Acute pancreatitis (AP) is a potentially life-threatening acute inflammatory condition of the pancreas. Gallstones are the most common cause of AP, being responsible for almost 60% of all cases.1 Notably, the prevalence of AP is increasing, paralleled by an increase in the prevalence of obesity,2 and is a chronic, low-grade inflammatory state characterized by high circulating levels of proinflammatory cytokines.3 A large amount of visceral fat surrounding the pancreas may easily promote the inflammation in AP4 and, in fact, Acute Physiology and Chronic Health Evaluation-O (APACHE-O), a new severity scoring system that includes obesity as an independent predictor for AP outcome, has been proposed.5 Obesity is also associated with higher levels of inflammatory markers,6 which could potentially be of predictive value.
In cases of AP, pancreatic injury leads to a massive release of pancreatic lipase, which causes digestion of peripancreatic adipose tissue,7 which is then infiltrated by significant quantities of monocytes. This leads to the synthesis and secretion of adipose-specific proteins termed adipocytokines,8 which serve as new predictive markers.7 Among these proteins, adiponectin,9 known as adipocyte complement-related protein of 30 kDa (ACRP30), is unique because it is anti-inflammatory in nature and has been consistently shown to be down-regulated in those who are obese.10 The net effect of adiponectin is to increase the inflammatory response in states of obesity. Serum adiponectin levels in patients with AP are inversely correlated with BMI11 and, to date, there are no data available addressing concentration levels of serum adiponectin in patients with acute biliary pancreatitis; therefore, the current study investigates the ability of serum adiponectin to predict clinical severity of acute biliary pancreatitis and its potential use as an early marker of BMI.
Materials and methods
A group of 102 patients with an initial attack of acute biliary pancreatitis between January 2010 and April 2013 were included in this prospective study. They were admitted to the surgical ward at King Fahd Hospital, Saudi Arabia, which is the only tertiary referral centre for the large geographic area. The study protocols were approved by our institution’s Ethics Committee and institutional review board and all patients provided written informed consent. All patients underwent a routine medical history and clinical examination and the diagnosis of AP was based on the results of these investigations as well as the presentation of typical clinical features. These investigations showed an elevated serum amylase level of at least three times the normal upper limit, which peaks approximately 24 hours after the onset of an attack, and/or serum lipase was more sensitive and specific in the diagnosis of pancreatitis. Gallstone pancreatitis was diagnosed based on findings of gallstone or bile duct dilatation using abdominal ultrasonography and acute biliary pancreatitis was diagnosed based on the results of biochemical tests, including serum alanine transaminase levels of ≥ 65 U/l, serum bilirubin levels of > 18.1 Umol/l and alkaline phosphatase levels of > 137 U/l. A combination of ultrasonography and blood tests yielded a sensitivity of 95–98%.12 None of the patients had a history of drug consumption known to cause pancreatitis and none had ever consumed alcohol. Lipid profile and serum calcium were tested to exclude hyperlipidaemia- or hypercalcaemia-induced pancreatitis. Double contrast-enhanced computed tomography (CT) of the abdomen was used to assess the extent of pancreatic necrosis 48 hours after the onset of symptoms to predict pancreatic parenchymal injury. We graded the severity of pancreatitis into five distinct groups, A–E,13 by using the Balthazar grading for severity. Clinical severity evaluations were carried out using Ranson scoring and APACHEII within 48 hours after admission. Atlanta criteria were adopted and mild AP was defined if there were no criteria to suggest severe AP.14 Although severe AP was defined as the presence of at least one criterion, pancreatic necrosis was defined by > 30% pancreatic parenchymal necrosis, the presence of a pancreatic abscess and/or pseudocyst, systolic blood pressure < 90 mmHg, Po2 ≤ 60 mmHg, creatinine > 2 mg/dl after rehydration, gastrointestinal bleeding of > 500 ml in 24 hours, or death.15 Early local complications were detected by double contrast-enhanced CT of the abdomen. Furthermore, late complications such as pancreatic pseudocysts, thrombosis and abscesses were followed up with CT, if required, based on clinical presentation at the outpatient clinic 4–8 weeks after admission.
We excluded patients with recurrent pancreatitis, diabetes mellitus, hyperlipidaemia or hypertension, presence of morbidities that could possibly affect patient weight (e.g. hypothyroidism, pregnancy, alcohol consumption), a history of episodes of idiopathic and chronic pancreatitis, drug-induced and pancreatic divisum, pancreatitis induced by endoscopic retrograde cholangiopancreatography (ERCP), hyperlipidaemia- or hypercalcaemia-induced pancreatitis and patients who had been showing symptoms for > 2 days. This exclusion was based on medical history and previous examination.
Obesity was classified by BMI, which was measured by weight in kilograms divided by height in metres squared (kg/m2), as recommended by the World Health Organization.16 Waist circumference was measured in the standing position at the height of the umbilicus, while the subject breathed out gently. When marked fat accumulation caused the umbilicus to sag downwards, measurements were taken at a height midway between the lower margin of the ribs and the anterior iliac spine using a circumference measuring tape (Seca 201, CE marked; Seca GmbH & Co. KG., Hamburg, Germany). Venous blood samples were collected and, after sampling, the tubes were immediately centrifuged at 1.5 g for 10 minutes. Aliquots of serum were stored at −20°C and routine haematology and biochemistry tests were performed. All biochemical measurements were carried out by the same team of laboratory technicians and samples were collected after the patient’s first attack of AP at the time of hospital admission. We enrolled a control group of 102 age-, sex- and BMI-matched individuals because it has been reported that serum adiponectin concentration decreases with obesity17 and is negatively correlated with BMI or body fat.11 Plasma samples of adiponectin were taken from control subjects who were also similar in term of comorbidities, especially any relating to cardiovascular problems. Adiponectin plasma samples for both the test patients and control group were stored at −80°C and then diluted 100-fold, as recommended by the manufacturer (Assaypro, Saint Charles, MO, USA). The analysis was performed by one biochemical laboratory technician in a blinded fashion and adiponectin concentrations for both groups were then measured by using enzyme-linked immunosorbent assay (ELISA).
Enzyme-linked immunosorbent assay assay conditions
The serum was quantitatively assayed for adiponectin using a sandwich enzyme immunoassay. Standards and samples were sandwiched by immobilized antibodies and biotinylated polyclonal antibodies specific for adiponectin. Samples were analysed using the Bio-plex suspension array system (Bio-Rad Laboratories Inc., Hercules, CA, USA), which included a fluorescent reader, and Bio-plex Manager analytic software version 6.1.1 (Bio-Rad Laboratories Inc., Hercules, CA, USA) at the Taibah University Biochemical Department. The mean value for duplicate readings was calculated for each standard and sample.
The data were analysed using Statistical Product and Service Solutions, version 15 (SPSS Inc., Chicago, IL, USA). Means and standard deviations were used to describe continuous variables and serum adiponectin levels and additional continuous variables were analysed using correlation tests. Receiver operating characteristic (ROC) curves were used to evaluate the ability of the serum adiponectin level to predict the severity of AP and the relationship with BMI. A chi-squared test was used to calculate the association between classified variables and to estimate the relationship between levels of adiponectin and different variables. A two-tailed P-value of less than 0.05 was considered to be statistically significant and differences between the study groups were tested using a paired t-test.
A total of 102 patients (42 men and 60 women) were included in the study. All were diagnosed with acute gallbladder stone-induced pancreatitis. The mean age of the participants was 45 years and the mean BMI value was 30.5 kg/m2 (obese, class I). BMI cut-off values varied from 14.9 to 44.5 kg/m2 in men and 14.9 to 46.1 kg/m2 in women and waist circumference varied from 35 to 127 cm for both men and women with a mean value of 94.9 cm. The patient characteristics are summarized in Table 1. During the first 48 hours after admission, 22 patients (21.6%) suffered from severe AP but the majority experienced mild pancreatitis (80 patients, 78.4%), all of whom were diagnosed based on the Atlanta criteria. Local complications (acute fluid collection, pancreatic necrosis, pancreatic pseudocyst, pancreatic abscess and thrombosis) and systemic complications [shock, pulmonary insufficiency, renal failure, gastrointestinal bleeding, disseminated intravascular coagulation (DIC) and severe metabolic disturbances] occurred in 22 (21.6%) and 19 (18.6%) patients, respectively. The observed mortality rate was 3.9% and we classified the complications as either early or late.
SD, standard deviation.
Early complications (shock, pulmonary insufficiency, renal failure, gastrointestinal bleeding, DIC, severe metabolic disturbance and acute fluid collection) manifest at the onset or within the first 2–3 days, whereas late complications (pancreatic necrosis, pseudocyst, pancreatic abscess and thrombosis) usually manifest months or years following the resolution of an acute attack.
We investigated the relationship between severity of pancreatitis and risk factors such as age, sex, BMI, waist circumference and body weight. Our results showed that only BMI significantly correlated with severity of acute biliary pancreatitis (P = 0.007) (Table 2 and Figure 1). Mean serum adiponectin levels were significantly lower in obese patients (median = 6.10 ng/ml, range = 3.10–10.70 ng/ml); there was a negative relationship between BMI and adiponectin levels with a Pearson correlation coefficient of < 0.45 (P = 0.001) (Figure 2) and the ROC and the area under the curve (AUC) = 0.986 (Figure 3). Serum adiponectin correlated with BMI, but it did not correlate with weight or waist circumference.
After controlling for age, sex and BMI, the mean adiponectin value in AP patients at time of hospital admission was 6.0 ng/ml, compared with the control group with a mean value of 10.2 ng/ml (P = 0.000) (Table 3 and Figure 4). Serum adiponectin was significantly lower in patients with AP than in control subjects ( P < 0.0001), who were identical regarding BMI (Table 3 and Figure 4). We investigated the local and systemic complications and their relation with BMI and reported that acute fluid collection and pancreatic pseudocysts correlated significantly with obesity ( P = 0.043 and P = 0.000, respectively) (Table 4). In addition, a CT severity index of ≥ 5 was noticeable in obese patients ( P = 0.032) (Table 4).
SD, standard deviation.
We studied the relationship between severity of acute biliary pancreatitis and plasma adiponectin. The mean adiponectin value at admission as 5.8 ng/ml in the severe acute biliary pancreatitis group and 6.1 ng/ml in the mild AP group (P = 0.569); therefore, adiponectin failed to serve as predictive marker of clinical severity. In addition, ROC and AUC were 0.460 (P = 0.569) (see Figure 3).
The main factors involved in acute biliary pancreatitis are pancreatic hyperstimulation and pancreatic duct obstruction by gallstone migration. The exact mechanisms by which they initiate AP are unknown18 but some authors reserve the term microlithiasis for small concretions in gallbladder or biliary trees that are < 3 mm and cannot be seen using conventional ultrasonography.19 A large amount of visceral fat surrounding the pancreas may easily exacerbate the inflammation in AP4 and this hyperinflammatory state may enhance gallstone formation and also contribute to papillitis as well as sphincter dysfunction.20 Interestingly, pancreatic microcirculation is lower in obese patients than in those who are not obese, which increases the risk of ischaemic injury and subsequent local infections.21 The pancreas is highly vulnerable to ischaemic damage and, therefore, pancreatic ischaemia due to hypoperfusion may also be a critical factor involved in the progression from oedema to necrosis.22
Obesity may also be linked to the severity of AP. It is generally accepted that obesity is associated with an increased risk of AP development; therefore, greater abdominal adiposity and higher waist circumference often accompany severe pancreatitis. Patients with an android fat distribution and higher waist circumference measurements are at greater risk of developing severe AP.23 A study by De Waele et al.24 demonstrated that obesity is correlated with severity of AP, local complications and mortality in patients with AP and these results were confirmed via meta-analysis carried out by Wang et al.25 Furthermore, our study confirmed that a BMI of > 30 kg/m2 is associated with severe pancreatitis ( P = 0.007) (see Table 2).
Central fat distribution was revealed as a more definitive risk factor than the actual amount of fat and it is, therefore, plausible that the inflammatory properties of intrapancreatic fat may be involved in triggering the onset of AP. In line with these data, obesity may intensify the immune response, which is able to exacerbate pancreatic injury. Adiponectin, also known as ACRP30, is secreted from peripancreatic adipose tissue. It inhibits the expression of adhesion molecules and is believed to have anti-atherosclerotic26 and anti-inflammatory properties27 and has been consistently shown to be down-regulated in those who are obese.10 The net effect of adiponectin is to increase the inflammatory response in states of obesity. The serum adiponectin level in patients with AP is inversely correlated with BMI,11 which was confirmed by our study (see Figure 2). Studies have indicated that decreased plasma adiponectin concentration is associated with obesity, insulin resistance,17 essential hypertension,28 inflammation and atherosclerosis,29 and acute myocardial infarction.30 A strong inverse association between adiponectin and systemic inflammation has been demonstrated,31 and Tukiainen et al.32 examined serum levels of adiponectin in 24 patients with AP and, after matching patients according to their age, sex, BMI and aetiology, the plasma levels of adiponectin and leptin at admission did not correlate with disease severity, suggesting that the adipokines do not affect the course of AP,32 but the sample size was small: only 24 patients.
Pancreatitis induced by gallbladder stones has a lower adiponectin value and these patients have a higher BMI, which may suggest that low adiponectin levels were related to obesity. In our study, the sample size was three times larger than in previous studies and was designed with adequate power and strict exclusion criteria to avoid multiple bias and to eliminate any confounding variables that affect patients’ weight, adiponectin level, BMI and the clinical sequelae of initial AP; however, the weak point of the study is that it seems almost impossible to select an obese population without hyperlipidaemia, hypertension or other comorbidities that are strongly associated with obesity. Part of the population is probably undiagnosed for multiple risk factors, but our exclusion of participants was based on medical history and on previous examinations.
In our sample, those with ERCP-induced pancreatitis were excluded because the serum adiponectin level may play a role in contributing to papillitis and sphincter dysfunction and could be a pancreatitis risk factor.33,34 All biochemical measurements were carried out by the same team of laboratory technicians using the same methods throughout the study period. In addition, one biochemical laboratory technician performed the ELISA blinded. The control subjects were age, sex and BMI matched to the patients with AP because it has been reported that serum adiponectin concentration decreases in obesity and is negatively correlated with BMI and body fat.11 Android fat distribution surrounding the pancreas can be detected by serum adiponectin level; however, this cannot be measured by routine BMI. We noticed that our acute biliary pancreatitis patients showed hypoadiponectinaemia after matching BMI with control subjects, which suggested that adiponectin level is low in the acute inflammatory phase of biliary pancreatitis and, therefore, serum adiponectin can be anti-inflammatory in cases of obesity. In addition, evaluating hypoadiponectinaemia may be a useful tool for the early assessment of insulin resistance and cardiovascular pathology in pancreatitis patients; however, before introducing adiponectin into clinical environment, these results must be confirmed by larger and multicentre studies and such investigations are warranted.
We have opened the gate to study adiponectin level in obese patients and to compare the values before and after a weight reduction programme or bariatric surgery. The current single-institution study of AP, and especially severe AP, is limited by low patient numbers. The results are also limited to the characteristics of the local study population, but the use of new pancreatic injury markers of inflammatory response is very promising. Adiponectin has been studied mainly in research settings and the clinical usefulness remains to be determined.
Obesity is a clear risk factor for severe acute biliary pancreatitis. We describe here, for first time, that serum adiponectin concentration is decreased in patients with acute gallbladder stone-induced pancreatitis as marker for high BMI at hospital admission, but adiponectin failed to serve as predictive marker of clinical severity in acute biliary pancreatitis.