Purpose: The purpose of this work was to investigate the natural history of
pancreatic necrosis on contrast-enhanced CT hi patients managed nonoperati
vely.
Method: A computer-based radiology information search revealed 32 patients
with pancreatic necrosis who had had serial contrast-enhanced CT scans and
were managed nonoperatively. There were 23 men and 9 women with a mean age
of 51 years. One hundred forty-five contrast-enhanced CT scans were retrosp
ectively reviewed for the location and extent of necrosis. The medical reco
rds of all patients were reviewed.
Results: The 32 patients bad a mean Ranson clinical grade of 5.8 (range 3-8
). Eighteen of these 32 patients were managed nonoperatively, and 14 patien
ts required a necrosectomy after initial nonoperative management. In the 32
patients, the location of necrosis was in the head (3), body (6), tail (2)
, head/body (2), head/body/tail (9), body/tail (9), and head/tail (1). Exte
nt of necrosis was 0-25% (9), 26-50% (6), 51-75% (6), and 76-100% (11). The
extent of necrosis remained stable during follow-up in 22 (69%) patients a
nd increased during follow-up in 10 (31%). Necrosectomy was performed in si
x (60%) patients in whom there was an increase in necrosis and eight (36%)
patients in whom necrosis was stable. No patient had restoration of normal
enhancement in an area that was previously necrotic. There were five patien
ts who were managed nonoperatively (mean follow-up 318 days) in whom the ne
crosis eventually resorbed, forming a focal parenchymal cleft reminiscent o
f a scar. Five of the 32 patients died.
Conclusion: Pancreatic necrosis as demonstrated by CT tends to remain stabl
e in most patients treated nonoperatively, If the extent of necrosis increa
ses, patients are more likely to require a necrosectomy. In some patients m
anaged nonoperatively, the pancreatic necrosis will resorb, resulting in a
fat-replaced cleft reminiscent of a scar.