Another diagnostic modality that has become invaluable in the workup and management of pancreatic disease is endoscopic ultrasonography (EUS). EUS is highly operator dependent and requires a skilled and experienced gastroenterologist to interpret the study in real time but has the ability to characterize pancreatic masses, cysts, and parenchymal architecture along with fine needle aspiration for cytopathologic evaluation (Fig. 36.3a–d). Utility is greatly increased by on-site determination of sample adequacy and preliminary diagnosis by an expert pathologist that is immediately available. In fact, this modality has replaced percutaneous needle biopsy as the preferred method for obtaining tissue diagnosis of pancreatic malignancy. EUS is a very safe and minimally invasive procedure that can also provide palliation for patients with surgically unresectable disease by performing celiac plexus neurolysis. EUS is highly accurate regarding T and N staging for pancreatic adenocarcinoma, pancreatic neuroendocrine tumors, and distal common bile duct cholangiocarcinoma [4]. Many surgeons argue that for resectable disease, EUS is unnecessary because tissue diagnosis is not mandatory for a patient with a pancreatic mass who is otherwise a good surgical candidate. The greatest value of EUS is that it permits high resolution imaging with simultaneous sampling. Therefore it is very useful in the evaluation of pancreatic cystic lesions, which are becoming more frequently incidentally found by the increasing use of highly sensitive cross-sectional imaging for other abdominal pathology. Information obtained by EUS can include surrounding inflammatory or fibrotic changes consistent with chronic pancreatitis, communication or involvement of the main pancreatic duct consistent with intraductal papillary mucinous neoplasm (IPMN), debris or mural nodules within the cyst wall, or fluid aspiration with tumor marker and enzyme evaluation, presence or absence of mucin, as well as biochemical fluid analysis [5].

Improved pancreatic imaging has led to lower rates of nontherapeutic laparotomies for pancreatic malignancies and pancreatic resections for benign indications. Additionally, accurate preoperative diagnostic imaging will assist the surgeon in identifying borderline or locally advanced disease prior to surgical exploration. Unresectable vascular involvement should be determined prior to entering the operating room because intraoperative surgical assessment of the extent of disease is difficult to determine even for experienced pancreatic surgeons due to the desmoplastic and inflammatory changes that often accompany benign and malignant pancreatic disease conditions. For the most part, nontherapeutic operations due to unresectable disease at experienced centers are mostly limited to patients with small volume peritoneal or surface liver disease, which is difficult for any imaging modality to find. Diagnostic laparoscopy prior to surgical resection still has a role for patients with malignant pancreatic disease, often most useful for patient presenting with back pain, weight loss, significantly elevated tumor markers, and lesions in the body/tail of the pancreas. Typically, however, the yield of this modality often inversely correlates with the quality of preoperative imaging.

Borderline resectable pancreatic tumors have received increased attention in the past several years. While at most centers patients with resectable disease will proceed directly to surgery and those with locally advanced or metastatic disease will undergo palliative treatments, patients with borderline tumors will be considered candidates for neoadjuvant treatment with chemotherapy, external beam radiation, or proton beam therapy. The clinical advantages of neoadjuvant treatment for these patients will include allowing for a time period of observing tumor biology and response to the treatment given. If very aggressive or metastatic disease is noted during restaging protocols, the patient is likely spared an operation that would have been associated with very little chance of success or benefit. Additionally, preoperative treatment for patients with borderline disease may downsize the tumor, allowing for a less aggressive vascular resection and fewer postoperative complications as well as helping to decrease the risk of an R1 margin positive resection, which has been shown to have a negative impact on long-term survival. Pancreatic surgeons, however, have only recently come to define borderline resectable pancreatic tumors, and this definition varies based on the use of a single institutional classification or a consensus definition proposed through international societies and associations [6]. Regardless of what definition is used, for the most part, patients with borderline resectable pancreatic tumors are preferentially given neoadjuvant preoperative treatments and have a reasonable rate of eventually proceeding to surgery. On the other hand, patients with locally advanced disease, often characterized by circumferential involvement of the superior mesenteric artery or long segment occlusive disease of the portomesenteric venous system with non-reconstructible anatomy, are very rarely if ever downstaged to resectable disease (Fig. 36.4a, b).

Preoperative administration of neoadjuvant treatment for resectable pancreatic adenocarcinoma is becoming more frequently performed but has yet to significantly improve long-term survival of patients. Selection criteria, treatment protocols, chemotherapy agents, and the use of radiation vary widely between institutions and trials making meaningful analysis difficult [7]. However, it is important to note that patients undergoing treatment prior to pancreatic surgery have not experienced an increased rate of complications or mortality after surgery. Neoadjuvant treatment protocols clearly have the advantage of selection bias by avoiding surgical intervention for those with aggressive disease or those patients with poor performance during treatment administration. Perhaps the greatest value of this treatment, until more effective agents are discovered, is in resource allocation of the surgical procedure to those patients who will benefit the most.

Elderly patients, who often have more medical comorbidities than their younger counterparts, are being more frequently evaluated for surgical treatment of pancreatic pathology. Due to advances in perioperative anesthesia, postoperative care, and the increasing safety of the procedure, these patients are undergoing pancreatic surgery often without a significantly increased complication or mortality rate over younger patients. Studies have demonstrated that elderly patients can safely undergo pancreatic surgery and that age alone is not a contraindication [8]. Pancreatic surgeons continue to push the limits for safe surgery for patients with cardiac, pulmonary, or other medical disease that were once thought to be associated with too much risk for postoperative complications.

Preoperative biliary stenting performed for obstruction within the head of the pancreas had previously been introduced as a method to decrease the risk of complications after PD as a way to improve liver dysfunction. However, multiple studies including a large multicenter randomized trial comparing preoperative biliary drainage versus proceeding directly to surgery indicate that this practice is not necessary and may actually increase the risk of infectious complications for these patients [9]. Generally, preoperative biliary drainage is used only for patients who are not proceeding directly to surgical resection within a reasonable time period and an endoscopic route is greatly preferred over a percutaneous route.

Pancreatic cystic lesions are becoming more frequently identified by the widespread use of cross-sectional imaging and are more frequently being referred for evaluation (Fig. 36.5a–d). Many of these lesions are mucinous cysts with the possibility of harboring preneoplastic or invasive disease. Therefore, multidisciplinary recommendations based on available literature have given treatment algorithms (Modified Sendai Criteria) for the evaluation, surgical treatment, and follow-up for these lesions [10]. In short, surgical resection is recommended for all fit patients with main duct intraductal papillary mucinous neoplasm (MD-IPMN) and mucinous cystic neoplasms (MCN). Branch duct intraductal papillary mucinous neoplasm (BD-IPMN) with high-risk characteristics including obstructive jaundice, enhancing mural nodules, dysplastic cytology, or main duct involvement should undergo surgical resection. Other worrisome features include size greater than 3 cm, thickened cyst wall, nonenhanced nodule, borderline main duct dilation, focal stricture, cyst growth, and other abdominal symptoms. Surgical treatment is often recommended for these patients, and if not, very aggressive close follow-up is recommended as the risk of malignancy is considerably higher for these patients over those who have none of these worrisome features.