CLINICAL NOTE

Article in PDF format - JOP Home page

Evaluation of the Bilio-Pancreatic Region Using Endoscopic Ultrasonography in Patients Referred with and without Abdominal Pain and CA 19-9 Serum Level Elevation

Jose Luis Ulla Rocha¹, Maria Victoria Alvarez Sanchez¹, Javier Paz Esquete², Estela Fernandez Salgado¹, Carlos Alvarez Alvarez³, Manuel Javier Vazquez Sanluis¹, Luis Ledo Barro¹, Enrique Vazquez Astray¹

Services of ¹Digestive Disease, ²Preventive Medicine, and ³Pathology, Complexo Hospitalario de Pontevedra. Pontevedra, Spain

Context When assessing the bilio-pancreatic region, collating the findings of serum CA 19-9 values together with findings from various imaging tests - especially endoscopic ultrasonography - is not a simple issue in daily clinical practice.

Aim To assess the usefulness of endoscopic ultrasonography in an Endoscopic Ultrasonography Unit in two situations: patients with asymptomatic elevation of serum CA 19-9 and patients who presented with abdominal pain plus elevation of CA 19-9.

Methods A retrospective study of those patients who underwent radial endoscopic ultrasonography between October 2004 and September 2005 in our institution, considering an elevation of CA 19-9 (equal to or greater than 37 U/mL) with or without symptoms. In each case, the parameters recorded were: levels of CA 19-9 one week before EUS, results from other imaging techniques (US, helical CT), and final diagnosis according to pathological and/or clinical evolution criteria. Patients with previous attacks of acute pancreatitis and also those who presented with bile duct dilation or space-occupying lesions in image studies (US and CT) were excluded. Twenty-two patients met the inclusion criteria.

Results Asymptomatic elevation of CA 19-9 was found in 15 patients while 7 patients had elevated CA 19-9 levels as well as pain of uncertain origin. The results of EUS in the asymptomatic patients were: chronic pancreatitis in 7 patients, no pancreatic alterations in 3 patients, and renal cysts, choledocholithiasis, michrolithiasis and liver cirrhosis in one patient, respectively. In patients with abdominal pain, EUS showed chronic pancreatitis in 6 cases and adenocarcinoma of the tail of the pancreas in the remaining patient.

Conclusions When EUS was indicated for the asymptomatic elevation of CA 19-9, the main findings were benign diseases. EUS was useful in studying patients with idiopathic abdominal pain and a slight elevation of CA 19-9 since it allowed us to detect chronic pancreatitis and even early adenocarcinoma of the pancreatic tail.

Radial endoscopic ultrasonography has become a technique used worldwide [1] and is very useful in bilio-pancreatic diseases [2]. Its ongoing development has led to a change in the clinical management of patients with these pathologies [3]. However, this technique should be more frequently used by endoscopists, due to its clinical usefulness in the correct management of patients [4]. It has also been shown that EUS is effective in the study of benign pathologies such as acute pancreatitis [5], chronic pancreatitis [6], and biliary lithiasis [7]. EUS permits the detection of small intrapancreatic tumors [8], and some authors advocate the use of this technique for an initial study of suspected small intrapancreatic tumors, avoiding the use of other diagnosing techniques [9]. This consideration is very important as this neoplasia has a poor prognosis [10]. CA 19-9 has been identified as a tumoral marker [11] having a high sensitivity [12] in patients with pancreatic neoplasia, but having a low specificity. CA 19-9 levels are also elevated in other benign pathologies [13] such as digestive pathologies (including acute pancreatitis, pancreatic pseudocyst, cholangitis, cholecystitis, hepatitis, or cirrhosis), respiratory processes (including bronchiectasis, tuberculosis, asbestosis, bronchial asthma, fibrosis of the lung and idiopathic interstitial pneumonia), renal cysts and rheumatologic disorders (such as Sjögren’s syndrome, rheumatoid arthritis, Sharp’s syndrome, arteritis of giant cells, dermatopolymyositis and rheumatic polymyalgia). This marker is also elevated in other malignant pathologies of the gastrointestinal tract [14]. Considering the usefulness of EUS in assessing the bilio-pancreatic area both in benign and malignant pathologies, our aims were to determine the most frequent pathologies in patients with an asymptomatic elevation of CA 19-9 and to assess the usefulness of EUS in diagnosing patients with idiopathic abdominal pain and elevation of CA19-9.

A retrospective descriptive study was performed and data on patients who came to our Gastroenterology Department from October 1^st, 2004 to September 30^th, 2005 were collected in a setting involving a 600 bed hospital which serves 250,000 inhabitants.

All patients who were sent to our department for EUS on the basis of suspected bilio-pancreatic disease and who presented an elevation of CA 19-9 (equal to or greater than 37 U/mL) in the week before EUS were included in the study. Patients with acute pancreatitis, with a dilated bile duct or a bilio-pancreatic space-filling lesion on US or helical CT were excluded. Twenty-two patients met the inclusion criteria.

The patients were subdivided into two categories: patients with an asymptomatic elevation of serum CA 19-9 and patients with abdominal pain of unknown etiology and an elevation of CA 19-9.

Patients included in our study underwent EUS and helical CT. EUS was performed with an Olympus UM160 endoscope. The bilio-pancreatic area was explored using frequencies of 5 and 7.5 MHz which allowed for a deeper exploration, but sectorial EUS fine needle aspiration and endoscopic retrograde cholangiopancreatography were not performed.

Serum CA 19-9 was evaluated in the week before EUS using the ARCHITECT 12000 CA 19-9 XR kit (Abbott, Abbott Park, North Chicago, IL, USA). Serum bilirubin values were also recorded since it is widely known that high bilirubin levels might interfere with the determination of CA 19-9 [15]. The field of the specialist who had previously seen the patients and the findings of both the abdominal ultrasonography (US) and the helical computerized tomography (CT) were recorded. Finally, the results of the EUS examination are described.

The study was approved by the Ethics Committee of the Complexo Hospitalario de Pontevedra, Pontevedra, Spain. All the patients were managed according to the usual accepted standards of clinical practice.

Out of the 22 patients enrolled in this study, 15 patients had an asymptomatic elevation of CA 19-9 while 7 patients had elevated CA 19-9 levels as well as abdominal pain.

The EUS findings and the final attributable cause of the elevation of CA 19-9 in the 15 patients who were referred due to an asymptomatic elevation of CA 19-9 are shown in Table 1. None of them showed bilio-pancreatic neoplastic pathology at EUS. In these patients, chronic pancreatitis was considered when they showed the following parameters: lobulation of the pancreatic border (Figure 1), presence of intrapancreatic echogenic foci and irregularity of Wirsung’s duct 2) parenchymal criteria plus 1 ductal criteria) [16]. Liver cirrhosis was diagnosed when lobulation of the hepatic lobe border was present, visualized from the stomach and the duodenal bulb [17]. Choledocholithiasis was diagnosed when hyperechogenic imaging inside the choledochus showed posterior acoustic shadowing [18]. Microlithiasis (Figure 2) was diagnosed when an intravesicular starry-sky pattern was observed. Renal cysts were visualized as anechoic images with dorsal acoustic accentuation inside the renal parenchyma. In the other referred patients, we attributed the elevation of CA 19-9 to other pathologies such as Sjögren’s syndrome or drug-induced hepatitis with eosinophilia. In one patient, we could not establish a definitive etiology of the asymptomatic elevation of serum CA 19-9.

Figure 1. Chronic pancreatitis.

Figure 2. Microlithiasis.

The patients referred mild epigastric and left hypochondrium abdominal pain. No abdominal lesions were detected at US and helical CT (Table 1).

An early pancreatic neoplasia (a 12 mm mass in the pancreatic tail) was detected at EUS in one patient with a serum CA 19-9 level of 53.4 U/mL (Figure 3). In this patient, an MRI also failed to show any pancreatic lesions. The patients underwent surgery; the tumor was classified as pT1, N0, M0 and the pathological study confirmed the diagnosis of ductal adenocarcinoma with perineural infiltration in the pancreatic exocrine portion and an adjacent desmoplastic reaction (Figure 4).

Figure 3. Pancreatic tail adenocarcinoma.

Figure 4. Pathologic findings of pancreatic tail cancer.

Asymptomatic elevation of serum CA 19-9 is not considered an indication for EUS [19]. However, because this tumor marker is used frequently, its positive results usually lead to further investigation, which might include EUS.

No case of pancreatic neoplasia was found in patients who were referred with an asymptomatic elevation of CA 19-9, which seems to be in agreement with previous published studies [20]. Furthermore, it has been reported in the literature that the sensitivity of EUS is better than that of helical CT in the detection of early morphological changes in chronic pancreatitis [21].

Pancreatic gland changes related to chronic pancreatitis were detected in about 50% of the patients. It has been reported that there may be a slight increase in serum CA 19-9 levels in patients with chronic pancreatic disease without an associated neoplasia [22]. A cut-off level of 300 U/mL has been established to differentiate this entity from an associated neoplasia, although the determination of this marker in pancreatic juice may be more accurate [23]. Cases diagnosed in our study as chronic pancreatitis had values of CA 19-9 which ranged from 38 to 144 U/mL. The use of EUS together with other biochemical parameters has been pointed out for an early diagnosis of chronic pancreatitis , considering the high alcohol intake in our country [24]. Liver cirrhosis may also slightly increase the levels of CA 19-9 and, therefore, 50% of the patients with chronic pancreatitis had also presented a morphology suggestive of cirrhosis on previous helical CT. At this point, a correct clinical history is necessary in order to exclude other extrapancreatic causes in the elevation of CA 19-9, such as rheumatic disease or kidney cysts which have been described as other causes of an altered serum CA 19-9 level.

Less aggressive procedures such as US are preferably used as a first step in investigating the bilio-pancreatic area. Recently, EUS has become an important addition in the study of these patients [25]. For a long time, ERCP was considered to be the gold standard in establishing a final diagnosis of chronic pancreatitis (Cambridge classification) [26]. Nowadays, EUS can be compared to ERCP alone or associated with analysis of pure pancreatic juice obtained after secretin stimulation [27].

In spite of the small number of cases, EUS disclosed diseases which were not seen using other imaging methods and ruled out pancreatic cancer in all of them. This observation suggested that EUS should be used more frequently which implies agreement with former studies where the importance of pancreatic cancer exclusion was stressed [9].

Our study shows no cases of neoplastic pathology and it shows that the most prevalent pathology is chronic pancreatitis in patients with an asymptomatic elevation of CA 19-9.

On the other hand, a frequent clinical problem is patients suffering from abdominal pain where a bilio-pancreatic origin is suspected. Less aggressive procedures such as US are preferably used as a first step in the diagnosis. Recently, EUS has become an important addition in the study of these patients [25]. The agreement between CT and endoscopic ultrasound studies revealed EUS to be more accurate in diagnosing benign and malignant pathologies [28].

Nevertheless, with the forthcoming multi-detector CT, this difference may disappear [29], although it should be considered that the cost of EUS is lower than that of helical CT [30].

In our study, 86% of the patients with abdominal pain (6/7) had chronic pancreatitis with CA 19-9 levels ranging from 38 to 153 U/mL. Clinical symptoms showed mild epigastric pain. Considerations related to the value of EUS diagnosis in chronic pancreatitis patients are similar to those that are shown above. In none of the cases was serum IgG4 determined to exclude autoimmune pancreatitis [31] since, in all cases, an elevated alcoholic intake was prevalent in the population studied [24].

It is important to point out the detection of an early adenocarcinoma of the pancreatic tail in a 75-year-old woman with two negative imaging studies (US and helical CT), a patient with mild abdominal pain irradiating to the back and a slight CA 19-9 elevation (53.4 U/mL). MRI was also performed on this patient, since this technique allows a non-invasive assessment of the pancreas [32]. Notwithstanding the fact that no pancreatic pathology was demonstrated and the fact that the results of the EUS were suspicious, a distal pancreatectomy was performed, showing the presence of a well-differentiated 12 mm adenocarcinoma with a free surgical edge and no adenopathies in the anatomopathologic exam. Currently, this patient is asymptomatic and shows no signs of recurrence. Nevertheless, it has been reported that some patients cannot synthesize the antigen (negative Lewis’ genotype) and, therefore, they show low levels of CA 19-9 in the presence of an adenocarcinoma. Consequently, the normality of the serum level of CA 19-9 cannot exclude the presence of a pancreatic adenocarcinoma [33].

Our conclusions, according to the study objectives, are as follows: 1) in patients with an asymptomatic elevation of CA 19-9, the most prevalent pathology was chronic pancreatitis and no case showed a neoplastic pathology; 2) in patients with unexplained abdominal pain and an elevation of serum CA19-9, EUS is helpful inasmuch as chronic pancreatitis is the most frequent disease due to high alcohol intake. EUS with a slight elevation of CA 19-9 allowed us to detect an early pancreatic tail adenocarcinoma in a patient who had been suffering from a backache for several months and had notable weight and appetite loss. The combination of clinical data, a careful interpretation of serum CA19-9 determination and EUS leads to the correct management of bilio-pancreatic pathologies.

Article in PDF format

Received November 30^th, 2006 - Accepted February 8^th, 2007

Keywords Abdominal Pain; CA-19-9 Antigen; Endosonography; Pancreatitis, Chronic; Pancreatic Neoplasms

Correspondence
Jose Luis Ulla Rocha
Digestive Disease Service
Complexo Hospitalario de Pontevedra
C/Mourente sn Pontevedra 36071
Spain
Phone :+34-98.680.0000
Fax:+34-98.680.0000
E-mail: ullarojl@terra.es