AJR 2002; 178:841-846
© American Roentgen Ray Society
Gallium Uptake in Complicated Pancreatitis
A Predictor of Infection
Jeffrey H. West1,2,
Stephen B. Vogel3 and
Walter E. Drane1
1
Department of Radiology, Shands Hospital at the University of Florida, 1600
S.W. Archer Rd., Gainesville, FL 32610.
2
Present address: Mori, Bean, and Brooks, Department of Radiology, Baptist
Medical Center, 800 Prudential Dr., Jacksonville, FL 32207.
3
Department of Surgery, Shands Hospital at the University of Florida,
Gainesville, FL 32610.
Received August 28, 2001;
accepted after revision October 2, 2001.
Address correspondence to W. E. Drane.
Abstract
OBJECTIVE. A retrospective evaluation was performed of the use of
gallium imaging in patients with known severe pancreatitis to detect infection
in pancreatic and peripancreatic fluid collections.
MATERIALS AND METHODS. Gallium-67 single-photon emission computed
tomography (SPECT) studies were retrospectively reviewed in patients with
complicated pancreatitis. Only patients who had undergone interventional
procedures within 10 days of the scanning were included in our analysis. A
total of 23 scans from 20 patients were reviewed. SPECT imaging was typically
performed 48-72 hr after injection of the gallium. All studies were correlated
with conventional CT findings. Findings from subsequent interventions (results
of aspiration, Gram stains, or cultures) were used as evidence of
infection.
RESULTS. Twenty patients underwent either percutaneous or surgical
drainage within 10 days of their gallium scanning. One patient underwent
gallium scanning on three different occasions and underwent three different
interventional procedures after each of the gallium scans, bringing the total
number of cases in our study to 23. Of these 23 cases, 18 patients (78%) with
gallium scans showing positive findings for infection had infected fluid; five
patients (22%) with negative findings for infection on gallium scans had
sterile fluid (p < 0.00001). No false-positive scans were found
among our study cases, and we found no correlation between the uptake of
gallium and the presence or absence of pancreatic necrosis.
CONCLUSION. Gallium does not actively accumulate in all patients
with severe pancreatitis, and gallium uptake does not correlate with the
presence or absence of necrosis. In patients with severe pancreatitis
complicated by fluid collections or inflammatory masses, gallium SPECT is a
useful predictor of infection and can be used to help guide subsequent
intervention. Gallium SPECT allows targeting sites of infected fluid in
patients with multiple fluid collections and potentially obviates intervention
in patients with sterile fluid collections.
Introduction
The management of severe pancreatitis can be difficult, particularly in
cases complicated by necrosis, acute fluid collections, and pseudocysts.
Patients with severe pancreatitis are often critically ill, requiring
aggressive fluid replacement and monitoring for multiorgan failure and
hemodynamic instability. The development of pancreatic infection, especially
in patients with necrotizing pancreatitis, is a major indicator of increased
morbidity and mortality [1].
The overall rate of infection in pancreatic tissue in patients with acute
pancreatitis has been estimated to range from 7% to 12%. In the subset of
patients with necrotizing pancreatitis, the rate of infection increases to
30-70% of patients [2,
3]. Most patients with
pancreatic infection (i.e., those with infected pancreatic necrosis or
pancreatic abscess) have signs and symptoms of infection, including fever,
leukocytosis, and tachycardia, but these same symptoms can be seen in patients
with sterile pancreatitis [3,
4].
CT has become the imaging modality of choice for patients with severe acute
pancreatitis. Accurate estimates of the extent of necrosis can be based on the
lack of enhancement of the pancreas on dynamic CT scanning
[5,6,7].
CT also allows accurate determination of the number and extent of acute fluid
collections [7,
8]. The Balthazar grading
system [9] has been developed
to identify patients in whom complications most often occur. However, it has
not been possible to consistently differentiate infected fluid collections
from those that are sterile on the basis of morphologic changes seen on CT.
Ultimately, sampling of these collections either by percutaneous
imaging-guided techniques or surgery is performed to establish the diagnosis
of infection [10].
Gallium single-photon emission computed tomography (SPECT) has the
potential to serve as a useful adjunct to CT in differentiating infected from
sterile fluid collections associated with severe pancreatitis. Use of gallium
in the imaging of patients with pancreatitis and its associated complications
has been evolving over the past 20 years. Originally thought to able to reveal
acute pancreatitis [11,
12], gallium scanning was
subsequently deemed inappropriate because of a large number of false-negatives
[13]. In addition, using
gallium to identify inflammatory lesions in the abdomen has been debated, with
the argument made that excretion into the gastrointestinal tract makes it
difficult to distinguish abnormal distribution from normal distribution of
gallium [13,
14]. We postulated that
gallium SPECT would differentiate the normal bowel from an inflammatory
process when interpreted in combination with CT and that gallium would
accumulate in fluid collections or necrosis that was infected, but not in
uninfected pancreatitis or its associated sterile fluid collections.
Materials and Methods
Our institution is a tertiary care center to which numerous patients with
complicated pancreatitis are referred. The patients in our study had undergone
various diagnostic studies and therapies before being transferred to our
institution for further treatment. All gallium-67 scans obtained in patients
with complicated pancreatitis over a 5-year period were retrospectively
reviewed. For inclusion in our study, the patient had to have undergone a
definitive diagnostic procedure—either percutaneous or surgical drainage
or aspiration—within 10 days of performance of the gallium scanning. A
total of 20 patients with 23 scans—11 men and nine women whose age range
was from 18 to 75 years—met this criteria. One patient had undergone
gallium scanning on three different occasions and had undergone three
different interventional procedures after each of the gallium scans, bringing
the total number of cases in our study to 23. We also reviewed 23 correlative
CT scans obtained during a time period ranging from the same day to 9 days
after the gallium scans were acquired (average elapsed time, 3 days). The
gallium scans had all been prospectively interpreted, first without and then
with CT correlation, and all studies were retrospectively reviewed by two
observers.
After IV administration of 222 MBq of gallium-67, whole-body planar images
were acquired with SPECT imaging of the abdomen (and, if necessary, the pelvis
as well). No bowel preparation was given. The gallium scans were obtained
using a dual-head, extra large field-of-view gamma camera equipped with
medium-energy collimators. Imaging was performed over three photopeaks (93,
184, and 296 keV) of gallium. Whole-body planar scans involved a 20-min
whole-body acquisition. SPECT imaging was performed using a 128 x 128
acquisition matrix (pixel size = 5.28 mm). Each head rotated 180° for 60
stops at 3° per stop. Imaging time was 25 sec per stop. Images were
reconstructed using a filtered back projection with a Hamming filter (cut off,
0.7 cycles/cm). Images were reconstructed in the standard transverse, coronal,
and sagittal planes with reconstruction of weighted reprojection images
(1/exponential) in 10° intervals around the body. These latter images were
viewed as a cine series.
Most patients had whole-body imaging performed at 24, 48, or 72 hr after
gallium injection with SPECT imaging usually performed at 48 or 72 hr. Because
most patients were critically ill and not always able to come to our imaging
suites, imaging times varied, ranging from 24 to 144 hr after gallium
injection for whole-body imaging and from 24 to 72 hr for SPECT imaging.
CT scans were obtained on Advantage HiSpeed scanners (General Electric
Medical Systems, Milwaukee, WI). Over the years of this study, protocols used
for pancreatitis varied slightly, particularly in the advent of helical
scanning. However, each CT study involved both pre- and post-IV contrast
imaging after administration of a 150-mL bolus of iodinated contrast material
to assess the degree of pancreatic enhancement. One CT scan was obtained
without use of contrast material.
The findings of gallium scans were categorized as positive or negative for
infection. A study with positive findings was defined as one in which any
accumulation of gallium greater than background activity was located outside
the expected physiologic distribution of gallium (i.e., the liver, spleen,
bowel, or skeleton). The CT scans were categorized retrospectively on the
basis of Balthazar criteria (A-E) and on the presence of necrosis as evidenced
by a lack of CT contrast enhancement.
Correlation with subsequent intervention (percutaneous sampling, drainage,
or surgery) was performed within 10 days of the gallium scanning. Fluid
collections were considered infected if frank pus or a positive finding for a
Gram stain or culture was obtained. Fisher's exact test was used for
statistical analysis of the correlation between gallium results and the
ultimate determination of infection and for analysis of gallium results versus
the CT findings of necrosis.
Results
Of the 23 gallium scans we studied, 18 scans (78%) had positive gallium
accumulation in the pancreas or in fluid collections in a peripancreatic
location. Five scans (22%) showed no major gallium accumulation within the
pancreatic bed or in sites of abdominal fluid collections identified on CT,
despite clinical or laboratory evidence of pancreatitis.
Of the 23 CT scans, 21 (91%) were classified as Balthazar E. One (4%) was
classified as Balthazar D. One (4%) was classified as Balthazar C. Fourteen
patients had CT evidence of necrosis manifested as having at least 30% of the
pancreas show no enhancement. Pancreatic enhancement remained intact in eight
patients. One patient had an unenhanced CT scan.
Table 1 shows the gallium
findings versus the Balthazar classification.
All 23 cases (in 20 patients) required intervention using either CT-guided
aspiration or drainage, surgery, or both within 10 days of the performance of
gallium scanning. In these cases, 18 had infected fluid, and five had sterile
collections (Table 2). Observers correctly predicted the presence of infection in 18 of the 18 cases
on the basis of the gallium scan (sensitivity, 100%). A representative case is
shown in Figure
1A,1B,1C.
Figure
2A,2B,2C
shows a patient with a fairly complicated case, in which the gallium SPECT
image showed a mesenteric abscess (with gallium accumulation around the
hepatic flexure of the colon, corresponding to pericolonic inflammation on
CT). The patient also had a large fluid collection in the pancreatic bed,
which was negative for gallium and proved to be uninfected at surgery.
Predictions of no infection made on the basis of the gallium scan were correct
in five of the five cases with sterile collections (specificity, 100%). An
example of a scan with true-negative findings is shown in Figure
3A,3B.
With the use of gallium SPECT and a correlative CT scan, we were able to
differentiate bowel from the fluid collections or pseudocysts in all cases
(p < 0.00001).
View larger version (110K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 1B. —47-year-old man with severe pancreatitis who had
true-positive finding for infection on gallium study. Gallium SPECT image
shows intense activity in fluid collection (arrows) and expected
activity in liver and spine.
|
|
View larger version (32K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 2B. —53-year-old man with severe pancreatitis and spiking fevers.
Axial gallium SPECT image shows no activity in pancreatic head, which was
confirmed as uninfected at surgery. Activity around right colon is evident,
and study was interpreted prospectively as colon activity and developing
abscess.
|
|
View larger version (135K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 3A. —38-year-old man with severe pancreatitis who had
true-negative finding for infection on gallium study. CT scan reveals large
fluid collection in pancreatic bed that extends into left anterior pararenal
space.
|
|
View larger version (46K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 3B. —38-year-old man with severe pancreatitis who had
true-negative finding for infection on gallium study. Axial gallium SPECT
image shows lack of gallium uptake in fluid collection. Activity seen in
liver, spleen, and spine is expected. Subsequent percutaneous drainage
confirmed absence of infection.
|
|
Figure
4A,4B
illustrates well the value of management using gallium. This patient was found
to have a fluid collection in the pancreatic bed and was injected with gallium
upon completion of the diagnostic CT. Before the gallium study was completed,
a CT-guided aspiration and drain placement were performed. We completed our
gallium study 48 hr after the drainage and reported the fluid collection as
negative for gallium but the drainage catheter course as intensely
gallium-avid. Follow-up CT performed 24 hr after our gallium study showed the
catheter had retracted and was no longer in the fluid collection. The culture
from the fluid aspiration was negative. The catheter was cultured and revealed
Candida organisms. No correlation was found between the presence or
absence of necrosis and gallium avidity
(Table 3).
View larger version (124K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 4A. —55-year-old woman with severe pancreatitis and spiking fevers
who developed catheter-introduced infection. On CT scan (not shown), fluid
collection in pancreatic bed was observed. Drainage was performed before
gallium scan was completed. Fusion image of CT scan and gallium study shows
intense gallium activity along drainage catheter tract. Catheter has become
dislodged and infected.
|
|
View larger version (110K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 4B. —55-year-old woman with severe pancreatitis and spiking fevers
who developed catheter-introduced infection. On CT scan (not shown), fluid
collection in pancreatic bed was observed. Drainage was performed before
gallium scan was completed. Fusion image of CT scan and gallium study obtained
at different level of transparency to allow visualization of catheter in the
gallium collection. Fluid collection shows no gallium uptake. Result of
culture of fluid was negative; result of culture of catheter was positive for
Candida organisms.
|
|
Discussion
Gallium-67 has been used to localize infection and inflammation for many
years [15]. With all the
different methods for infection imaging available today
(111In—labeled WBC, 99mTc-labeled WBC, and even
FDG), the use of gallium for imaging infections has declined, but its accuracy
has been documented over decades of use, and when combined with SPECT imaging,
it remains a powerful tool. Because of its high sensitivity, gallium has been
used in evaluations of patients with fevers of unknown origin. It has also
been valuable in detecting longstanding infections, such as chronic
osteomyelitis, and in assessing the response to therapy
[16]. The exact mechanism of
gallium accumulation remains unclear. However, increased blood flow and
capillary permeability, incorporation and transportation within leukocytes,
and bacterial ingestion have been postulated to be possible factors in the
process [17]. Gallium
resembles the ferric ion and binds to several proteins, including transferrin,
lactoferrin, and ferritin. Bacteria contain siderophores in their membranes
that assist in trapping and absorbing the ferric ion, and so gallium may bind
to the siderophores present at the site of bacterial infection
[17].
In cases of severe acute pancreatitis, particularly if complicated by
necrosis, the presence of infection results in increased morbidity and
mortality. Thus, it becomes critical to identify those patients in whom
infection is present so aggressive treatment can be initiated. Although CT has
been shown to accurately depict the degree of pancreatic necrosis, the
bacteriologic status of acute fluid collections or pancreatic necrosis must be
determined with invasive aspiration, drainage, or surgical
débridement.
As discussed earlier, there has been much confusion for decades over
whether gallium accumulates in cases of uninfected pancreatitis. In our study
population, our data indicate that the uptake of gallium indicates infection,
not sterile inflammation or necrosis. SPECT is an essential tool with which to
delineate normal bowel activity from sites of infection. We have shown in our
study that gallium has a high sensitivity and specificity as a predictor of
infection in pancreatic fluid collections. The clinical use of gallium SPECT
in complicated pancreatitis is a more complex issue. In patients with
complicated pancreatitis, clinical indications point to draining large sterile
fluid collections. Obviously, drainage of infected fluid is indicated as well.
In a patient with sepsis who has multiple fluid collections, gallium SPECT is
clearly useful in identifying the "culprit" fluid collection so
that rapid intervention may be accurately performed without infecting sterile
fluid collections. In our clinical experience, we have found that not all
fluid collections need an invasive procedure. In addition, if the collection
is sterile, an aspiration or placement of a drainage catheter could introduce
infection. We have included a patient in Figure
4A,4B
to serve as an example of such an occurrence. The presence of infected
pancreatic necrosis in a patient who receives no intervention is associated
with a high likelihood of mortality
[18], a fact which suggests
that a patient with necrosis that is identified on dynamic contrast-enhanced
CT and with a positive finding on the gallium scan would need prompt drainage
or débridement. The cases in our study came from a larger series of
cases that were prospectively acquired to study gallium's potential role as a
predictor of the need for surgical intervention in patients with complicated
pancreatitis. We are continuing our work on a larger role for gallium SPECT in
the treatment of complicated pancreatitis.
We find it interesting that the presence or absence of necrosis
(Table 3) did not affect the
avidity for gallium. The pathophysiology of necrosis, in part, involves
disruption of capillary blood flow to the pancreatic parenchyma
[19]. One might be concerned
that gallium may not accumulate in infected necrosis because the delivery of
the radionuclide to the pancreatic bed would be impaired. We did not find this
to be the case.
The use of SPECT was essential in the interpretation of the gallium scans.
Previous reports have debated the usefulness of gallium in evaluating
abdominal processes because of its normal excretion via the bowel. Planar
studies alone in our patient population would have made the interpretation of
the scans difficult. Gallium is taken up intensely in hypertrophic gastritis,
a fairly common condition in critically ill patients. Focal inflammation of
the bowel as a result of pancreatitis itself and focal adynamic ileus also
contribute to the problem of persistently intense gallium uptake in bowel.
Combining SPECT imaging with CT correlation allows better anatomic
localization of the gallium activity and prevents the misinterpretation of
bowel activity as infection in most cases. By interpreting the gallium SPECT
scan with the CT scan, we allow the gallium, in effect, to act as a CT
contrast agent for infection. Although not every case had image fusion
performed (Figs.
1A,1B,1C
and
4A,4B),
the use of this technique may even further facilitate the accurate
interpretation of these studies. Figure
5A,5B,5C
illustrates the advantages of fusing the gallium and CT data sets.
View larger version (137K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 5A. —62-year-old woman with severe pancreatitis and clinical signs
of sepsis. These images show importance of SPECT and CT fusion images in
establishing anatomic location of gallium. CT scan shows rim-enhancing fluid
collection (arrow) in left pararenal space and inflammatory changes
around antral wall and lesser sac.
|
|
View larger version (111K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 5B. —62-year-old woman with severe pancreatitis and clinical signs
of sepsis. These images show importance of SPECT and CT fusion images in
establishing anatomic location of gallium. Axial gallium SPECT image shows
increased uptake in left anterior pararenal space and anteriorly in mid
abdomen.
|
|
View larger version (121K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 5C. —62-year-old woman with severe pancreatitis and clinical signs
of sepsis. These images show importance of SPECT and CT fusion images in
establishing anatomic location of gallium. Fusion image shows gallium in fluid
collection, a finding confirmed as infection at CT-guided aspiration. Anterior
gallium accumulation is located in stomach and lesser sac and is most likely
due to surrounding inflammation from pancreatitis.
|
|
In the future, serial gallium scans may be helpful in assessing the
effectiveness of noninvasive medical therapy. For example, if a patient
without necrosis but with numerous fluid collections has a positive finding on
a gallium scan and is treated with high doses of antibiotics instead of
drainage, could a follow-up gallium scan be used to assess the effectiveness
of the antibiotic therapy by depicting gallium avidity? A technique based on
this rationale has been successfully applied in patients with chronic
osteomyelitis [16]. However,
our study did not address this issue for patients with complicated
pancreatitis. Further work needs to be done to fully explore this
question.
In conclusion, in patients with severe pancreatitis complicated by fluid
collections (Balthazar classifications D and E), gallium uptake is a useful
predictor for the presence of infection. It may serve as a valuable adjunct to
CT in the evaluation of these critically ill patients and may help to guide
subsequent intervention.
References
-
Beger HG, Rau B, Mayer J, Pralle U. Natural course of acute
pancreatitis. World J Surg
1997;21:130
-135[Medline]
-
Beger HG, Bittner R, Block S, Buchler M. Bacterial contamination of
pancreatic necrosis. Gastroenterology
1986;91:433
-438[Medline]
-
Bradley EL, Allen K. A prospective longitudinal study of
observation versus surgical intervention in the management of necrotizing
pancreatitis. Am J Surg
1991;161:19
-25[Medline]
-
Karimgani I, Porter KA, Langevin RE, Banks PA. Prognostic factors
in sterile pancreatic necrosis. Gastroenterology
1992;103:1636
-1640[Medline]
-
Johnson CD, Stephens DH, Sarr MG. CT of acute pancreatitis:
correlation between lack of contrast enhancement and pancreatic necrosis.
AJR
1991;156:93
-95[Abstract/Free Full Text]
-
Paulson EK, Vitellas KM, Keogan MT, Low VHS, Nelson RC. Acute
pancreatitis complicated by gland necrosis: spectrum of findings on
contrast-enhanced CT. AJR
1999;172:609
-613[Free Full Text]
-
Balthazar EJ, Freeny PC, vanSonnenburg E. Imaging and intervention
in acute pancreatitis. Radiology
1994;193:297
-306[Abstract/Free Full Text]
-
Siegelman SS, Copeland BE, Saba GP, Cameron JL, Sanders RC,
Zerhouni EA. CT of fluid collections associated with pancreatitis.
AJR
1980;134:1121
-1132[Abstract]
-
Balthazar EJ, Ranson JHC, Naidich DP, Megibow AJ, Caccavale R,
Cooper MM. Acute pancreatitis: prognostic value of CT.
Radiology
1985;156:767
-772[Abstract/Free Full Text]
-
Gerzof SG, Banks PA, Robbins AH, et al. Early diagnosis of
pancreatic infection by computed tomography-guided aspiration.
Gastroenterology
1987;93:1315
-1320[Medline]
-
Aburano T, Yokoyama K, Hisada K, Kakuma K, Ichiyanagi K. The role
of Ga-67 citrate imaging in pancreatitis. Clin Nuc Med
1988;13:808
-811[Medline]
-
Myerson PJ, Berg GR, Spencer RP, Myerson D. Gallium-67 spread to
the anterior pararenal space in pancreatitis: case report. J Nucl
Med 1977;18:893
-895[Abstract/Free Full Text]
-
Maderazo EG, Hickingbotham NB, Woronick CL, Sziklas JJ. The
influence of various factors on the accuracy of gallium-67 imaging for occult
infection. J Nucl Med
1988;29:608
-615[Abstract/Free Full Text]
-
Hoffer P. Gallium and infection. J Nucl Med
1980;21:484
-488[Free Full Text]
-
Lavender JP, Lowe J, Baker JR, Burn JI, Chaudhri MA. Gallium-67
citrate scanning in neoplastic and inflammatory lesions. Br J
Radiol 1971;44:361
-366[Abstract/Free Full Text]
-
Alazraki N, Fierer J, Resnick D. Chronic osteomyelitis: monitoring
by 99mTc phosphate and 67Ga citrate imaging.
AJR
1985;145:767
-771[Abstract/Free Full Text]
-
Tsan MF. Mechanism of gallium-67 accumulation in inflammatory
lesions. J Nucl Med
1985;26:88
-92[Abstract/Free Full Text]
-
Baron TH, Morgan DE. Acute necrotizing pancreatitis. N
Engl J Med 1999;340:1412
-1417[Free Full Text]
-
Klar E, Messmer K, Warshaw AL, Herfarth C. Pancreatic ischemia in
experimental acute pancreatitis: mechanism, significance, and therapy.
Br J Surg
1990;77:1205
-1210[Medline]
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
This article has been cited by other articles:
|
|
|
|
 
M. E. Spieth, B. S. Gauger, and W. Drane
Gallium Uptake in Complicated Pancreatitis
Am. J. Roentgenol.,
April 1, 2003;
180(4):
1177 - 1178.
[Full Text]
[PDF]
|
|
|
Top
Abstract
Introduction
