Article Text
PDF
Abstract
Objective Most studies have assessed the impact of elevated serum amylase levels in clinical practice, but only a few have investigated the significance of low serum amylase. We therefore, aimed to review the literature to understand the conditions associated with low serum amylase and its clinical relevance.
Method This systematic review was performed in accordance with the criteria established in the Preferred Reporting Items for Systematic Reviews and Meta-Analyses guidelines. The search was conducted on Medline and Embase databases until November 2022. After identifying relevant titles, abstracts were read and data of eligible articles retrieved. The conditions associated with low serum amylase were evaluated. The quality of the studies was assessed using the Newcastle-Ottawa Score.
Results Our search strategy identified 19 studies including a total of 15 097 patients for systematic review. All the studies were observational including two studies which used secretin-induced test. The main conditions associated with low serum amylase were diabetes mellitus (n=9), metabolic syndrome (n=3), chronic pancreatitis (CP) (n=3), non-alcoholic fatty liver disease (n=2) and obesity (n=1). Low serum amylase showed a high specificity (94%) with low sensitivity (38.7%–59%) in diagnosing chronic pancreatitis.
Conclusion This systematic review revealed a unique insight into the relevance of low serum amylase in clinical practice. Low serum amylase can be a useful adjunct test in the assessment of patients with CP, pancreatic exocrine insufficiency, diabetes mellitus and metabolic syndrome.
- chronic pancreatitis
- diabetes mellitus
- exocrine pancreatic function
Data availability statement
There are no data in this work.
Statistics from Altmetric.com
WHAT IS ALREADY KNOWN ON THIS TOPIC
The clinical relevance of low serum amylase in clinical practice is still poorly understood.
Most studies have assessed the impact of elevated serum amylase levels in clinical practice.
Few studies have investigated the significance of low serum amylase.
WHAT THIS STUDY ADDS
This systematic review represents a comprehensive summary of conditions associated with low serum amylase.
The most common conditions associated with low serum amylase were diabetes mellitus, metabolic syndrome, obesity and chronic pancreatitis.
HOW THIS STUDY MIGHT AFFECT RESEARCH, PRACTICE OR POLICY
Amylase has been implicated in an existing endocrine–exocrine crosstalk.
Low amylase can be part of a metabolic process making it a useful metabolic marker in cases like obesity, metabolic syndrome and diabetes mellitus. It can be used as an exocrine marker and trigger investigation for also be used as an exocrine marker and trigger investigation for pancreatic exocrine insufficiency.
Introduction
The pancreas is a unique organ with dual functions: endocrine and exocrine functions. It is estimated that 84% of the pancreas is exocrine, 10% extracellular, 4% ductal cells and blood vessels and only 2% is endocrine.1
There are two main isotypes of amylase: pancreatic amylase and salivary amylase. It is estimated that 45% of total serum amylase is produced by the pancreas, whereas the salivary glands contribute to the rest.2 3 Salivary amylase initiates carbohydrate digestion in the oral cavity, a process which is continued in the small intestine by the pancreatic amylase. The catabolic activity of salivary amylase ceases when it enters the acidic environment of the stomach.4 Most pancreatic enzyme stimulation occurs via the intestinal phase when the chyme enters the duodenum. During the intestinal phase, duodenal I-cells produce cholecystokinin (CCK) which in turn stimulates pancreatic acinar cells to produce enzymes including amylase.5 6 Pancreatic amylase is a glycoside hydrolase enzyme that functions to break down complex carbohydrate into simple sugar. There is increasing evidence that insulin can also affect amylase secretion through islet–acinar cell axis. Insulin binds its receptor on acinar cells to stimulate amylase secretion.7
Serum amylase test is easy to perform, inexpensive and widely available. The vast majority of studies have assessed the impact of elevated serum amylase levels in clinical practice,8 however, only a few studies have investigated the significance of low serum amylase. Elevated serum amylase is used to diagnose acute pancreatitis with a sensitivity of 78%.9 Other clinical causes of serum hyperamylasaemia include bowel perforation, chronic alcoholism, anorexia nervosa, bulimia nervosa, chronic kidney disease, mumps, carcinomas, opiate use and macroamylasaemia.3 10
The significance of low serum amylase in clinical practice is still unclear. Therefore, we conducted a systematic review of the literature to understand the associated conditions.
Methodology
This systematic review was performed in accordance with the criteria established in the Preferred Reporting Items for Systematic Reviews and Meta-Analyses guidelines. This study was registered with the University of York International Prospective Register of Systematic Reviews (PROSPERO; Registration number CRD42022362415).
Search strategy
Studies were identified by performing a literature search on Medline and Embase databases, the last search was conducted on 1 November 2022. Search terms were ((((low serum amylase (Title/Abstract)) OR (low serum pancreatic amylase (Title/Abstract))) OR (low serum amylase(Title/Abstract))) OR (hypoamylasemia(Title/Abstract))) OR (amylase deficiency(Title/Abstract)).
Further studies were added by manual search to retrieve relevant data. The papers were restricted to English language. Two authors (MJ) and (SAG) screened all titles, abstracts and full texts for relevance to the study. Disagreements were resolved by consensus or by consulting with a third investigator (KN).
Inclusion and exclusion criteria
Inclusion criteria were: (1) studies that included adult patients with low serum amylase as a main aim of their study and (2) papers in English language. Exclusion criteria: articles written in a language other than English, and studies conducted in patients younger than 18 years of age .
Data extraction
Data extracted from each eligible study included: (1) study authors, (2) year of publication, (3) study type, (4) number of patients, (5) age, (6) sex, (7) body mass index (BMI), (8) tests performed (serum amylase or specific pancreatic amylase) and (9) conditions associated with low serum amylase. Missing data entries were marked with N/A (not available). Due to the nature of this study, ethical approval was not required or obtained.
Main outcome
The primary aim of this study was to conduct a literature search and systemic review assessing the conditions associated with low serum amylase.
Assessment of methodological quality
The quality of the studies was assessed using the Newcastle–Ottawa Score for the observational studies.
Results
Search result
The initial search identified 204 papers. Initial screening excluded 90 papers which were duplicates and conference abstracts. After reading the titles, abstracts and text, 13 papers were selected in the review process. A further six papers were selected from the references of the selected studies. Our search strategy identified 19 eligible studies including a total of 15 097 patients for systematic review. Our study selection process is illustrated in figure 1.
Preferred Reporting Items for Systematic Reviews and Meta-Analyses diagram showing the selection of included studies.
Characteristics of eligible studies
The characteristics of the selected studies, tests used, that is, serum total amylase or specific pancreatic amylase, and the conditions associated with low serum amylase are summarised in table 1. All the studies were observational except for two studies which used secretin-induced test.11 12
Characteristics of eligible studies with Newcastle-Ottawa Score provided
Definition of low serum amylase
Low serum amylase was defined according tlocal laboratory reference range. Selected studies used either hypoamylasaemia (amylase level below the normal reference range) or lower amylase level defined as the group with lower amylase level usually in comparison to another group or controls and not necessary to be in the hypoamylasaemia range. Salivary amylase will be mentioned if it was used.
Conditions associated with low serum amylase
Chronic pancreatitis
Three studies included in this systematic review assessed the association between low serum amylase level and chronic pancreatitis (CP).13–15 A recent prospective study by Olesen et al assessed the utility of pancreatic amylase for the diagnosis of CP.13 An amylase <17.3 U/L had a specificity of 94% and sensitivity of 59% for the diagnosis of CP. Amylase was strongly associated with duration of CP, diabetes mellitus (DM) and pancreatic exocrine insufficiency (PEI). Amylase level was lower in patients with PEI compared with exocrine-sufficient CP. Also, amylase level was lower in patients with DM compared with patients without DM . Similarly, Oh et al Similarly Oh et al found lower levels of pancreatic amylase in CP compared with healthy controls. Levels below the normal range were highly specific for CP. Calcified CP had lower levels compared with non-calcified CP.14 Authors suggested using low value of amylase with a cut-off of 32 U/L to rule out CP. Kwon et al showed similar results with lower amylase level in calcified CP.15 Amylase level of <27.5 U/L showed high specificity 94.1% but lower sensitivity of 38.7%.
Diabetes mellitus
Most papers in our review conducted studies on patients with DM including both type 1 DM (T1DM) and type 2 DM (T2DM). Dozio et al 16 investigated the exocrine function in patients with T1DM and compared with healthy controls. The study confirmed low faecal elastase (FEL-1), low serum pancreatic amylase and lipase in T1DM. The reductions in FEL-1, serum pancreatic amylase and lipase correlated with the reduction of fasting and urinary C-peptide. Madole et al compared serum amylase in patients with T1DM and T2DM to healthy individuals. Amylase levels were significantly lower in T1DM and T2DM with the lowest levels observed in T1DM.1 In their study, amylase was used as a surrogate marker for pancreatic exocrine function. Muneyuki et al study showed that low serum amylase is associated with impaired insulin action in asymptomatic patients who were not treated for DM.17 Those with low serum amylase had higher BMI and plasma glucose after a 75 g oral glucose tolerance test compared with those with normal to high serum amylase. In their analysis, low serum amylase was correlated with BMI in univariate analysis but in multivariate analysis, low serum amylase was correlated with low insulin secretion as well as high insulin resistance. The alcohol intake was not different between the groups (low vs normal-to-high serum amylase). Their study was relatively small with predominantly male patients. A large epidemiological study (n=2425) by Nakajima et al showed low serum amylase was associated with most cardiometabolic risk factors including obesity, diabetes and metabolic syndrome (MetS). The association of amylase with metabolic abnormalities was independent of smoking. A subset of patients who had previous check-up (5 years ago) had undergone retrospective analysis and it was found that those who had low serum amylase at that stage developed severe metabolic abnormalities during follow-up.18 Swislocki et al investigated the effect of secretin stimulation in patients with T1DM, T2DM and controls.19 There was no rise in amylase production in T1DM compared with twofold increases in T2DM and controls. Their study was small in size and all patients were male. Dandona et al used pancreatic trypsin and amylase as markers for pancreatic exocrine function.20 Trypsin and amylase were markedly reduced in patients with insulin-dependent DM and only slightly impaired in those on sulphonylurea. Frier et al measured fasting C-peptide in 18 juvenile-onset DM and post IV glucagon injection.12 The exocrine function was assessed by measuring pancreatic bicarbonate, amylase and trypsin response to IV secretin and CCK. Amylase levels were positively correlated with C-peptide in preserved islet B cells.
Obesity and metabolic syndrome
Kondo et al recruited healthy male students and grouped them according to their ideal body weight using Broca’s formula (Height (cm) − 100) × 0.9 kg.11 Their study showed lower amylase level in the obese group. Lower amylase level was also observed post secretin test in the obese group. Zhan et al found lower total and pancreatic amylase in patients with MetS compared with healthy controls.21 Furthermore, amylase level was negatively correlated with BMI. In the receiver operating characteristic curve (ROC curve) analysis, the best cut-off-value for pancreatic amylase was defined as 58 U/L for which MetS prevalence significantly increased. Yao et al found that NAFLD with MetS had lower amylase levels compared with NAFLD without MetS.22 Similar results were found when NAFLD with DM and NAFLD without DM were compared.
Other conditions
Nakajima et al found the severity of NAFLD increases with decrement in serum amylase in asymptomatic patients.23 Approximately 42% of patients had NAFLD (of all grades) within the lowest quartile of serum amylase. Curd et al found that 5.5% of hospital serum amylase levels were below the reference range.10 Conditions associated with low serum amylase were DM, cystic fibrosis, hypertriglyceridaemia and the use of antibiotics gentamicin.
Discussion
This systematic review represents a comprehensive summary of conditions associated with low serum amylase. The most common conditions associated with low serum amylase were DM, MetS, obesity and CP.
In CP, low serum amylase is thought to be due to parenchymal destruction leading to reduced acinar secretion.24 25 CP is a fibroinflammatory process and according to this process the acinar cells are gradually replaced with fibrotic tissue.26 The process can be accelerated further depending on various genetic and environmental factors such as smoking and alcohol. Therefore, the loss of exocrine function can develop at any point during CP depending on the aforementioned factors.
Obesity can cause reduced insulin action and increased resistance which is often associated with excess adipose tissue. Reduced amylase is observed in obesity and a potential cause is the existing physiological mechanism that works to reduce carbohydrate absorption from gut by reducing amylase secretion in obesity.27 Therefore, pancreas may play a role in energy balance and glucose homeostasis. Other potential reasons are the increased free fatty acid in obesity which work to impair insulin receptor activity.4 Low serum amylase has been associated with acute pancreatitis secondary to hypertriglyceridaemia supporting the hypothesis of it being a metabolic marker.28 MetS is a disorder characterised by certain features including central obesity, dyslipidaemia, high blood pressure and high glucose levels.21 Kondo et al showed that low serum amylase can be reversed with weight loss in overweight patients. Four patients managed to lose an average of 10 kg during a 12-month follow-up period and serum amylase increased significantly. Serum amylase and trypsin were lower in obese subjects than in lean subjects, and were significantly and inversely correlated with body weight.11
The pathophysiology of low serum amylase in patients with diabetes has been gathering interests. Although it is not fully understood, there is strong evidence of an existing endocrine–exocrine crosstalk mediated via paracrine and incretin effects.1 The mechanism of exocrine impairment in DM is linked to the lack of insulinotropic effect of insulin on acinar cells,12 20 figure 2. Also, diabetic neuropathy may impair enteropancreatic reflexes and can lead to exocrine dysfunction.1 The correlation between low serum amylase level and C-peptide in T1DM may represent the loss of insulinotropic effect on acinar cells. However, patients with obesity, MetS and T2DM usually have high insulin levels and are also found to have low serum amylase. Therefore, it is likely the pathogenesis of exocrine impairment is different between T1DM and T2DM with some overlap.18 Other factors like haemoglobin A1C, dose of insulin or age of DM onset showed no impact on pancreatic exocrine secretion.20 Glucagon has been suggested to play a role in the endocrine–exocrine crosstalk. Glucagon has an inhibitory effect on the exocrine secretions. Therefore, the deficiency of insulin and excess of glucagon in patients with DM may lead to decrease amylase secretion.29
Pathophysiology of impaired amylase secretion in DM. (A) normal pancreatic exocrine system including ducal and acinar cells and (B) the endocrine–exocrine crosstalk in patients where insulin and glucagon have receptors on acinar cells, DM can cause loss of insulinotropic effect and increased inhibitory effect of glucagon resulting in decreased amylase secretion. Other explanations of reduced exocrine function in DM including (C) leucocyte infiltration in exocrine pancreas and (D) atrophy of acinar cells. DM, diabetes mellitus.
T1DM is a chronic autoimmune disease in which β-cells within pancreatic islets are selectively destroyed resulting in low insulin production.30 Several studies including radiological and postmortem showed evidence of acinar atrophy in patients with long-standing DM.19 The findings in this systematic review showed a potential crosstalk between the endocrine and exocrine systems.16 This crosstalk has been demonstrated by either low levels of amylase or low FEL-1. Pancreatic and acinar atrophy and altered amylase expression are some of the findings in patients with DM.16 In Dozio et al study, the correlation between exocrine function and residual C-peptide reinforced the theory of a parallel function of endocrine and exocrine pancreas.16 Also, the lack of insulinotropic effect on the acinar cell is thought to cause exocrine impairment.31 32
Insulin resistance is defined as loss or decreased sensitivity of target organ to the effect of insulin.1 Insulin resistance in T2DM may play a role in decreased amylase secretion. Zhuang et al investigated the relationship between serum amylase level and islet β-cell function in T2DM. Their study included those with lifestyle treatment only and normal amylase as inclusion criteria and found that serum amylase in the normal range were positively associated with integrated islet β-cell function.2 Another study found low serum amylase correlated with hyperglycaemic levels in patients with T2DM.33
Genetics is likely to play a role in the production and activation of amylase. Human amylases are encoded by salivary amylase genes (AMY1A, AMY1B and AMY1C), and pancreatic genes (AMY2A and AMY2B).21 Copy number variation (CNV) is common in human amylase genes. CNV refers to a phenomenon where a copy of a specific segment of DNA varies among different individuals’ genome.34 In humans, the variation of CNV is wider in AMY1 compared with AMY2.35 36 Several studies showed that CNV of AMY1 was associated with obesity and MetS.37–39 The studies on the association of CNV and metabolic risks are conflicting. Mejía-Benítez et al evaluated the number of AMY1 copies in 597 Mexican children and found high number of AMY1 copies was associated with reduced risk of obesity and those >10 AMY1 copies were normal weight.40 Viljakainen et al showed low AMY1 in females to be associated with obesity; however, the result in males were completely different in which obese male patients had highest number of AMY1 copies.41 In Zhan et al study, low serum amylase but not AMY1, AMY2A and AMY2B copy numbers were significantly lower in patients with MetS .21
There are several limitations in this study. We were unable to maintain a homogeneity with regards to the amylase test used in the studies; some studies used serum total amylase and others used serum pancreatic amylase or both. Some studies used amylase as a surrogate for exocrine function without correlation with an established exocrine tests like FEL-1. The use of amylase alone as an exocrine test is yet to be elucidated. Other anthropometric and biochemical markers would have been useful to assess the biochemical and metabolic state of patients. Our aim was to search for papers that focused on low serum amylase, therefore, we acknowledge we may have missed papers with findings of low serum amylase that were not included in the title or abstract. Finally, the term low or lower amylase level used in the studies may not reflect the actual term hypoamylasaemia (amylase level below normal reference range). Due to expected variations in reported amylase range, we recommend following local laboratory’s cut-off level of low amylase.
Clinical implication of low serum amylase
Our findings showed that low amylase can serve a dual function: as a metabolic marker in certain cases like obesity, MetS and DM; and as an exocrine marker. The diagnoses in our findings were already established, therefore, using amylase alone to make a diagnosis is not supported by our study. However, amylase is an inexpensive, widely available and easy to perform test. More importantly, the diagnosis of CP or PEI is not based on a blood test, therefire, amylase in a patient with suspected symptoms can be used as an adjunct test in assessment and guide diagnosis. In certain circumstances, for example, where FEL-1 is borderline with low suspicion of CP or PEI, amylase can be used to exclude them due to high specificity. Another example is patients with obesity, DM or MetS with gastrointestinal symptoms which may have under-recognised CP or PEI; a low amylase can be used as adjunct test and guide further referral to specialist for assessment and investigation.
Data availability statement
There are no data in this work.
Ethics statements
Patient consent for publication
Ethics approval
Not applicable.
References
Footnotes
Twitter @DrM_Jalal, @helpatologist
Contributors MJ: conceived the idea for the review, conducted literature search, interpretation of data, wrote review protocol and first draft of the paper and acts as a guarantor. SAG: conducted literature review and interpretation of data. NT: interpretation of data and first draft. KN: conceived the idea for the review, and first draft of the paper. All authors gave their approval for the final version of the work to be published.
Funding The authors have not declared a specific grant for this research from any funding agency in the public, commercial or not-for-profit sectors.
Competing interests None declared.
Provenance and peer review Not commissioned; externally peer reviewed.
Linked Articles
- Highlights from this issue