76 7 .4, Emil J. Baithazar, M.D. John H.C Ranson, BM., B.Ch. David P. Naidich, M.D. Alec J. Megibow, M.D. I’ Robert Caccavale, M.D. Matthew M. Cooper, M.D. Acute Pancreatitis: Prognostic Value of CT1 In 83 patients with acute pancreatitis, the initial computed tomographic (CT) ex- A aminations were classified by degree of disease severity (grades A-E) and were correlated with the clinical follow-up, objective prognostic signs, and compl ca- .tions and death. The length of hospital- iz tion correlated well with the severity of the initial CT findings. Abscesses oc- .4 curred in 21.6% of the entire group, com- pared with 60.0% of grade E patients. Pleural effusions were also more common in grade E patients. Grades A and B pa- ,tients did no have abscesses, and none died, regardless of the number of prog- F nostic signs. Abscesses were seen in 80.0% I ‘of patients with six to eight prognostic signs, compared with 12.5% of those with zero to two. The use of prognostic signs with initial CT findings results in im- proved prognostic accuracy. Early CT ex- amination of patients with acute pancrea- titis is a useful prognostic indicator of morbidity and mortality. ‘Index terms: Pancreas, computed tomography, 77.1211 #{149} ancreatitis, 77.291 Radiology 1985; 156:767-772 From the Departments of Radiology (E.J.B., D.P.N., A.J.M.) and Surgery (J.H.C.R., R.C., M.M.C.), New York University Medical Center, Bellevue Hospital Medical Center, New York City. Received January 10, 1985 accepted and revision requested March 18, 1985; revision received April 3. 1985. c RSNA, 1985 T HE degree, duration, and type of treatment of acute pancreatitis are based on the early evaluation of the initial attack’s severity. Until recently, this evaluation relied mainly on te presence on absence of varied clinical parameters such as tachycardia, fever, dyspnea, oligunia, protracted ileus, and tense abdomen. Several methods of a more objective evaluation have been reported (1-7) that potentially improve prognostic ability and prediction of com- plications. Among them, the statistical analysis of early objective measurements of multiple risk factors, described by Ranson (2, 3), has received wide attention and has been considered a reliable prognostic indicator of the diseases’s severity. These objective prog- nostic signs (grave signs or risk factors) have significantly im- proved the initial assessment ba ed on clinical crit ria alone and are used as guidelines in the decision-making process of selecting proper medical or surgical treatment in our institution. Since morbidity and mortality depend in great measure on the local pancreatic and penipancreatic complications (i.e., abscess, pseudocyst, hemorrhage), computed tomographic (CT) examina- tion could play an important role in the initial assessment of the severity of acute pancneatitis. For this reason, in the past 4 years we have embarked on a comprehensive study designed to assess the prognostic value of the initial CT examination in patients with acute pancreatitis. Our objectives are (a) to describe, classify, and analyze the early CT findings in acute pancreatitis; nd (b ) to assess their predictive value based on correlation of early CT findings with clinical and objective prognostic signs. M TERIAL AND METHODS Our study is based on a detailed analysis of CT, c inical, and laboratory findings of 83 patients with acute pancreatitis admitted to our institution in the past 4 years. There were 63 men and 20 women, aged 17-79 years, with a mean age of 45 years. The clinical diagnosis was based on typical symptoms such as nausea, vomiting, abdominal pain, and elevation of serum amylase levels above 200 Somogyi unit . The etiology of pancreatitis was chronic alcohol abuse in 51 patients, cholelithiasis in 11, gallstones and alcohol in five, hyperlipidemia in two, and miscellaneous or unknown in 14. There were no cases of traumatic pancreatitis included in his series. We used the previously reported objective prognostic signs (2, 3, 6, 7), listed in Table 1, to assess the severity of the attack and its possible compli- cations. All patients were initially treated by nasogastric suction, intrave- nous fluid, and supportive therapy. We drained infected fluid collections (abscesses) in 18 patients (21.7%), some upon initial evaluation and others as complications developed. The clinical course, complications, treatment, and response to treatment were recorded for all individuals, until death or discharge from the hospital. C T examinations were performed on a GE 8800 scanner (Milwaukee) using standard technical parameters. Diluted 2% barium sulfate (E-Z-CAT, E-Z-EM, Westbury, N.Y.) was used as oral contrast material, and a rapid intravenous drip infusion of 30% diatrizoate meglumine (Reno-M-DIP [Squibb]) was started immediately before scanning unless contraindicated. Bolus injections were not used in this study. A total of 152 CT scans were obtain d, either as a single examination or as consecutive, follow-up examinations approximately every 2 weeks. The
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
CT scan of en la rged body and ta il o f the pan creas (a ) w ith assoc ia ted flu id co llection in left an te rio r para rena l space (b ) (a rrow s) (g rade D ).
r Figure 5
a. b.
CT scan show ing la rge flu id co llec tions in the lesser sac and an ter io r pa rarena l sp ace in p atie n t w ith g rad e E pancreati tis . N ote com p res sion
. w ith par tia l ob stru ction o f th e du odenum and sligh t th icken in g of ga llb ladd er w all (a rrow s) .
a. CT scan sh ow ing increased den sity of the p eripan cre atic retrop eri ton eal fa t associated w ith ex tra lum in al air (arrow ) in p atien t w ith
pem ipancrea tic ab scess .
b . B ila te ra l, ill-de fined , re troperitonea l flu id co llec tions w ith m u ltip le g as bu bb les in pa tie n t w ith absces s (g rad e E ).
r
I
.I
A
Figur e 6
77 0 #{149}adio logy Sep tem ber 1985
(10 .8% ), th e in f lamm atory pro cess in -
vo lved exc lusiv ely or p redom inan tly
the head o f the pancreas (F ig . 7 ) ; in
f iv e, th e b ody and ta il; and in one ,
o n ly th e tail o f the pancreas. Sw elling
o f on ly the head of the pancreas w as
p resen t in th ree of the 1 1 pa tien ts
w ith g allston e p an crea titis (2 7 .3% )
bu t in on ly s ix cases o f a ll o the r types
o f pancrea titis (8 .3% ). Tw o pa tien ts
w ith h is to ries o f p rev ious pan crea titis
h ad p an crea tic d uc ta l ca lcifica tio ns
d em onstra ted o n CT scan s.
T h e p a tien ts w ere d iv ided accord -
ing to th e f iv e grades , an d the m ela -
tionsh ips b etw een d iffe ren t g rades
and the c lin ical course and prognos tic
s igns w ere an a ly zed . T h ere w ere 12
pa tien ts (1 4 .5% ) in grad e A , 19 (22 .9% )
in g rade B , 17 (20 .5% ) in grad e C , 12
(1 4 .5% ) in grad e D , and 23 (2 7 .7% ) in
grade E.
CT and C lin ica l C ourse
The re la tion sh ip be tw een ear ly CT
find in gs and c lin ica l course is sum -
m anized in T ab le 3 . T he av erag e n um -
ben of fas ting d ay s (n o th ing by
m ou th ) and days in th e ho sp ita l com e-la ted rough ly w ith the severity o f the
in itial C T find ing s. E xcep tio ns to the
genera l trend , how ever, o ccu rred ,
w ith som e pa tien ts in g rade B requ ir-
ing 4 w eeks of hosp ita lization and
som e in g rade D requ ir in g less than 2
w eek s of trea tm en t. N o pa tien t w ith
grad e A pancrea titis w as ser io usly ill,
and a ll five p atien ts w ho d ied because
of local com plications (abscesses ) m i-tially h ad grade D or E pancm ea titis.