Kouchi K, Hishiki T, Saito T, Mitsunaga T, Takenouchi A, Tsuyuguchi T, Yamaguchi T, Ohnuma N: Endoscopic sphincterotomy is a useful preoperative management for refractory pancreatitis associated with pancreaticobiliary maljunction. J Pediatr Surg. 2008, 43: 495-499. 10.1016/j.jpedsurg.2007.10.071.View ArticlePubMedGoogle Scholar
- Elman R, Arneson N, Graham EA: Value of blood amylase estimation in the diagnosis of pancreatic disease; clinical study. Arch Surg. 1929, 19: 943-10.1001/archsurg.1929.01150060005001.View ArticleGoogle Scholar
- Garrison R: Amylase. Emerg Med Clin North Am. 1986, 4: 315-327.PubMedGoogle Scholar
- Agarwal N, Pitchumoni CS, Sivaprasad AV: Evaluating tests for acute pancreatitis. Am J Gastroenterol. 1990, 85: 356-366.PubMedGoogle Scholar
- Ventrucci M, Gullo L, Daniele C, Bartolucci C, Priori P, Platé L, Bonora G, Labò G: Comparative study of serum pancreatic isoamylase, lipase, and trypsin-like immunoreactivity in pancreatic disease. Digestion. 1983, 28: 114-121. 10.1159/000198973.View ArticlePubMedGoogle Scholar
- Clavieny PA, Burgan S, Moosa AR: Serum enzymes and other laboratory tests in acute pancreatitis. Br J Surg. 1989, 76: 1234-1243. 10.1002/bjs.1800761205.View ArticleGoogle Scholar
- Kandula L, Lowe ME: Etiology and outcome of acute pancreatitis in infants and toddlers. J Pediatr. 2008, 152: 106-110. 10.1016/j.jpeds.2007.05.050.View ArticlePubMedGoogle Scholar
- Swensson EE, Maull KI: Clinical significance of elevated serum and urine amylase levels in patients with appendicitis. Am J Surg. 1981, 142: 667-670. 10.1016/0002-9610(81)90308-1.View ArticlePubMedGoogle Scholar
- Hall ER, Howard JM, Jordan GL, Witt R: A study of serum amylase concentration in patients with acute cholecystitis. Ann Surg. 1956, 143: 517-519. 10.1097/00000658-195604000-00013.View ArticlePubMedPubMed CentralGoogle Scholar
- Rogers FA: Elevated serum amylase: a review and an analysis of findings in 1,000 cases of perforated peptic ulcer. Ann Surg. 1961, 153: 228-240.View ArticlePubMedPubMed CentralGoogle Scholar
- Räty S, Sand J, Nordback I: Detection of postoperative pancreatitis after pancreatic surgery by urine trypsinogen strip test. Br J Surg. 2007, 94: 64-69. 10.1002/bjs.5572.View ArticlePubMedGoogle Scholar
- Andersen AM, Novovic S, Ersbøll AK, Jorgensen LN, Hansen MB: Urinary trypsinogen-2 dipstick in acute pancreatitis. Pancreas. 2010, 39: 26-30.View ArticlePubMedGoogle Scholar
- Gambill EE, Mason HL: One-hour value for urinary amylase in 96 patients with pancreatitis. JAMA. 1963, 186: 24-28. 10.1001/jama.1963.03710010058009.View ArticlePubMedGoogle Scholar
- Warshaw A, Fuller A: Specificity of increased renal clearance of amylase in diagnosis of acute pancreatitis. N Engl J Med. 1975, 292: 325-330. 10.1056/NEJM197502132920701.View ArticlePubMedGoogle Scholar
- Levine RI, Gleuser FL, Berk JE: Enhancement of the amylase-creatinine clearance ratio it disorders other than acute pancreatitis. N Engl J Med. 1975, 292: 329-332. 10.1056/NEJM197502132920702.View ArticlePubMedGoogle Scholar
- Levitt MD, Johnson SG: Is the Cam/Ccr ratio of value for the diagnosis of pancreatitis?. Gastroenterology. 1978, 75: 118-119.PubMedGoogle Scholar
- McMahon MJ, Playforth MJ, Rashid SA, Cooper EH: The amylase-to-creatinine clearance ratio—a non-specific response to acute illness?. Br J Surg. 1982, 69: 29-32. 10.1002/bjs.1800690110.View ArticlePubMedGoogle Scholar
- Eckfeldt JH, Levitt MD: Diagnostic enzymes for pancreatic disease. Clin Lab Med. 1989, 9: 731-743.PubMedGoogle Scholar
- Werner M, Heilbron DC, Maruhn D, Atoba M: Patterns of urinary enzyme excretion in healthy subjects. Clin Chim Acta. 1970, 29: 437-449. 10.1016/0009-8981(70)90014-8.View ArticlePubMedGoogle Scholar
- Maruhn D, Strozyk K, Gielow L, Bock KD: Diurnal variations of urinary enzyme excretion. Clin Chim Acta. 1977, 75: 427-433. 10.1016/0009-8981(77)90362-X.View ArticlePubMedGoogle Scholar
- Blainey JD, Northam BE: Amylase excretion by the human kidney. Clin Sci. 1967, 32: 377-383.PubMedGoogle Scholar
- Duane WC, Frerichs R, Levitt MD: Distribution, turnover, and mechanism of renal excretion of amylase in the baboon. J Clin Invest. 1971, 50: 156-165. 10.1172/JCI106469.View ArticlePubMedPubMed CentralGoogle Scholar
- Klumpp TG, Neale AV: Gastric and duodenal contents of normal infants and children; duodenal enzyme activities and gastric reactions. Am J Dis Child. 1930, 40: 1215-1229. 10.1001/archpedi.1930.01940060055005.View ArticleGoogle Scholar
- Delachaume-Salem E, Sarles H: Normal human pancreatic secretion in relation to age. Biol Gastroenterol (Paris). 1970, 2: 135-146.Google Scholar
- Otsuki M, Yuu H, Saeki S, Baba S: The characteristics of amylase activity and the isoamylase pattern in serum and urine of infants and children. Eur J Pediatr. 1977, 125: 175-180. 10.1007/BF00480594.View ArticlePubMedGoogle Scholar
- Sjölund K, Häggmark A, Ihse I, Skude G, Kärnström U, Wikander M: Selective deficiency of pancreatic amylase. Gut. 1991, 32: 546-548. 10.1136/gut.32.5.546.View ArticlePubMedPubMed CentralGoogle Scholar
- Skude G, Kollberg H: Serum isoamylases in cystic fibrosis. Acta Paediatr Scand. 1976, 65: 145-149. 10.1111/j.1651-2227.1976.tb16527.x.View ArticlePubMedGoogle Scholar
- Tye JG, Karn RC, Merritt AD: Differential expression of salivary (Amy1) and pancreatic (Amy2) human amylase loci in prenatal and postnatal development. J Med Genet. 1976, 13: 96-102. 10.1136/jmg.13.2.96.View ArticlePubMedPubMed CentralGoogle Scholar
- Calkins WG: A study of urinary amylase excretion in patients with acute pancreatitis. Am J Gastroenterol. 1968, 49: 415-424.Google Scholar
- The pre-publication history for this paper can be accessed here:http://www.biomedcentral.com/1471-2431/13/205/prepub
This article is published under license to BioMed Central Ltd. This is an open access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.