Role of Dynamic Perfusion CT in Pancreatic Adenocarcinoma

Abstract

Objective

The aim of this study is to evaluate the utility of perfusion CT (PCT) in the diagnosis of pancreatic ductal adenocarcinoma (PDAC).

Materials and Methods

In this ethically approved prospective study, PCT was performed in 71 patients with histologically proven PDAC. Perfusion parameters studied included blood flow (BF), blood volume (BV), permeability surface area product (PS), time to peak (TTP), peak enhancement intensity (PEI), and mean transit time (MTT). Forty-four controls with no pancreatic pathology were also studied.

Results

Out of 71 patients, 43 (60.56%) were males and 28 (39.44%) were females, with male:female ratio of 1.54, and the mean age of the patients was 50.62 ± 13.86 years. The mean size of the masses was 4.47 ± 2.43 cm (range: 1.3–12 cm). Among the perfusion parameters, BF and BV were found to be the most reliable for the diagnosis of pancreatic adenocarcinoma. They were reduced in PDAC (BF: 19.54 ± 19.85 mL/100 mL/min and BV: 5.14 ± 3.82 mL/100 mL) as compared with normal controls (BF: 96.91 ± 39.09 mL/100 mL/min and BV: 30.45 ± 12.90 mL/100 mL) and the difference was statistically significant (p = 0.0001). Based on ROC analysis, cut-off values of 55.25 mL/100 mL/min for BF and 14.42 mL/100 mL for BV yielded optimal sensitivity (97.2% for BF and 98.6% for BV) and specificity (91% for BF and 91% for BV) for the diagnosis of pancreatic adenocarcinoma.

Conclusion

PCT parameters are useful in making an imaging diagnosis of PDAC and useful tool to detect isodense pancreatic masses. Approximated values of BF and BV perfusion parameters may serve as independent diagnostic predictors for the diagnosis of PDAC.

 

MissingLinkText

Keywords

pancreas - perfusion - computed tomography - adenocarcinoma

Introduction

Pancreatic cancer is the third most common tumor of the gastrointestinal tract, has a poor prognosis, and is the third leading cause of cancer-related deaths worldwide.[1] [2] The incidence of pancreatic cancer has increased in Western countries over the last few decades.[1] [2] Pancreatic ductal adenocarcinoma (PDAC) is the most common primary malignancy of the pancreas, constituting ∼80% of malignant pancreatic tumors, and is most commonly observed in the head of the pancreas. It is an aggressive disease with an overall 5-year survival rate of less than 5%, and curative surgery is possible in only 20% of patients at the time of diagnosis.[1] [2] More than 95% of patients with PDAC present at an advanced stage at the time of diagnosis, either in the form of locally advanced disease (∼40%) or distant metastases (∼40–50%).[3] [4] Contrast-enhanced computed tomography (CECT) is the current standard imaging modality used for the diagnosis, characterization, and staging of PDAC. On CT, PDAC is seen as a hypodense lesion in all phases of the multiphase CT scan. Assessment of exact size, vascular invasion, lymph node involvement, and local and distant metastases is necessary for staging and planning surgery. CECT has limited accuracy in lesion detection and characterization, especially for isodense lesions, which constitute ∼11% of cases. Focal acute pancreatitis, paraduodenal pancreatitis, and mass-forming chronic pancreatitis may also mimic pancreatic cancer and are important differential diagnoses. Smaller tumors and well-differentiated pancreatic adenocarcinomas lacking characteristic CT imaging features may be missed via CECT.[5] [6] [7] [8]

There is a need for a better noninvasive imaging technique to characterize, detect, and improve the accuracy of diagnosing and grading pancreatic adenocarcinoma. Perfusion CT (PCT) may be helpful for identifying these masses with improved accuracy and can prevent unnecessary surgery.[9] [10]

PCT is an advancement of multidetector CT (MDCT) technology and has been widely used in oncologic imaging over the last few years.[10] [11] [12] [13] [14] [15] [16] PCT is a technique that provides quantitative and qualitative information regarding tissue hemodynamics noninvasively based on contrast kinetics in tissues after intravenous administration of a nonionic iodinated contrast agent. Moreover, perfusion parameters are significantly correlated with angiogenesis markers and constitute the basis for the use of perfusion parameters in neoplastic conditions for grading, progression, prognostication, predicting the response to therapy, and differentiating benign from malignant lesions, in addition to diagnosing and characterizing the tumor.[10] [11] [12] [13] [14] [15] [16] The PCT concentration has also been used to predict the development of pancreatic necrosis in the early stage of severe acute pancreatitis.[17] [18] [19] The purpose of this study was to assess the utility of PCT for diagnosing PDAC.

 

MissingLinkText

Materials and Methods

Patients

This prospective ethically approved study (IRC-PA-215/2022) was performed after providing informed consent from patients from June 2022 to August 2024. PCT was performed for patients with suspected pancreatic masses on previous imaging or endoscopic studies. A total of 110 patients with pancreatic masses underwent PCT, 75 of whom were diagnosed with PDAC by histopathological examination (the gold standard for diagnosis). After the histopathological diagnosis of PDAC, the tumors were classified as low-grade tumors (defined as well- or moderate-differentiated lesions with low cytological atypia) and high-grade tumors (poorly differentiated lesions with several cytological atypia). Pancreatic masses other than pancreatic adenocarcinoma were excluded from the study. Four patients were excluded from the study because of the suboptimal quality of PCT. Finally, a perfusion study of 71 patients aged >18 years with histologically proven PDAC before the start of treatment was performed; these patients were included in the study. Patients who did not provide consent for the study, patients aged less than 18 years, had a history of contrast allergy, patients with impaired renal function test results (creatinine level >1.4 mg/dL), patients who were already receiving chemotherapy, or those with features of obstructive acute pancreatitis were excluded from the study. Forty-four patients who underwent PCT for pathology unrelated to the pancreas were also included as the control group.

 

MissingLinkText

Perfusion CT Protocol

PCT was performed on a 128-slice dual-source MDCT scanner (Somatom Definition Flash, Siemens, Erlangen, Germany). After the scout image (120 kV and 35 mAs), a limited noncontrast scan (120 kVp, 210 mAs, slice thickness 0.625 mm, pitch 1.2, rotation time 0.5 seconds) of the upper abdomen was performed to locate the pancreas. Dynamic acquisition was then performed using a split protocol in free breathing for 95 seconds (first three scans every 3 seconds, next 5 scans every 2 seconds, next 5 scans every 4 seconds, and last 7 scans every 8 seconds) after the administration of 40 mL of nonionic iodinated contrast agent (350 mg iodine/dL) via a power injector into the antecubital vein at a rate of 5 mL/second, followed by a saline chase (20 mL at a rate of 5 mL/second). The perfusion scanning parameters were 100kVp, 100 mAs, a slice thickness of 0.625 mm, a 300 mm field of view, a z-axis coverage of 7 to 10 cm, a pitch of 0.6, and a rotation time of 0.28 seconds. After completion of the PCT scan, all the images were transferred to an external workstation (Multi-Modality Workplace, Siemens) for further analysis. The effective radiation dose for the CT perfusion study was 7.6 ± 1.34 mSv. Portal venous phase CT (120 kVp, 220 mAs, slice thickness 0.625 mm, FOV 300 mm, pitch 1.2, and rotation time 0.5 seconds) of the whole abdomen was obtained with a delay of 45 seconds after giving an additional 40 mL IV contrast at a rate of 3.5 mL/second after the perfusion study in all patients with PDAC for the purpose of staging.

 

MissingLinkText

Perfusion CT Postprocessing

Three radiologists with more than 10 years of experience, unaware of the final histopathological diagnosis of pancreatic masses, analyzed the perfusion images and color-coded the perfusion maps along with the grayscale images. PCT images were postprocessed to obtain color-coded maps of perfusion parameters, including BF (blood flow, mL/100 mL/min), BV (blood volume, mL/100 mL), PEI (peak enhancement intensity, HU), PS (permeability surface area product, mL/100 mL/min), MTT (mean transit time, second), and TTP (time to peak, second). Numerical values of the abovementioned parameters were also obtained in selected regions of interest using a dedicated workstation (Syngo Multimodality Workplace [MMWP], Siemens Healthcare, Germany) with the vendor-provided pancreas module of Volume Body Perfusion CT (VPCT Body) software with automatic 4D motion and noise correction, followed by automatic bone removal by segmentation. A reference arterial input curve was obtained by placing a region of interest (ROI) in the aorta, followed by blood vessel segmentation and calculation of perfusion maps. Parametric perfusion maps were generated using the double-compartment Patlak model for analysis.

In the control group, oval-shaped ROIs (area 10–20 mm²) were manually drawn on MIP images in the region of the head, body, and tail of the pancreas; these ROIs were automatically applied to other parametric perfusion maps. In the case of PDAC, the ROI was drawn in the center and periphery of the mass lesion, avoiding vessels and ducts. An ROI was also drawn in the normal, spared region of the pancreas of patients with PDAC. The perfusion parameters of the PDAC patients and the control patients were compared.

 

MissingLinkText

Statistical Analysis

The data were analyzed using SPSS version 20.0 software, and the results are expressed as the mean ± standard deviation for continuous data and the median for noncontinuous data. For categorical variables, differences were compared between groups using the chi-square test and Fisher's exact test. Differences in continuous data within the groups were analyzed by paired t-tests, while differences between the groups were analyzed by independent t-tests, one-way ANOVA, and Mann–Whitney tests. p-Values less than 0.05 were considered to indicate statistical significance. Receiver operating characteristic (ROC) curves were generated to calculate cutoff values for quantifiable perfusion parameters between PDAC and normal pancreas tissues.

 

MissingLinkText

 

MissingLinkText

Results

PCT was performed on 44 control patients to study normal pancreatic perfusion. Of the 44 control patients, 30 (68.18%) were males and 14 (31.82%) were females, with male:female ratio of 2.14. The mean age of the individuals in the control group was 45.16 ± 12.52 years, and the age ranged from 20 to 75 years. The pancreatic BF, BV, PEI, PS, MTT, and TTP did not significantly differ among the different parts of the pancreas (p > 0.05). The range of normal pancreatic BF was 84.74 to 113.21 mL/100 mL/minute, BV was 26.44 to 34.93 mL/100 mL, PEI was 72.84 to 152.82 HU, PS was 13.87 to 35.13 mL/100 mL/minute, MTT was 10.24 to 16.82 seconds, and TTP was 12.98 to 16.39 seconds ([Fig. 1]).

Fig. 1 Axial (A) CECT image in the portal venous phase of a 57-year-old female showing a homogenously enhancing normal healthy pancreas with early and high peaks on the time density curve with a strong wash-in followed by early washout (B). On perfusion CT color-coded maps, the normal pancreas shows homogenous blood flow (C) and blood volume (D), which are greater than those of the liver. On the color scale, red represents the highest perfusion parameter, and purple represents the lowest.

Among the 71 patients with PDAC, 43 (60.56%) were males and 28 (39.44%) were females with male:female ratio of 1.54. The mean age of the patients was 50.62 ± 13.86 years, and the age ranged from 32 to 95 years. There was no significant difference in age or sex between the normal control group and the PDAC patient group (p > 0.05).

Forty-three masses (60.56%) were located in the head and uncinate process, 10 (14.08%) in the body, 3 (4.23%) in the tail, and 15 (21.13%) in the body and tail of the pancreas. The mean size of the pancreatic adenocarcinoma masses was 4.47 ± 2.43 cm (range: 1.3–12 cm). About 4 (5.6%) out of 71 patients with pancreatic adenocarcinoma had imaging features suggestive of chronic pancreatitis. Approximately 76.06% of the patients (54) had distant metastasis or locally advanced disease at the time of the study. Sixty-one patients (85.92%) had a hypodense mass visible on grayscale PCT images, while in 10 patients (14.08%), the mass could not be detected on grayscale PCT images. However, all 71 adenocarcinoma masses were detected on the color-coded PCT maps ([Figs. 2]–[5]). The characteristics of all ten isodense pancreatic masses are shown in [Table 1].

Fig. 2 Noncontrast axial (A) image of a 62-year-old male patient with pancreatic adenocarcinoma showing bulky head of pancreas appearing an isodense mass in the pancreatic head (thick white arrow) on contrast-enhanced pancreatic phase axial (B) and contrast-enhanced venous phase coronal (C) images causing abrupt cessation of CBD (blue star) and MPD (thin blue arrow) with upstream dilatation. On the perfusion CT color-coded map, the mass showed a reduced BF (D) and BV (E) (thick white arrow) compared with those of the normal spared pancreas. The mass is difficult to visualize on CECT images but is easily observed on perfusion CT images as a perfusion defect area.

Fig. 3 Arterial phase (A), pancreatic phase (B), venous phase (C), axial and venous phase (D) coronal images of a 57-year-old female patient with pancreatic adenocarcinoma showing an isodense mass in the pancreatic head (thick white arrow) in all phases causing abrupt cessation of CBD (blue star) and MPD (thin blue arrow) with upstream dilatation in image (D). On the perfusion CT color-coded map, the mass showed a reduced blood flow (E) and blood volume (F) (thick white arrow) compared with those of the normal spared pancreas. The mass is difficult to visualize on multiphase CT images but is easily observed on perfusion CT images as a perfusion defect area.

Fig. 4 Axial CECT images in the portal venous phase (A) of a 65-year-old female patient with pancreatic adenocarcinoma showing a large hypodense mass in the pancreatic body and tail with mild peritumoral desmoplastic reaction (thin white arrow) and an abrupt decrease in the MPD with proximal dilatation. On perfusion CT color-coded maps, the mass shows a reduced blood flow (B), blood volume (C), and permeability surface area product (D) as shown by the thick white arrow compared with the surrounding normal pancreas.

Fig. 5 Axial CECT images in the arterial phase (A) and venous phase (B) of a 52-year-old female patient with small pancreatic adenocarcinoma showing a small, very aggressive hypodense mass in the tail of the pancreas (thick white arrow) encasing splenic vessels with perisplenic and perigastric collaterals. The mass is showing marked perilesional desmoplastic reaction with infiltration of the left adrenal gland and an ill-defined right lobe hepatic metastasis. On perfusion CT color-coded maps, the mass shows a reduced blood flow (C) and blood volume (D) as shown by the thick blue arrow compared with the surrounding normal pancreas.

Table 1 Characteristics of pancreatic isodense masses observed on perfusion color-coded images

Number	Age (years)/Sex	Location in the pancreas	Size	Pancreatic phase CECT image (isodense lesion)	Color-coded perfusion maps (seen as a perfusion defect)
1.	65/M	Head	3.2 cm	Not seen	Seen
2.	62/M	Head	2.8 cm	Not seen	Seen
3.	32/M	Head	1.3 cm	Not seen	Seen
4.	55/F	Uncinate process	3.4 cm	Not seen	Seen
5.	57/F	Head	3.7 cm	Not seen	Seen
6.	38/F	Head	2.6 cm	Not seen	Seen
7.	67/M	Head	2.5 cm	Not seen	Seen
8.	49/F	Head and uncinate process	4.5 cm	Not seen	Seen
9.	36/M	Head	3.7 cm	Not seen	Seen
10.	44/M	Head	2.9 cm	Not seen	Seen

PEI, BF, BV, PS, MTT, and TTP significantly differ between pancreatic adenocarcinoma and normal pancreas tissues (p < 0.0001; [Table 2]). Significant differences in all perfusion parameters were noted between tumor tissue and normal spared pancreatic tissue in the pancreatic adenocarcinoma group (p < 0.0001; [Table 2]). As we moved from the center to the periphery of the PDAC lesion, perfusion parameters increased from the center to the periphery of the lesion. The perfusion parameters also significantly differed between the peripheral region of the lesion and the normal pancreas (p < 0.01). The perfusion parameters of the normal spared pancreatic tissue were similar to those of the normal control tissue (p > 0.05), as shown in [Table 2].

Table 2 Comparison of perfusion parameters between pancreatic adenocarcinoma patients and normal controls

Parameters	Normal controls (Group A)	Pancreatic adenocarcinoma (Group B)	Normal spared pancreas of the pancreatic adenocarcinoma group (Group C)	p-Value (A vs. B)	p-Value (B vs. C)	p-Value (A vs. C)
PEI (HU)	116.33 ± 17.79	74.26 ± 18.78	119.03 ± 19.97	0.0001	0.0001	0.15
BF (mL/100 mL/min)	96.91 ± 39.09	19.54 ± 19.85	95.39 ± 28.43	0.0001	0.0001	0.92
BV (mL/100 mL)	30.45 ± 12.90	5.14 ± 3.82	33.07 ± 9.87	0.0001	0.0001	0.28
PS (mL/100 mL/min)	23.16 ± 31.02	10.89 ± 6.88	22.49 ± 22.82	0.0001	0.0001	0.84
MTT (s)	12.88 ± 6.11	22.73 ± 10.14	12.39 ± 6.70	0.0001	0.0001	0.56
TTP (s)	14.11 ± 3.60	23.82 ± 6.99	15.36 ± 3.71	0.0001	0.0001	0.26

Abbreviations: MTT, mean transit time; TTP, time to peak.

Notes: The results are expressed as the mean ± standard deviation. p-Value <0.05 was considered to indicate statistical significance.

In patients with pancreatic adenocarcinoma, the BF was 79.83% lower, the BV was 83.12% lower, the PEI was 35.55% lower, and the PS was 52.98% lower than those in healthy controls (p < 0.0001), while the MTT was 43.33% greater and the TTP was 40.76% greater in adenocarcinoma patients than in normal controls (p < 0.0001). The normal pancreatic parenchyma showed 4.96 times higher BF, 5.92 times higher BV, 1.55 times higher PEI, and 2.13 times higher PS than pancreatic adenocarcinoma (p < 0.0001). The pancreatic adenocarcinoma showed 1.76 times higher MTT and 1.69 times higher TTP than the normal pancreas (p < 0.0001).

Out of the 71 pancreatic adenocarcinoma masses, 32 (45.07%) masses showed features of low-grade malignancy, while 39 (54.93%) masses showed features of high-grade malignancy on pathological examinations. High-grade pancreatic adenocarcinoma showed significantly lower values of BF, BV, and PEI in comparison to low-grade pancreatic adenocarcinoma (p < 0.01). PS was lower in high-grade pancreatic adenocarcinoma in comparison to low-grade pancreatic adenocarcinoma (p = 0.8). MTT and TTP were higher in high-grade pancreatic adenocarcinoma in comparison to low-grade pancreatic adenocarcinoma (p > 0.05).

Comparison of Isodense PDAC and Hypodense PDAC

Out of 71 adenocarcinoma patients, we have separately analyzed the characteristics of isodense PDAC (10) and hypodense PDAC (61) masses. Among 10 isodense masses, 6 (60%) were male and 4 (40%) were female with mean age of 50.50 ± 7.88 years, whereas in hypodense masses, 37 (60.66%) were male and 24 (39.34%) were female with mean age of 50.63 ± 14.58 years (p = 0.423 for age and 0.682 for sex). The mean size of isodense masses was 3.06 ± 1.63 cm, while the mean size of hypodense masses was 4.70 ± 2.63 cm (p = 0.442). Among the perfusion parameters, PEI, BF, and BV were significantly higher in isodense masses in comparison to hypodense masses (p < 0.0001). PS was higher in isodense mass than in the hypodense mass (p = 0.779). TTP and MTT were lower in isodense mass in comparison to the hypodense mass (p = 0.219 for TTP and 0.906 for MTT). Both the isodense and hypodense masses showed reduced BF, BV, and PS than the normal controls, while higher TTP and MTT were observed in both masses in comparison to normal controls as shown in [Table 3]. There is no significant difference in PEI between the isodense mass and normal control (p = 0.998). In isodense masses, the BF was 32.22% higher, the BV was 25.13% higher, the PEI was 63.65% higher, the PS was 11.28% higher, the MTT was 1.77% lower, and the TTP was 11.19% lower than the hypodense masses.

Table 3 Comparison of perfusion parameters between isodense mass, hypodense mass, and normal control

Parameters	Isodense mass (Group A)	Hypodense mass (Group B)	Normal controls (Group C)	p-Value (A vs. B)	p-Value (A vs. C)	p-Value (B vs. C)
PEI (HU)	115.96 ± 16.03	70.86 ± 17.29	116.33 ± 17.79	0.0001	0.998	0.0001
BF (mL/100 mL/min)	47.95 ± 21.16	14.88 ± 9.66	96.91 ± 39.09	0.0001	0.001	0.0001
BV (mL/100 mL)	10.63 ± 2.38	4.23 ± 3.21	30.45 ± 12.90	0.0001	0.001	0.0001
PS (mL/100 mL/min)	12.54 ± 4.31	11.12 ± 6.15	23.16 ± 31.02	0.779	0.001	0.0001
MTT (s)	22.67 ± 10.36	23.08 ± 6.35	12.88 ± 6.11	0.219	0.001	0.0001
TTP (s)	23.40 ± 7.23	26.35 ± 4.84	14.11 ± 3.60	0.906	0.001	0.0001

Notes: The results are expressed as the mean ± standard deviation. p-Value <0.05 was considered to indicate statistical significance.

Based on the ROC curve ([Fig. 6A, B]), the cutoff values of BF, BV, PEI, PS, MTT, and TTP for differentiating PDAC from normal pancreas tissue were chosen, and the area under the curve (AUC) values for BF, BV, PEI, PS, MTT, and TTP were 0.976, 0.992, 0.943, 0.715, 0.763, and 0.849, respectively. A cutoff value of 55.25 mL/100 mL/minute for BF had 97.2% sensitivity and 91% specificity for differentiating pancreatic adenocarcinoma from normal tissue, while for BV, a cutoff value of 14.42 mL/100 mL had 98.6% sensitivity and 91% specificity. A cutoff value of 38.27 mL/100 mL/minute for PS (98.6% sensitivity and 81.8% specificity), 96.89 HU for PEI (90.1% sensitivity and 91% specificity), 3.76 seconds for MTT (98.6% sensitivity and 97.7% specificity), and 8.04 seconds for TTP (98.6% sensitivity and 97.7% specificity) could be used to differentiate pancreatic adenocarcinoma from normal pancreas tissue.

Fig. 6 (A) Receiver operating characteristic (ROC) curve for perfusion parameters showing excellent area under the curve (AUC) for blood flow (0.976), blood volume (0.992), peak enhancement intensity (0.943), and good area under the curve (AUC) for permeability surface area product (0.715). (B) ROC curves for MTT and TTP reveal AUC of 0.763 for MTT and 0.849 for TTP (good). MTT, mean transit time; TTP, time to peak.

Based on the ROC curve, the cutoff values of BF, BV, PEI, PS, MTT, and TTP for differentiating isodense PDAC from hypodense PDAC were chosen, and the AUC values for BF, BV, PEI, PS, MTT, and TTP were 0.972 (excellent), 0.932 (excellent), 0.867(good), 0.470, 0.482, and 0.370, respectively. A cutoff value of 27.13 mL/100 ml/min for BF (90% sensitivity and 98% specificity), a cutoff value of 9.7 mL/100 mL for BV (70% sensitivity and 98% specificity), a cutoff value of 1.2 mL/100 mL/minute for PS (90% sensitivity and 98.4% specificity), 95.79 HU for PEI (90% sensitivity and 90.7% specificity), 34.56 seconds for MTT (90% sensitivity and 90.2% specificity), and 33.13 seconds for TTP (90% sensitivity and 88.5% specificity) could be used to differentiate isodense pancreatic adenocarcinoma from hypodense pancreatic adenocarcinoma.

 

MissingLinkText

 

MissingLinkText

Discussion

PCT is a novel and emerging noninvasive technique that provides information regarding tumor microvascularity in vivo and has been used mainly in oncoimaging over the last few decades.[20] Following the administration of intravenous contrast agent and repetitive scanning of the field of interest, information about the perfusion parameters of the tumor microvasculature can be obtained. PCT can provide valuable information about tumor characterization, local infiltration, nodal involvement, vascular invasion, good delineation of tumor margin along with desmoplastic reaction and soft tissue cuffing, prognosis, extent of angiogenesis, prediction of response to therapy, and monitoring of treatment response (chemotherapy or chemoradiotherapy). The two widely used analytic methods for postprocessing perfusion studies are compartmental analysis (single or double compartment) and deconvolutional analysis for measuring tissue perfusion from PCT data. Double compartment analysis assumes that both the extravascular and intravascular space are separate compartments and is used for the measurement of capillary permeability due to the extravasation of contrast agent from the intravascular compartment into the extravascular compartment across the capillary basement membrane during the second phase (after 40 seconds).[9] PCT is also used in acute pancreatitis patients to detect perfusion defects for the prediction of pancreatic necrosis.[17] [18] [19]

PCT was performed on 44 normal controls in our study via pathology unrelated to the pancreas to study pancreatic perfusion and obtain normal perfusion parameter values. Moreover, perfusion parameters (PEI, BF, BV, PS, MTT, and TTP) did not significantly differ among the different regions of the pancreas (head, body, and tail), which is in accordance with the findings of other studies.[9] [11] [21] [22] [23] [24] [25] [26] Significantly greater values of BF in the head were observed than in the body and tail of the pancreas, while the other perfusion parameters did not significantly differ according to the methods used in the study by Klauss et al.[27] [28] However, the absolute perfusion values in our study differed from those in other published studies because perfusion parameters vary with the scanner, mathematical model, and protocol used.

To improve the survival rate of patients with PDAC, early and timely detection of the lesion with diagnosis of pancreatic adenocarcinoma followed by curative surgery is an important determining factor.[28] More than 95% of the patients with pancreatic adenocarcinoma presented at an advanced stage, either in the form of locally advanced disease (vascular or perineural invasion) or distant metastasis. Prognosis and therapeutic approaches depend on the resectability of the lesion at the clinical presentation. About 30% of the resected patients die of the disease within 1 year of surgery, and recurrence is early in these patients. The early deaths in these patients are due to either incomplete pre- or intraoperative staging or aggressive behavior of the lesion. Thus, there is a need to identify such patients at risk of early deaths before curative surgery.[3] [4] MDCT is the imaging modality of choice for both diagnosing and staging pancreatic tumors and determining tumor resectability. Compared with normal pancreatic parenchyma, pancreatic adenocarcinomas are hypodense lesions on CECT in most cases[29]; however, ∼11% of pancreatic cancers are isodense in comparison to the normal surrounding pancreatic parenchyma.[5] Therefore, there is a need for another imaging technique that allows better detection of these lesions, and PCT may be used for this purpose. PCT of the pancreas was performed while the patients were breathing freely, and the patients were instructed to perform regular shallow breathing. With advancements in CT technology, whole-organ pancreatic perfusion or perfusion studies of whole lesions involving the pancreas can be performed with a craniocaudal anatomical coverage of 16 cm via a 320-slice CT scanner.[22] [30] PCT evaluates lesions over a period of time (various phases of contrast distribution) and can help in the differentiation of various pancreatic masses by exploiting the differences in perfusion characteristics, whereas conventional CECT is a snapshot technique that merely reflects the tissue density in a particular phase (in HU).

In our study, out of 71 patients with pancreatic adenocarcinoma, 10 (14.08%) had an isodense mass that remained isodense to the pancreatic parenchyma in all phases of the dynamic study and was not observed on conventional CECT images. However, all these tumors were seen as areas of perfusion defects on the color-coded perfusion maps. The remaining 61 (85.92%) patients had hypodense tumors in all phases of the dynamic study and on conventional CECT images. They were seen as perfusion defects on color-coded perfusion maps. The lesion that can be missed on conventional CECT can also be seen on the perfusion maps, reducing the interobserver variation. All the pancreatic adenocarcinoma masses had significantly lower PEI, BF, BV, and PS values than in the normal controls, while the MTT and TTP values were significantly greater in the pancreatic adenocarcinoma group than in the normal control group. The lower BF, BV, and PEI values in adenocarcinomas than in normal pancreases are likely due to the lower microvascular density of tumor tissue and the resulting hypodensity on conventional CECT images. Additionally, the associated fibrosis and necrosis can lead to lower BF, BV, and PEI values in adenocarcinomas than in normal pancreases. The perfusion of the tissue is directly determined by the tissue microvascular density. A tissue with higher microvascular density will exhibit greater perfusion characteristics, while a tissue with lower microvascular density will exhibit lower perfusion characteristics.[31] This is the reason for the increase in perfusion values from the center of the adenocarcinoma to the periphery of the lesion. As widespread endothelitis is caused by tumor infiltration of blood vessels, the PS is also decreased in adenocarcinomas compared with that in normal pancreases. Similar results (significantly decreased PEI, BF, BV, and PS) were also reported by many studies[9] [11] [21] [22] [27] [28] [32]; however, due to the use of different scanners, mathematical models, and protocols, the actual perfusion values among different studies cannot be compared. However, other studies have shown equivocal results.[17] [33] [34]

In the literature, there is no standard consensus regarding the use of PSs in patients with adenocarcinoma. An increase in the PS was also observed in some studies.[23] [24] [25] [34] Increased permeability is expected in pancreatic adenocarcinoma due to the decreased integrity of the tumor vascular endothelium and inflammatory vascular dilatation observed via histological studies. The value of PS is also directly affected by the BF and BV. Decreased BF and BV values were observed in pancreatic adenocarcinoma patients, which in turn decreased the PS concentration in tumor tissue.[21] [35] Out of all perfusion parameters in our study, BF and BV are the most important parameters for pancreatic disease, while BF and PS are important perfusion parameters, as shown by Purdie et al and this may indicate malignancy and prognosis of the disease.[36] [37] [38]

The time density curve revealed a lower peak of enhancement and longer time to peak in pancreatic adenocarcinoma patients than in normal controls. These changes were attributed to the lower BF and BV values in pancreatic adenocarcinoma tissue than in normal healthy pancreatic tissue from controls, which also explains the decreased enhancement of tumor tissue on CECT.

Based on the ROC curve, the cutoff values of the BF, BV, PS, PEI, MTT, and TTP perfusion parameters that had the best performance in terms of sensitivity and specificity for differentiating pancreatic adenocarcinoma patients from normal controls were 55.25 mL/100 mL/minute, 14.42 mL/100 mL, 38.27 mL/100 mL/minute, 96.89 HU, 3.76 seconds, and 8.04 seconds, respectively. The areas under the curve for BF, BV, PS, PEI, MTT, and TTP were 0.976, 0.992, 0.715, 0.943, 0.763, and 0.849, respectively. Among the six perfusion parameters, BF and BV performed the best in differentiating pancreatic adenocarcinoma patients from normal controls (p < 0.01). A study performed by Zaborienė et al[32] also showed that BF and BV perfusion parameters were the most reliable parameters for the diagnosis of pancreatic adenocarcinoma, similar to the results of our study. If these perfusion parameter values can be further validated and found to be reproducible in larger studies, they could become useful additional paradigms for the diagnosis of pancreatic adenocarcinoma.

PCT can help in direct and preoperative characterization of high-grade PDAC, leading to identification of patients with high risk of early deaths and can provide each patient with the best treatment strategy. In our study, perfusion parameters like PEI, BF, and BV were statistically higher in high-grade adenocarcinoma in comparison to low-grade carcinoma, and these perfusion parameters were used to differentiate high-grade tumors from the low-grade tumors. Like our study, a study done by D'Onofrio et al[14] also showed perfusion parameters like PEI and BV were used to differentiate between high-grade tumors from low-grade tumors. Considering the results of our study, PCT can be helpful in the clinical and laboratory data in identifying the high-grade malignancy along with duration of disease symptoms and CA 19–9 level. These findings ultimately lead to the identification of patients with early deaths.

Many studies have shown that PCT was helpful in differentiating mass-forming chronic pancreatitis from pancreatic adenocarcinoma. Both these lesions showed reduced perfusion parameters in comparison to the normal pancreas, but there was a greater extent of reduction of perfusion parameters in pancreatic adenocarcinoma than in mass-forming chronic pancreatitis. PCT can be helpful in differentiating different pancreatic lesions out of which cystic pancreatic tumor and adenocarcinoma showed least perfusion than normal pancreas, mass-forming chronic pancreatitis showed higher value of perfusion parameters than adenocarcinoma but less than normal pancreas, and inflammatory pancreatic lesions showed higher perfusion than pancreatic carcinoma, cystic pancreatic tumors, and mass-forming chronic pancreatitis but less than normal pancreas with higher value of perfusion parameters seen in neuroendocrine tumor among all pancreatic masses even higher than normal pancreas.[9] [11] [15]

The perfusion values cannot be directly compared with published values because they depend on the type of scanner, the mathematical model, and the software used. Therefore, the main limitations of the study are related to the perfusion methods, high radiation dose, and the need for an intravenous contrast agent. The high radiation dose of PCT in the future can be reduced with the use of iterative reconstruction, and further potential dose reduction of more than 30% can be achieved with the use of new detector technologies. Another limitation of our study is that it was a single-center study, and the number of patients was small. The results reported in the study need to be tested in a larger cohort of patients and in multicenter studies.

 

MissingLinkText

Conclusion

In the present study, we showed that PCT is a promising newer tool providing perfusion parameters of the pancreatic cancer based on contrast kinetics of tissue after contrast administration using double-compartment Patlak model for analysis. The basic CT perfusion parameters provide information regarding the diagnosis and grading of the pancreatic cancer. So, perfusion parameters (particularly BF and BV) can be an important additional parameter for diagnosing pancreatic adenocarcinoma with good accuracy and are useful tools for detecting isodense pancreatic masses.

 

MissingLinkText

 

MissingLinkText

Conflict of Interest

None declared.

Data Availability Statement

Stored in the institute and can be reproduced when needed by the corresponding author's request.

Authors' Contributions

A.K.Y.: Written main manuscript, data collection, data interpretation, manuscript design and concept, prepared the figures.

N.Y.: Statistical analysis, manuscript editing, and research design.

R.P.Y.: Data Editing and manuscript correction.

S.T.: Data collection, data interpretation, manuscript design and concept, prepared the figures.

B.D.: Data collection, data interpretation, manuscript design and concept, prepared the figures.

N.P.: Data analysis and manuscript editing.

All authors reviewed the manuscript.

All authors contributed to the study from the start of the study till writing the manuscript.

Ethical Approval

This study was performed after approval from the Nepal Health Research Council (NHRC), Institutional Research Committee of Birat Medical College Teaching Hospital, Biratnagar, Morang, Nepal. The reference number was IRC-PA-215/2022. The study was performed after the ethics declaration, following standard norms and in accordance with the Declaration of Helsinki.

Patients' Consent

All patients had given consent to participate in the study.

• References

• 1 Ueno H, Kosuge T. Adjuvant treatments for resectable pancreatic cancer. J Hepatobiliary Pancreat Surg 2008; 15 (05) 468-472
• 2 Loos M, Kleeff J, Friess H, Büchler MW. Surgical treatment of pancreatic cancer. Ann N Y Acad Sci 2008; 1138: 169-180
• 3 Czito BG, Willett CG, Clark JW, Fernandez Del Castillo C. Current perspectives on locally advanced pancreatic cancer. Oncology (Williston Park) 2000; 14 (11) 1535-1545 , discussion 1546, 1549–1552
• 4 Willett CG, Czito BG, Bendell JC, Ryan DP. Locally advanced pancreatic cancer. J Clin Oncol 2005; 23 (20) 4538-4544
• 5 Prokesch RW, Chow LC, Beaulieu CF, Bammer R, Jeffrey Jr RB. Isoattenuating pancreatic adenocarcinoma at multi-detector row CT: secondary signs. Radiology 2002; 224 (03) 764-768
• 6 Zakharova OP, Karmazanovsky GG, Egorov VI. Pancreatic adenocarcinoma: Outstanding problems. World J Gastrointest Surg 2012; 4 (05) 104-113
• 7 Bond-Smith G, Banga N, Hammond TM, Imber CJ. Pancreatic adenocarcinoma. BMJ 2012; 344 (344) e2476
• 8 Schneeweiß S, Horger M, Grözinger A. et al. CT-perfusion measurements in pancreatic carcinoma with different kinetic models: Is there a chance for tumour grading based on functional parameters?. Cancer Imaging 2016; 16 (01) 43
• 9 Delrue L, Blanckaert P, Mertens D, Van Meerbeeck S, Ceelen W, Duyck P. Tissue perfusion in pathologies of the pancreas: assessment using 128-slice computed tomography. Abdom Imaging 2012; 37 (04) 595-601
• 10 Hartwig W, Schneider L, Diener MK, Bergmann F, Büchler MW, Werner J. Preoperative tissue diagnosis for tumours of the pancreas. Br J Surg 2009; 96 (01) 5-20
• 11 Yadav AK, Sharma R, Kandasamy D. et al. Perfusion CT - Can it resolve the pancreatic carcinoma versus mass forming chronic pancreatitis conundrum?. Pancreatology 2016; 16 (06) 979-987
• 12 Miles KA, Charnsangavej C, Lee FT, Fishman EK, Horton K, Lee TY. Application of CT in the investigation of angiogenesis in oncology. Acad Radiol 2000; 7 (10) 840-850
• 13 Mazzei FG, Mazzei MA, Cioffi Squitieri N. et al. CT perfusion in the characterization of renal lesions: an added value to multiphasic CT. BioMed Res Int 2014; 2014: 10
• 14 D'Onofrio M, Gallotti A, Mantovani W. et al. Perfusion CT can predict tumoral grading of pancreatic adenocarcinoma. Eur J Radiol 2013; 82 (02) 227-233
• 15 Yadav AK, Nadarajah J, Kandasamy D, Sharma R, Gamanagatti S, Garg PK. Volume perfusion computed tomography as an aid to diagnosis of pancreatic tuberculosis. Trop Gastroenterol 2014; 35 (04) 264-266
• 16 Kambadakone AR, Sahani DV. Body perfusion CT: technique, clinical applications, and advances. Radiol Clin North Am 2009; 47 (01) 161-178
• 17 Tsuji Y, Yamamoto H, Yazumi S. et al. Perfusion computerized tomography can predict pancreatic necrosis in early stages of severe acute pancreatitis. Clin Gastroenterol Hepatol 2007; 5 (12) 1484-1492
• 18 Tsuji Y, Hamaguchi K, Watanabe Y. et al. Perfusion CT is superior to angiography in predicting pancreatic necrosis in patients with severe acute pancreatitis. J Gastroenterol 2010; 45 (11) 1155-1162
• 19 Yadav AK, Sharma R, Kandasamy D. et al. Perfusion CT: can it predict the development of pancreatic necrosis in early stage of severe acute pancreatitis?. Abdom Imaging 2015; 40 (03) 488-499
• 20 Miura F, Takada T, Amano H, Yoshida M, Furui S, Takeshita K. Diagnosis of pancreatic cancer. HPB (Oxford) 2006; 8 (05) 337-342
• 21 Xu J, Liang Z, Hao S. et al. Pancreatic adenocarcinoma: dynamic 64-slice helical CT with perfusion imaging. Abdom Imaging 2009; 34 (06) 759-766
• 22 Kandel S, Kloeters C, Meyer H, Hein P, Hilbig A, Rogalla P. Whole-organ perfusion of the pancreas using dynamic volume CT in patients with primary pancreas carcinoma: acquisition technique, post-processing and initial results. Eur Radiol 2009; 19 (11) 2641-2646
• 23 Lu N, Guo QY, Ren Y, Bi CL, Chen LY. The value of 64 slices CT perfusion in diagnosis of pancreatic cancer. J China Clin Med Imaging 2007; 18: 554-558
• 24 Ye R, Guo SL, Zhou HQ. et al. Study of 64 slices CT perfusion imaging in pancreatic cancer. Chin J Med Imaging Technol 2007; 23: 1362-1365
• 25 Zhao XM, Zhou CW, Wu N. et al. Study of multiple slices pancreatic CT perfusion. Zhonghua Fang She Xue Za Zhi 2003; 37: 845-849
• 26 Liang ZH, Feng XY, Zhu RJ, Jin C. Study of multiple slices CT perfusion imaging in normal pancreas. Shengwu Yixue Gongcheng Yu Linchuang 2006; 10: 218-222
• 27 Klauss M, Stiller W, Fritz F. et al. Computed tomography perfusion analysis of pancreatic carcinoma. J Comput Assist Tomogr 2012; 36 (02) 237-242
• 28 Klauss M, Stiller W, Pahn G. et al. Dual-energy perfusion-CT of pancreatic adenocarcinoma. Eur J Radiol 2013; 82 (02) 208-214
• 29 Prokesch RW, Schima W, Chow LC, Jeffrey RB. Multidetector CT of pancreatic adenocarcinoma: diagnostic advances and therapeutic relevance. Eur Radiol 2003; 13 (09) 2147-2154
• 30 Wittkamp G, Buerke B, Dziewas R. et al. Whole brain perfused blood volume CT: visualization of infarcted tissue compared to quantitative perfusion CT. Acad Radiol 2010; 17 (04) 427-432
• 31 Bluemke DA, Cameron JL, Hruban RH. et al. Potentially resectable pancreatic adenocarcinoma: spiral CT assessment with surgical and pathologic correlation. Radiology 1995; 197 (02) 381-385
• 32 Zaborienė I, Barauskas G, Gulbinas A, Ignatavičius P, Lukoševičius S, Žvinienė K. Dynamic perfusion CT - a promising tool to diagnose pancreatic ductal adenocarcinoma. Open Med (Wars) 2021; 16 (01) 284-292
• 33 Eastwood JD, Lev MH, Provenzale JM. Perfusion CT with iodinated contrast material. AJR Am J Roentgenol 2003; 180 (01) 3-12
• 34 Cáceres AV, Zwingenberger AL, Hardam E, Lucena JM, Schwarz T. Helical computed tomographic angiography of the normal canine pancreas. Vet Radiol Ultrasound 2006; 47 (03) 270-278
• 35 Lu N, Feng XY, Hao SJ. et al. 64-slice CT perfusion imaging of pancreatic adenocarcinoma and mass-forming chronic pancreatitis. Acad Radiol 2011; 18 (01) 81-88
• 36 Mirecka J, Libura J, Libura M. et al. Morphological parameters of the angiogenic response in pancreatic ductal adenocarcinoma – correlation with histological grading and clinical data. Folia Histochem Cytobiol 2001; 39 (04) 335-340
• 37 Purdie TG, Henderson E, Lee TY. Functional CT imaging of angiogenesis in rabbit VX2 soft-tissue tumour. Phys Med Biol 2001; 46 (12) 3161-3175
• 38 Linder S, Blåsjö M, von Rosen A, Parrado C, Falkmer UG, Falkmer S. Pattern of distribution and prognostic value of angiogenesis in pancreatic duct carcinoma: a semiquantitative immunohistochemical study of 45 patients. Pancreas 2001; 22 (03) 240-247

Publication History

Article published online:
12 August 2025