Utility of Contrast-Enhanced Computed Tomography for Differentiating Pancreatic Ductal Adenocarcinoma from Inflammatory Mass in Chronic Calcific Pancreatitis

Abstract Objective  The aim of this study was to identify the most useful contrast-enhanced computed tomography (CECT) features for differentiating pancreatic ductal adenocarcinoma (PDAC) from mass-forming chronic pancreatitis (MFCP) in chronic calcific pancreatitis (CCP). Methods  In total, 101 patients with CCP and focal pancreatic mass formed the study group. Sixteen qualitative and four quantitative parameters were analyzed. Qualitative parameters included size, site, margin, intralesional hypodensity, collateral duct sign, abrupt pancreatic duct (PD) cutoff, upstream PD dilatation, distal pancreatic atrophy, double duct sign, enhancement pattern, contrast attenuation, peripancreatic inflammation, vascular involvement, regional nodes, and metastasis. Quantitative parameters included duct-to-body ratio, common bile duct (CBD) diameter, main pancreatic duct (MPD) diameter, and carcinoembryonic antigen 19-9 (CA19-9). Sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV), and accuracy were calculated for prediction of PDAC. A receiver operating characteristic (ROC) curve analysis was performed and the area under ROC curve (AUROC) was calculated to determine diagnostic accuracy to assess the optimal cutoff. Results  PDAC was confirmed in 48 patients and MFCP in 53 patients. A duct-to-body ratio greater than 0.48 had 95.5% sensitivity, 83.3% specificity, 80.8% PPV, 96.2% NPV, and 88.5% accuracy for predicting PDAC. A CBD diameter cutoff ≥9.5 mm had an accuracy of 75% ( p  < 0.019) and an MPD cutoff ≥6.25 mm had an accuracy of 67.8% ( p  = 0.008) for predicting PDAC. On binary logistic regression, the duct-to-body ratio was found to be the significant independent factor associated with malignancy. Conclusion  A duct-to-body ratio greater than 0.48, intralesional hypodensity, and abrupt duct cutoff are the most helpful computed tomography (CT) features for distinguishing PDAC from MFCP in CCP. On binary logistic regression, the duct-to-body ratio was found to be a significant independent factor. Interspersed normal parenchyma was observed as a very specific sign of MFCP. Intraparenchymal hypodensity has high specificity, but further validation is needed.


Introduction
Chronic calcific pancreatitis (CCP) is characterized by pancreatic duct (PD) dilatation, pancreatic atrophy, and coarse pancreatic ductal calcification. 1 There in an increased risk of pancreatic ductal adenocarcinoma (PDAC) in CCP. 2,3In approximately 30% of patients with CCP, fibrosis and inflammation of the glandular elements manifest as focal pancreatic enlargement that may simulate a mass, an entity known as mass-forming chronic pancreatitis (MFCP) and is a benign entity. 4Differentiating PDAC from MFCP is difficult due to their overlapping clinical and imaging features. 5owever, the differentiation between PDAC and MFCP is important, as surgical resection is the standard of care for PDAC and the only effective means of cure, whereas MFCP is managed conservatively. 4Incorrect diagnosis of MFCP as PDAC can result in an unnecessary surgical resection and conversely PDAC diagnosed as MFCP may delay surgical. 2oninvasive imaging may play a crucial role in early diagnosis of cancer to differentiate between these two entities.
Multiphasic contrast-enhanced computed tomography (CECT) is the most common initial imaging modality used for evaluation of patients with focal mass lesions in chronic pancreatitis.A literature search shows a few studies differentiating between benign and malignant lesions in CCP. 6ith CECT, PDAC in CCP shows variable enhancement characteristics.PDAC appearance can vary from a hypodense to isodense to hyperdense mass. 7MFCP also has variable attenuation based on the extent of fibrosis and inflammatory component and shows increased enhancement compared to PDAC but could appear isodense or hypodense. 7The imaging features of PDAC and MFCP overlap, making it difficult to differentiate the two entities based on enhancement characteristics alone.2][13] Perfusion CT and dual-energy CT are useful advanced technologies that may be helpful in differentiating MFCP from PDAC using perfusion characteristics and iodine uptake within the lesions. 14,15However, perfusion CT and dual-energy CT techniques are not widely available.
The purpose of this study is to identify the most useful CT features on multiphasic CECT that would help distinguish PDAC and MFCP in the setting of CCP.

Study Population
This retrospective study included 101 consecutive patients with CCP who underwent a CECT and had a diagnosis of focal pancreatic mass that was histologically confirmed.The study period was from January 2015 to December 2020 at our institute.The inclusion criteria were (1) CT imaging features of CCP, (2) presence of a focal pancreatic mass, and (3) pathological confirmation of the mass.The pathological specimens were obtained with either endoscopic ultrasound-guided fine needle aspiration (EUS-FNA) or surgical resection.Pathological interpretation was performed by the institutional pathologist with 20 years of experience.Patients without a dedicated multiphasic CECT from our institute were excluded from the study.

Imaging Technique
All the patients were imaged using either a 64-slice multidetector row CT (MDCT; General Electric [GE] Lightspeed, GE Healthcare, Milwaukee, WI, United States) or 128-slice Maxima (GE Healthcare).All the patients received 600 mL of Gastrografin (GE Healthcare) as oral contrast.The protocol consisted of a noncontrast CT performed initially followed by CECT in two phases: pancreatic parenchymal phase at 40 to 50 seconds and portal venous phase at 65 to 70 seconds after the start of injection.A total of 80 to 100 mL (1.5 mL/kg body weight) of 350 mg iodine/mL (Iohexol [Omnipaque], GE Healthcare, Shanghai, China) or Iopromide (Ultravist, Bayer Zydus Pharma Ltd, Whippany, NJ, United States) was administered intravenously at a flow rate of 4 mL/s using a dual head pressure injector (Medrad Stellant CT Injector System, Indianola, PA, United States).The images were obtained from the dome of the diaphragm to the iliac crest in all three phases (noncontrast, pancreatic phase, and portal venous phase).
The CT parameters were current modulation, 5-mm slice thickness with 0.625-mm reconstruction, 35 to 50 cm field of view and 512 Â 512 matrix with sagittal and coronal reconstructions.

Image Analysis
Two abdominal radiologists with 21 and 11 years of experience reviewed the images in consensus.The images were reviewed on advantage GE workstation 4.2 and picture archiving and communication system (PACS).Qualitative parameters included size, site of involvement, margin, intralesional hypodensity, collateral duct sign, abrupt PD cutoff, upstream PD dilatation, distal pancreatic atrophy, double duct sign, enhancement pattern, contrast attenuation, Conclusion A duct-to-body ratio greater than 0.48, intralesional hypodensity, and abrupt duct cutoff are the most helpful computed tomography (CT) features for distinguishing PDAC from MFCP in CCP.On binary logistic regression, the duct-to-body ratio was found to be a significant independent factor.Interspersed normal parenchyma was observed as a very specific sign of MFCP.Intraparenchymal hypodensity has high specificity, but further validation is needed.peripancreatic inflammation, vascular involvement, regional nodes, and metastasis.Quantitative parameters included DBR, common bile duct (CBD) diameter, main pancreatic duct (MPD) diameter, and carcinoembryonic antigen 19-9 (CA19-9).The final diagnosis of PDAC or MFCP was based on histopathology.
All the qualitative and quantitative parameters were studied in detail to differentiate between benign and malignant masses (►Table 1).
Intralesional hypodensity was defined as a lack of contrast enhancement surrounded by an irregular inner wall in the postcontrast phase.][18] They appeared as unilocular or multilocular cystic areas with septa with or without enhancement.
Intralesional interspersed parenchyma refers to the presence of islands of normal pancreatic tissue within the mass.
DBR and MPD diameter were determined just distal to the mass to within 1 cm and pancreatic parenchymal thickness measured perpendicular to the longitudinal axis of the parenchyma at the same point.The CBD diameter was measured at the suprapancreatic level.
The enhancement characteristics analyzed were categorized as homogeneous or heterogeneous depending on the pattern of enhancement and isodense/hyperdense/hypodense compared to the adjacent pancreatic parenchyma in the portal venous phase.

Statistical Analysis
All the data collected were coded and entered in a Microsoft Excel sheet, which was rechecked and analyzed using SPSS statistical software version 22. Quantitative variables were summarized using mean and standard deviation (SD).Categorical variables were represented using frequency and percentage.Independent sample test and Mann-Whitney U test were used to test the statistical significance of difference between the means of variables among different independent groups.Pearson's chi-squared test and Fisher's exact test were used for comparing the categorical variables between groups.The receiver operating characteristic (ROC) curves were generated for comparing the area under the curve (AUC) depicted by different parameters for prediction of malignancy and for finding out the diagnostic characteristics.Binary logistic regression was performed to determine independent factors associated with malignancy.A p-value of less than 0.05 was considered statistically significant.

Results
The study population comprised 70 males and 31 females with age ranging from 15 to 84 years.PDAC was confirmed in 48 patients and MFCP in 53 patients.The mean age of patients with PDAC was significantly higher than that of the patients with MFCP (53.5 AE 10.4 vs. 46.4AE 13.2 years, p < 0.003).Both PDAC and MFCP were common in males as compared to females.Among the PDAC, 60.4% of patients were males, while 39.6% were females.
Most of the benign and malignant masses measured between 2 and 5 cm with no statistically significant difference between size groups of PDCA compared to MFCP (p ¼ 0.948).The most common location for both MFCP and PDAC was the head (61.5 and 81.3%, respectively) of pancreas followed by the body (34.6 and 18.8%, respectively) with no significant differences (p ¼ 0.06).
CBD dilatation, double duct sign, attenuation characteristics, and regional lymphadenopathy did not contribute to the significant difference between PDAC and MFCP.CBD dilatation was seen in masses located in the head/body lesions extending to the head (45% in PDAC vs 43.8 % in MFCP; p ¼ 0.8).

Discussion
There is an increased risk of PDAC in CCP as evidenced by various studies with cumulative risk increasing over time from less than 2% at 10 years to 4% at 20 years following diagnosis of chronic pancreatitis. 3We therefore evaluated the CT findings to identify the most useful features for diagnosing malignancy.In this retrospective study, we found Utility of Contrast-Enhanced Computed Tomography Pratap et al.
that DBR, intraparenchymal hypodensity, and abrupt PD cutoff were the most significant radiological features with higher accuracy for predicting PDAC in CCP.
In a prior study, Eloubeidi et al showed that a DBR of greater than 0.34 strongly favors the diagnosis of PDAC. 13In their study, the PPV, NPV, sensitivity, specificity, and accuracy of pancreatic DBR !0.34 for detecting malignancy were 87, 99, 94, 97, and 97%, respectively.When the presence of PD dilatation alone was considered, the PPV and accuracy was only 50 and 83%, respectively, for the detection of pancreatic cancer.Another study by Karasawa et al showed that a DBR of greater than 0.5 indicated malignancy. 19Our study results with DBR of greater than 0.48 are comparable to these studies.On binary logistic regression, the DBR was found to be the significant independent factor associated with malignancy similar to those obtained by Karasawa et al 19 (►Fig.1).
Intralesional hypodensity had a high accuracy, sensitivity, and NPV in identifying malignant lesions, although there were overlaps.The diagnosis of intrapancreatic fluid-containing cysts indicated inflammatory mass.Distinguishing intralesional hypodensity from walled-off necrosis is important, although at times difficult.In our study, intralesional hypodensity was more frequent in PDAC compared to MFCP, which was consistent with the study by Ren et al 17 where this finding was seen in 59.6% of PDAC as compared to 28.6% of MFCP cases (►Fig. 2).
In our study, MPD abrupt cutoff had moderate accuracy of 68% but a high sensitivity of 96.2% for predicting malignancy.Gangi et al 20 has reported that MPD abrupt cutoff and upstream MPD dilatation were the earliest finding of PDAC, and this was detected more than 18 months before the PDAC diagnosis.Abrupt duct cutoff can be identified in malignant lesions, whereas progressive narrowing with smooth tapering of ducts is seen in benign cases. 16The combination of abrupt PD cutoff and DBR increased the diagnostic accuracy (►Figs.3 and 4).CA19-9, a diagnostic and prognostic marker of malignancy, is also elevated in inflammatory lesions of the pancreas.In our study, a cutoff of 106 for CA19-9 had good specificity but with low sensitivity.There are several studies proposing different cutoffs. 21,22In a study by Bedi et al, CA19-9 using a cutoff value of 37 U/mL had 68% sensitivity and 70% specificity. 21Higher positivity rates were obtained using other cutoff values such as 100, 200, and 300 U/mL and reaching 100% specificity for PDAC using 300U/mL.
Ruan et al 18 showed larger lesions with lobulated margins were predominantly observed in PDAC than in MFCP (83.33 vs. 12.5%) as in our study.It is difficult to assess the margins if the lesions are small and contour nondeforming (<2 cm). 23isible remnants of normal pancreatic tissue (interspersed normal parenchyma) within lesions were a very helpful finding and could be used to distinguish MFCP from PDAC, 18 and our study showed that this finding had a PPV of 100% (►Fig. 5).
The presence of dilated collateral or side branches of the PD coursing throughout the mass is a finding that can be seen in inflammatory masses. 24A collateral duct sign is better appreciated in magnetic resonance imaging (MRI) than in CT.In our study, the collateral duct sign was shown to have poor specificity in predicting MFCP.This may be due to the fact that our study included only CECT studies.
We found that the enhancement pattern and contrast attenuation of mass were not statistically significant to differentiate between PDAC and MFCP, which may be due to a variable extent of necrosis or fibrosis in these lesions as suggested by Johnson et al. 25 Graph 1 Receiver operating characteristic (ROC) curve of duct-to-body ratio for predicting malignancy.Peripancreatic inflammation and pseudocysts had low sensitivity, specificity, and accuracy in differentiating between benign and malignant lesions as pseudocysts could be seen in malignancy due to obstructive pancreatitis. 18,26scular involvement may help in differentiation although the vessels can be involved in both MFCP and PDAC.Venous narrowing of the splenoportal confluence with collateral formation is seen in both PDAC and MFCP.Hence, venous  encasement did not help in differentiating between benign and malignant lesions.Our study showed vascular involvement in the majority of malignant lesions (64.2%; p ¼ 0.002).However, 33% of MFCP also showed vascular involvement on imaging decreasing the specificity (►Fig.6).
Our study has some limitations.This was a retrospective design as our inclusion criteria were histological confirmation of the pancreatic mass.Some of the parameters assessed could not be applied to all masses, for example, DBR could not be applied to masses present in the body and tail masses; CBD dilatation was studied only for head masses.Intralesional hypodensity which is one of the most helpful features in differentiating benign from malignant lesions was seen only in 58% of total cases studied.It is possible that MRI may be more useful in demonstrating several of these signs and the advantage of multiplanar imaging capability.This study focused only on the CECT features as it was the most commonly performed imaging modality in our study population.The utility of MRI should be evaluated in future studies.

Conclusion
Differentiating PDAC from MFCP in CCP continues to be a challenge.A DBR greater than 0.48, intralesional hypodensity, abrupt PD cutoff, and CA19-9 value of 106 IU/mL were the most useful quantitative CT features for distinguishing PDAC from MFCP.The DBR was found to be the significant independent factor associated with malignancy.Interspersed parenchyma is a very specific sign for MFCP.Further characterization of intralesional hypodensity may be possible with advanced imaging applications, large-scale data analysis, and radiomics.

Compliance with Ethical Principles
All procedures performed were in accordance with the ethical standards of the institutional research committee and with the 1964 Helsinki declaration and its later amendments or comparable ethical standards.Informed consent was obtained from all patients for being included in the study.

Graph 2 Graph 3
Receiver operating characteristic (ROC) curve of main pancreatic duct (MPD) diameter for predicting malignancy.Receiver operating characteristic (ROC) curve of common bile duct (CBD) diameter for predicting malignancy.

Fig. 1
Fig. 1 (a) Contrast-enhanced computed tomography (CECT) showing a malignant head mass with calcification (arrow) and dilated main pancreatic duct (MPD) with distal pancreatic atrophy.The shorter line indicates the MPD diameter and the longer long indicates the transverse diameter of pancreatic body.The duct-to-body ratio (DBR) was 0.92.(b) CECT showing benign pancreatic head mass (asterix and arrow) with a DBR of 0.27.

Fig. 2
Fig. 2 (a) Axial contrast-enhanced computed tomography (CECT) showing a malignant mass (arrow) in the pancreatic body region with intralesional hypodensity and calcific speck with distal parenchymal atrophy.Underlying chronic calcific pancreatitis (CCP) with parenchymal calcifications noted in the head and tail region.(b) Axial CECT showing a malignant pancreatic mass (arrow) with intralesional hypodensity in a background of atrophic pancreatitis.(c) CECT showing a malignant head mass (arrow) with intraparenchymal hypodensity and calcification.(d) CECT showing an inflammatory mass with intrapancreatic pseudocyst (dot).

Fig. 3
Fig.3Contrast-enhanced computed tomography (CECT) axial oblique section showing total atrophy of the pancreas with dilated pancreatic duct, showing abrupt cutoff (arrow) in caliber at the level of the pancreatic head mass.The lesion is homogeneous and isodense with a lobulated contour (asterisk).

Fig. 4
Fig. 4 (a) Coronal contrast-enhanced computed tomography (CECT) showing a malignant pancreatic head mass with abrupt duct cutoff and duct-to-body ratio >0.9.Also demonstrated is a double-duct obstruction.Asterix indicates dilated CBD and dot the MPD.(b) CECT showing an inflammatory pancreatic head mass with calcification with smooth narrowing of the main pancreatic duct (MPD) (dot).

Fig. 5
Fig. 5 (a) Axial contrast-enhanced computed tomography (CECT) demonstrating an inflammatory head mass with interspersed normal pancreatic parenchyma (arrow) and collateral duct sign.(b) CECT showing an inflammatory pancreatic mass with interspersed normal pancreatic parenchyma (dot) in the head region.

Fig. 6
Fig. 6 (a) Contrast-enhanced computed tomography (CECT) in the arterial phase demonstrating a malignant mass encasing the splenic artery with wall irregularity (arrow).(b) CECT in the portal venous phase demonstrating a malignant mass infiltrating the splenic vein (dot) with collaterals.(c) CECT showing a malignant head mass infiltrating the superior mesenteric vessels.

Table 1
Comparison of demographic and qualitative CT characteristics of PDAC and MFCP in CCP Journal of Gastrointestinal and Abdominal Radiology ISGAR © 2023.The Author(s).

Table 2
Accuracy of different CECT parameters for predicting pancreatic ductal carcinoma in CCP

Table 5
Area under the ROC curves for predicting PDAC in CCP Abbreviations: CBD, common bile duct; CCP, chronic calcific pancreatitis; MPD, main pancreatic duct; PDAC, pancreatic ductal adenocarcinoma; ROC; receiver operating characteristic.a Statistically significant.