Pancreatic parenchymal atrophy, abbreviated PPA, is defined on computed tomography as a narrowing of the pancreatic tissue below a line connecting the cephalic and caudal margins of the lesion. It is not a single disease but the end point of many different insults. Free radicals generated by alcohol damage acinar cells, the workhorses of the pancreas, triggering inflammation and fibrosis; alcohol and its metabolites also directly injure these cells, promoting lipid accumulation and cell loss. Alcohol and other substances raise protein concentrations in pancreatic juice, forming plugs that obstruct the ducts. Acute pancreatitis causes acinar cell necrosis, which heals as scar tissue. In animal models, both obstruction of pancreatic duct flow and ischemia of the parenchyma produced atrophy, and impaired blood flow drove the gland to replace itself with fat. Precancerous lesions called pancreatic intraepithelial neoplasia, or PanIN, can block the flow of juice and set off the same cascade of fibrosis and shrinkage.

The review classifies PPA into three morphological types with distinct clinical meanings. Focal PPA, or FPPA, appears as localized or segmental wasting of tissue adjacent to a duct narrowing, or sometimes without any visible narrowing at all. Upstream PPA, or UPPA, describes global atrophy of the tail-side pancreas downstream of a main duct stricture. Diffuse PPA, or DPPA, involves moderate to severe replacement of the entire gland by fat and is linked to metabolic factors and chronic inflammation rather than to early malignancy. The authors excluded DPPA from their surgical analysis and focused on the focal and upstream forms, because these are the patterns most tightly associated with early pancreatic cancer. FPPA is frequently confused with the generalized atrophy that follows pancreatitis or with pancreatic steatosis, and clinicians must integrate multimodal imaging, changes over time, and the patient’s medical history to tell them apart.

Under the microscope, PPA is a dynamic imbalance between the regeneration and loss of pancreatic epithelial cells. Acinar cells make up more than 90 percent of the pancreatic parenchyma, and their loss is the defining feature. The vacant space is filled by adipocytes and fibroblasts, which is why the authors describe atrophy and fat replacement as two sides of the same coin. Pathologists have traditionally defined pancreatic fat replacement as replacement of more than 25 percent of the parenchyma by adipocytes, though recent reports have proposed a lower threshold of 10 percent. Even in severe cases, the lobular architecture of the gland is generally preserved, and in extreme fat replacement only the ducts and islets remain embedded in adipose tissue. The literature offers a confusing array of names for this process, including pancreatic lipomatosis, fatty infiltration, lipomatous pseudohypertrophy, and pancreatic steatosis, and no consensus definition yet exists.

The epidemiology is striking. Pancreatic fat replacement affects an estimated 16 percent of the global population, most of whom never develop symptoms. The average fat content of the pancreas in people in their eighties has been reported at 35 percent, showing that the process advances with age. Diffuse fat replacement is associated with acute and chronic pancreatitis, pancreatic tumors, diabetes, and arteriosclerosis, and with the rare genetic disorder Shwachman–Diamond syndrome, which produces primary lipomatous pseudohypertrophy alongside aplastic anemia and exocrine dysfunction. Localized fat replacement, however, is a different story: it is frequently linked to PanIN and to intraductal papillary mucinous neoplasms. High-grade PanIN is often invisible as a mass on scans; what reveals it is precisely this indirect signature of localized fat replacement, which can prompt further investigation with pancreatic duct cytology or endoscopic ultrasound-guided biopsy.

The evidence connecting atrophy to early cancer is compelling. In a recent large-scale study, suspicious findings of pancreatic cancer were visible on prediagnostic CT scans in 47.8 percent of patients, and the most characteristic finding was progressive focal parenchymal atrophy appearing roughly 2.7 years before diagnosis, followed by progressive main pancreatic duct dilatation about 1.1 years before diagnosis. Focal atrophy occurred more often in the body and tail of the pancreas than in the head. A systematic review found that the pooled prevalence of intrapancreatic fat deposition in people with pancreatic cancer or premalignant lesions was 52 percent, and the presence of such lesions significantly increased the risk of fat deposition. Focal pancreatic steatosis on CT may even indicate pancreatic carcinoma in situ, the earliest stage of the disease, when abnormal cells are still confined to the duct lining.

Detecting these changes requires a deliberate imaging strategy. The authors identify contrast-enhanced CT with three-directional imaging as the ideal first-line modality, with MRI, magnetic resonance cholangiopancreatography, and endoscopic ultrasound serving as sensitive tools for mapping the distribution of atrophic and fatty lesions. Endoscopic ultrasound offers sufficient resolution to detect changes of carcinoma in situ, which is often accompanied by chronic pancreatitis and fatty infiltration in the background parenchyma. Intraductal ultrasonography can simultaneously visualize lesions inside the duct and the surrounding anatomy to guide resection. Cytological evaluation of pancreatic juice obtained by endoscopic retrograde cholangiopancreatography can confirm the presence of cancer cells, but the sensitivity of this technique is limited and the risk of post-procedure pancreatitis is real, so the authors recommend reserving it for selected patients in whom the result would change management.

The surgical implications are the heart of the review. In one key study, patients with focal or upstream atrophy showed significantly longer intraductal lateral cancer extension than those without atrophy, with a median extension of 20.0 millimeters versus 5.0 millimeters. Two patients in that series had positive resection margins despite wide resection of the atrophic area, and three developed recurrence in the remnant pancreas. Because cancer can spread along the duct more than twice the length of the visible atrophy, the authors recommend that surgeons choose the transection line carefully and rely on frozen-section histopathology to guide additional resection. Total pancreatectomy should be avoided when feasible, but additional pancreatectomy is justified for positive margins or for areas of chronic pancreatitis and fat infiltration. For lesions in the head or tail, pancreatoduodenectomy or distal pancreatectomy with appropriate margins carries a low risk of positive margins; for body lesions, an extended pancreatoduodenectomy is technically easier to extend intraoperatively.

Not every atrophic pancreas demands radical surgery. When no tumor is detectable and carcinoma in situ is suspected, limited procedures such as middle pancreatectomy or spleen-preserving distal pancreatectomy are considered acceptable, particularly since minimally invasive techniques have improved their safety. However, the authors issue a sharp warning: clinically T1 pancreatic cancers, even those smaller than 10 millimeters, can already show regional lymph node metastasis or invasion of the periarterial neural plexuses, so limited resection without lymph node dissection must be avoided once invasive cancer is suspected. Patients with carcinoma in situ face a real risk of recurrence in the remnant pancreas and should be monitored every six months for more than five years, with repeated pancreatectomy considered if recurrence appears. Prognosis falls steeply with invasion: the predicted five-year overall survival is 76 percent for carcinoma in situ, 36 percent for tumors under 1 centimeter, and just 17 percent for tumors of 1.1 to 2.0 centimeters.

The review’s authors acknowledge that major questions remain unresolved. Most studies of PPA are retrospective and rely on CT definitions, and no research has yet determined whether atrophy is a consequence of cancer or a contributor to carcinogenesis itself. The evidence linking the length of intraductal cancer extension to the choice of treatment is scarce, and nationwide studies are needed. Still, the practical message is clear and potentially transformative: a focal shadow of a shrinking pancreas on a routine scan, appearing years before any mass, may be the first and only chance to catch pancreatic cancer at its most curable stage. For clinicians, recognizing that pattern, and knowing how to operate on it, could mean the difference between a 76 percent five-year survival and a 17 percent one.

Subject of Research: Pancreatic parenchymal atrophy as an early imaging marker of pancreatic ductal adenocarcinoma and its surgical management

Article Title: Pancreatic Atrophy: A Narrative Review and Surgical Interpretation

Article References: Fujino, R., Okada, K.-I., Masugi, Y., Iwasaki, E., & Mori, M. (2026). Pancreatic Atrophy: A Narrative Review and Surgical Interpretation. Annals of Gastroenterological Surgery, 10(5), 1419-1427. https://doi.org/10.1002/ags3.70230

Image Credits: AI Generated

DOI: 10.1002/ags3.70230

Keywords: pancreatic cancer, pancreatic parenchymal atrophy, pancreatic ductal adenocarcinoma, computed tomography, pancreatic fat replacement, PanIN, carcinoma in situ, pancreatectomy, endoscopic ultrasound, surgical margins, early detection, gastrointestinal surgery

The clinical backdrop explains why the innovation matters. Pancreatic ductal adenocarcinoma, the most common and most lethal form of pancreatic cancer, has seen its five-year overall survival improve from roughly 4 percent to 13 percent over recent decades, according to data cited in the report. Yet the prognosis remains poor, and typical presentations include weight loss, jaundice, abdominal pain, new or worsening diabetes, and steatorrhea, the fatty stools that signal impaired digestion. Tumors of the pancreatic head sit in a crowded anatomical neighborhood, wrapped around the portomesenteric venous confluence where the splenic and superior mesenteric veins merge into the portal vein. As the tumor grows, it can compress or invade these vessels, producing portal hypertension, fluid accumulation in the abdomen known as ascites, and swelling of the bowel wall that interferes with nutrition. Under European Society for Medical Oncology guidelines, contrast-enhanced computed tomography is the mainstay of diagnosis, mapping tumor size, vascular involvement, and metastatic spread, while biopsy is indicated before chemotherapy, with endoscopic ultrasound-guided fine-needle biopsy preferred over CT-guided approaches.

The patient in the report arrived with a four-month history of weight loss, mild steatorrhoea, and newly developed ascites. Imaging revealed a mass in the pancreatic head that encased both the superior mesenteric vein and the superior mesenteric artery, obstructed the portomesenteric venous confluence, and had spawned a network of venous collaterals alongside thickening of the bowel wall, particularly the duodenum. Apposing thrombus, or clot, was also present within the affected veins. He was judged ineligible for surgery, making a tissue diagnosis essential before palliative chemotherapy could begin. The first endoscopic ultrasound-guided biopsy attempt had to be abandoned because interposed collateral vessels blocked a safe needle path. A second attempt returned tissue that showed only signs of chronic pancreatitis, an inconclusive result that left the team without proof of malignancy and the patient without access to cancer treatment.

At this point the multidisciplinary team chose a dual-purpose intervention. To relieve the ascites and bowel oedema, they planned a transjugular portosystemic shunt with portal vein recanalization, an approach abbreviated PVR-TIPS. In a standard TIPS procedure, a track is created through the liver connecting the hepatic vein to the portal vein, shunting blood to decompress the portal system. Here, the team extended the concept: after creating the shunt under fluoroscopic guidance, they passed catheters and guidewires through the blocked portomesenteric venous confluence and advanced them into the superior mesenteric vein, reopening the obstructed venous highway from within. This recanalization simultaneously created a working channel that led directly past the tumor-bearing portion of the pancreatic head.

What followed was the technically novel step. Because there was still no histological proof of cancer, the team decided to sample the tumor through the vein they had just reopened. An 8 French gastroscope biopsy forceps, an instrument normally used inside the stomach, was pushed under fluoroscopic guidance through the 10 French TIPS sheath and advanced into the tumor-bearing segment of the pancreatic head. Three tissue samples were obtained from inside the vessel. The interventional sequence was then completed with the placement of a 12-millimeter bare metal stent across the externally compressed venous segment, restoring flow through the portomesenteric confluence. The clinical response was rapid: bowel oedema resolved within hours, ascites decreased markedly within four weeks, and histology confirmed pancreatic ductal adenocarcinoma. Palliative chemotherapy was initiated just one week after the procedure.

The report situates this case within a growing but still limited literature on venous decompression in pancreatic cancer. Transhepatic portomesenteric venous stenting is already an established palliative treatment for locally advanced disease. In the largest published series, covering 129 patients of whom 119 had pancreatic cancer, technical success reached 97 percent with only four adverse events, and at 20 months of follow-up primary stent patency was 80 percent, with symptom relief in most patients. A separate series of 40 pancreatic cancer patients treated with direct percutaneous transhepatic portomesenteric venous stenting showed that the approach can do more than relieve symptoms: it enabled patients with tumor thrombus in the portal vein to receive chemotherapy, and 15 of the 40 went on to undergo the Whipple procedure, the complex surgical removal of the pancreatic head, without vascular resection.

Biopsy through blood vessels, by contrast, has a narrower track record. Non-targeted transvenous biopsies are routine for diffuse liver and kidney disease, but targeted transvenous sampling of specific lesions remains rare. Where it has been used, operators typically combine fluoroscopy with intravascular ultrasound for image guidance, and studies suggest the technique is valuable precisely for targets that lack a safe percutaneous window, the corridor of tissue a needle must cross to reach a lesion without injuring vessels or organs. Previous intraportal approaches have been percutaneous rather than transjugular: one recent report described a percutaneous transhepatic portal access using gastroscope biopsy forceps to biopsy portal vein invasion by a pancreatic neuroendocrine tumor, and a similar transhepatic route has been used in hepatocellular carcinoma with tumor thrombus. A transbiliary route, passing instruments through the bile ducts, has also been described for sampling inaccessible pancreatic head masses, though it is equally uncommon.

The transjugular route described in the new report offers distinct advantages. Because the shunt and stenting were performed anyway for palliative decompression, the biopsy added no separate access route, allowing venous decompression and histological confirmation to be accomplished in one sitting. The authors also argue that an endovascular transportal approach may reduce the inherent bleeding risk compared with percutaneous or endoscopic ultrasound-guided biopsy, in the same way that transjugular liver and kidney biopsies are favored in coagulopathic patients, since any bleeding occurs into the vessel itself rather than into the peritoneal cavity. Nevertheless, the technique is not without hazards: vessel injury remains a concern, and adjunct imaging such as cone-beam computed tomography is recommended to minimize damage to non-target structures and to improve sample quality. The single-case nature of the evidence means broader safety and efficacy data will be needed before the approach enters routine practice.

Even so, the case signals a meaningful expansion of the interventional toolkit for pancreatic cancer, a disease in which the window for effective treatment is often agonizingly narrow. For patients whose tumors are wrapped in collateral vessels, whose clot or anatomy blocks standard needle paths, or whose earlier biopsies returned falsely reassuring inflammation, the ability to combine shunt creation, vein recanalization, stenting, and tumor sampling through a single transjugular access could shorten the path from obstruction to diagnosis to chemotherapy. The report’s conclusion is measured but clear: endovascular transportal biopsy is a feasible method for obtaining histopathological diagnosis in pancreatic head tumors, even after failed endoscopic ultrasound-guided biopsy, and it can be seamlessly combined with other interventional measures such as portal vein recanalization and stenting. As interventional oncology continues to blur the line between diagnosis and treatment, this vein-borne route to one of medicine’s hardest-to-reach tumors may find a growing role.

Subject of Research: Endovascular transportal biopsy of pancreatic head tumors via the portal vein prior to stenting

Article Title: Transportal endovascular biopsy prior to portal vein stenting in pancreatic cancer

Article References: Mahnken, A. H. (2026). Transportal endovascular biopsy prior to portal vein stenting in pancreatic cancer. CVIR Oncology, 2(1), Article 12. https://doi.org/10.1007/s44343-026-00046-2

Image Credits: AI Generated

DOI: 10.1007/s44343-026-00046-2

Keywords: pancreatic cancer, endovascular biopsy, portal vein stenting, transjugular portosystemic shunt, interventional radiology, pancreatic ductal adenocarcinoma, portal vein recanalization, endoscopic ultrasound, ascites, venous decompression, biopsy techniques, palliative care