PREDICTING EXOCRINE INSUFFICIENCY FOLLOWING PANCREATIC RESECTION

We read with interest the article, “Reduced Pancreatic Parenchymal Thickness Indicates Exocrine Pancreatic Insufficiency After Pancreatoduodenectomy” by Dr. Nakamura et al.[1] Their work presents a potentially potent tool by which the pancreatic surgeon may treat their patients following surgery to alleviate the effects of exocrine pancreatic insufficiency (EPI). Importantly the metric they investigated, pancreatic parenchymal thickness on pre- and postoperative computed tomography (CT) scans, is readily available and easily translatable to current clinical practice making their results immediately applicable. Of particular interest was their finding that the preoperative pancreatic duct to parenchyma ratio was correlated to postoperative EPI with a cutoff of 14.5 mm being 67.6% sensitive and 72.2% specific.[1]

While it is intuitive that EPI, clinically defined as the onset of steatorrhea, is directly related to the extent of resection, the volume of residual pancreas needed to prevent postoperative EPI is unknown and probably highly variable; the extent of the underlying disease process and preoperative pancreatic function may play a role as exemplified by parenchymal fibrosis seen in chronic pancreatitis. Generally, 90–95% of pancreatic enzyme excretion must be lost before clinical signs of EPI develop.[2] Additionally, the concurrent resection of the duodenum or stomach affects the neurohormonal axis of the foregut with alterations in gastrin, cholecystokinin, and pancreatic polypeptide which exert trophic and secretogogue effects on the pancreas. Loss of these enteric hormones affects the rate of atrophy of the pancreatic remnant as well as its digestive function.[3]

Due to these factors, it is difficult to predict the onset of EPI and current clinical practice dictates waiting for clinical signs of steatorrhea or malabsorption to develop before starting patients on oral pancreatic enzyme supplementation. Biochemical tests such as fecal fat content, β-1 Elastase, and N-benzoyl-l-tyrosil-para-aminobenzoic-acid (NBT-PABA) can be cumbersome and/or costly to perform routinely and researchers have failed to show consistent relationships between pre- and postoperative testing that would aid in accurately predicting level of postoperative exocrine pancreatic function.[4] Furthermore, it has been recognized for some time that while most patients who have undergone PD have significantly abnormal fecal fat content, many will not have clinical signs of steatorrhea or require oral pancreatic enzyme supplementation.[5]

The use of CT to evaluate the thickness of the remnant pancreas as a marker of pancreatic exocrine insufficiency has been reported by Lemaire et al.[6] They found that all patients undergoing PD with pancreaticogastric anastomosis had atrophy of the pancreatic gland on CT and marked increase in fecal fat and fecal-1 elastase. However, they were unable to correlate the degree of parenchymal atrophy to the degree of exocrine insufficiency by biochemical tests.

Tajima et al utilized dynamic MRI to monitor the development of pancreatic fibrosis following PD and found that grading the time-signal intensity curve (MRI-tic) served as a predictor of impending exocrine insufficiency. Patients exhibiting an increase in the pancreatic fibrosis grade by MRI-tic analysis tended to experience significant deterioration of pancreatic exocrine function within approximately one year.[7]

Finally, a provocative test utilizing IV secretin administration coupled with magnetic resonance cholangiopancreatography (secretin-MRCP) has been used to evaluate the exocrine function of the pancreas.[8] In a study of postoperative PD patients investigators determined that while there was a significant correlation between pancreatic atrophy and patency of pancreaticogastric anastomosis, secretin-MRCP imaging alone was not sufficiently sensitive to diagnose exocrine insufficiency.[9]

Dr. Nakamura and colleagues addressed some of the weaknesses of previous studies by correlating preoperative parenchymal thickness to postoperative EPI with a benchmark value (14.5 mm) below which there was increased risk for EPI. They also established a cutoff for diagnosis of postoperative EPI using postoperative parenchymal thickness (13 mm). It would be interesting to obtain ¹³CO₂ values for preoperative patients as this would establish a baseline in patients with pre-existing pancreatic disease-particularly for the two patients with chronic pancreatitis. Inclusion of patients with preoperative EPI may falsely increase the sensitivity of the test by overestimating the number of patients with %CD-7h of less than 5% that have reduced parenchymal thickness postoperatively (i.e.: falsely elevated true positive).

Although the authors comment on the adequacy of a single cross-sectional measurement of pancreatic parenchyma thickness in predicting postoperative EPI, the use of volumetric analysis may increase the specificity of the test by more accurately determining the amount of residual parenchyma. This may be particularly true in cases where there is focal dilation of the pancreatic duct leading to underestimation of the parenchymal thickness while there is relatively normal exocrine pancreatic function thus leading to a false positive test result.

The use of abdominal imaging to predict the functional status of the pancreas appears to be a developing modality with significant promise as a clinically relevant and applicable test. Perhaps the greatest potential lies in the use of preoperative imaging to predict the onset of postoperative EPI following pancreatic resection or to be able to plan the extent of resection to minimize the risk of EPI while performing an oncologically sound resection.