Jou rnal of the Korean Radiologica l Society, 1994; 30(5) : 859-864

Acute Pancreatitis: CT Grading and Its Significance1

11 Young Kim, M.D. , Min SOO Kim, M.D., Hae Ryung Park, M .D. ,

Pyo Nyun Kim, M.D. , Byoung Ho Lee, M.D. , Moon Ho Lee, M .D .2, Sun Joo Kim, M.D.2

Purpose: To assessthe prognosticvalue of computed tomography (CT) in acute pancreatitis, initial CT examinations were classified into 5 grades in terms of the degree of disease severity and CT findings were correlated with clinical cou rse, objective prog nostic sig ns, and compl ications.

Material and Methods : In 29 consecutive patients with acute pancreatitis, the CT examination and clinical course were reviewed without knowledge of each other finding. 80th Ranson ’s prognostic signs and modified Glascow prognostic signs were used for evaluation of acute pancreatitis.

Results: Of the 29 patients, 25 patients recovered with medical treatment while 4 patients (14%) developed pseudocysts and the other 4 patients (14%) developed abscesses. CTfindings ofthe acute pancreatitis are as follows; 10 patients (35%) in grade Äwith normal pancreas, 2 patients (7%) in grade 8 with pancreatic en ­largement, 2 patients (7%) in grade C with intrinsic pancreatic abnormalities associated with haziness, 3 patients (10%) in grade D with single, ill-defined fluid collection, and 12 patients (41%) in grade E with extensive and two or multiple fluid collections or presence of gas in or adjacent to the pancreas. Correlation co­efficient is 0.46 (p=0.012) between CT findings and Ranson ’ s prognostic signs, 462 (p=0.0115) between CTfindings and modified Glascow prognostic signs .

Conclusion: CT gradings of acute pancreatitis may be useful in predicting clini ­cal course.

Index Words : Pancreas, CT Pancreatitis

Accurate diagnosis of acute pancreatitis is import­ant, since the degree, duration , and type of treatment of acute pancreatitis are based on the early evaluation of severity of the initial attacks. Computed Tomography (CT) plays a role in the evaluation with pancreatic en­largement, formations of phlegmon and abscess for­mation. On the cl inical base , several methods of a more objective evaluation have been reported that poten­tially improve prognostic ability and prediction of complications (1 -7). Among them , Ranson ’s criteria and modified Glascow criteria have been used in the evaluation of prognostic indicator 01 the disease se­verity (8 -9)

CT examination plays an important role in the initial

' Depa이ment of Radiol ogy Col lege of Medicine , Soonchunhyang University 2Department ofMedic ine Coll ege of Medicine, Soonchunhyang University Received November 6, 1993; Accepted March 7,1994 Address reprint requests to: 11 Young Kim , M.D., Department of Radiology Col­lege of Medicine, Soonchunhyang University. # 23-20 Bongmyung-dong Chuan , 330-1 00 Korea Tel. (0417) 565 - 3711 Fax. (0417) 565 -3721

859

assessment of the severity of acute pancreatitis , since morbidity and mortality depend in great measure on the local pancreatic and peripancreatic complications (i. e. , abscess , pseudocyst, and hemorrhage) . For this reason , CT findings of acute pancreatitis were analyzed and compared with clinical outcome retro­spectively. The objectives of this study are to describe, classify , and analyze the early CT findings in acute and relapsing pancreatitis and to assess their predictive value. Th is study was also performed to compare re­lationship between CT findings and Ranson ’s critería and relationsh ip between CT findings and modified Glascow criteri a.

MATERIALS and METHODS

This study was based on a analysis of CT, clinical , and laboratory findings of 29 pat ients with acute pan­creatitis for the past 2 years. There were 24 men and 5 women , aged 26 -76 years (mean = 48)

The cl inical diagnosis was based on symptoms such

Journal of the Korean Radiological Society, 1994; 30(5) : 859-864

as nausea, vomiting , abdominal pain , and elevation of serum amylase levels above 200 Smoggy units. The etiology of pancreatitis was chronic alcohol abuse in fourteen patients , cholelithiases and gallstones in four patients , hyperlipidemia in one case , and miscel­laneous and unknown in ten patients

We used the previously reported objective prognos­tic signs (2, 8, 9) listed in Table 1 and Table 2 to assess the severity and possible complications. The recorded clinical course , complications , treatment, and respon­se to treatment for all individuals during hospitaliza­tion was reviewed.

CT examinations were pertormed on a Hitachi -W70 o scanner (Tokyo , Japan) using standard technical parameters. Patients received 500 mL of diluted diatrizoate meglumine (Gastrografin ; Bristol -Myers Squibb , Princeton , NJ) as oral contrast material , and a rapid intravenous drip infusion of 100 -150 mL of 68. 3% meglumine ioglicate (RayvistR300 , Schering (Korea) Ltd.) was started immediately before scan­ning , except 7 patients with contraindication such as renal failure. A total of 36 CT scans were reviewed in this study. The first CT scanning was pertormed within the fi rst 7 hospital days in all patients. CT scans were interpreted by two radiologists with consensus without

Table 1. Ranson ’s Early Prognostic Signs 01 Acute Pancreatitis

At Admission or Diagnosis During Initial48 Hours

Age over 55 years Hematocrit lall greater than 10 percentage points

White blood cell count over Blood urea nitrogen level rise 16 x 103 /ul (16 X10' 1) morethan5mg/dl (1.79mmoll l)

Blood glucose level over Serum calcium level below 200mg /dl(11 mmolll) 8mg /dl(2 mmolll)

Serum lactic acid dehydro- Arterial P02below 60mm Hg genase level over 350lUII (7.98 kPa) Serum glutamic-oxaloacetic Base delicit greater than transaminase level over 4 mEqll(4 mm이11)

250lU /dl

score 1 -2 : mild 3 -5: moderate 6-8 ‘ severe

Estimated Iluid sequestration morethan 6,000 ml(6Iiter)

Table2. M。이lied Glascow Criteria within 48 hrs 01 Admission

age ) 55 yrs WBC ) 16 ,OOO/mm 3

glucose ) 180 mg/dl *

LDH ) 600 IU /L

score 0-2 : mild

Pa02( 60mmHg albumin ( 3. 2 g/dl

Calcium ( 8 mg/dl

BUN ) 45 mg/dl **

*: no previous Hx 01 DM 3 - 5: moderate .. : no response to hydration 6-8 : severe

prior knowledge of clinical findings or objective prog­nostic signs.

We classified the type of pancreatic inflammation seen on CT into five categories. This classification was based on an overall assessment of size , contour , and density of the gl and and peripancreatic abnormal ities. Pancreatic lesion was not measured in this assess­ment. We used Balthazar classification (8) : grade A , normal pancreas; grade B, focal or diffuse enlarge­ment of the pancreas ; grade C, intrinsic pancreatic ab­normalities associated with haziness and streaky densities representing inflammatory changes in the peripancreatic fat; grade 0 , single, ill -defined fluid collection (phlegmon); grade E, two or multiple , poorly defined fluid collections or presence of gas in or adjac­ent to the pancreas. According to the above grade, we classified the severity of pancreatitis as prognostic sign of pancreatitis. The relationships between differ­ent grades and the cl i nical course and prognostic signs were analyzed

RESULTS

Of the reviewed 29 patients , 25 recovered with medi­cal treatment while 4 patient (13.7 %) developed sepsis and underwent surgical drainage of abscess. Two patients with abscess recovered atter surgery but the other two patients died due to sepsis. The relationship 。f the Ranson ’s objective prognostic signs to the clini­cal course and modified Glascow prognostic signs to the clinical course are shown in Table 3 and 4

CT findings that may correlated with the severity of acute pancreatitis were reviewed. In our evaluation , we noted grade A in 10 of 29 (34.5%) cases , grade B in 2 of 29 (7%) cases (Fig. 1, 2) , grade C in 2 of 29 cases (7 %) (Fig. 3) , grade 0 in 3 of 29 cases (10 %) (Fig. 4) ,

Table 3. Rel ationship between Ranson’ s Prognostic Sign and Clinical Course

Prognostic Patients Average Days Abscess Abscess

Signs Hospital ized Deaths*

0-2 15 12.8 3-5 10 25.8 6-8 4 37.5 3 2

Table 4. Relationship between Glascow Prognostic Sign and Clinical Course

Prognostic Patients Average Days Abscess Abscess Signs Hospitalized Death

0-2 19 15.7

3-5 7 24.1 6-8 3 47.7 3 2

- 860 -

and grade E i n 12 of 29 cases (41.4 %), (Fig. 5). Among

the patients with grade E, there were two cases with

Fig. 1 . CT scan of normal pancreas in patients with clinical pan­

creatitis (grade A)

Fig . 2. Diffuse enlargement of the pancreas without peripan­

creatic inflammatory changes (grade B)

Fig. 3. Haziness and increased density are noted in pancreas

tail portion and anterior superior aspect of the left kidney (grade C)

11 Young Kim, et al: Acute Pancreatiti s

peripancreatic abscesses showing retroperitoneal air

density (Fig. 6)

Fig. 4. Enlarged pancreas with associated fluid collection in an­

terior aspect of peripancreatic space (grade D)

Fig. 5. CT scan shows enlarged pancreas with peripancreatic

low density in patient with grade E acute pancreatitis. Confirmed

hemorrhagic pancreatitis

Fig. 6. CT scan shows extensive fluid collection with multiple

curvilinear air (arrows) in right anterior and posterior pararenal

space in patient with peripancreatic abscess

- 861 -

Table 6. Relationship between CT Findings and Ranson ’s Prog­nostic Si gns

Journal of the Korean Rad iological Society, 1994; 30(5) : 859- 864

Average Prognostic

Si gns/patient

。。

ζ)

꺼a

n。

0

3

2

2

3

Patients by Number 01 Prognostic Signs Grade

8 7

2

6

2

5 4

3

3 2

4

1, ,, 。J

0

5 A

B

C

D

E

Table 7. Relationship between CT Findings and Modilied GI ascow Prognostic Sign Fig. 7. Retroperitoneal air density in patient with gas lorming

peripancreatic abscess (arrows) Average Prognostic

Si gns/patient 8 6

5

5

5

6

nU

。J

‘l nι 끼J

Patients by Number 01 Prognostic Signs Grade

7 6 5 4 3 2 0

4 5 A[

R]

P)

nu

I」

Table 5. Relationship between Early CT Findings and Clinical Course

Abscess

Deaths

Abscess Average Days Hospitalized

Patients Grade

2 2 3

2

2

Table 8. Relationship between CT Findings and Ranson ’s Prog­nostic Signs and Development 01 Abscess and Death

Table 9. Relationship between CT Findings and Modilied Glascow Prognostic Signs and Development 01 Abscess and Death

Patients Grade Prognostic Regression Abscesses* Abscess sign (%) Death*

2(17 %) 4(33 %)

R=0.46 (P=O.012)

12 A&B 1.3

5 C&D 2.2

12 E 3.8

* Percentages in parentheses

Patients* Grade Prognostic Regression Abscesses*

Abscess sign (%) Death

12(41.5) A& B 1.1 R=0.462

5(17) C&D 1.6 P=O.0115 12(41.5) E 3.6

*Percentages in parentheses

2(17 %)

pancreatitis and its complications were well estab­lished in the literature (1 0 -20). A comprehensive evalu­ation of the prognostic value of the initial CT examin­ation based on clinical follow - up, surgical findings , and correlation with prognostic signs had been reported by Balthazar et a l. (8, 21) . However , there were m any prognostic signs using cl inical evaluation of acute pan­creatitis such as Ranson ’s prognostic signs and modi­fied Glascow prognostic signs (8, 9). This study

The role of CT scanning in initial diagnosis of acute attempted to compare both prognostic signs with CT - 862

4(33 %)

2

The relationship between early CT findings and cl ini­cal course is summarized in Table 5. Length of hospitalization correlated roughly with the severity of initial CT findings. Exceptions to the general trend , however, occurred with one patient in grade A for 19 days of hospitalization and one patient in grade 0 re­quiring less than 10 days of treatmen t. No patient with grade A was seriously ill , and the two patients who died of local complications (abscesses) initially had grade E pancreatitis

Fluid collections were initially seen in 15 patients in grade 0 and E . There was one case of hemorrhagic pancreatitis requiring peritoneal dialysis. Complicated pseudocysts were noted in 4 patients of grade E on fol­low up CT examination.

The relationship between early CT findings and prognostic signs is shown in Table 6, 7. The relation­ship between the number of prognostic signs and grades of pancreatitis varies widely in patients with zero to seven prognostic signs

The numbers of patients with abscess or those that died were analyzed as a function of combined CT findings and prognostic signs, and the results were shown in Table 8 , 9.

4

n3

꺼J

m

띠 쟁 얘 %ι

m 2

2

3

다

A

B

C

D

E

DISCUSSION

finding for the value of prognostic sign and CT scan­ning not only in the initial diagnosis of pancreatitis , but also as a prognostic indicator of the diseases severity and expected complications

Although the clinical course was consistent with the grade of pancreatitis , some grade A patients may not have had pancreatitis at al l. Therefore , the exact per­centage of patients with acute pancreatitis and a nor­mal CT scan is difficult to assess. This percentage dependends mainly on the severity of acute pancrea­titis and the time of the examination and should be expected to vary from series to series (8). However, Hill et al reported 28 % of normal CT in acute pancreatitis (22). In our data , 32 % was normal on CT scanning in acute pancreatitis. Those cases with grade A acute pancreatitis showed good prognosis. Diffuse enlarge­ment of the pancreas with grade B was classified as mild pancreatitis and fairly well correlated with clinical outcome. There was a clear correlation between the severity of pancreatitis determined on the in itial CT examination and the clinical course. We noted a steady trend toward an increased day of hospital ization in patients with more severe grades of pancreatitis (Table 5). Abscesses occurred in 4 patients (13.7%) , and there were all classified as grade E. Surgical inter­vention was indicated when abscess was detected on CT scan. There was strong relationship between the in­itial presence of peripancreatic fluid collection and the development of abscess and pseudocyst. Four cases of the abscess were developed in grade E. Three cases of the pseudocyst were developed in grade E and one case in grade D

Pancreatitis with findings such as hemorrhagic pan­creatitis , necrotizing pancreatitis , pancreatic absces­ses were classified grade E in this study. These acute pancreatitis could be included in grade E and these grading were correlated with clinical prognostic signs on this study. Sigelman et al reported that extravasated pancreatic secretions and development of large peripancreatic fluid collections were considered an es­cape mechanism , leading to a beneficial decom­pression of the pancreatic duct system (23). In our study , based on short - term CT and clinical f이 low-up

evaluation , we did not success to detect any advan­tages of large fl u id collections.

Fluid collections containing air may develop second­ary to enteric fistulas and may not always indicate an abscess (24 , 25). In 0

- 863

11 Young Kim, et a/ : Acute Pancreatitis

signs. We found that using prognostic signs and CT findings led to better estimation of the risk of death in this series. None of grade A , B , C , and D patient died re­gardless of prognostic signs , these patients had zero to four signs. On the other hand, the mortality of patients with grade E correlated well with the increasing num­ber of prognostic signs

We concluded that initial CT examination in cases of acute pancreatitis was helpful in establishing or confirming the clinical diagnosis as well as in depicting associated abnormalities. CT could also be used as an early indicator of the diseases severity and expected morbidity and mortality. There were no significant dif­ference between Ranson ’s prognostic signs and modi­fied Glascow prognostic signs for the predicting prog­nosis. We found a good correlation between the grades in CT appearance and the clinical course , development of abscesses , pseudocysts and death. The use of ob­jective prognostic sign in initial CT findings improved the original prognostic estimation and identified patients in whom life threatening complications may develop.

REFERENCES

1. Jacobs ML, Daggett WM , Civetta JM , et al. Acute pancreatitis analysis of factors influencing survival. Ann Surg 1977 ; 185 43-51

2. Ranson JHC , Pasternak BS. Statistical methods for qual ifying the severity ofcl in ical acute pancreatitis. J Surg Res 1977 ; 22: 79-91

3. Berry AT, Taylor TV, Davies GC‘ Diagnostic tests and prognostic indicators in acute pancreatitis. J R Co// Surg Edinb 1982 ; 27

345-352

4. Ranson JHC , Spencer FC. The role of peritoneallavage in severe acute pancreatitis. Ann Surg 1978 ; 187: 565-575

5. Ranson JHC, Rifkind KM , Turner Jw. Prognostic signs and nonoperative peritoneal lavage in acute pancreatitis. Surg

Gyneco/ Obstet 1976 ; 143 : 209-219

6. Hill MC, Barkin J, Isikoff MB , et al. Acute pancreatitis : clinical vs CTfindings. AJR 1982; 139 : 263-269

7. Silverstein W, Isikoff MB ,Hill MC ‘ Barkin J. Diagnostic imaging of acute pancreatitis : prospective study using CT and sonography AJR 1981 ; 137: 497-502

8. Balthazar EJ, Ranson JH , Naidich DP , Megibow AJ , Caccavale R,

Cooper MM. Acute pancreatitis: Prognostic value of CT. Radi­

%gy1985 ‘ 156: 767-772 9. Blamey SL , Imrie CW, Neill JO, Gilmour WH , Carter DC

Prognositic factors in acute pancreatitis. Gut 1984; 25

1340-1346

10. Mendez G Jr ‘ Isikoff MB , Hil l MC. CT of pancreatiti s: interim as­sessment. AJR 1980 ; 135 : 463-469

11 . Wiliford ME, Foster WL Jr., Hal vorsen RA, Thompson WM. Pan­creatic pseudocyst comparative evaluation of sonography and computed tomography. AJR 1983 ; 140 : 53-57

12. Ranson JHC. Etiological and prognostic factors in human acute pancreatitis : a review. Am J Gastroentero/1982 ; 9 : 633-638

13. Jeffrey RB , Federle MP, Cello JP , Crass RA. Early computed tomograph ic scanning in acute severe pancreatitis. Surg

GynecolObstet1982; 154: 170-174

Journal of the Korean Radiological Society, 1994 : 30(5) : 859-864

14. Pringot J, Dardenne AN, Lousse JP, et al. Contribution of Springer, 1981 ; 72-77

computed tomography in the diagnosis of severe acute pancrea- 20. Kivisaari L, Somer K, Standertskjold-Nordenstam CG , titis. In : Hollender LF , ed. 'Controversies in acute pancreatitis. Schroeder T, Kivilaakso E, Lempinen M. A new method lor diag-

Berlin : Springer , 1981 ; 64-71 nosis 01 acute hemorrhagic-necrotizing pancreatitis using con-

15. Dembner AG, Jaffee CC , Simeone J, Walsh J. A new computed trast-enhanced CT. Gastrointest Radiof1984 ; 9: 27-30

tomographic sign 01 pancreatitis‘ AJR 1979 ; 133 : 477-479 21 ‘ Balthazar EJ‘ Robinson DL, Megibow AJ , Ranson JHC‘ Acute

16. Jeffrey RB , Federle MP, Laing FC. Computed tomography 01 pancreatitis : Value 01 CT in establishing prognosis. Radiofogy

mesenteric involvement in lulminant pancreatitis. Radiofogy 1990 ; 174 : 331-336

1983 ; 147: 185-188 22. Hill MC, Dach JL, Barkin J, et al. Role 01 percutaneous aspiration

17. Federle MP, Jeffrey RB, Crass RA, Dalsem VV. Computed tom- in diagnosis olpancreatic abscess. AJR 1983; 141 : 1 035-1038

。graphy 01 pancreatic abscesses. AJR 1981 ; 136 ‘ 879-882 23. Siegelman SS, Copeland BE, Saba GP, et al. CT 01 Iluid

18. Segall , Epstein B, Lawson HL, et al. The syndrome 01 pancreatic collections associated with pancreatitis. AJR 1980 ; 134

pseudocysts and Iluid collections‘ Gastrointest Radio/1984 ; 9 : 1121-1132

115-122 24. Alexander ES, Clark RA, Federle MP. Pancreatic gas: indication

19. Dammann HG , Grabbe E, Eichluss HP, Flasholl D. Computed 01 pancreaticlistula. AJR1982 ;139 :1089-1093

tomography and clinical severity 01 acute pancreatitis. In 25. Torres WE, Clements JL Jr., Sones PJ , Knopl DR. Gas in the pan-

Hollender LF, ed. Controversies in acute pancreatitis. Berline: creatic bed without abscess. AJR 1981 ; 137 : 1131-1133

대 한방사선 의 학회 지 1994; 30(5) : 859-864

급성 훼장염 :CT 등급과 임상적 의의

김일영 · 김민수 · 박해령 • 김표년 · 이병호 .01문호1. 김선주1

순천향대학교 의과대학 밤사섣과학교실 내과학교실

목 적 :급성 훼장염환자에서의 CT소견에 의한 예후판정의 유용성을 알아보기 위해 춰|장엽의 CT소견을 그 정도에 따라 5

단계로 분류하고 이를 환자의 예후， 추정지표， 합병증등과 비교하였다.

대상 및 방법 :29명의 환자의 CT소견과 임상소견을 각각 분석하였으며 입상소견을 분석하는데는 Ranson’s 예후지표와 변

형한 Glascow 예후지표를 사용하였다.

결 과 :29명의 환자는 25명은 내과적 치료에 의해 회복되었으며 . 4명 ( 14%)에서 훼가성낭종이 발생하였으며 또 다른 4엽

(14%)에서 훼농앙이 발생하였다. 급성 훼장염이 CT소견은 다음과 같았다 :10명 (35%)이 grade A이면서 정상 춰|장 소견이

였고.2명 (7%) 01 grade 801 면서 훼장이 커졌으며.2염 01 (7%) grade cOI 며 훼장내 이상 음영이 관잘되었으며.3명 ( 10%)

이 grade 0로 용액저류가 관찰되었으며. 12명 (41%)01 grade E로 광범위하거나 여러개의 용액이 훼장 주위에 관찰되었다.

CT소견과 Ranson ’s 예후지표와는 0.46 (p =0.012)의 상관계수를 나타내고 Glascow 예후지표와는 0.462 (p=0.0115)의 상관

계수를나타냈다.

결 론 :CT에 의한 급성춰|장엽의 분류는 환자의 예후추정에 유용할것으로 사료된다.

864-